ELECTRIMACS 2019

The real-time simulation is a valid help to test electrical systems when a physical device is not available. This is significantly evident when used in hardware and software co-simulation environment, where it is possible to connect the emulator to a real subsystem to test or validate it. In this paper, a model of the photovoltaic system is presented that can be implemented within a hardware simulator to be able to interface it with a real circuit, the hardware simulator used is the National Instruments RIO system.

This paper explains several industrial cases involving the HIL simulation of MW-range drives and inverters using CPU cores with FPGAs to compute model equations. The use of HIL simulators is common today in the industry to accelerate design cycles, mitigate financial and human risks and support software updates throughout the product life cycle.The first case presented is a 2-level inverter scheme in which increasing power specifications are met by adding parallel IGBT-modules. The second case is a multi-level motor drive with low harmonic injection on the AC-side. The third case is a modular multi-level converter in a grid application. We also discuss a new T-type inverter model that uses an industry PV-to-grid power converter.In each case, all power system modelling was done using Simulink and SimPowerSystems in conjunction with the SSN solver from the ARTEMiS blockset in addition to code generation for CPU execution at time steps in the 20–50 μm range, with an exception for MMC models on FPGA. In all cases the firing accuracy of the IGBTs remains in the nanosecond range using time-stamping techniques and an FPGA board. In the case of the parallel 2-level inverters, there is significant difficulty regarding the small firing delays (typically <500 ns) between modules that create circulating currents. These circulating currents are rendered correctly on the HIL bench.Also discussed in the paper are the various optimisations, solvers and methods that enable these performances.

The purpose of this paper is to propose and implement a novel rotor position detection method for permanent magnet synchronous machines (PMSMs) based on linear Halls, which are embedded inside of stator of PMSMs. A three-phase 9-slots/8-poles PMSM is exampled to verify the method. Firstly, a special point located in stator yoke (back-iron) is found by two-dimensional finite element analysis (2D-FEA), where the open-circuit flux-density due to permanent magnets versus rotor position (B PM) shows a high amplitude and good linearity, while the armature-reaction flux-density (B armature) due to armature currents exhibits a low amplitude and good linearity versus armature currents. Then, an analytical model is built and the analytical relationship between armature currents and the B armature is derived. Based on the analytical mode, B PM can be obtained by separating the B armature from the synthetic magnetic field (B Synthetic). Thereafter, the resultant B PM can be used to detect the rotor position information with differential-type piecewise-linear analytical method. The feasibility of the proposed detection method is verified by co-simulations and experiments. The simulation results show that the novel linear Hall-based angle sensor can achieve the accuracy equivalent to 3000-line. The experimental results indicate that compared with an encoder, the maximum error of electric angle position at different speeds is less than 0.3%.

This paper proposes a discrete-time control strategy of grid-side currents of a three-phase grid-connected converter with LCL filter. The implemented control system, a RMRAC (Robust Model Reference Adaptive Control) by state feedback, presents robustness to parametric uncertainties and rejection of periodic disturbances. To demonstrate the controller performance, numerical simulation results, considering parameters of a real plant, are presented.

This paper presents an electrothermal characterization of a prototype double-sided cooling power module. The junction temperature T j is an important parameter of power devices. Different methods exist for junction temperature measurement. In this work, an electrical method based on temperature sensitive electrical parameter (TSEP) is conducted to estimate the junction temperature of the power module. A 3D thermal model was built to better comprehend thermal behavior within the module. A comparison between simulation and measurement results is performed and analyzed. Results have shown that 3D numerical modeling help understanding several manufacturing defects (soldering, sintering, die defaults, etc.).

This paper is situated in the framework of future hybrid electric aircraft in which embedded weight minimization and maximization of power efficiency are the key challenges to address fuel reduction and environmental constraints. In the first part, the integrated design process aiming the overall power train optimization is described. The second part presents models specifically oriented towards the integrated design. Finally, a sensitivity analysis is carried out at the power train system level to study the influence of both electric components-specific powers and efficiencies on the Maximum Take Off Weight (MTOW) and on the fuel burn of the hybrid propulsion aircraft.

A doubly fed induction generator (DFIG) driven by a variable speed wind turbine (WT) is presented in this paper. The DFIG is partially interfaced with its rotor via a back-to-back converter. This one is supplied via a three-level inverter in the rotor side and controlled with a flexible algorithm based on DTC technique, to ensure mastery of this generator. The main aim of this contribution is to analyze the performances and robustness of the proposed control technique. The aim of this structure is to obtain at the generator output AC sine waveforms signals with a constant frequency and a low THD, as well as minimum output voltages ripples, regardless of the variation of the wind speed. Indeed, the main objective of this work is the performance analysis of the DTC applied to a three-level inverter in the rotor side of the DFIG considering some constraints that reflect the real operation of the wind turbine generator, such as the randomness behavior of the wind speed, allowing all operation modes of this generator. These operation modes are carried out in a successive and continuous manner, while showing synchronous and overspeed modes. Simulation results, performed under Matlab/Simulink, are presented and analyzed.

