Skip to main content
main-content

Über dieses Buch

Hydraulic machinery such as turbines and pumps is widely used around the world. Related topics concerning design, operation and maintenance are of relevant interest. In this context, cavitation is a phenomenon to be taken into account, and this was treated in the XVIII IAHR Symposium on Hydraulic Machinery and Cavitation which took place in Valencia, Spain, 16th-19th September, 1996 and which was hosted by the Polytechnic University of Valencia.
The proceedings of the Symposium have been published in two volumes. In this first volume, the papers included cover the following topics: Hydraulic Turbines, Analysis and Design Hydraulic Pumps Hydraulic Elements, Dynamic Characterization and Hydraulic Behaviour Cavitation and Sand Erosion
In the second volume, the papers included cover the following topics: Hydraulic Transients and Control Systems Related to Hydraulic Machinery and Plants Oscillatory and Vibration Problems in Hydraulic Machinery and Power Stations Experimental Investigations related to Hydraulic Machinery and its Applications Practical Applications of the Hydraulic Machinery Monitoring, Predictive Maintenance and Refurbishment
The 119 papers presented at the Symposium, from research groups, consulting companies and manufacturers, constitute an important collection for investigators, engineers and technicians who are interested in updated information on hydraulic machinery. This book is intended to be a reference text comprising the latest innovations on this subject.

Inhaltsverzeichnis

Frontmatter

Invited Lectures

Frontmatter

The Hydroelectricity in the World — Present and Future

Only about 15% of the exploitable hydro power potential in the world has so far been utilized. However, in the industrialized countries in Europe and North America the majority of the hydro power has been developed for electricity production.

Hermod Brekke

Analysis of Transients Caused by Hydraulic Machinery

The analysis of transients in systems having hydraulic machinery is necessary during the design of these systems. It may be needed to upgrade, modify, or retrofit an existing system or if the mode of operation is changed. Sometimes, such analyses are done to determine the cause of incidents, accidents, and failures; and to develop procedures or system modifications to prevent their re-occurrence.

M. Hanif Chaudhry

Some Present Trends in Hydraulic Machinery Research

An overview is given of the recent developments in the field of research about the dynamic behaviour of hydraulic machines, both as concerns their reaction to external disturbances (passive behaviour) and their role as a source of excitation for the rest of the installation (active behaviour). The interest of such investigations for manufacturers and users of hydraulic machines is outlined. The case of the draft tube surge caused by the interaction with the elbow of the cavitated vortex rope at the outlet of a Francis turbine operating at partial load is treated in some detail. Hints are offered as to the likely future trends of research about these topics.

Michelle Fanelli

Rapid Prototyping of Hydraulic Machinery

The paper presents a philosophy of ”Rapid Prototyping of Hydraulic Machinery” and describes the main tools for the efficient tailor-made design and optimization.

R. Schilling, N. Riedel, R. Bader, T. Aschenbrenner, CH. Weber, A. Fernandez

Fluid Transients in Flexible Piping Systems

A Perspective on Recent Developments

Between 1970 and 1990 a substantial amount of research activity focused on understanding the mechanical interaction between unsteady flow in piping and the resulting vibration of the pipe elements and support structures. As a result, several analytical and numerical techniques were developed and reported. A primary motivation for this work emanated from the nuclear power industry, where a number of water hammer incidents and resulting pipe motion occurred. In the 1980’s, analytical methods were advanced to the point where dynamically coupled numerical predictions proved to be attainable without extensive complication.

D. C. Wiggert

Symposium Contributors

Frontmatter

Hydraulic Turbines. Analysis and Design

A Decision Aid System for Hydraulic Power Station Refurbishment Procedure

Francis and Kaplan Turbine

Maintenance policies for hydrogenerating equipment vary considerably from one operator to another, from a least cost strategy to the greatest care and tightest scheduling. However, whatever policy is adopted, sooner or later maintenance is no longer enough to uphold acceptable levels of performance or safety. When that happens, renovation or refurbishment takes over.

L. Bellet, E. Parkinson, F. Avellan, T. Cousot, E. Laperrousaz

A Three Dimensional Spiral Casing Navier-Stokes Flow Simulation

A comparative study

Navier-Stokes flow simulations are now commonly used in the process of hydraulic turbomachinery design. Comparison with experiments and careful understanding of the behavior of such tools is however very important to the correct interpretation of the results. Following this trend, Navier-Stokes flow simulations, using three solvers, applied to a pump turbine spiral casing in turbine mode, are presented and discussed in this paper.

Ch. Bruttin, J.-L. Kueny, B. Boyer, K. Héon, T. C. Vu, E. Parkinson

Tip Clearance Flow in Turbomachines — Experimental Flow Analyses

To analyse the tip leakage flow, it is considered as necessary the set up of an experimental database. For that, a test’s water tunnel, without rotation, that represents the blade-to-blade canal of a rocket inducer, has been constructed and qualified at Centre de Recherches et d’Essais de Machines Hydrauliques de Grenoble (CREMHyG — France). For these configurations the flow evolution in the water tunnel has been analysed, and specially the vortex of the leading edge by the 3D velocity measurements, performed by LDV. The tip clearance flow analysis has been carried out by velocity measurements: to the inlet and the exit of the tip clearance by LDV and into the tip clearance by PTV. The different levels of the turbulence kinetic energy has been measured at the different positions. The analysis of the phenomenon has been thus achieved.

Gabriel Dan Ciocan, Jean Louis Kueny

Numerical Prediction of Hydraulic Losses in the Spiral Casing of a Francis Turbine

Power plants that have been in operation for several decades show a considerable potential for improvement in hydraulic performance by bringing the design up to modern standards. This situation has led to numerous rehabilitation projects for existing hydraulic power plants.

Peter Drtina, Andreas Sebestyen

Improvements of a Graphical Method for Calculation of Flow on a Pelton Bucket

A graphical method for calculation of particle paths over Pelton buckets is presented. This method involves a laborious graphical derivation. This paper presents a new method for deriving particle velocities and accelerations. This method is based on cubic splines, and is calculated on a computer. A comparison with a graphical method is performed.

Morten Hana

Flow Behavior and Performance of Draft Tubes for Bulb Turbines

To cope with the warming global environment, the development of the ultra-low head energy is active by welcoming as the clean and cool energy resources. Bulb turbines, which are suitable to the ultra-low head, have the very high velocity head at the runner outlet, relative to the net head. Since the role of the draft tube on the hydraulic performance is extremely high, the bulb turbine normally uses the straight draft tube which is composed of the inlet annular diffuser with the hub followed by the diffuser having the transitional cross section from conical to rectangular shape. The performance of such a draft tube can not be predicted successfully by usual data for two-dimensional, conical and/or annular diffusers. Thus, the pressure recovery of the straight draft tube for the bulb turbine and the internal flow in the.S-shaped draft tube for the tubular turbine have been investigated [1]–[3].

T. Kanemoto, M. Uno, M. Nemoto, T. Kashiwabara

Performance Analysis of Draft Tube for Gamm Francis Turbine

To precisely analyze the hydraulic losses in the various components of a Francis turbine, it is necessary to know the internal flow velocity distributions in the respective components. Since, however, the hydraulic energy loss Δψ and the performance characteristics such as energy coefficient ψ and discharge coefficient ø etc. are all the one-dimensional information for a whole turbine, it is sufficient to know the sectional mean velocities along a single representative mid-streamline through the turbine. Thus, the one-dimensional flow theory is decisively useful for the loss analysis corresponding to the performance diagrams. Performance diagrams acquired by the precise model test are so reliable that utilizing the diagrams to the loss analysis is highly recommendable. Two of the authors have presented, applying the one-dimensional flow theory, a new algorithm of extracting the various component losses in bulb turbines from the performance diagrams measured with the model tests [1].

T. Kubota, F. Han, F. Avellan

Modelling Complex Draft-Tube Flows using Near-Wall Turbulence Closures

This paper presents a finite-volume method for simulating flows through complex hydroturbine draft-tube configurations using near-wall turbulence closures. The method employs the artificial-compressibility pressure-velocity coupling approach in conjunction with multigrid acceleration for fast convergence on very fine grids. Calculations are carried out for a draft tube with two downstream piers on a computational mesh consisting of 1.2×106 nodes. Comparisons of the computed results with measurements demonstrate the ability of the method to capture most experimental trends with reasonable accuracy. Calculated three-dimensional particle traces reveal very complex flow features in the vicinity of the piers, including horse-shoe and longitudinal vortices and regions of flow reversal.

Y. Ventikos, F. Sotiropoulos, V. C. Patel

Fluid Flow Interactions in Hydraulic Machinery

The paper deals with the fluid flow interactions between rotor and stator as well as rotor and side spaces of hydraulic machinery. It is shown that the flow through the rotor strongly influences the leakage flow through the side space, the friction torque and the axial forces acting on the rotor.

T. Aschenbrenner, N. Riedel, R. Schilling

Numerical Optimization of High Head Pump-Turbines

A Challange for Mechanical and Hydraulic Design

In order to cope with the technical challenges of high head pump-turbines, new methods like simultaneous engineering and numerical studies are required. In this paper the application of simultaneous engineering is shown with an example, as well as typical pump-turbine concepts and design features.

H. Buchmaier, B. Quaschnowitz, W. Moser, D. Klemm

Numerical Hill Chart Prediction by Means of CFD Stage Simulation for a Complete Francis Turbine

A full stage simulation method is used to calculate the first numerically predicted hill chart of a high specific speed Francis turbine. The numerical method simulates rotating and non-rotating components and their mutual interactions within a single computation. An efficiency hill chart is determined numerically and compared to experimental results from a model test whereby good agreement is obtained.

Helmut Keck, Peter Drtina, Mirjam Sick

Development of a New Generation of High Head Pump-Turbines, Guangzhou Ii

For the Guangzhou Pumped Storage Plant in the People’s Republic of China we had to develop the Pump-Turbines of the second phase considering the very tight delivery and erection schedule. The hydraulic, as well as the mechanical design, the manufacture of a new complete homologous model machine and the detailed model tests had to be performed within a very short time.

