Day-Ahead Optimal Scheduling of Time-Series Multi-Scenario Considering Source-Load Uncertainty

In order to solve the problems of high-risk scenario leakage, poor scheduling plan economy and optimal scheduling plan determination in the scenario analysis method in the day-ahead scheduling optimization considering source and load uncertainty, the time series multi-scenario considering spatiotemporal correlation is divided into normal and extreme time series scenario through confidence interval. The typical time series scenario is obtained by the improved K-means method. A two-stage optimization scheduling model based on time series scenarios including a scheduling plan optimization layer and a dispatching plan optimization layer was established. When solving the model, the typical normal time series scenario is first optimized and solved, and the optimal scheduling plan for the day ahead is obtained with the goal of maximizing the sum of the positive reserve and negative reserve of the system in different time periods. On this basis, the typical extreme time series scenarios are optimized to obtain dispatching plans. The example results verify the feasibility and economy of the proposed day-ahead scheduling optimization method.