This paper deals with the implementation of a sensorless control strategy devoted to Switched Reluctance Motor Drives used in the traction drive of a small truck. The sensorless technique operates at low and zero speed. In the proposed approach an additional high frequency magnetic field is injected into the machine and a suitable demodulation algorithm is exploited to extract useful information on the rotor position and speed. The feasibility of the implementation is verified by simulations.

Recently, offshore wind farms have attracted more and more attention because of their greater energy capacity. To get the best performances of a wind farm park, a technical and economic compromise between energy yields and overall investment must be established. In this paper, a study was done on Borssele I & II offshore wind farm with HVAC and HVDC transmission technologies to compare their performances with different transmission distances.

This paper deals with the analysis and simulation of a current sensing technique based on the estimation of the current flowing in the low-side MOSFETs of an inverter. In this case the power MOSFET is utilized as a current sensor, by estimating its internal on-state resistance. Due to the temperature dependency of the internal on-state resistance, the temperature of MOSFET die has been also estimated by measuring the internal body diode forward voltage of the power MOSFET. The proposed low-side MOSFET current sensing method will be used as current sensing in three-phase inverters for automotive applications.

The paper presents a procedure to achieve an input-state feedback linearization on a bidirectional Boost DC/DC converter connected to a passive load. The system linearization is achieved by a proper state-space/output transformation performed on a non-dimensional form of the analytical model. The resulting system is then controlled through a standard linear regulator. An online load estimation technique is also provided to overcome the transformations parameter dependency. The proposed approach has been numerically tested and compared with a standard two-loop controller.

In this paper, we present a two-dimensional (2-D) analytical model of conventional switched reluctance machines (SRMs). This model has been applied to an 8/6 conventional SRM supplied by conventional excitation (viz., standard asymmetric H-bridge). The goal is to determine the electromagnetic performances. The proposed analytical model is based on solving the partial differential equations (PDEs) due to Maxwell’s equations in each domain of the studied machine (viz., air-gap, rotor and stator slots). A parametric study by using the developed analytical model has been compared with that obtained by numerical computations in linear and non-linear conditions. The results showed that the analytical and numerical results are in good agreements in linear conditions. However, in non-linear conditions, the developed model overestimates the performances. Indeed, to predesign the machine, this model can be incorporated in optimization environments where savings in computation time are needed.

This contribution presents a study of the effect of synchronous motors losses parameters variations on the machines behaviour. In particular, the effects on efficiency maps are investigated. The goal is to identify parameters sets allowing improving energy efficiency for a given application.

This paper proposes an in-depth analysis from the control point of view of dynamic models of a modular multilevel converter (MMC) for high-voltage direct current (HV-DC) application. Firstly, a generic method of analysis is presented for a natural arm-level state-space model. Its structural analysis highlights the decoupled nature of the MMC. Secondly, the well-known sum and difference of the upper and lower arm state and control variables is considered to obtain a (Σ∕Δ) model. This transformation leads to a coupling between state and control variables and to an increase of the system complexity. Using the analysis results of the natural model and the (Σ∕Δ) model, an original arm-modular control is finally proposed. The simulation results show the effectiveness of the proposed control, which is simpler to design compared to a conventional (Σ∕Δ) control.

With the proliferation of renewable energy sources and the adoption of several policies to reduce environmental risks caused by traditional polluting sources, the concept of microgrids, especially DC microgrids, is currently gaining interest. In fact, most renewable energy sources (RESs) and loads are inherently DC type. Moreover, DC microgrids offer many merits over AC ones in terms of ease of control and efficiency. While most of researches address the control hierarchy and strategy in DC microgrids, this paper focuses on the modeling and simulation aspect. A typical configuration of an islanded DC microgrid is modeled in MATLAB/Simulink, and a primary-level control strategy is adopted where two approaches of converters modeling are tested: instantaneous and average model. The two approaches of modeling are compared in terms of precision of losses modeling, dynamic response of the system, simulation time, and computational burden. Simulation tests are conducted, and the results show that, despite its accuracy, the instantaneous model can be applied only for short-term simulations due to many limitations, whereas average converter modeling presents a better solution for long-time simulations, since it ensures a tradeoff between model accuracy and simulation time, which makes the application of the three levels of hierarchical control in DC Microgrids valid in one simulation model.