D. Klemm, E.-U. Jaeger, C. Hauff

Development of Integrated Cae Tools for Design Assessment and Analysis of Hydraulic Turbines

Recent advances in numerical analysis can lead to substantial improvements in turbine design. In light of the present tendency to increase the unit power and operating speed of these machines at better cost, electric utilities have to develop their expertise in the field of numerical simulation with a view to improve technical decisions, validate conceptions and, in the long term, enhance plant reliability. Applications of numerical simulation range from analysis of problems specific to hydraulic machines such as manoeuvre margin estimation of turbines in operation, correction of design flaws, design validation and technical assistance for planners, designers and operators. This paper describes an integrated system of Computer Aided Engineering (CAE) tools developed to assess designs, optimize retrofits and provide insight for problem analysis.This system is described using examples to illustrate the use and limitations of the different tools. Selected applications are presented such as the flow simulation in a water intake, the geometric measurements of Francis runners, computed and measured pressure and stress distributions on runner blades, vibration analysis of a Francis turbine, draft tubes flow validation, spiral casing flow and friction losses due to the roughness in a hydraulic tunnel.

Bernard Massé, Henri Pastorel, Robert Magnan

Design and Analysis of a Two Stage Pump Turbine

Nowadays the design of all turbomachinary components can hardly be conceived without taking numerical flow modelisation into account. This is due to the advances in numerical methods and in computer technology.

Farid Mazzouji, Maryse Francois, Frank Hebrard, Jean Bernard Houdeline, Daniele Bazin

Study on High Speed and High Head Reversible Pump -Turbine

In this paper, improvement of hydraulic performance, particularly pressure fluctuation and cavitation, in order to speed up the runner revolution for a pump-turbine of 500m head class, is numerically and experimentally studied. High speed design of a pump-turbine offers an economical merit, because the size of the pump-turbine and generator-motor can be made smaller and the efficiency is improved. However, there are some problemes to be solved in order to develop a high speed pump-turbine. The first one is the increase of pressure fluctuation of runner outlet at partial load turbine operation. The second one is the deterioration of cavitation performance at pump operation. The third one is the increase of stress and vibration for runner. In order to overcome these problemes, the optimization of runner shape is carried out at design stage by using numerical simulation and its results are experimentally verified.

T. Nakamura, H. Nishizawa, M. Yasuda, T. Suzuki, H. Tanaka

Analysis of the Performance of a Bulb Turbine Using 3-D Viscous Numerical Techniques

In this paper, a high-accuracy 3-D viscous numerical technique, which has been developed previously by the authors, is introduced to analyze the flows through a bulb turbine passing through the guide vanes, runner and draft tube. The techniques of predicting the interaction between the guide vanes(stationary) and runner vanes(rotating), the tip clearance, and a new turbulent model which was previously developed by the authors, are introduced to this analysis. In order to check the shortcomings of the k-ε two equations turbulent model, the limitation of the k-ε turbulent model in the application to vortex flows are discussed by various swirling rates in this paper. The purpose of this paper is to attempt to show that 3-D numerical techniques have been developed to analyze the flows through the all flow channels, guide vanes, runner, and draft tube. The flow patterns, separation phenomenon and performance of bulb turbine can be simulated depending on the characteristics of discharge rate and head only. The simulations are carried out under a series of discharge rate and rotating speed conditions. That is, when the runner opening is fixed and when the guide vane openings are changed to several opening angles to determine the On-cam operating conditions following the discharge rate and rotational rate.

Y. Qian, R. Suzuki, C. Arakawa

Analysis of the Inlet Reverse Flows in a Pump Turbine Using 3-D Viscous Numerical Techniques

The purpose of this paper is to apply advanced technology to predict the phenomenon of reverse flows created at the inlet during pump operation of Francis type pump-turbines. The authors introduce 3-D viscous numerical techniques, which were developed for hydro-turbine runners several years previous, to predict and analyze the recirculation flows at the inlet during pump operation. The motions of the separation flows are predicted by reducing the discharge rate from design to off-design operation conditions. In order to accurately capture the strong vortex, a new boundary condition of turbulent parameters on the wall is introduced in this simulation. The computational results are found to be in reasonable agreement with the experimental results. The limitations of the k- ε two equations turbulent model in the application of vortex flows are discussed in this paper. Finally the authors present a basic scale for use with Navier-Stokes simulation techniques with k- ε two equations model when using the model as design tool in the case of reverse flows.

Y. Qian, R. Suzuki, C. Arakawa

Importance of Interaction between Turbine Components in Flow Field Simulation

The application of viscous flow prediction methods, i.e. Navier — Stokes methods, has drastically increased during the development of hydraulic machines in the course of the last several years. This proceeding supports the design process through a detailed knowledge of the viscous flow behaviour in all components of the hydraulic machine. Usually, a broad operation range is investigated, covering the optimum conditions, too.

S. Riedelbauch, D. Klemm, C. Hauff

From Components to Complete Turbine Numerical Simulation

In the quest for turbine improvement, computational fluid dynamics (CFD) is the emerging tool. The fast development of computers combined with the new capabilities of turbulent flow codes make possible the calculation of an entire turbine unit. Until recently, fluid simulations for turbines were done considering the various components separately. This convenient way of analysis can however present some major drawbacks. As components can be strongly coupled, the unknown introduced by assumed boundary conditions needs to be clarified. One consequence of a complete turbine numerical simulation is to avoid assumptions of internal boundary conditions.

M. Sabourin, Y. Labrecque, V. de Henau

Validation of a Stage Calculation in a Francis Turbine

This paper describes the verification of a simulation method for complete hydraulic turbine systems, from spiral casing through distributor and runner to the outlet draft tube. The method solves the steady 3D Reynolds-averaged Navier Stokes equations with a mixing plane at the interfaces between the components, so that the flow in each frame of reference is steady. The steady-state interactions between the components are taken into account but unsteady interactions are neglected. The paper verificates the method by comparison of computations with detailed flow measurements from a high specific speed Francis turbine model. Computations in which each component is computed in isolation are compared to computations using the mixing interface between components to demonstrate the advantages of the new technique.

Mirjam Sick, Michael V. Casey, Paul F. Galpin

Simulation of Flow Through Francis Turbine by Les Method

The traditional approach of Francis turbine design which is based on the steady potential flow theory and heavily dependent on model testing and engineering experience, has come a long way in producing efficient and relatively cavitation free turbines. But further improvement of performance for design and off design operating conditions will be extremely difficult with the traditional method because it will depend more on those phenomena such as, boundary layer separation, vortex dynamics, interactions between different components, vibrations, etc., which are not predictable with conventional approach and difficult to measure in physical models.

Charles C. S. Song, Xiangying Chen, Toshiaki Ikohagi, Johshiro Sato, Katsumasa Shinmei, Kiyohito Tani

Simulation of Flow Through Pump-Turbine

For years Francis — type reversible pump — turbines have been applied to many pumped storage power plants over the world. So far, they have been used under peak shaving generating operation with AFC (automatic frequency control) function during daytime, and plain pumping — energy consuming — operation at night time. Recently, an adjustable speed pump — turbine was realized in Japan. It begins to add a remarkable AFC function in pumping mode to the conventional pump-turbines.

Charles C. S. Song, Changsi Chen, Toshiaki Ikohagi, Johshiro Sato, Katsumasa Shinmei, Kiyohito Tani

The Scale Effect in Kaplan Turbines. New Relationships for the Calculation of Scalable and Non-Scalable Hydraulic Losses and of the Coefficient V

In this paper new directions are opened concerning the scale effect at Kaplan turbines. The original relationships are developed for the calculation of coefficients δM, δns. and V in the whole domain of operating of Kaplan turbines. Their use needs only the hill diagram of the turbine. Applying a new method there were obtained new relations, for the coefficients δMo, δnso. and Vo corresponding to the optimum operating regime of a Kaplan turbine.

I. M. Anton

Analysis of Losses in Hydraulic Turbines

The paper gives a brief discussion on the influence of the different losses in a hydraulic turbine. Of special interest are the friction losses which decreases with Re. This leads to a scale up of the turbine efficiency from model to prototype as expressed in the IEC code according to the theory by prof. Spurk [1].The importance of homology in geometry and flow vector fields for the validity of the efficiency scale effect will be discussed.The shift of the best efficiency point (BEP) from model to prototype caused by flow friction loss and disk friction loss will be discussed.Losses from guide vane end leakages and non-scalable losses will be described.A brief discussion on the problemes with reliable scale up formulas for Pelton turbines will be included

Hermod Brekke

Scaling-up Head-Discharge Characteristics from Model to Prototype

Often the acceptance tests of an hydraulic turbine are based on model tests. So prototype characteristics have to be determined from model results. The scaling-up for head, discharge, output and efficiency is made according to IEC standards: output is corrected by efficiency head and discharge are not corrected. But on-site measurements for different types of hydraulic turbines are not in good agreement with this evaluation [1,2,3].

Michel Couston, Robert Philibert

Recent Development of Studies on Scale Effect

In 1986, a sub-committee was organized in Japan Society of Mechanical Engineers (JSME) to standardize a new scale effect formula by studying the outstanding problems of scale effect such as dependence of V (loss distribution coefficient) on specific speed, roughness effect and so on. Then the sub-committee established a new performance conversion method and published it as JSME S 008 in 1989 [11. After its publication, the second sub-committee has been organized to revise the conversion method by collecting the results of further study.

T. Ida, T. Kubota, J. Kurokawa, H. Tanaka

Prediction of Scalable Loss in Francis Runners of Different Specific Speed

Hydraulic losses are composed of scalable loss and non-scalable loss. For the determination of scale-up formulae in hydraulic turbines, it is important to predict the scalable loss, that is the friction losses in a runner and fixed channels, as precisely as possible. However, it is still difficult to determine them both theoretically and experimentally, partly because the boundary layer is too thin to prepare sufficient mesh size for CFD, and partly because the scalable loss cannot be separated experimentally from the measured total hydraulic loss.

