Dynamic reactive power optimization considering load uncertainty and period optimization

Aiming at the uncertainty of load and fully combining its timing characteristics, a dynamic reactive power optimization model considering load uncertainty and time optimization is proposed. The model fully considers the probability model of load forecasting error, and uses the scenario reduction method combining improved K-means clustering and simultaneous backward reduction (SBR) to obtain load samples and their probabilities. The expected values of load samples are normalized by statistical indicators to form a comprehensive load trend sequence. At the same time, in order to overcome the ' dimension disaster ' problem in the process of dynamic reactive power optimization, an adaptive method based on characteristic trend is designed to divide the time period. Finally, considering the scene probability, with the goal of minimizing network loss, the voltage over-limit and generator reactive power over-limit are included in the penalty function, and the dynamic reactive power optimization model is established and solved by genetic algorithm based on elite preservation strategy. Simulation results verify the correctness and effectiveness of the proposed model.