The fast field cycling (FFC) experimental technique allows to overcome a technical difficulty associated with the nuclear magnetic resonance (NMR) signal-to-noise ratio (SNR) at low frequency spin-lattice relaxation measurements when using conventional NMR spectrometers. Constituting a step forward than the classical analog approaches, in this paper, a digital control system for an FFC-NMR relaxometer power supply was developed. The hardware and software were designed to allow for the modulation of the Zeeman field as required by this technique. Experimental results show that under digital control the system performs fast transitions between the high and low magnetic flux density levels, i.e., the switching times obtained are in the millisecond range, and, assures a good stability of the field during the steady states. Comparative proton relaxometry measurements in two compounds (liquid crystal 5CB and ionic liquid [BMIM]BF4) were made to assess the digital control system performance.

At least two different actuators work in cooperation in regenerative braking for electric and hybrid vehicles. Torque blending is an important area, which is responsible for better manoeuvrability, reduced braking distance, improved riding comfort, etc. In this paper, a control method for electric vehicle blended antilock braking system based on fuzzy logic is promoted. The principle prioritizes usage of electric motor actuators to maximize recuperation energy during deceleration process. Moreover, for supreme efficiency it considers the battery’s state of charge for switching between electric motor and conventional electrohydraulic brakes. To demonstrate the functionality of the controller under changing dynamic conditions, a hardware-in-the-loop simulation with real electrohydraulic brakes test bed is utilized. In particular, the experiment is designed to exceed the state-of-charge threshold during braking operation, what leads to immediate switch between regenerative and friction brake modes.

A new concept of AC motor drive is proposed in this paper. It allows to boost voltage of an AC machine without supplementary components. The main idea is to wisely connect the neutral point of the AC machine to the DC power supply. With some modifications on the control algorithm, the proposed solution allows to supply the AC machine with a higher voltage than with classical inverters. The concept is general for different AC machines and different topologies of inverters. The case study of an induction motor driven by a three-phase two-level inverter is illustrated. In steady state, a factor gain up of 1.7 of maximum RMS voltage can be obtained with the proposed solution compared to a classical scheme. Experimental validation on an induction motor test bench shows the effectiveness of the proposed concept.

The present paper focuses on the study of the generated power smoothing performances of hybrid renewable energy sources dedicated to the application of a micro-grid. The latter is constituted of a variable speed wind turbine (VSWT) based on a DFIG, a PV system and a hybrid energy storage system. The DFIG rotor side is powered by a three-level inverter and controlled by a direct reactive power control (DRPC) technique. This topology is chosen to improve the quality of the output power and current injected into the grid. The hybrid storage system is made of a combination of battery banks (BBs) and supercapacitors (SCs). The main purpose of the work is to manage the energy used by the system and to control the active and reactive powers flowing through the micro-grid. Also, the use of the DFIG with its converter as a local reactive power compensator is considered. In addition, to mitigate fluctuations due to random changes in wind speed and solar radiation, an energy management algorithm is introduced. The latter must also ensure the smoothing of the output power, the control of the discharge depth of the batteries, and the continuity of service. Simulation results, performed under Matlab/Simulink, are presented and analyzed.

The reliability and service continuity in the distribution grid are the basic requirements, but are threatened by various grid faults. Compared to the conventional power transformer, a smart transformer (ST) can flexibly switch between current control mode and voltage control mode, and thus has high capability on power flow control and to mesh the electric grid in secondary substations. Consequently, the ST can significantly improve the reliability and service continuity in secondary networks under grid faults. This paper presents the possible solutions to manage faults in MV and LV distribution grid, respectively, and the corresponding control strategies to improve system performances.

In the present work, the authors present a comparative study of an electrical power distribution system. The work aims to study efficient and cost-effective ways to manage power in high-energy-demanding plants. By doing this, power losses can be reduced with economic and environmental benefits. As a case study, a galvanization plant is taking under consideration. Such plants indeed need a high quantity of electrical energy for the galvanizing process; hence, also a slight increase of the overall efficiency can lead to consistent economic benefits. The proposed plant in addition has a PV solar roof which is meant to integrate the grid for the process supply. In the work, the present situation is analyzed and a different solution is hence proposed. The plant indeed is now supplied with a medium voltage (MV) connection and transformed into three-phase 400 V AC low voltage (LV). Galvanization pools and auxiliary utilities are supplied with this voltage. Concerning the PV solar roof, the strips output 800 V DC. In order to exploit the PV produced energy, a DC-AC inverter connects the PV plant to the AC 400 V line. The presented structure involves several power losses which can be reduced by exploiting a more efficient power management strategy.

This paper presents a strategy to control the active and reactive power in the point of common connection (PCC) of a wind generation system operating in islanded mode. A full back-to-back voltage source converter (VSC) is connected between each wind generator and the PCC. The control scheme considers voltage and frequency regulation for each VSC. The voltage and frequency references are obtained from P-V and Q-f droop characteristics of the generators, where Q and P are the reactive and active power supplied by each VSC to the PCC. Proportional-integral (PI) controllers process the voltage and frequency errors and set reference currents (in d-q frame) to be imposed by the converter. The strategy has been validated by mean of simulations, and results are presented showing the performance of the control strategy proposed.