T. Kitahora, J. Kurokawa, M. Matumoto, R. Suzuki

Scale Effect of Jet Interference in Multinozzle Pelton Turbines

The IEC model acceptance test code stipulates that there is no scale effect in Pelton turbines[1]. One of the primary reasons why the scale effect in Pelton turbines is ignored, so far, is that the complicated unsteady free water-sheet flow on Pelton buckets prevents to numerically analyze the loss mechanism. Another reason is that the positive scale effect in the closed conduit flow from the turbine inlet to the nozzle outlet tends to compensate for the negative scale effect in free water-sheet flow on the buckets.

Yuji Nakanishi, Takashi Kubota

Further Development of Step-Up Formula Considering Surface Roughness

The paper is based on the proposal for a new step-up formula made by the author at the IAHR Symposium 1994 in Beijing. In its first part an upgrading of proposed formula to also consider influence of surface roughness is derived. In a second step it is demonstrated that the scalable loss So which is originally a constant can be extended to a function of flow and specific hydraulic energy. Efficiency step-up at a constant flow coefficient and geometry automatically involves a change of specific hydraulic energy, so a formula to describe this effect is also given in the paper, which is concluded by a set-up for the power shift which can be derived from the above. Summarizing the paper offers a complete set of formulae describing a method for the transposition of characteristics from model to prototype.

Alois Nichtawitz

Numerical Simulation of Jet in a Pelton Turbine

In case of predicting the performance of prototype Pelton turbine from a model test, a scale effect has been ignored so far. However, it is getting clear that there is an outstanding deviation from a classical similarity laws. The direction of the deviation is completely inverse against the case of reaction turbines, namely, the performance of prototype is worse than the predicted one. In order to analyze this negative scale effect, the behavior of jets should be considered more precisely. In this study, the unsteady jet issuing from a Pelton turbine nozzle was calculated numerically using HSMAC method under the assumption of axisymmetric flow. The calculation shows that the needle considerably affects the velocity distribution in the jet and the velocity distribution in the jet changes by the head.

Tomoyasu Nonoshita, Yoichiro Matsumoto, Takashi Kubota, Hideo Ohashi

An Assessment of the Loss Distribution in Francis Turbines

Scalable loss in Francis model turbine is analyzed for four runners with various number of blades which are tested in the same model turbine. The flow in the stay vane, the guide vane and the runners are analyzed by a 3D Euler flow analysis code, and then three methods of boundary layer calculations are carried out in order to obtain scalable loss. As for spiral casing and draft tube, scalable loss is obtained by the analogy of pipe flow.From the comparison between runners, the difference in peak efficiency seems to be mainly caused by non-scalable losses. Therefore scalable loss itself is most suitable to be prescribed in the scale-up formula than loss distribution coefficient.Problems on the effect of number of blades might be solved by using scalable loss in scale-up formulae.

R. Suzuki, Y. Qian, T. Kitahora, J. Kurokawa

Hydraulic Pumps

Unsteady Flow Calculation in a Centrifugal Pump Using a Finite Element Method

In order to predict rotor-stator interactions in hydraulic turbomachinery, a new multidomain method was implemented in a fmite element code developed in the Research Division of Electricité de France. This code deals with 2D or 3D laminar or turbulent flows in complex geometries. This method was used to study the unsteady flow in a simplified centrifugal pump model. The model consists in a 420 mm diameter unshrouded centrifugal impeller with seven untwisted constant thickness backswept blades and a radial vaned diffuser with twelve vanes and a six percent vaneless radial gap in which important unsteady interactions are expected. Unsteady numerical results are compared with the experimental results published by M. UBALDI et al. and also with a multistage calculation on the same geometry using a mixing plane to communicate circumferential average of flow properties.

P. F. Bert, J. F. Combes, J. L. Kueny

Steady and Unsteady Flow Pattern Between Stay and Guide Vanes in a Pump-Turbine

Due to a collaboration INPG, EDF and NEYRPIC an experimental database concerning the flow in stay and guide vanes, on an industrial test pattern of a turbine — pump has been achieved. The structure of the flow, be it steady or unsteady. has also been determined. An analysis of these flows as well as their evolution according to the functioning point have been determined. These data have been used as boundary lavers and validation data for a steady numerical calculation and are available for the unsteady calculations.

Gabriel Dan, Jean Louis, André Luiz Amarante

Self-Sustained Oscillation of Gas-Liquid Flow in a Centrifugal Pump with Semi-Open Impeller

When a centrifugal pump is operated at very low discharge with low suction head, the air dissolving in water merges to become bubble and stays inside an impeller channel. The staying bubbles accumulate to form a large bubble zone and decrease pumping head. This phenomena is called “air locking”, and often arises when air-rich water is used in a pumping system, such as a water supply system of a building.

J. Kurokawa, J. Matsui, H. Takada, T. Hirayama

Measurements in the Dynamic Pressure Field of the Volute of a Centrifugal Pump

This paper presents an experimental investigation into the dynamic pressure field existing in the volute of an industrial centrifugal pump in order to characterize the interaction phenomena between impeller and volute. For that purpose, pressure signals were obtained simultaneously at different points of the volute casing by means of two miniature fast-response pressure transducers. Particular attention was paid to the pressure fluctuations at the passing blade frequency, regarding both amplitude and phase delay relative to a reference point. The analysis of the dependence of the pressure fluctuations on both flow-rate and position along the volute clearly indicates the leading role played by the tongue in the impeller-volute interaction and the increase of the amplitude of the dynamic forces in off-design conditions.

J. L. Parrondo, J. Fernández, C. Santolaria, J. González

Functional Modelling of Pump Volute Geometry

The task of hydraulic design is usually undertaken using traditional techniques, coupled with many simplifying assumptions. This can lead to long design lead-times and under-utilisation of performance potential. New computational approaches are now required. In this resolve, some considerations of linking intelligent computer technology with fundamental aspects of hydraulic design, in particular pump volute geometry, are discussed. The paper firstly reviews volute flowfield measurements and the effects of geometry, and goes on to discuss some recent experiences in developing an experimental system to provide a functional modelling facility.

P. R. Thackray, R. D. James

Analysis of Flow Measurements in the Impeller and Vaned Diffuser of a Centrifugal Pump Operating at Part Load

In a recent european contract, four impellers with different specific speeds have been equiped with transparent front shrouds in order to obtain optical access to the flow between the blades. The presented results are the last analysis of internal flow measurements in one of these impellers (Ns32) fitted with a vaned diffuser. They are especially related with the interaction between impeller and diffuser, e.g. the phenomenon of flow separation, wakes, and backflow inside the pump. Unsteady velocity measurements were carried out in the hydraulic laboratory of CETIM with the help of laser doppler velocimetry technique. The data fields have been treated by the CNAM-LEMFI in order to give instantaneous view of the flow between the blades. From these pictures, a three dimensionnal movie has been built, that allows to explain some details in the behaviour of centrifugal impeller between 0.2Qn and Qn.

Michel Toussaint, François Hureau

Influence of the Blade Roughness on the Hydraulic Performance of a Mixed-Flow Pump. A Viscous Analysis

The aim of this work is to quantify the blade roughness influence on the hydraulic performances of a mixed-flow pump. In this way the flow in a hydrodynamic tunnel equipped with a blade profile was analysed by measurements performed by the CREMHyG and numerical simulations with the code N3S (developed by ‘Electricité de France’). The roughness parameter z0 used in the simplified rough wall law was calibrated from the measured roughness for three Reynolds numbers and three rougimesses. The application to the pump flow was obtained for the scale model conditions (0.41). For the prototype pump conditions, a calibration was suggested but has to be refined by some other simulations.

S. Undreiner, E. Dueymes

Improvement of Performance of Centrifugal Pumps Based on Computational and Theoretical Methods and Experimental Design

The problems of quality performance improvement of centrifugal pumps are actual and long-range problems. This type of pumps is widely used in all branches of national economy. Improvement of various technological processes requires modernization of old centrifugal pumps or manufacture of new ones able to provide improved performance.

A. F. Vinokurov, A. V. Volkov, G. M. Morgunov, S. N. Pankratov

Liquid-particulate Two-Phase Flow in Centrifugal Impeller by Turbulent Simulation

In the present work, turbulent liquid-particulate two-phase flows at the particle dilute concentration through a centrifugal impeller has been simulated by using the K- ε -Ap turbulence model. In the present numerical treatment, the finite volume method, based on the SIMPLEC algorithm, is applied in a body fitted coordinate system. The liquidparticle two-phase flow turbulence model used in this study can correctly predict essential features of this flow in centrifugal impellers at dilute concentrations.

Wu Yu-lin, Dai Jing, Mei Zu-yan, Oba Risaburo, Ikohagi Toshiaki

Hydraulic Elements. Dynamic Characterization and Hydraulic Behaviour

Application of the Method of Kinetic Balance for Flow Passages Forming

Definition of flow field boundaries is a common problem in hydraulic engineering. Their shape should enssure stable fluid flow, without separation and transient phenomena. Such problems are often solved using either the acquired experience or a large number of expensive experiments, with many trials, finally leading to the proper solution.

M. Benišek, S. Čantrak, B. Ignjatović, D. Pokrajac

Dynamics of Large Hydrogenerators

Experimental and Theoretical Aspects to Calculate Guide Bearings Dynamic Coefficients

Based on experimental data, this paper shows that the mathematical models of large hydrogenerators, used to preview the dynamic behavior at the design phase or as a predictive maintenance tool at the operation phase, must include the effects of the stiffness and damping forces of the oil film in the guide bearings.Based on an analytical study of the Reynolds equation, this paper presents a simplified procedure to determine the oil film stiffness and damping coefficients. This procedure is applied to the Itaipu generating units guide bearings and some practical aspects are examined.

