A Novel Algorithm for Wind, Solar Irradiation, and Load Joint Scenario Generation: MC-AE-PSO-Kmeans

This paper introduces a novel algorithm MC-AE-PSO-Kmeans for generating typical scenarios of regional renewable energy generation and regional load, integrating machine learning and intelligent optimization techniques. The proposed approach begins with a statistical analysis of the original dataset to extract essential statistical features for simulation generation. Dimensionality reduction is then applied to identify the most important features, enhancing the method’s stability and reliability. Finally, clustering reduction is performed using intelligent optimization to derive typical scenarios. The method demonstrates high integration, strong adaptability, and excellent adjustable performance, requiring only modifications to the original dataset, the number of Monte Carlo generated scenarios, and the number of typical scenario reductions to quickly obtain results. Experimental tests have yielded four typical scenarios and determined the probability of generating similar scenarios in reality, showcasing the effectiveness and practicality of the proposed method.