In the automotive domain, the electrical and electronic items are playing day after day a more central role. Since most of these units are in charge of safety-relevant functionalities, a strict development process is required. The ISO26262 automotive functional safety standard describes a mandatory process to design, validate and verify item designs. The aim of this work is to describe a suitable way to overcome some safety life cycle issues. The description starts from the concept phase, with the Hazard Analysis and Risk Assessment, in where the safety goals are defined, and an Automotive Safety Integrated Level is assigned to each of them. After that preliminary phase, it will be shown how it is possible to check the reliability of the obtained hardware design keeping into account the failure detection and mitigation capabilities of both hardware and software. To achieve this goal, a simulation-based Failure Mode and Effect Analysis assessment technique is applied to assess the hardware design’s possible sources of failures and to analyse detection, isolation and mitigation capabilities. To achieve this result, the hardware model and the embedded software have been implemented using the Model-Based Software Design approach. This approach has been demonstrated on an electrical vehicle powertrain design.

A fault localization method for single-phase to ground short-circuit (SC) faults in low voltage (LV) smart distribution grids is presented in this paper. Both the use of rms voltage phase measurements and an analysis of symmetrical components of the voltage were investigated and compared in this study. Phase measurements were found to be more suitable for single-phase to ground faults. The described method is a three-step process beginning with the identification of the faulty branch, followed by the localization of the sector in which the fault occurred and concluding with the estimation of the fault distance from the beginning of the feeder. Fault resistance values of 0.1, 1, 5, 10, 50, 100, 500 and 1000 Ω were tested. An heterogeneity analysis was performed to test the effect of the use of various conductors on the method. Faults in all three phases were implemented and simulated on a real case of a semi-rural LV distribution network of Portugal, provided by Efacec. Finally, the method presented an average estimation accuracy of 89.33% and an increased accuracy of 93.11% for low impedance faults (up to 10 Ω of fault resistance).

This paper presents an example of collaboration between two different air quality monitoring systems, one developed for indoor usage, the other one used in some regions of Italy as an example of citizens’ collaborative work for monitoring the air quality in smart cities. The exchange of information between the two systems (the inner one and the external one) allows making a weighted decision for improving the inner air quality. By evaluating both indoor and outdoor air quality levels, a reasoner decides the best policy to be automatically adopted to improve, or at least not worsen, the indoor air quality.

In this paper we developed a prototype device for smart monitoring and fault detection of a stand-alone photovoltaic system (SAPVS), using an Internet of Things approach. An electronic sensing board has been designed and a web-based application has been developed in order to monitor the data (current and voltages delivered by the SAPVS, as well as air temperature and solar irradiance) in real time. The prototype has been tested experimentally at the Renewable Energy Laboratory of Jijel University, Algeria. The experimental results show the capability of the prototype to monitor data, detecting and signalling out possible faults based on the output PV power, and inform users via website about the state of the system. The faults that have been investigated are: open circuit, shading effect and dust accumulation on PV modules. The prototype is cost-effective and very easy to be implemented without any additional circuits and efforts.

This study presents a new fault detection approach for Permanent Magnet Synchronous Motors (PMSMs), based on the Park Current Vector (PCV), which addresses the total loss position/speed information fault in electric motor drives. Two PMSMs were considered: a 53 kW PMSM was adopted for simulations, whereas experimental validation was performed using a 1 kW PMSM. Effective position/speed sensor fault detection was achieved using the proposed PCV-based approach, in which validity and feasibility are proved by simulation and experiment results.

electrochemical hydrogen pump (EHP) is a promising technology capable of extracting hydrogen from miscellaneous gas mixture and compresses it to very high pressures. The basic working principles are similar to that of a proton exchange membrane (PEM) fuel cell. Consequently, its performance is heavily dependent on humidity level of the membrane. Unlike PEM fuel cells where water is generated as a by-product, in the case of EHP the humidity has to be delivered via external humidifier. Therefore, it is paramount to have accurate information regarding the humidity in order to achieve optimal exploitation. Inaccessibility of the membranes makes it almost impossible to perform direct humidity measurements. Addressing this issue, this paper presents a method for online estimation of humidity levels based on the parameters of an equivalent circuit model (ECM). The parameter estimation is performed through a combination of evolutionary algorithm and simplex optimization. The method is evaluated on a market ready EHP device with capacity of pumping 1.4118 stl/min of H2.

Lithium-ion battery ageing modelling and prediction is one of the most relevant topics in the energy storage research field. The development and assessment of reliable solutions are not straightforward, because of the necessity to acquire information on the cell ageing processes by employing very time-consuming tests. During these tests the cells are subjected to different profiles, usually based on the repetition of several charge/discharge cycles, in order to reproduce the ageing effects in laboratory. This paper aims at accelerating the advancement in this research field by discussing a dataset containing three different ageing tests and making it available to be used by other research groups. The tests are accurately described and a preliminary analysis of the obtained results is carried out.