G. C. Brito, H. I. Weber, A. G. A. Fuerst

Instabilities in a Flow-Control Valve

This paper is about some instabilities that have been observed during the testing procedure of a flow-control valve. It was observed that the complete closure of the valve was very often reached after a transient state with oscillating behaviour of its mobile element whose displacement was recorded by means of proximity sensors, facing on opposite sides the moving body of the valve. The outputs have been collected by two signal conditioning devices and sent to a dual channel spectrum analyser. Measurements have also been performed through a servo-accelerometer which has been fixed to the pipe. The influence of the pipe length on the behaviour of the system was studied. The oscillations recorded by the proximity sensors are fairly well reproduced by a series of square waves, with a number of low amplitude, strongly damped, higher frequency contents. During a single test the frequency of the oscillations is not a constant but increases in its central part. The phenomenon cannot be related to one of the natural frequencies of the piping system, that are much higher. The observed frequency depends on the length of the adduction pipe.

A. Cigada, A. Guadagnini, E. Orsi

Study of Stayvane Vibration by Hydroelastic Model

Usual method described to avoid stay vane cracking in hydraulic turbines consists in checking that frequency of vortex shedding is inferior to the first natural frequency of the vane.

Jean-Loup Deniau

Study of Dynamic Behaviour of Non-Return Valves

Although fluid carrying industrial installations have long been insured against non-return valves malfunctions, attention must be drawn to the risk of operating failures involved by the use of valves not well suited to the installation they are fitted to.

P. François

Optimum Hydraulic Design of Two-Way Inlet Conduit of Wangyuhe Pumping Station

The numerical analysis of 3-D turbulence flow through the two-way inlet conduit of is executed. In the light of the analysis the conduit is finally optimized by considering all geometrical parameters which influence the conduit shape. The hydraulic characteristics of two pump systems with different inlet conduit are compared with model tests, which show that the better characteristics are achieved for that with the optimized conduit.

Linguang Lu, Jiren Zhou, Rentian Zhang

Flow Analysis for the Intake of Low-Head Hydro Power Plants

In low-head hydro power plants often severe flow problems arise at the inlet region. Since model tests are mostly too expensive, especially for small power plants, numerical flow analysis is introduced. In this paper the numerical analysis of the flow behavior is shown for the intake region of two small power plants. For the first power plant the problem was that due to a disturbed flow in the trash rack, causing severe head losses, the power output was too low. By suggesting a relative simple modification of the geometry the problem could be cured. In the second case investigated there were oscillation problems initiated by a vortex shedding at the separation pier. Again, by changing the shape of the pier based on the results of the flow analysis, this problem could be avoided.

A. Ruprecht, M. Maihöfer, E. Göde

Cavitation and Sand Erosion

Efficiency Alteration of Francis Turbines by Travelling Bubble Cavitation

Experimental and theoretical study

The setting level of a hydraulic machine, specially for low head machines, is decided with respect to the possible alteration of the efficiency due to the cavity development. This alteration can easily be noticed by following the evolution of the efficiency η as a function of the Thoma number σ leading to the so-called η—σ cavitation curves. Observation of the cavity extent in the flow passage of the runner allows to associate the drop of efficiency with a particular type of cavity development.However, depending on the type of cavities this drop cannot be very easily explained. Obviously, for a leading edge attached cavity corresponding to high head operating points, the presence of the vapour phase on the blade suction side limits the pressure at the vapour tensile strength value which causes the flow alteration. In the case of travelling bubble cavitation, corresponding to the outlet cavitation at the nominal head, previous experiments with a 2-D NACA profile show that the modification of the mean pressure field is mainly due to the bubble dynamics. The aim of this paper is to present the results and the analysis of two experiments intending to explain the influence of the nuclei content on the mean pressure field correction due to the bubble dynamics.

Ch. Arn, Ph. Dupont, F. Avellan

Cavitation Erosion Prediction on Francis Turbines-Part 1 Measurements on the Prototype

In the process of developing tools for cavitation erosion prediction of prototypes from model tests, 4 on board aggressiveness evaluation methods were tested on a severely eroded blade of a 266 MW Francis turbine. These are pressure, pit counting, DECER electrochemical and vibration measurements. All methods provided coherent results on the blade mounted measurements. The test program provided understanding of the heterogeneous erosion distribution of the prototype blades and quantitative data for comparison in subsequent tests on the model of the machine.

P. Bourdon, M. Farhat, R. Simoneau, F. Pereira, P. Dupont, F. Avellan, J.-M. Dorey

Determination of Critical Cavitation Limit in the Pressure Control Devices

This paper describes a method for predicting the cavitation intensity of hydraulic devices by means of the analysis of recorded pressure signal. Infact, it seems possible to get in detail the behaviour of the observed variable through the determination of some frequencies of the pressure fluctuations produced by flow turbulence and by the cavitation impulses. It has been found that it is possible to define the numerical value of the critical cavitation limit by means of the determination of a global index as the root mean square of the pressure fluctuations and in which way the spectral signal analysis proves this approach.

A. Castorani, G. de Martino, U. Fratino

Stability of Air Cavities in Tip Vortices

A model is presented to interpret the behavior of air cavities formed in the tip vortex issued from hydrofoils in a water tunnel, when air is injected. These cavities have the ability to move upstream and reach a stable position near the hydrofoil. From detailed measurements of the velocity components the conditions for the cavities to appear are examined, and compared to the model results. The proposed model, that is based on the equilibrium balance of the cavity, and takes into account that it is closed downstream by a reentrant jet developing in a Lamb vortex, can explain aspects of the behavior not contemplated by another models.

A. Crespo, F. Castro, F. Manuel, D. H. Fruman

Cavitation Erosion Prediction on Francis Turbines-Part 3 Methodologies of Prediction

In the frame of a joint research programme between EDF, Hydro-Québec and IMHEF, different methods are investigated to predict cavitation erosion on Francis turbines from model. They are based on measurement of pitting, pressure fluctuations and acceleration. The measurement techniques have been detailed in Part 1 and Part 2. The present article describes essentially the theoretical and practical aspects of the methods and discusses the results obtained until now from the model and prototype tests. The first analysis shows that the methods proposed are suitable to measure cavitation aggressiveness on model and on prototype, and that the level on the model is several orders of magnitude smaller than on the prototype. To adjust transposition laws, a more complete set of data is needed.

J. M. Dorey, E. Laperrousaz, F. Avellan, P. Dupont, R. Simoneau, P. Bourdon

Cavitation Erosion Prediction on Francis Turbines. Part 2 Model Tests and Flow Analysis

Different measurement techniques have been used to detect cavitation on a Francis turbine model. The results are compared to those obtained on the prototype and presented in the first of this series of articles. The runner model used for that study is build on the basis of a geometrical recovery of one of most eroded blade of the prototype. The results of the different measurements are presented and commented by comparison with prototype measurements. This comparison leads to a proposal of the physics which should be involved in transposition laws for the prediction of prototype erosion from cavitation model tests. The consequences of such scaling laws, as well as their application to the prototype and model results, are part of the third facet of this work.

Ph. Dupont, J.-F. Caron, F. Avellan, P. Bourdon, P. Lavigne, M. Farhat, R. Simoneau, J.-M. Dorey, A. Archer, E. Laperrousaz, M. Couston

Impact of Vapour Production and Cavity Dynamics on the Estimation of Thermal Effects in Cavitation

Vapour production through cavitation extracts heat from the fluid surrounding the cavity and creates a temperature difference between liquid and vapour. This thermal effect is particularly significant in cryogenic liquids. Estimates of the temperature difference can be made provided: a) the rate of vapour production required to sustain a given cavity, and b) an appropriate model for the heat exchange at the interface of the cavity are known. The vapour production is usually estimated by assuming that it is equal to the non condensable gas flow rates necessary to sustain ventilated cavities of equal geometry. It has been shown that experimental results previously obtained can be quite satisfactorily predicted if the interface is assimilated to a rough flat plate through which the amount of heat necessary to generate the vapour is being fed. In order to obtain data for cavities developed over walls whose geometry and pressure gradient are analogous to those of turbopump inducers, and to achieve a better precision, tests were conducted in a specially designed cavitation loop operating with Freon 114. It is shown that the experimental results are well predicted by the proposed model.

Daniel H. Fruman, Jean-Luc Reboud, Benoît Stutz

Aeration Versus Cavitation in Dam Spillways: Self-Aeration and Artificial Aeration (Aerators)

This paper describes the most important aspects of the effect that flow aeration has in combating the damage caused by cavitation. An explanation is also given of how to characterise these flows when the air enters the flow in a natural way, as well as giving an account of some of the artificial aeration devices that are placed on spillways when natural aeration proves to be insufficient.

Ramón Gutierrez Serret

Leading Edge Cavitation in a Centrifugal Pump: Numerical Predictions Compared with Model Tests

The aim of this paper is to present the results obtained with a 3-D numerical method allowing the prediction of the cavitation behaviour of a centrifugal pump and to compare this prediction to model tests.The proposed method consists in assuming the cavity interface as a free surface boundary of the computation domain and in computing the single phase flow. The unknown shape of the interface is predicted by an iterative procedure matching the cavity surface to a constant pressure boundary (p v ). The originality of the presented method is that the adaptation process is done apart from the flow calculation, allowing to use any available code.

R. Hirschi, Ph. Dupont, F. Avellan, J.-N. Favre, J.-F. Guelich, W. Handloser

The Relation between Erosion Ripples on the Wetted Surface of Hydraulic Turbine and Instability Waves in the Turbulent Boundary Layer

The perturbation theory is used to analyze the hydrodynamic instability of the viscous sublayer in the turbulent flow near the wall region. Based on the relation between the viscous sublayer and the coherent structure in the turbulent boundary layer, a simplified turbulent coherent structure model, or one for the unstable wavelike motion of the Kelvin-Helmholtz vortex layer with approximated continuous distribution under the action of perturbation, is proposed. This model has been used to study the motion of dilute particles in the turbulent boundary layer, and it has been found that the coherent structure plays a dominant role in the motion of dilute particles.

Shehua Huang, Liangjun Cheng

Acoustic Method and Its Applications on Measuring and Judging Cavitation of Hydraulic Turbine

In past 15 years, acoustic methods on judging cavitation has been adopted in vacuum model tests and prototype observations in hydraulic structures in the institute.The researches has been made on the physical machanism and acoustic characters of cavitation, and its measuring and judging methods. This paper will give some of the researches related with hydraulic turbines.

