Andrzej Wetula

Photovoltaic (PV) inverters are key elements in modern distributed generation systems. One of the critical requirements for their reliable operation is islanding detection, as defined by IEEE 1547.1-2020 and PN-EN 62116 standards. This study investigates the performance of 28 commercially available, low-voltage, prosumer-dedicated PV inverters under unintentional islanding conditions, analyzing detection time, voltage amplitude, frequency deviations, and rate of change of frequency (ROCOF). Experimental results confirm that while most units comply with the 2 s disconnection limit, significant variability in reaction times was observed, with some units exceeding standard thresholds. Moreover, deviations in voltage and frequency indicate that in many cases, power quality standards were not maintained, potentially leading to safety and stability concerns. In the second stage of tests, a parallel operation of inverters was examined, revealing additional disturbances, such as voltage swells and beat phenomena caused by phase-locked loop (PLL) desynchronization. A simplified theoretical model confirmed the possibility of resonance and oscillatory interactions between parallel inverters during islanding. These findings highlight the need for improved coordination and control strategies for distributed generation based on inverters, particularly in microgrid applications.

The paper presents a method for selecting the locations of field sensors under a medium- or high-voltage line or substation busbars, in a contactless voltage measurement system. The proposed method uses the condition number of a distance matrix, correlated with the capacitance matrix of a system, as an optimization criterion. As a robust optimization algorithm was expected to be necessary for this task, genetic algorithm and particle swarm optimization algorithm have been tested, both in regular and hybrid versions. The proposed method was tested in simulations, using four power line geometries based on real-life pylons. Optimization results were juxtaposed with reference values coming from a sensor placement that would most probably be selected by a human operator when not using optimization. The proposed method offers significantly better (although still not good) conditioning of a system equation compared to reference placements. The results also provide an interesting insight into the influence of popular line geometries on numerical properties (and thus one component of uncertainty) of a contactless measurement system.

In recent years, the Steer-by-Wire (SBW) technology has been gaining popularity and replacing classical steering systems. It plays the most crucial role in autonomous cars where the vehicle must perform maneuvers on its own without driver's intervention. One of the key components of this system is the steering wheel angle sensor (SAS). Its reliability and performance may affect driver's life and health. The purpose of this paper is to show a test system to comprehensively evaluate the performance of the steering wheel angle sensor in the SBW system during real-world maneuvers and show how SAS parameters such as accuracy of angle, angular speed etc. affect car trajectory resulting in hit cones. For this purpose, a test system was built, with the use of virtual test drives based on CarMaker software, CANoe and VTSystem hardware. In order to evaluate its performance, the errors introduced by the system were determined. Additionally, using the realised test system, three commercial steering wheel angle sensors were tested and compared during a virtual test drive. Their errors were determined, as well as their performance in the SBW technology and the consistency of the obtained results with the parameters declared by the manufacturer were verified as well.

Measurements of medium and high voltages in a power grid are normally performed with large and bulky voltage transformers or capacitive dividers. Besides installation problems, these devices operate in a relatively narrow frequency band, which limits their usability in modern systems that are saturated with power electronic devices. A sensor that can be installed directly on a wire and can operate without a galvanic connection to the ground may be used as an alternative voltage measurement device. This type of voltage sensor can complement current sensors installed on a wire, forming a complete power acquisition system. This paper presents such a sensor. Our sensor is built using two dielectric elements with different permeability coefficients. A finite element method simulation is used to estimate the parameters of a constructed sensor. Besides simulations, a laboratory model of a sensor was built and tested in a medium-voltage substation. Our results provide a proof of concept for the presented sensor. Some errors in voltage reconstruction have been traced to an oversimplified data acquisition and transmission system, which has to be improved during the further development of the sensor.

The paper presents an experimental voltage divider circuit with the auto-calibration function during normal operation of the divider in the place of its installation. The results of first tests have been presented and on their basis the metrological properties of the system have been determined. Experiments proved that the circuit can successfully track changes of the divider parameters and correct their influence on the measurement results in a wide range of frequencies. Also, a major drawback, namely a systematic error of auto-identification results has been revealed. On the basis of additional tests the cause of the systematic error has been deduced to be the leakage of current between the elements of the divider through internal parasitic capacitances. This has led to the assumptions for another divider circuit design allowing for experimental verification of the hypothesis.

In the paper a method for the estimation of uncertainty of harmonic impedance measurement is proposed and validated. The validation is conducted using signals originating from the laboratory model of a grid of a known harmonic impedance, by the analysis of the histograms of estimated harmonic impedances. The results show that the proposed method can be used for the evaluation of the quality of harmonic impedance measurement results, independently for each of analyzed harmonics.

This paper presents a method for contactless measurement of instantaneous voltage and waveform reconstruction for use in a medium voltage (MV) indoor substation. Voltage waveforms are reconstructed by artificial neural network (ANN) using the signals originating from electric field sensors located under MV bus. The method is validated by the experiment in a typical indoor MV substation. Depending on the selection of ANN architecture and data used for training the root mean squared error of waveforms, reconstruction as low as 0.3% to 2.1% can be achieved in a steady state. The practical advice on the application of the proposed measurement method in a production environment is also given.

This paper describes a new method of power-grid signal variation measurement, consisting of a demodulation algorithm and a set of measures (indices). The measures that will be presented are the abstract functionals, operating on the complex envelope of a signal. This paper presents a demodulation algorithm, definitions of proposed functionals, results of simulations, as well as laboratory experiments.