This work focuses towards online control strategy for detecting fuel and oxidant starvation and predicting an optimal stoichiometry for operation under different fuel compositions using the electrochemical impedance spectroscopy (EIS) parameter extraction method. The tests involve three fuel compositions, namely dry hydrogen, dry reformate ( , , and ) and wet reformate ( , , and ). The characterization of anode and cathode stoichiometry (both low and high) is carried out with each fuel composition by measuring electrochemical impedance spectroscopy (EIS) and current–voltage (IV ) curves. The results suggest positive effects of humidified gas on the fuel cell stack performance. The changes in the mass transport resistance due to excess gas or gas starvation both on the anode and cathode could only be deduced using the EIS method. Online EIS measurement seems useful in deducing the optimal stoichiometry as the IV curves are unable to show the changes in the mass transport. Thus, to operate the fuel cell stack under an optimal fuel and oxidant utilization, an online EIS with parameter extraction algorithm can be helpful. This would ensure a better fuel and oxidant utilization and improve the system efficiency.

In this paper, a procedure is proposed to optimally design a boost converter (including its input filter and its control) for embedded applications, where power density and high efficiency are crucial criteria. The load consists of a constant power load (CPL), which represents the most stringent case in term of stability issues. This optimization problem can be treated as a multi-objective one, which aims to maximize compactness and efficiency of the studied system. In order to find a trade-off between compactness and efficiency, the two objective functions (volume and power losses), a genetic algorithm was implemented to generate the most convenient design solutions (Pareto front). The operating conditions include both steady and fast output power load transients (step-up and step-down). Since CPL transients may lead to large deviations of the output voltage from its nominal value, the control strategy is discussed as well and included in the design procedure. The presence of a differential input filter is likely to interact with the constant power load and cause some instability issues. A constraint function excludes solutions that do not respect the feasibility, the magnetic saturation or the stability constraints.

Energy harvesting from ambient sources is an interesting opportunity for wireless and self-powered electronics, increasing research efforts toward the development of new devices. Among all energy sources, vibrations seem particularly convenient for this kind of application. Piezoelectric resonant systems, though offering configurations well suited to recover energy from vibrations, suffer from narrow operational bands, and for this reason new solution to enhance performances at off-design excitation conditions is sought. In this paper a piezoelectric resonant energy harvester is developed, focusing the attention on both ceramics production method and support material choice in order to maximize the oscillation amplitudes, and consequently the energy output. The device produced at ISTEC laboratories is then compared with a commercial product under harmonic excitations. Results relative to power output show that the in-house assembled device has better performance than the commercial one in the considered tested conditions, both in absolute terms and with respect to the active piezoelectric volume of the two devices.

This paper discusses an alternative realization approach of the reconfigurable full-bridge/voltage doubler rectifier for implementation in hybrid DC–DC converters, i.e., the power electronic converters that can adaptively change the power circuit topology for optimization of performance or enabling the post-fault operation. The proposed rectifier is characterized by the reduced number of semiconductor components and the ability of independent use of its capacitors for forming the series resonant tank and output voltage filtering. The paper explains the derivation and operation principle of the semi-active reconfigurable rectifier and discusses the experimental results obtained by the help of a 350 W laboratory prototype. Finally, the fully controlled version of the proposed rectifier is presented, which features superior control and reconfiguration possibilities in both directions of the power flow and, therefore, could be used as a versatile power electronic building block for hybrid DC–DC converters.

Over the last decade, research on technologies to exploit tidal current kinetic energy for renewable electricity generation has had a significant growth. However, as to date, there is not a consensus worldwide on standard Power Take-Off (PTO) systems, due to the current immaturity of tidal energy converter technologies. In most cases, mechanical/electrical power conversion follows well-proven technologies derived by the mature wind-energy sector. However, the peculiarities of tidal energy resource impose ad hoc technology solutions. In this paper, different generator topologies and recent developments for marine tidal energy systems are reviewed and compared. The aim is to provide an overall perspective and identify areas for further development. Among considered technologies, the direct-drive permanent magnet synchronous generator by the full-rated frequency converter (FFC) represents an appealing solution, for reduced system complexity and maintenance requirements and possibility to develop smart Maximum Power Point Tracking (MPPT) strategies.

The energy yield of a photovoltaic plant depends on the performance of each of its single photovoltaic modules. In this paper, a methodology is presented that allows calculating the losses produced in a module and identifying if it is operating properly according to given climatic conditions. The proposed model is based on the comparison between the theoretical performance ratio of a module and the resulting experimental one under given climatic conditions. The results obtained allow estimating the losses that affect a module nonoperating properly. By means of real-time monitoring at module level, the herein described experimental system together with the proposed methodology allows to quantify the module losses. The immediate corrective actions will avoid further losses in the generated energy.