Kehuang Liu, Chun Yang

Numerical Simulation for Dilute Sandy Water Flow in Plane Cascade

The flow of dilute sandy water (sand carrying capacity less than 1.0kg/m3 ) in plane cascade is numerically solved. The collision efficiency is defined. The influence of the Reynolds number based on the characteristic length of the cascade, the angle of attack and the sand diameter on the collision efficiency are discussed. The trajectories and impinging action of sand particles moving in plane cascade flow are calculated by including the effects of the wall, the boundary layer, the wake, the sand particle size and the collision of the sand particle-wall.

X. B. Liu, Q. C. Zeng, L. D. Zhang

Review of Research on Abrasion and Cavitation of Silt-Laden Flows Through Hydraulic Turbines in China

Silt content is very high in many rivers in China. A large number of hydraulic turbines working in these rivers encounters serious problems of silt abrasion and aggravated cavitaion. In lighter cases machine performances are impaired and in serious cases structural damages may result. In recent years, Chinese hydraulic engineers are faced with the serious challenge of solving the problems of silt-laden flow in the many hydro projects on the Huanghe (Yellow River) and the Changjiang (Yangtze River) and many other silt-laden rivers. Attention has been focused on the cavitation erosion rate in the sediment carrying flow, that is, the effects of silt concentration on the cavitation behavior and intensity of cavitation erosion. This report covers the major results of field observation and analysis of extensive laboratory research in recent years.

Zu-yan Mei, Yu-lin Wu

Effect of the Leading Edge Design on Sheet Cavitation Around a Blade Section

An experimental study of the cavitation behaviour of two-dimensional hydrofoils simulating a section of an inducer blade has been performed in the CREMHyG large cavitation tunnel. Two leading edge shapes, chosen at the request of the SEP to approach rocket engine inducer designs, have been tested with respect to the development of sheet cavitation. Pressure distributions along the foil suction side and the tunnel walls have been measured for different cavit lengths. Cavity shapes have been provided by visualisations under laser sheet light and image processing. Laser anemometry (LDA) around the leading edge and total pressure measurements along the foil suction side have been performed to characterise the effects of cavitation on the liquid flow. A numerical model of cavitating flows has been improved using those experimental results. Numerical predictions have been performed, showing a good agreement with experiments on a large range of flow conditions.

Jean-Luc Reboud, Claude Rebattet, Philippe Morel

Hydraulic Transients and Control Systems Related with Hydraulic Machinery and Plants

Optimal Closure of a Valve for Minimzing Water Hammer

Due to its relevance, many authors have paid a lot of attention to determine the optimal closure of a valve. As some of the results have in practice difficult implementation, the problem remains open.This paper deals with the different methods of analysis, and highlight their hypoihcois, results and reliability. By the simultaneous use of the rigid and the elastic /uod,|u, the influence of the inertia and the elasticity of the system, regarding the optimum oloauro, are considered. FiuxUy, some practical conclusions are pointed out.

J. Abreu, E. Cabrera, J. Garcia-Serra, P. A. Lopez

Qualitative Flow Visualizations during Fast Start-up of Centrifugal Pumps

The main uses of pumps in hydraulic plants require quasi-steady state operation. However, some applications such as rocket engines include severe transient phases. The approach assuming that at any time, all flow conditions are identical to those occurring during steady-state working is no more sufficient. Thus several laboratories have started to study the behaviour of turbomachines working under unsteady conditions ([SAI82], [TSU82]).

J. P. Barrand, A. Picavet

Analysis of a Numerical Model for the Oscillatory Properties of a Francis Turbine Group

The present work represents an assessment of the authors’ several years of study on the application of the transfer matrix method to hydropower plant components. It is part of the strong research effort promoted by ENEL-DSR-CRIS on the oscillatory properties of hydropower groups equipped with Francis turbines, particularly important in presence of cavitating vortex rope (CVR).

A. Cattanei, A. Capozza, P. Molinaro

Transients Analysis and Dynamic Criteria for Hpp Exploatation

Transient regimes in hydro power plants and hydraulic systems are: normal operating conditions, emergency operating regimes and catastrophic cases. The list of analyses which must be carried out and some backgrounds needed for these computations are discussed. This list is vital for economical design, construction and operation of hydraulic systems.The paper deals with characteristics of turbomachines in pump, turbine and dissipation operating zones, Some methods of four quadrant representation are briefly described; Suter and modified Suter curves are particularly pointed out. ‘The methods of interpolation and extrapolation of performance curves is shown.Hydraulic transients analysis (waterhammer, hydraulic oscillations and stability), extreme values of stresses and the pressure fluctuations along the system and other characteristics important for the safety of the plant, should be determined. Comparing machine vibrations to the recommended statistical data, exposure rate can be estimated and probability of vibration appearance considered.

A. Gajić, S. Pejović, LJ. Krsmanović, Z. Stojanović

Modelling a Protection Device in a Low Pressure Lifting System

A water hammer protection device in lifting systems has been experimentally studied, especially low pressure ones [1][2][3][7] (see figures n°1 and n°2, pages n°5 and n°6 of the paper). Water hammer, in the experimental tests of the analyzed device, has been caused by the fast water flow interception with a spherical valve located just downstream from the pump of the hydraulic system. Such a device is composed of a generic check valve and a concentrated head loss located at one end of a bypass; the latter must be installed between the upstream reservoir and the delivery pipe of the hydraulic system, just downstream from the section in which water flow cutoff has been generated, with the valve close to the reservoir. The valve allows water to flow from the reservoir to the delivery pipe through the bypass. The positive experimental results regarding the efficiency of this device bypass-check valve-concentrated head loss in protecting low pressure lifting systems from water hammer, have been the starting point for studies which have been directed towards the creation of a numerical model of the hydraulic system. From the trials presented in [1][3], the authors have noted that a numerical model, simulating water hammer in a system equipped with the studying protection device, must use theoretical studies from the cavitation and models reproducing it. Previously [2], a simulating model, without considering the possibility that cavitation occurs, was realized. As the new experimental trials show [1][3], having been carried out starting from steady velocities much higher than older ones [2], the previous simulating model can he applied only when cavitation does not occur.

Orazio Giustolisi, Michele Mastrorilli

Dynamic Compression of Entrapped Air Pockets by Elastic Water Columns

The problem of dynamic compression of entrapped air pockets in a pipeline has a great technical importance when designing water pressured pipelines.

R. Guarga, A. Acosta, E. Lorenzo

Generalization of Pump Station Boundary Condition in Hydraulic Transient Simulation

In this paper, a generalized treatment of the equations describing the behavior of a pumping stat ion is presented. It enables the simulation of many reasonable combinations of elements within a pumping station, by using a single routine and one set of state data suitably maintained. Thus, it represents a unified algorithm massively solved that easily accommodates to virtually every condition in a pumping station. Besides, the interpolating method it uses allows to incorporate manufacturers information into dimensionless curves helping plausibly to produce more real results. The interest of the paper is mainly computational. Nevertheless, the scheme we propose enables the joint consideration of a wide range of elements in a pumping station without the need of including short reaches between them. In this way, it is claimed to represent a good improvement for existing packages to model transient analysis in networks.

J. Izquierdo, P. Iglesias, V. Espert, V. Fuertes

Analysis of Unsteady Characteristics of Flows Through a Centrifugal-Pump Impeller by an Advanced Vortex Method

In the present study, the applicability of the vortex methods is extended for the analysis of unsteady characteristics and occurrence of unsymmetrical flow through a centrifugal impeller of a pump. In this numerical investigation, two-dimensional unsteady features of the whole flow-field are solved without introducing the periodicity of the blade-to-blade stream, by using an advanced vortex method. As an example of applications of the present method, unsteady flows through a centrifugal impeller consisting of simple configuration of blades were calculated, and the results show how the separation of boundary layer develops into a strong vortex structure and how the vortex is periodically convected downstream and produces an oscillatory flow in a blade-to-blade passage.

K. Kamemoto, H. Kurasawa, H. Matsumoto, Y. Yokoi

Expert System for Analysis of Pumped Storage Schemes

In Pumped Storage Schemes (PSS) it is necessary to know the hydrodynamic personality of the whole installation, in extensive and transient period of operation, in order to prevent inadequate pressure surges and to provide the operational optimization for energy production.This paper presents a general computer model that provides the engineer with the means to simulate steady, transient and extensive period operation of the installation, considering also oscillatory excitations that induce flow instabilities. To perform the analysis, the complete hydraulic characteristics of the turbine-pump are based on Suter variables, adjusted by Fourier series, and PM feedback controller provides the necessary variation of parameters for stability analysis. Part load operation of machines that induce flow pulsation can be analyzed.Examples are presented to illustrate the use of the expert system and to show the influence of parameter variations on the installation components.

E. Koelle, J. G. P. Andrade, E. Luvizotto

Model-Based Analysis of Active PID-Control of Transient Flow in Hydraulic Networks

The transient-flow response to the action of automatic valves driven by the working fluid and under PID-control is modelled and simulated so as to investigate its potential and capabilities as an alternative arrangement for the control and suppression of hydraulic transients in pipeline systems.

José Carlos Lauria, Edmundo Koelle

Simulation of transients in Pressurized Hydraulic Systems with Visual Tools

In contrast with classical methods used to analyze different transient phenomena due to the start-up, shut-off or normal operation on the pressurized hydraulic systems, in which hydraulic machinery is usually present, a different approach based on the use of a visual tool is presented. This tool uses the object oriented language to build the system model by means of block diagrams. Many additional features make a powerful tool to simulates dynamic systems in a very clear and easy way this new software product.

F. Martinez, J. Izquierdo, R. Pérez, A. Vela

Dynamic Behaviour of Governing Turbines Sharing the Same Electrical Grid

A mathematical model of a power generating system containing several turbines, with separate conduit systems, sharing the same electric grid are presented. The purpose of the model is to analyse the dynamic behaviour of the governing turbines during load admittance and load rejectance on the grid with focus on the mechanical and hydraulic side.

