5th International Colloquium on Transformer Research and Asset Management

Insulating system of a power transformer consists of an insulating liquid (mineral oil or ester liquid) and cellulose-based materials (paper and pressboard). Dielectric properties of an insulating material may be expressed in terms of relative permittivity and material conductivity. Material conductivity influences electric fields in power transformers exposed to DC voltages (e. g. HVDC transformers), whereas relative permittivity is more relevant in power transformers subjected to power frequency (50 or 60 Hz). It is well known that a higher discrepancy between relative permittivity of liquid-impregnated material and insulating liquid increases electric field in insulating liquid. It is therefore essential to know values of relative permittivity of cellulose-based materials in order to properly evaluate electric field strengths in the transformer. This paper presents both theoretical and practical approach to relative permittivity assessment of two types of cellulose material in two different impregnation states.

An unexplained difference between noise levels of 50 and 60 Hz transformers was presented in the previous published work of CIGRE workgroup A2.54 titled ‘No-load sound power levels for specification purposes derived from more than 1000 measurements–a representative figure for three-phase transformers’. The purpose of this paper is to provide further explanation of this phenomenon based on predictable and unpredictable effects, which are the source of the observed no-load noise difference between 50 and 60 Hz transformers.

Sound power level is a unique feature of large power transformers used to estimate noise effects on the environment. It is determined from the direct measurement of the sound pressure level or sound intensity level. In the paper, procedures for reproducibility estimation of sound power level and procedures for estimation of uncertainty of sound pressure level and sound intensity level are given for point-to-point and walk-around test procedures.

The aim of this paper is to demonstrate how stressed oil volume (SOV) theory can be used in experimental research to obtain better reliability of transformer insulation design criteria. For this purpose, combined stress research is taken as an example. In combined stress research winding edge geometry is investigated, where at the same time winding edge is exposed to axial lightning impulse stress due to potential difference between discs/coils and radial LI stress due to potential difference between windings. Transformer designers usually consider this axial and radial stress as two independent manifestations, evaluating each with its own design criteria. Combining the radial and axial stress into a single design criterion should lead to a more reliable insulation design. This paper shows how SOV theory can be used to analyse the influence of radial electric field on axial oil gap LI withstand stress, which is an essential step in combined stress design criteria development.

This paper describes impact of synthetic ester as insulation liquid on power transformer electrical and thermal design, material compatibility issues and differences in manufacturing process. Also, higher fire safety feature and biodegradability of synthetic esters are highlighted.

The application of online monitoring systems has been intensified over the past two decades and they are now frequently installed with new transformers or retrofitted to the existing ones. A lack of standardization, different asset management strategies and practices between utilities, as well as certain economic factors resulted in systems made to the user’s specifications having different level of complexity: from basic units which monitor only a few parameters, to more comprehensive systems that are monitoring a wide range of parameters and are delivering various diagnostic information related to the transformer’s health. Nowadays data from monitoring systems is in most cases available only in the substation, where data is shown using a locally installed computer, or directly on monitoring devices installed on the transformer where the user can analyze the status of the transformer unit. This makes use of the data impractical and thus prevents the value of the systems to be utilized for a fleet wide asset management. In this paper, future trends in the transformer online monitoring will be presented, based on the previous development of such systems, and according to the concept of the condition-based maintenance of power transformers. As a key functionality, the status of the power transformer will be presented using either a numeric or graphical representation, which will represent the general status of the equipment (health index). This will give the user guidance on how to rank transformers in their network, in order to decide on the schedule and to pinpoint the transformer component which needs to be maintained. For this purpose, the system will need to consider other available information such as results from off-line diagnostic tests, transformer design data and history of known faults. Such approach will enable asset management strategy where the user can implement condition-based maintenance, which will lead to the cost reduction and improvement of reliability and safety with regard to power transformers.

This paper demonstrates how to find optimal threshold values for the cooling system of power transformers in respect to minimize the total loss consumption and maximise the life span without adding extra maintenance costs. To reach this ambition the transient thermal behaviour of a power transformer is studied for different ambient temperature scenarios together with the moisture behaviour over a complete life cycle, an essential parameter to determine the aging condition of the cellulosic insulation.

