Power Quality Prediction, Early Warning, and Control for Points of Common Coupling with Wind Farms

Abstract

Wind farms can affect the power quality (PQ) of the power supply grid, with subsequent impacts on the safe and stable operation of other electrical equipment. A novel PQ prediction, early warning, and control approach for the common coupling points between wind farms and the network is proposed in this paper. We then quantify PQ problems and provide rational support measures. To obtain predicted PQ data, we first establish a trend analysis model. The model incorporates a distance-based cluster analysis, probability distribution analysis based on polynomial fitting, pattern matching based on similarity, and Monte Carlo random sampling. A data mining algorithm then uses the PQ early warning flow to analyze limit-exceeding and abnormal data, quantify their severity, and output early warning prompts. Finally, PQ decision support is applied to inform both the power suppliers and users of anomalous changes in PQ, and advise on corresponding countermeasures to reduce relevant losses. Case studies show that the proposed approach is effective and feasible, and it has now been applied to an actual PQ monitoring platform.

Keywords

• data mining
• decision support
• early warning
• power quality (PQ)
• trend analysis
• wind farm