Torbjørn K. Nielsen

Prediction of Natural Frequencies in a Hydro Power Plant Supplying an Electric Network by Itself Having a Known Load Type

Consider a hydro-electric power plant, according to Fig.1, whose pipe system consists of a penstock upstream and a tailwater tunnel downstream having different lengths L 0 and L u and hence water hammer-conditioned wave passage times T 10 = L 0 /a 0 ,T lu =L u /a u ,a 0 and a u being the celezities (sonic speeds). The penstock is starting and the tailwatertunnel is ending in a large basin having constant pressure head. The other ends of both the pipe sections are connected with the pressure-and suction-flange of a speed-and gate-controlled Francis turbine having its known hill-diagram. The turbine is driving an alternator (generator) supplying an electric network by itself having a known load type,e.g. resistors, machine tools, pumps and blowers. This set has a definite start up time T a . The turbine gate is adjusted by a servomotor, see Fig.1 via a valve activated by a speed deviation meter(here symbolized by a fly-ball pendulum), and fed back by the servomotor piston rod via a dash pot, ensuring a stiff connection of valve-activating lever end and servo motor piston under quick motion here considered. This integral proportional governor by its elastic feedback has a reset time T i , see [1].

J. Raabe

Modelling and Practical Analysis of the Transient Overspeed Effect of Small Francis Turbines

The transient overspeed effect of slow Francis turbine can be a major factor in the design of a small hydroelectric scheme with long penstocks or tunnels.After a full-load rejection, iit a small hydroelectric power plant equipped with low inertia Francis turbine, the runner speed will increase and the runaway conditions can be quickly attained. The hydrotransients induced by overspeed of reaction turbines depends on the turbine type. The specific speed indicates the main characterization of the turbine behaviour in what concerns the flow variation during runaway conditions.In order to improve the pressure transient analysis since the early phases of the design of the hydraulic circuit, for lack of information about the equipment to be installed, the authors present in this paper a simplified method for reaction turbines modelling that was validated by computational and experimental comparisons.

Helena Ramos, A. Betâmio De Almeida

Simulation of Turbine Governing in Time Domain

Approximation in frequency domain using the Limited Zero and Pole Method

In this work there have been developed rational approximations for the elastic pressure/flow transfer function in a pipe. The approximation method is called Limited Zero and Pole Method, and it approximates the function with a selected number of its own zeros and poles. The method was carried out by Li [3] for a system without friction. The approximation has been successfully extended by including a stationary friction term. Originally, the pressure/flow transfer function is a transcendent.’ function, but becomes algebraic and linear through the approximation, This rakes it possible to transfer the function from the frequency domain to the time domain, to use modern automatic control theory for linear systems. A state space formulation are established and the pressure and the flow are simulated in time domain. The simulations are compared with results computed by the well known Characteristics Method, described by Streeter & Wylie [1]. The results shows very good agreement.

Dag Birger Stuksrud

Parametrical:Modelling of Power Characteristics of the Francis and Kaplan Hydraulic Turbines

This paper proposes a new approaches on the mathematical modelling and numerical simulation of the steady-state running of Francis. Kaplan and propeller hydraulic turbines by using the new techniques of parametric identification from experimental data.

Mugur Florin Tolea, Jean Louis Kueny

Unsteady Frictions in Pipelines

Simulation in time domain using the modified Allievis equations

This paper presents a friction model to be used in a numerical program to get a solution of oscillatory flow. The model is in time domain. The model is build on Allievis equations, shown in Streeter & Wylie [2]. The intention of this work is to ma a model that simulate a transient flow with good accuracy. This is done using a diffusion term as the model for the unsteady friction. This has been done by B.Svingen [3] using a finite element method. The diffusion term added as an extra term in to the Allievis equation. The equations are solved by the method of characteristics. The model is compared with a laboratory test done by Eichinger [1], and shows good agreement with that experiment.

Vennatrø Roar

Oscillatory and Vibration Problems in Hydraulic Machinery and Power Stations

Swirl Flow in Conical Diffusers

In order to study swirling, confined, turbulent and non cavitating flow in conical diffusers, a series of 2D LDV and pitot measurements are described. A tangential entry swirl flow inducer was used to generate the swirl flow, LDV was used for the measurements of velocity distribution, and a three hole pitot tube was used for the pressure distribution measurements. The Reynolds number was constant at Re = 5,7 * 105 and the Swirinumber varied from S = 0,1 to S = 0,35. Numerical analysis with FLUENT’s RSM, RNG and K-ε turbulence models were carried out. The experimental and numerical results are compared and discussed.

Ole Gunnar Dahlhaug

Experimental Investigation of Vortex Core in Reverse Swirl Flow From Francis Runner

It is well known that the precession movement of the helical vortex core generated in the swirl flow from the runner exit of Francis turbine causes the pressure oscillation in the draft tube. This phenomenon which occurs in the part-load operation have been investigated in detail for a long time. At the part-load operation, flow from the runner exit has the swirl in the same direction as that of the runner rotation. On the other hand, when the turbine operates at over-load, flow from the runner exit has the swirl against the runner rotation and the characteristics of the pressure oscillation in such a situation as this is very different in quality from that at the part-load operation. Sometimes this phenomenon raises a hard problem in the actual plant. In the present paper, the behavior of the vortex core and the basic characteristics of pressure oscillation under the conditions of reverse swirl flow are investigated using a small Francis turbine test rig with a cone-type draft tube.

Yoshinobu Furuie, Hikaru Mita, Yutaka Hosoi

Hydraulic Oscillation Analysis Using the Fluid-Structure Interaction Model

The paper presents a fluid transients analysis in hydraulic systems including fluid-structure interaction (FSI). The method is based upon an improved model, that takes into account gravitational and hydraulic resistive forces. The results of calculation applying the FSI model are compared to results obtained by simplified models and to those recently published.

A. Gajić, S. Pejović, Z. Stojanović

Francis Turbine Surge: Discussion and Data Base

A system is considered as stable in sustained operating conditions if the response to unavoidable disturbances does not endanger its safety. Normal disturbances of hydraulic turbines are pressure fluctuations [3, 5, 10], influenced by machine design, operating conditions and by the dynamic response of the water conduits and rotating components. They may be associated with mechanical fluctuations of shaft torque, rotational speed, hydraulic load on guide vanes etc. as well as with vibrations.

Thierry Jacob, Jean-Eustache Prénat

Non-Stationary Flow in Reversible Francis Turbine Runner Due to Wakes Trailing the Guide Vanes

Experimental and numerical analysis are presented of the unsteady flow in a reversible Francis pump turbine runner due to the effect of wakes trailing the guide vanes. The dynamic fluctuating pressure has been measured with pressure transducers on the inlet part of the runner, blades in generating mode in two operational points. A numerical solver of three-dimensional turbulent flow based on weakly compressible Navier-Stokes equations is applied to predict the unsteady flow in the turbine runner. The effect of the guide varies upstream the runner is modelled with a sinusoidal inlet function which fluctuate the absolute inflow velocity.

Jo Jernsletten

The Swirling Inlet Flow Effects on the Pressure Recovery of a Low Head Water Turbine Draft Tube

In this paper, the experimental results concerning the swirling inlet flow effects on the pressure recovery of a divergent passage whose cross-section changes from annular to rectangular were discussed. A developed turbulent flow having a swirling flow component was introduced into the diffuser and the alterations of the flow patterns both in the velocity and pressure distributions in each test cross-sections were carefully examined. The influence of swirling flow component on the axial velocity profile as well as on the streamwise wall pressure distributions was identified for different inlet swirl conditions. To give a detail explanation to the flow behavior and pressure recovery, a flow visualization technique using a tuft grid and surface tuft methods was adopted. Furthermore, based on the measured velocity and pressure, the approximated energy profiles at the best efficiency point of the unit were presented.

K. Kikuyama, Y. Hasegawa, G. Augusto, K. Nishibori, S. Nakamura

Two Kinds of Whirl on Fixed-Blade Propeller Type Turbine

Two kinds of whirl have been identified on propeller-type generating units. In both kinds, the shaft whirl follows the same direction as the rotation of the machine and the source of the instability seems to be the interaction between the fluid and the structure. The first kind of whirl was observed at 0.33 Hz at plant #3 of Hydro-Québec’s Beauharnois powerhouse and, also, at 0.25 Hz at LG-1 and Carillon. It occurred below the peak efficiency operating range. The vibration at 0.33 Hz (0.21 times the rotation speed) affects not only the power output but also the vibration at the guide bearing and the pressure in the penstocks. The second kind of whirl was observed at around 5.14 Hz (3.6 times the rotation speed) at LG-1 and occurred at no load or low load. Measurements at the bearings combined with those of the runner displacements reveal that the shaft undergoes like the second bending mode. Proximity and pressure measurements taken where the blade passes on the discharge ring of the runner at LG-1 shows that the whirl at 5.14 Hz is due to a runner instability.

F. Léonard

Self-Excited Hydraulic Oscillations Due to Unstable Valve Behaviour. A Case Study

During the bottom outlet operation of a dam, a single-seal bureau-type valve experienced strong vibration, which in turn induced severe pressure oscillations in the conduit system, with peak to peak amplitudes of up to three times the static pressure.This paper presents the facts and the frequency domain analysis of the system. Special attention is paid to the gate characterization as internal exciter of the hydraulic oscillation. Analytical results are compared with pressure recordings in order to validate the hypotheses and procedure.

C. Mateos, T. Pérez-Andújar, M. Andreu, E. Cabrera

An Experimental Study on Fins, Their Role in Control of the Draft Tube Surging

Although the method based on installation of fins has been widely used to alleviate the draft tube surge at part-load operation of a Francis turbine, the corresponding mechanism is still unknown. Thus, the present study was made to clarify the effect of fins on the pressure surge experimentally by analyzing wall pressure fluctuations.