In this paper, an on-line transient overvoltage monitoring system (TOMS) for power transformers is used for measurement of overvoltages on the transformer bushing tap. The focus of the paper is on the analysis of transient overvoltages caused by lightning strikes recorded at the terminals of power transformer. Several recorded overvoltages are analyzed and their amplitudes and frequency spectrum are presented and compared with those referring to standard impulse voltages from IEC standard. Collected data include number, peak and duration of recorded transient overvoltages and can be used for the assessment of the transformer insulation condition and estimation of health index. Data recorded by TOMS are also of significant importance since the insulation system of power transformer and other equipment in the substation can be damaged by lightning or switching overvoltages.

Cigre working group A2.56 Power Transformer Efficiency conducted a survey among utilities worldwide to determine how maximum allowed losses are managed in their existing specifications of transformers—whether the losses are capitalized and how the load and no load, loss factors are calculated etc. Findings of that survey will be presented in the paper.

Power transformer efficiency measurements are a crucial part of the acceptance tests of a power transformer. These tests have increased importance since the EU Ecodesign Directive per 1 July 2015 has put efficiency limits to power transformers that are sold on the European market. In this paper we present a series of measures to assure that power transformer efficiency tests performed by power transformer manufacturers are accurate and reliable. This includes the development of more accurate industrial loss measure-ment systems (LMS), optimized LMS calibration approaches, and a reference system for on-site LMS calibration.

The long maintenance intervals of vacuum tap-changers require extensive testing. Since resistive vacuum tap-changers where introduced to the market in the 2000s there are very few of these that have been in service 10–15 years without any kind of maintenance and likely no experience at all from installations with longer service intervals. To really verify that a tap-changer can handle 30–40 years in difference load conditions and applications, testing beyond normal type testing is needed.

Distribution Transformer (DT) is an integral component of a distribution network. Electric utilities have invested efforts in reducing DTs failure rates. This paper presents a method for prediction of probability of DT failure by analyzing a correlation between weather data and historical DT failures data. Logistic regression prediction model is used in order to predict DT failure, and to extract the correlation between parameters of weather and DT failure. Accuracy of prediction is reliable, which is presented using evaluation metrics. This method not only has a vital significance for the maintenance of DTs, but also improves the economic efficiency and reliability of distribution network operation.

High voltage electrical devices such as power transformers are often stressed by the overvoltages that occur during switching operation and atmospheric discharges. Consequently, it is of particular interest to monitor these events in the power network. Nowadays, power transformers can be equipped with transient recorders that measure the voltages at the bushing measurement taps. It is possible to simulate events captured by these recorders, in EMTP. In this process, the crucial element that has to be modelled properly is the power transformer. In this paper, three wideband transformer models are presented: black-box, grey-box and white-box. These models are validated using the transmitted overvoltage measurements at high voltage laboratory. Then, the results of the simulation of the transient event recorded in 220/110 kV substation is presented and compared to the on site measurements results.

To study the transient interaction between the transformers and the power system is essential use very accurate transformer models considering the dependence of damping with frequency. In this work the fundamentals of transformer parameters calculation for high frequency transients using finite elements method (FEM) are reviewed and a promissory time-domain equivalent circuit is proposed based in the analysis of transient measurements results obtained by the CIGRE JWG A2/C4.52 in two transformers manufactured by WEG in Mexico.

Research tests were performed on a shell-form autotransformer with a more rigorous impulse test methodology than the traditional recurrent surge oscillograph method. These tests included non-standard terminal connections with open terminals, differing from those tests defined by the international standards related with impulse testing of power transformers. The test voltage responses were obtained using voltage transfer frequency sweep measurements that were converted into time domain waveforms. The measurement results were compared against simulations by a white-box model, demonstrating satisfactory accuracy of the transient calculation tool. In addition, the sensitivity of the measurements to measuring probes length was experimentally evaluated. Reducing the lengths with 1.5 m only affected frequency components around 1 MHz.

