Hybrid Prediction-Based Dynamic Constrained Multiobjective Evolutionary Algorithm for Multisource Complementary Scheduling

Under the critical requirements on reducing greenhouse gas emissions, solar, wind, and other clean energy have been significantly developed for power generation, and formed multisource complementary system. To rationally schedule them, inherent dynamic factors that may cause insufficient energy utilization, especially time-varying power demand, and solar and wind fluctuations, are fully mined. Taking minimum cost and carbon emission as objectives, a dynamic constrained multiobjective scheduling model is built for multisource complementary system. To solve this problem, a hybrid prediction strategy-based dynamic constrained multiobjective evolutionary algorithm is presented to transfer the cross-temporal knowledge by fully mining historical scheduling schemes, achieving the optimal output of each generation unit. Extensive numerical results indicate that the proposed method is superior to state-of-the-art competitors with performance indicators and scheduling results, showing satisfactory potential in promoting the low-carbon economic operation of power system.