One of the major problems in photovoltaic (PV) arrays operation is the mismatching effect due to partial shading; it produces significant drops in the power delivered by the PV system. The power losses due to partial shading for a given operating condition can be mitigated in part by changing the scheme connection of the PV array. This paper introduces a solution to reconfigure a 3×3 PV array by means of a graphic interface using Matlab and Simulink. A mathematical model is used to calculate the configuration which provides the maximum power under a given operating condition. Then, a connection matrix is identified and the connections on the PV array are implemented by using an ARDUINO DUE and a relay based reconfiguration board. The proposed solution is validated by comparing the power vs voltage (P − V ) characteristics of different configuration schemes obtained from experimental tests.

Photovoltaic (PV) systems are prone to deep, steep and frequent irradiance fluctuations, mainly originated from overpassing cloud shadows, which cause fluctuations in PV power production. These irradiance transitions have been modelled by using a mathematical function to study their behaviour in a systematic way. Although the used methods and obtained results seem to be reliable, the simulation model has not been verified in detail. In this paper, the accuracy of the used theoretical model for irradiance transitions has been verified experimentally. The results show that the simulation model is accurate enough to study the irradiance transitions caused by moving clouds and their effects on the operation of PV systems.

The optimization of power dispatching has been proved to be useful for reducing the operation energy cost of a microgrid based on photovoltaic source. However, the formulation of the optimization problem needs the weather forecast to predict photovoltaic generation. The current hourly forecast is always available and often lacks accuracy. Thus, this work proposes the optimization based on a clear sky model to predict the solar irradiance. This model has the advantage of simplicity, since it depends only on the geographical coordinates. The analyses have been done to compare the weather data during 5 months, and the validation of the proposed model is carried out by simulation. The results show the optimization results of the proposed model are slightly better than a common hourly forecast weather provided by a meteorological website.

Photovoltaic energy harvest in distributed maximum power point tracking systems has demonstrated to be superior to the traditional photovoltaic systems under mismatch conditions. The distributed architecture usually consists of series-connected DC/DC converters forming a string, dedicated to process the power of individual photovoltaic panels. However, the classical approach assumes an independent control of the DC/DC converters preventing them from knowing the operating condition of the other converters in the string. The adoption of centralized algorithms allows full control of the variables in distributed maximum power point tracking systems and hence further increases the energy harvest. This paper proposes a novel centralized control that matches distributed and central maximum power point tracking functions, as well as an innovative functionality that improves the dynamic performance in photovoltaic applications.

The aim of this paper is to provide a dynamic reconfiguration method for partially shaded photovoltaic arrays. The implemented strategy is able to increase power production of the array with respect to the initial topology in real time and with any shading pattern. The array is supposed to be made of strings of modules interconnected in parallel and each module is constituted by series-connected photovoltaic cells. Irradiance values are calculated through a closed-form relation given the operating point of the modules, their temperatures, and their equivalent circuit model. This procedure frees the system from the necessity of costly pyranometers. The implemented method has been validated in Matlab environment simulating random shading conditions and implemented on a low-cost 32-bit microcontroller with wireless connectivity capabilities. The results prove the efficiency of the proposed solution.

This paper presents a new maximum power point tracking (MPPT) algorithm based on the adaptive linear neuron concept. This algorithm is designed to extract the maximum power in single-stage grid-connected photovoltaic (PV) system configuration. In the considered system, a PV panel is directly connected to a grid through a three-phase pulse-width modulation inverter. The control is achieved in the synchronous dq frame, and the proposed MPPT estimates directly the optimal d-axis duty cycle component. Furthermore, in order to achieve a unity power factor operation, the q-axis reference current is set to zero. In this work, only one proportional-integral controller is used to maintain the reactive power to zero value. To verify the effectiveness of the proposed algorithm, the grid-connected system is implemented and simulated under MATLAB-Simulink software. The obtained results are compared to those achieved by the conventional perturb and observe based MPPT technique under fast and slow irradiance changes. The simulation results show that the proposed method leads to achieve incomparable performances such as unity efficiency and zero oscillations in the PV panel in both transient and steady-state operations.

This paper is aimed to assess the performance of distributed converters in large PV plants through the analysis of a case study represented by a 2 MW PV plant in Central Italy. The electrical layout of a 500 kW subfield has been modified performing the installation of DC/DC converters at string level in order to create an independent MPPT control for every string. This kind of performance analysis is usually carried out using data acquired by the plant data logger. Unfortunately, the presence of partial unavailability, monitoring system faults, shutdown for maintenance activities, etc. can create several issues in data processing. To support data elaboration, a novel behavioral modeling approach has been developed and exploited in this work. This novel approach, based on an integrated state-space average model, can improve the performance analysis ensuring a satisfactory accuracy but keeping a low computation effort. Validation is performed considering real operating scenarios in case study.

