The effect of distributed generation on the quality of power

Abstract

This document presents a study of the effects distributed generation (DG) has on the power quality (PQ) of electrical power networks. At the time the study was proposed the scheme of DG was speculative by nature and the initiative to employ it was eagerly anticipated. The aim of the study is to explore the possibilities of DG. More specifically, the impact DG has on PQ in power networks is best studied and described by practical measurements of real existing systems in accordance with the purposes and goals of this study. Unfortunately, such systems (termed scenarios) are uncommon given that the technologies associated with DG were not widely operational at the time of writing the dissertation. The alternative is to simulate the DG scenarios and lay a foundation for a well defined knowledge base sketching the behaviour of DG scenarios. This could aid with predictions and provide realistic boundaries for DG performance expectations. The effects are observed, documented. motivated and apt conclusions drawn and guidelines are proposed for the effective utilization of DG. In this document the development of three experimental simulation scenarios will be described. The reader will also encounter the proposed and adopted methods for simulation of the developed scenarios. It is shown that a strategic application of DG is accompanied by numerous beneficial influences. all of which are advantageous effects and ensures improved long-term scenario performance. These effects include, but are not limited to, the isolation of waveform disturbing loads (sources), the reduction of impedance paths leading to these non-linear loads. and the attenuation of harmonics by coincidental passive filtering paths in a network, to name an important few. Proof of these advantageous effects is found in the comprehensive study and investigation of accurately developed computer simulation scenario models and the generated results thereof. Finding correlations between the researched literature and the responses of the simulated scenarios support the validity of the simulation results and will aid in truthfully describing the observed effects. Precise conclusions are drawn and form the basis of the suggested strategy for PQ improving DG applications. A cost function is proposed. developed and presented. The generated data is scrutinized and after careful perusal key factors are identified that play a major part in the cost function. The identified crucial factors are incorporated in the assembly of three indicators that collectively quantify. the performance of the scenario being investigated. A cost function is thus applied to fittingly quantify the total performance of each of the three individual scenarios. The cost function is tested and shown to respond appropriately if any of the three indicators are intentionally disturbed - i.e. reset one indicator to a new value while keeping the other two constant. The cost function relays the value change returning a new performance index, reflecting the change in the concerned indicator. The generated cost function data is carefully studied in conjunction with the simulation results and further conclusions are drawn based on motivations supported by the researched literature. These final conclusions complement the initial findings and provide guidelines for the strategic application of DG maximizing the effect DG has on the performance of the optimized scenario. The observed effects are summarized and detailed presentations of each are made in cause-related categories. Recommendations are made for improvements to the present study and future work linked to the study is proposed.