M. Nishi, X. M. Wang, K. Yoshida, T. Takahashi, T. Tsukamoto

Model for Vortex Rope Dynamics in Francis Turbine Outlet

The present communication deals with the simulation of the vortex rope present in the outlet of a Francis turbine when operating at partial load regimes This approach is not devoted to the analysis of a complete turbomachinery system but rather to develop a rigorous mathematical analysis for a local phenomenon of practical relevance, We synthesise the problem calculating the three-dimensional unsteady motion of a vortex filament confined inside a cylindrical wall. In this first analysis we ignore cavitation effects in order to detect those phenomena which can be explained in teams of incompressible vortex dynamics The system rope-diffuser is assumed as isolated and we neglect any direct interaction with the turbine blades or the diffuser outlet.

G. Pedrizzetti, G. Angelico

Vortices Rotating in the Vaneless Space of a Kaplan Turbine Operating under Off-Cam High Swirl Flow Conditions

Vortices rotating in the vaneless space of a Kaplan turbine model have been observed under variety of off-cam sluicing conditions. The number of vortices was influenced by the discharge (guide vane opening) and the air admission as well. The phenomena observed could be described in terms known from wave analysis, like modes, stability limits, etc.

L. Půlpitel, A. Skoták, J. Koutník

Experimental Investigation of Frequency Characteristics of Draft Tube Pressure Pulsations for Francis Turbines

This paper presents an experimental investigation of frequency characteristics of draft tube pressure fluctuations for Francis turbines. The different types of vortex ropes in draft tubes were experimentally observed and categorized. The relationship between the surging frequency and the vortex pattern is discussed. The frequency characteristics of pressure pulsations at different measuring locations of the draft tube are also presented.

Qinghua Shi

On the Suppression of Coupled Liquid/Pipe Vibrations

The vibration of liquid-filled pipe systems can be caused by internal (pressure surges) or external (rotating machinery) sources, or a combination of both. Liquid pulsations can be diminished by devices like an air chamber, whereas pipe vibrations can be reduced by suitable supports. The performance of vibration suppression devices in the system under consideration can be investigated by numerical simulation. This is usually done with conventional waterhammer or pipe-stress computer codes. Present-day software also permits fully-coupled liquid-pipe analyses. The analysis of suppression devices and their interaction with the entire system is an important tool in design and trouble shooting.The dynamic behaviour of a simple system consisting of one pipe, one liquid column and one suppression device is studied herein by numerical simulation and physical experiment. Liquid-pipe interaction mechanisms are taken into account. A closed water-filled steel pipe is excited by external impact. The suppression device is a short piece of plastic pipe. The closed system prevents rigid-column behaviour of the liquid relative to the pipe, so that the local device does not decrease amplitudes of vibration. It does, however, shift the natural frequencies of the system, which is an important means of vibration control. This simple and cheap device may find its application in reducing fatigue problems and in noise reduction.

A. S. Tijsseling, A. E. Vardy

Unsteady Hydraulic Force on an Impeller Due to Rotor-Stator Interaction in a Diffuser Pump

Fluctuations in hydraulic forces on impeller may be caused by interaction between impeller and diffuser vanes/volute casing. Knowledge of the unsteady forces related to the rotor vibration is important to understand rotor dynamics of turbomachinery. There, however, is a limited quantity of data available on the unsteady forces on diffuser pump impeller (Brennen, 1994).

H. Tsukamoto, M. Uno, W. Qin, T. Teshima, K. Sakamoto, T. Okamura

Swirling Flow with Helical Vortex Core in a Draft Tube Predicted by a Vortex Method

A mathematical model is proposed in this paper to describe the swirling flow with spiral vortex core in a draft tube. In the model, it is assumed that the flow in the circular cross-section is given by the sum of an axisymmetric base flow and that induced by an infinite helical vortex filament in a pipe. The corresponding computer program is developed where a series of the image vortex filaments is considered so as to satisfy the boundary condition at the pipe wall, providing that the flow rate, swirl rate and circulation of incoming flow are prescribed as input values. The velocity distribution in the cross-section and the rotating frequency of vortex core are calculated and compared with experimental results in draft tubes. The reasonable agreement between them is obtained.

X. M. Wang, M. Nishi

Experimental Investigations Related with Hydraulic Machinary and Its Applications

Friction Loss of Rough Wall Passage in a Turbomachinery

An impeller or a flow passage in a turbomachinery is normally made from casting and the wall surface of the passage is usually not hydraulically smooth on the friction loss. When the friction loss of the passage is estimated by using the friction loss coefficient, it is necessary to consider the effect of the roughness of the passage wall. The friction loss coefficient of the passage is commonly obtained from the well known formulae based on the equivalent sand roughness Ks, However, the roughness of the surface of a prototype turbomachinery is expressed by the arithmetic average roughness Ra. Therefore, it is necessary to determine the correlation between Ks and Ra, or to obtain the formula of the loss coefficient based on Ra.

S. Akaike

Redesign of Sharp Heel Draft Tube — Results from Tests in Model and Prototype

Draft tubes with a sharp heel are quite common in hydro power stations constructed 50 years ago. This design reduced both construction time and investment. We have shown that there is potential for increasing unit performance by a moderate modification of such draft tubes. The redesign has been done with the aid of model tests in a turbine rig. The new design has been installed in a 50 MW hydro unit and an efficiency improvement in the order of 0.5% has been verified through accurate measurements.

N. Dahlbäck

Model and Prototype Draft Tube Pressure Pulsations

All producers of Francis turbines daily meet the problem of draft tube vortex appearance and its consequences [1,2,3,4]. Limitations in turbine operation because of the vortex appearance and undesirable consequences on hydraulic system are sufficient reason for its intensive researching. According to such unpleasant and sometimes only hard soluble consequences on the prototypes, the research is directed to model tests and numerical flow analysis with the aim to predict consequences on prototype.

Vladimir Kercan, Marin Bajd, Vesko Djelić, Andrej Lipej, Dragica Jošt

Fish Bypass System Impact upon Turbine Runner Performance at Rocky Reach Dam

Public Utility District No. 1 of Chelan County, Washington State, USA (Chelan) owns and operates the Rocky Reach and Rock Island Hydroelectric Projects located on the Columbia River. Several species of salmon and steelhead trout migrate through these projects. Downstream migrating juvenile fish pass through either the turbines, where they experience some mortality, or through migration-augmenting spill flows, which are marginally effective and very expensive. Chelan has developed a “Surface Collection” fish bypass system and “Fish-Friendly” turbines at its Rocky Reach Project to mitigate harmful effects to the downstream migration of the juvenile fish. The surface collection system will be installed within the Rocky Reach forebay, supported by the upstream face of the dam, and will encroach upon the generating unit intakes. This paper describes the modeling investigation used to define the effects the surface collector structure will have upon turbine operating characteristics and how that information will be used to balance bypass effectiveness with turbine performance.

Alfred Lang, Bill Christman

LDV Measurements in an Impeller-Generated Turbulent Jet Developing in a Weak Coflow

Results are presented from detailed measurements in an impeller-generated turbulent jet developing in a weak coflow. A laser doppler velocimeter was used to measure the mean velocities and normal stresses in three coordinate directions, together with two of the Reynolds shear stresses. An impeller operating at 1800 rpm was placed in a flume with a coflow having a velocity of 0.05 m/s upstream the impeller. The coflow velocity decreased uniformly along the impeller jet as water was entrained into the jet. Complete velocity measurements were made at selected downstream cross sections, including both the zone of flow establishment and the zone of established flow. The mean velocities and the normal stresses approximately displayed self-similar behavior 6D downstream the impeller, where D is the impeller diameter.

Per Petersson, Magnus Larson, Lennart Jönsson

Inline Radial Force Measurement of Turbine Runners

Radial forces are important parameters for the design of pump turbines and Francis turbines. Until now the forces were measured with special measuring arrangements. Establishing these arrangements and measuring itself was very time consuming. As the time between start of the project and going into service of the machines should most of the time be as short as possible, these tests were often canceled. Consequently important machine parameters were not available for the design of the rotating parts of the turbine. To overcome this discrepancy, ASTRÖ developed a new measuring system, which allows the change from normal testing to radial force measurement without any delay. The following paper shows details of the new solution as well as first measuring results.

Josef Riener, Arnold Egger, Gerald Schnur

Different Types and Locations of Part-Load Recirculations in Centrifugal Pumps Found from LDV Measurements

LDV measurements were carried out to analyse the flow field at part load operation of three test impellers having different values of specific speed. On the basis of the measured flow velocities, different types of flow recirculations were identified which can occur in the same impeller from different causes, at different locations and at different values of the flow rate.

B. Stoffel, K. Weiss

Turbulent 3D Flows near the Impeller of a Mixed-Flow Pump

Experimental Investigations for a Specific Speed of 120

Compared to other fluid machines, the flow patterns in a mixed-flow pump are highly complex, due to adverse pressure radial gradients, secondary flows and the flow separation effects in the impeller inlet and outlet.This paper deals with experimental investigations and their analysis. Special emphasis is laid on the three-dimensional flow fields at the inlet and the exit of a closed mixed-flow impeller to improve design and calculation methods.By means of a hot wire probe and an angular trigger blade-to-blade measurements of the flow field at the impeller inlet and outlet have been performed from the hub to the shroud within the entire operating range of the pump.Results of particular interest are the performance curve instability of a mixed-flow pump, the beginning of recirculation at the inlet and the exit of the impeller and the influence of inlet guide vanes in the pump suction side upon the pump performance.

Baogang Wang, Dieter-Heinz Hellmann

Practical Applications of the Hydraulic Machinery

Overload in Kaplan Turbines of Salto Grande Hydropower Complex

The Salto Grande hydropower complex, in industrial operation since 1982, is located on the reach of the Uruguay river which forms the international border between the Uruguay and Argentina. The complex includes two simetrical power houses located on the river banks and a surface spillway located in the middle of the river. Each powerhouse includes seven 135 MW generating units resulting in a total installed capacity of 1,890 MW. The turbines are of the vertical shaft Kaplan type, of 187,500 CV each, under a rated net head of 25.3 m, and a specified rotation at speed of 75 r.p.m. The runner diameter is 8.5 m, The Salto Grande Kaplan turbines had a power limit for each head, which was guaranteed by the supplier. This allowed power limit decreased remarkably as the head decreased, which provoked a direct energy loss when the river had abundance of water, In order to increase the generation, studies and tests were carried out with the turbines operating under overload conditions.