The performance of photovoltaic installation is highly affected by faults in single modules. Faults in photovoltaic arrays are difficult to detect, locate and diagnose due to the way in which modules are configured. Given that photovoltaic arrays are formed by modules in series, a fault in a single module affects the whole system. Therefore, the technology to detect and diagnose faults inside solar arrays is emerging, the present paper proposes several expressions that help to detect and diagnose failures using a proposed algorithm. The expressions were obtained by an inductive approach based on the analysis of simulation cases where different faults were tested. The array model used for the simulation was built in Spice software based on the five-parameter model of a solar module.

Recently, the Lambert W function has emerged as a valuable mathematical tool in photovoltaic (PV) modeling and other scientific fields. This increasing interest is because it can be used to reformulate the implicit equations of the single-diode PV model into explicit form. However, the computation of the Lambert W function itself is still not clear in the literature; some studies use the iterative built-in functions in MATLAB or other computational platforms, while others adopt their own approximation formulae. This paper takes a deeper look at the ways the Lambert W function is evaluated in PV models and carries out a comparative study to assess the most commonly used methods in terms of accuracy, computational cost, and application range. These alternatives are implemented in a modern computer and a typical microcontroller to evaluate their performance in both simulations and embedded applications. The analysis concludes that some series expansions are good options for PV modeling applications, requiring less execution time than the built-in MATLAB lambertw function and exhibiting negligible approximation error.

The lithium-ion capacitor is a recent energy storage component. Although it has been commercialized for several years, its hybridization still requires further investigation to characterize it. The literature has studied some of its characteristics focusing on experimentation at positive temperatures. This paper aims to enlarge the tests to include very low temperatures, showing the difference between Nyquist plots at 65 and −30 ˚C. It also presents the Ragone plot for several temperatures, with a comparison between three storage systems: a battery, a supercapacitor, and the lithium-ion capacitor. Finally, a model of the LIC is proposed, for low and high temperatures, with experimental validation.

This paper deals with the comparison analysis of the proposed six-phase interleaved boost converter (IBC) for fuel cell electric vehicle (FCEV) application. Silicon carbide (SiC) semiconductor and inverse coupled inductors have been used to improve converter’s performance. According to the comparison analysis, the proposed converter’s input current ripple has been reduced, while fuel cell stack’s life span can be extended. Then, the total volume and weight of magnetic component has been decreased due to the inverse coupled inductor. Furthermore, benefiting from the SiC semiconductor, high switching frequency has been selected and low switching loss has been obtained. The power loss of the proposed converter has been reduced, while the reliability, efficiency, thermal performance, and power density have been increased.

A battery system with a thermally optimized module design with regard to boundary conditions in automotive applications is developed. Measures for spatial and temporal temperature homogenization are realized. Highly thermal conductive pyrolytic graphite sheets as heat spreaders replace conventional metallic cooling sheets in a lightweight module design. Efficient space utilization with a novel phase change material for thermal peak-shaving enables benefits in thermal management and lifetime. Heat-conductive adhesives and elastomer-based gap filler sheets further reduce the thermal resistance and the rise in temperature. Measurements showed a maximum temperature difference between the cells of 4.3 K and a maximum thermal resistance between cells and coolant of 0.12 K/W. By integrating thermal solutions, the gravimetric and volumetric overhead was reduced by 25% and 10% compared to the state of the art.

Resilient microgrids have been the subject of growing interest from information and communications technology (ICT) service providers to assure service availability (and therefore revenue) even in the case of grid fault due to bad weather conditions, as an internal storage capacity as uninterruptible power supply is used. However, such storage equipment represents an unavoidable cost in terms of initial investment, maintenance, and operational efficiency. In this work, starting from a previous development of a prototype supply system for a landline station, the control algorithm of the storage devices was investigated to optimize the cost/benefit ratio. A fuzzy logic system controller was developed to exploit the revenue opportunities offered by the energy market, converting a landline station into an active system that exchanges power through the grid. Besides this, a fuel cell generator was integrated to achieve further benefits (system resiliency and battery size reduction). The simulation results indicated a well-reactive behavior for energy price, battery state of charge, and grid fault probability variations.

One of the main concerns regarding energy storage systems during their normal operation is the possibility to perform an accurate state-of-charge estimation. This cannot be done by simple ampere-hour counting, unless drift correction means are put in place to avoid accumulation of measurement errors over time. In this paper, a state-of-charge estimation algorithm is widely analysed and tested on a nickel manganese cobalt oxide (NMC) lithium cell. The procedure consists of the utilisation of an equivalent electrical network battery model and the implementation of a Luenberger technique for a runtime correction, from the measure of battery’s voltage and current. Although application of Luenberger-style estimation is not new in literature for application to batteries, new expressions of battery model parameters and more detailed simulations are shown, to imply much higher estimation accuracy than in the past. After setting the model parameters, different test cycles have been considered, to evaluate the robustness of the proposed technique.

