An intelligent evaluation method for steady-state power quality in distribution networks incorporating dynamic correlations

To accurately and efficiently evaluate the steady-state power quality at key nodes in a distribution network， a hierarchical intelligent evaluation method for steady-state power quality metrics incorporating dynamic correlations is proposed. First， the maximum information coefficient （MIC） is used to analyze the relationships among steady-state power quality metrics at these key nodes； then an integrated evaluation system for steady-state power quality is developed. Next， the CRITIC， an objective weighting method， along with the results from the correlation analysis， is employed to calculate the overall weights of each metric in the evaluation system. Finally， the aggregated evaluation score is used as the expected value， and an integrated intelligent evaluation model for steady-state power quality at key nodes is developed using a long short-term memory （LSTM）. The LSTM is optimized with the measured sample data. Simulation results demonstrate that the proposed method significantly reduces the complexity of the evaluation process while maintaining accuracy， thereby greatly enhancing evaluation efficiency.