Mabel Baccino, Enrique Bonnecarrere

Full Sized Tests on a French Main Coolant Pump under Two-Phase Flow

The French Safety Authorities required EDF to demonstrate the ability of the new main coolant pump to withstand two-phase flow conditions without damage. Therefore three full sized tests, simulating a bleeding flow on the primary system, were performed on a laboratory test loop at real operating conditions (temperature = 290 °C, pressure = 155 b, flowrate = 7 m3/s, electrical power 7 MW). The maximum value of the mean void fraction reached 75%. The outcome of the tests is very positive: the mechanical behaviour of the main coolant pump is good, even at high void fraction, The maximum vibration levels were below the limits fixed by the manufacturer. Correlations between the mechanical behaviour of the pump and the pressure pulsation in the test loop have been found.

J. C. Huchard, C. Bore, E. Dueymes

Computer Simulations of Dynamic Performance of 400 MW Adjustable Speed Pumped Storage Units

Two units of the world largest 400 MW adjustable speed pumped storage units for Ohkawachi Power Station, The Kansai Electric Power Inc. , Japan were commissioned in December 1993 and June 1995 respectively.They were designed to utilize the flywheel energy of the rotor to the maxim extent posseble so that they can change in stepwise at least 32 MW in the generate a and at least 80 MW in the pump mode within 0.2 second. The pump turbines have two kinds of local multivalued speed versus discharge characteristics, so-called S characteristics in the turbine operation zone and so-called hump characteristics in the pump operation zone, against which careful consideration are required to minimize their influences. The two pump turbines are connected to the same penstock and receive water hammer interferences from each other through the bifurcation.Since the adjustable speed machines were considered too difficult to be adjusted by trial and error in the field, a complete set of simulation studies corresponding to the series of the commissioning run on and adjustment know-hows were lished prior to the tests. As a result, the commissioning tests were completed very smoothly without anyparticular readjustments. This paper is to present some of the computer simulations as well as brief explanation of the sophisticated computer simulation programs for the adjustable speed pumped storage units. Computer simulation studies focused on electrical equipment such as the generator motors and the cycloconverters were done separately from those focused on the mechanical equipment. Hereinafter Discussed are only the latter.

Eizo Kita, Hiroto Nakagawa, Takao Kuwabara, Masataka Harada

Performance Comparison of Nuclear Reactor Recirculation Pumps Tested under Large Reynolds Number Difference

The reactor coolant recirculation pump mounted on advanced boiling water reactor (ABWR) is a single-stage mixed flow pump. On 10 pumps, hydraulic performance tests were performed under largely different Reynolds number conditions. In this paper, pump characteristics tested under low Reynolds number conditions are strictly converted, with regard to the influence of the variations of Reynolds number in particular, and they were compared with the results of the tests under high Reynolds number conditions. Through these tests, the methods for converting performances into those under different Reynolds number conditions have been established for the hydraulic performances of mixed flow pumps mounted directly on a vessel without upstream and downstream pipes and the validity of the applied precise conversion method has been verified.

K. Saiki, T. Iikura, K. Matsumoto, H. Komita, M. Kobayashi, T. Saito, H. Tanaka

Performance of Candu Heat Transport Pumps under Two-Phase Flow Conditions

Heat Transport (HT) Pumps in a CANDU (CANadian Deuterium Uranium) reactor circulate heavy water coolant at high temperatures and pressures to remove fuel decay heat from the reactor core. Under a Loss-of-Coolant-Accident (LOCA), HT pumps are required to maintain their operational capability and structural integrity. Licensing requirements stipulate that HT pumps continue to operate under specific conditions, such as two-phase flow, for a specific duration (17 minutes) During this time period, decay heat has to be removed continuously from the reactor core before safe reactor shutdown is initiated.This paper presents a summary of the simulated two-phase flow tests performed at the HT Pump supplier’s test plant (Sulzer Bingham Pumps, Inc.) on a full-scale CANDU 6 HT pump-motor set. Results demonstrate that the HT pumps maintained their performance and structural integrity under the worst two-phase flow conditions attainable on the test loop for the specific duration of 17 minutes. The pump hydrodynamic bearing, lubricated by the pumped fluid, did not suffer any notable damage. Despite the lack of external seal injection during two-phase flow conditions, the mechanical seals showed no signs of damage.

H. Samarasekera, A. N. Kumar

Interdependence of Draft Tube and Tailwater Flow in Bulb Turbine Power Plants

The advantages of bulb turbines result mainly from smaller dimensions and better efficiencies in comparison to the vertical-axis Kaplan type used earlier. Bulb turbines are characterized by high values of specific speed n q which is closely related to the unit discharge Q´1. Optimum results have been obtained with Q´1-values that rise in proportion with the specific speed. As the specific velocity head h c1/H at the runner exit increases with the square of the unit flow Q´1 an efficient deceleration of the flow in the draft tube is decisive for the performance of high specific speed turbines. Still the exit losses increase with the specific speed n q and may reach values of 5% or more of the available head.

CH. Schneider, W. Knapp, R. Schilling

Study of Hydraulic Transients Using the Bond Graphs Method

One of the latest modeling technique for dynamic systems studies came through the 1950’s. Created by H.M. Payner (Speranza, 1992) who, according to Thoma (1975), felt the urge to create methods and criteria able to generalize the accomplishment of state equations representative of a system. That is the “Bond Graphs” technique. It is based on the power flow among the system components and provides a generalized approach to the system state equations, allowing simulation and interaction of different power domains like: hydraulical with mechanical, or mechanical with electrical, and others. The “bond graphs” techiniques became a powerful tool used in studies and modeling of dynamical systems as seen in works by Karnopp and Rosenberg (1983 and 1975).

Geraldo Lúcio, Tiago Filho

Monitoring, Predictive Maintenance and Refurbishment

Numerical Flow Analysis of a Kaplan Turbine

This paper details the numerical analysis of the flow in a Kaplan turbine, The purpose of the research is related to the refurbishment of two hydropower plants on the Drava River in Slovenia. This project was begun at the beginning of 1994 and is now nearly finished.

Dragica Jošt, Andrej Lipej, Kazimir Oberdank, Mateja Jamnik, Boris Velenšek

18-Paths Acoustic Flowmeter and Transducer Protrusion

Hydro-Québec’s network includes several power plants of more than 1000 MW. Over the past 10 years, work was done in order to improve the accuracy of the turbine efficiency determination, which in turn, leads to a better management of the hydraulic resources and to a better income.

J.-M. Lévesque, J. Néron, C. M. Tran

Step-up in Rehabilitation: Not a Myth, a Science

New turbines, model tested, manufactured and site tested according to state of the art techniques, show consistently an efficiency step-up at least equal to IEC 995 prediction. Although some manufacturers have cast doubts about the capability to reproduce the step-up in rehabilitation, we provide evidence that it has been achieved on well conducted projects. We successively analyze major factors, on the model and on the prototype, that influence the step-up. We finally review some examples to illustrate the desired attitudes that will maximize all the benefits of the rehabilitation.

B. Mahé, V. de Henau, M. Sabourin

Super System: A Hydroelectric Unit Condition Monitoring System in Operation at Hydro-Quebec

The knowledge of the status and the dynamic behavior of a hydroelectric unit at any time is a major asset to optimize and to timely plan maintenance activities. That knowledge can become available through a user-friendly on-line monitoring system capable of processing and storing a large amount of data, retrievable in various meaningful configurations.After eight year of research and development, Hydro-Québec started the implementation of a monitoring system at the end of 1993. This implementation is in progress reaching all Hydro-Québec’s hydraulic units rated 100 MW and up and most of the smaller capacity units. The system is presently operational on more than (65) hydroelectric units in Brisay, LG1 (La Grande-1), LG2, LG3, LG4 , LA1 (LAFORGE-1) and Manic-5 plants.Our objective in this paper is to show how machine condition monitoring can help solving maintenance problems. We will first describe what are the preventive maintenance goals that led to the decision of implementing such a system; how the design criteria has been chosen and what are the developments breakthroughs realized to reach these maintenance goals. Then, we will use recently acquired real life data to illustrate how data available from a monitoring system can help in the diagnostic of the machines’ condition. Finally, we will discuss some problems encountered during implementation and their solutions.

Y. Mossoba

Application of Rotor Response Analysis to Fault Detection in Hydro Powerplants

In this article some considerations obtained during the utilization of rotor response analysis tecniques in hydraulic powerplants are discussed. An applied research work was carried out in two hydraulic turbines analysing the rotor response both theoretically and experimentally. A developed mathematical model was used to simulate the rotordynamic behaviour of Francis and Kaplan turbines. The main dynamical effects that appear during the operation of the machines are discussed too. A series of measurements were carried out in the turbines using impact hammers to determine the modal behaviour of the units. The tests were carried out with the machine still and in operation. Some results and the comparison with the theory is presented in this article. The improved theoretical model was used for a sensitivity analysis of the different bearings to the main excitations that take place during the machine operation. From this analysis, the best measuring points for condition monitoring were determined.

L. P. Nascimento, E. Egusquiza

The Placement Exploration of Diagnostic Measuring Points on Operating Unit Equipment

With the development of diagnostic technology in abroad, the research and application of failure diagnostic technology on operating unit equipment of water power station in China attract many departments have been rapidly developed in recent years,Its realization and expansion has very important siguificance and effect to the unit safe and economic operation, cost decrease, system simplification, operating level improvement of entire power network and science adminstration. The Ministry of Electrical Power Industry has envolled the study of diagnostic technology as key item of science and technology development plan for year 2000.

Liu Xiao-Ting
Weitere Informationen