Microgrids represent a promising energetic scenario applicable in different contexts, especially in residential clusters. In this paper, authors propose a novel control logic to implement a coordinated management of generators, loads, and hybrid energy storage systems (HESS) in a microgrid by means of a hierarchical smart converter architecture. The innovative algorithm is embedded in a master converter. It allows the online management of energetic fluxes in cooperation with slave converters distributed among the microgrid resources. They carry out a smart coordination of microgrid generation, absorption, and battery-supercapacitor storage systems with the aim to improve the availability of the storage systems for providing ancillary services to the power grid. The effectiveness of the control is tested applying the smart converter master-slave architecture, including the combined management BESS-supercapacitor algorithm, to a grid-connected residential microgrid.

The main scope of the study is the characterization of the capacitive and resistive behavior of two supercapacitor cells and one hybrid supercapacitor available on the market, through potentiostatic electrochemical impedance spectroscopy (PEIS). The PEIS tests were performed by applying to all cells the same voltage perturbation in the same frequency range. In a first phase, the instrumentation used for the acquisitions was optimized, with particular care to the connections between the potentiostat and the supercapacitor cell. The Nyquist diagrams obtained for each sample are compared and capacitance/frequency graphs are deduced. The technological differences between various devices are then discussed in relation to the results. The characterization of the sample cells and the collected data are used to propose the corresponding models conceived for circuit simulation. These models are based on simple electronic components available in the standard circuit simulation software tools.

This study concerns a grid-tied photovoltaic (PV) generator related to a hybrid storage system composed of lithium NCA battery (energy source) and Maxwell supercapacitor (power source). Two supervision algorithms have been proposed for energy management system (EMS): a Boolean and a fuzzy logic EMS. Moreover, a comparative study between both supervision algorithms based on levelized cost of energy (LCOE) and the lifespan of the storage system has been suggested. The economic analysis is done with two different planned PV power production profiles: one with a “clear sky” bell curve and a second with an ideal forecast. The supervisor based on Boolean method is simple and easy to understand, while the fuzzy logic method offers more flexibility in supervision. It improves the battery lifespan and system performance a little and reduces significantly the system penalties. The simulation results show for all scenarios the achievement of the planned aims in terms of respect of the production program taking into account the constraints of the electrical network manager with an LCOE below 130 €/MWh.

In this paper we want to show some of the most recent results obtained in our laboratory concerning the fabrication of lithium sulfur batteries. For their construction we used two different binders and two carbons with different surface areas, deposited directly on the separator. Sulfur was introduced mixed with the electrolyte in the form of polysulfide. The particular cell configuration has allowed to obtain stable specific capacities after numerous charge and discharge cycles of more than 800 and 1200 mA h/g and low cell resistances.

The paper deals with the design and verification of a monitoring system for the analysis of the useful life of a water filter. The filter is made of pressed graphene nanoplatelets, obtained from commercial graphite with a low-cost fabrication procedure. The state of the filter is monitored by measuring the electrical impedance at the port of a suitable circuit embedding the graphene filter. It has demonstrated a good sensitivity of the impedance with respect to the saturation level of the pollutants into the filter. The technique is also shown to provide a high level of reproducibility and stability with environmental conditions.

The risk reduction, in case of catastrophic events, is strongly conditioned by the possibility of performing mechanical actuation in extreme conditions and without electric energy. Many devices, which implement security features and fulfill these specifications, are on the market, however the capability to transfer data and information through reliable technologies is also required in order to monitor and coordinate the safety systems. A possible candidate that could provide a communication link, even in the absence of electricity, is LoRa technology, even if it has some problems and limitations mainly due to the topological configuration of the LoRaWAN communication network and the increasing number of devices and users. This paper reports a proposal to improve the reliability of LoRa data transfer in very severe environmental conditions and on large areas. Some accurate simulations conducted on a LoRa network, make it possible to verify the effectiveness of the proposed communication strategy.

Li-ion batteries had a great development in recent years, and their use has grown massively because of their higher energy and power density with respect to traditional ones.However, their high energy density implies great danger in the event of malfunctions or failures, due to the emission of toxic and highly flammable substances. In the worst case, thermal runaway can occur. It is a chain reaction where unwanted reactions take place that leads to an uncontrolled and unstoppable increase in temperature. It can cause uncontrolled combustion and then explosion with great danger.In order to identify the conditions that lead to the thermal runaway and to limit its occurrence, thermal stability of Li-ion batteries is here investigated. Thermal abuse tests are performed on lithium nickel manganese cobalt oxide cells from Panasonic in an ISO 5660 cone calorimeter.Heat release rate is measured by changing the state of charge (SoC) of the cells and the radiant power of the cone calorimeter. The relationship between the SoC and the onset of the heat release is clearly revealed.