Search

Improvement of photovoltaic (PV) power plant performance under partial shading conditions aiming to increase electrical energy generation is in the focus of this research. This paper proposes the performance optimization of PV power plant under partial shading conditions by new PV string reconfiguration technique. The Matlab/Simulink model is used to simulate the operation of the PV string under partial shading conditions and with the proposed recombination technique. Simulated operational conditions have experimentally been verified by string characteristics measurement on the existing roof-top PV system at the Faculty of Technical Sciences in Novi Sad. Simulation and experimental results showed a high degree of matching, while the obtained values proved that proposed method leads to output power increase and higher PV system generation in PV string operation under partial shading.

Regarding the possibility of harmonic generation, personal computers (PCs) are very significant within the category of non-linear low power loads. In modern distribution systems, the number of PCs simultaneously connected to distribution grid constantly increases. This PC clustering can negatively affect the quality of supply. The paper deals with mathematical modeling of the voltage total harmonic distortion (THDU) caused by operation of PC cluster. The proposed model is based on the power quality measurements carried out in a computer centre and computer simulations. It takes into account the THDU dependence on the PC cluster size and grid stiffness. Model parameters are derived in the least squares sense. The influences of cable cross-section and pre-existing THDU of the phase voltage are also discussed. The accuracy of the proposed model is confirmed by additional measurements performed in a commercial building.

This paper presents a novel space vector selection scheme applicable for the control of dual three-phase induction motor drives supplied from a six-phase voltage source inverter (VSI). The vector selection method is based on the vector space decomposition technique (VSD). Unique vector selection pattern simplifies problems related to complicated implementation of standard VSD in commercially available digital signals processors (DSP). The proposed vector selection scheme is verified through a theoretical analysis, computer simulations and practical experimental results conducted on a dual three-phase test rig prototype with control algorithm implemented in Texas Instrument?s TMS320F2808 DSP.

The novel hybrid method of the Discrete Fourier Transform (DFT) and the Enhanced Phase-Locked Loop (EPLL) has been presented. The original well-known methods and the Hybrid method have been analysed and tested in several simulations and experiments. The methods are implemented on the Digital Signal Processor and compared in a real time power quality application. Although it requires a bit higher computational effort, the Hybrid method compared to original methods shows advantages for on-line harmonic analysis.

Harmonics current emission from personal computers (PCs) is of special interest to power quality researchers, due to their wide spread application and the fact that they are usually concentrated in large groups (clusters). The input current total harmonic distortion (THDI) dependence on number of connected PCs in a cluster (NPC) is affected by network parameters and load side parameters of each cluster's PC simultaneously. As the impact of grid parameters on the THDI=f(NPC) dependence is analyzed in the author's previous papers, special emphasis is given on the impact of the DC side smoothing capacitor (C) in PCs power supply unit. For engineering application it is convenient to present the THDI=f(NPC) function with simple mathematical expressions which include these effects. In the paper authors are proposing an improved mathematical expression to presents impact of load side (DC capacitance) parameters. To achieve this goal, results of the measurements of harmonic emission from a large computer center with 167 clustered PCs and the computer simulation for even larger number of PCs are used.