Distribution Network Balance Method under High Proportion of Distributed Power Penetration Scenario

Abstract

With the rapid development of various subjects represented by distributed generation, how to implement power balance in a distribution network has become a research direction. The utilization efficiency of a distribution network can be improved by the bidirectional power flow of flexible resources. However, the traditional distribution network power balance method relies on historical data, making it challenging to determine the principle of new subject participation. It does not meet the requirements of bidirectional power flow balance and loses critical information such as complementary load characteristics. We believe that distribution network power prediction needs to transform from single load prediction to multi-resource prediction, from quantitative prediction to quantitative plus characteristic prediction. In comparison to the traditional methods, curve superposition can be better compatible with the research results of different new load forecasting topics and meet the needs of efficiency and precise network planning.