Relevant international instrument transformer standards specify internal arc testing to prove the transformer behavior under internal fault conditions. However, the test is defined in a way that does not recognize that it is possible to limit and reduce the total fault energy. For such instances, currently defined testing is mostly inapplicable. The purpose of this paper is to provide a theoretical background on behavior of units with sectioned active part under fault conditions, aided by numerical analysis of different fault scenarios on a variety of inductive voltage transformers, spanning from 72.5–550 kV. Numerical calculations are verified with several tests performed on a 123 kV inductive voltage transformer. This paper is a part of a continuous broad research with an aim to develop and specify adequate routine, type and special testing sequence for qualifying the paper-oil insulation systems that limit internal arc energy. Furthermore, the idea is to prove performance of such systems by introducing criteria that exceed the practices of current standards and aim to improve the instrument transformer fault statistics.

This paper introduces a dynamic thermo-hydraulic network model for transformer applications. The model includes the dynamic performance of temperatures, moisture levels, oil flows and the temperature dependent losses, separated in individual windings and individual positions. Therefore, interactions between single windings and also the core are covered by this model which is not the case in standard models. The temperature and moisture values are applied to calculate the insulation aging at different positions. In IEC 60076-7 (2018) the influence of moisture and oxygen can be considered roughly for aging calculations while the determination of the local moisture levels is not part of the standard. Here this paper gives assistance to determine such local moisture levels during the life time of a transformer. The thermo-hydraulic model includes also the dynamic properties of moisture transport in solid and liquid insulation to determine the local moisture levels and DP-values in the insulation. Simulations point out the difference between the aging model of IEC and the model developed by the transformer manufacturer. In addition, the moisture generation during cellulose aging and the change of moisture absorption capability by aging are parts of the simulation studies in the paper.

Although their primary purpose is not line discharging, inductive voltage and combined transformers are recognized as suitable solution for line discharge due to a typically very fast decay of trapped charge. Discharge capability of instrument transformers is not defined by relevant standards. Instead, it is often specifically defined by technical specifications of individual network utilities. In this paper authors have analysed discharge capability of inductive voltage and combined transformers. The focus of this paper will be placed on two main aspects of line discharge activity: thermal and mechanical capability. The aim of this document is to prove that discharge capability of inductive and combined transformers can be simulated with adequate precision. Furthermore, the proposed approach provides an indispensable tool for determining and optimizing relevant electrical and mechanical transformer parameters in order to achieve requested discharge capability even when technical specifications or project requirements exceed discharge capability of a standard instrument transformer design. Lastly, the paper will present relevant testing procedures, which verify discharge capability of inductive voltage and combined transformers.

In this paper the influence continuous transposition of the conductor has on winding RLC parameters and electric field distribution is analysed. Study is made on two types of windings, one winding with and one without conductor transposition. 3-D finite element method (FEM) based software is used for electric field computation and the computation of RLC matrices of the windings.

In generator step-up power transformer units (GSU), low voltage windings carry sufficiently high currents which generate high magnetic field levels. Therefore, when it comes to 3D finite element method (FEM) calculation of the stray losses of such transformer units, it is not enough to model low voltage windings just as two cylinders. Low voltage leads should also be modelled, and it is necessary to know how to accurately model and connect leads to the windings. Modelling windings and leads in 3D FEM is very demanding and calculation is time-consuming, so an appropriate model of leads can therefore reduce the mesh size and solver time. In this paper, multiple modelling approaches are tested and compared with the model without leads.

Due to the large number of thin laminations in the transformer core, homogenization of the core is necessary. Thus, new material of anisotropic characteristics is calculated. Anisotropic electrical conductivity tensor then allows large eddy current loops to be taken into account directly. In the case of low-frequency time-harmonic generator current, the influence of narrow eddy current loops on the distribution of the magnetic flux density within a domain can be taken into account by using additional manually calculated anisotropic magnetic reluctivity tensor. Therefore, no modification of the standard weak formulation of the magnetodynamics problem is required. The only code implementation is required in postprocessing to account for losses due to narrow eddy current loops. A real-size model is used as a numerical example.