基于解集动态分析含风电接入的多目标机组组合研究

Multi-objective unit commitment with wind farms based on dynamic analysis of solution set

风电的大规模接入给多目标节能减排发电调度带来了新的机遇和挑战。由于风电场出力具有随机性,采用置信区间简化风电场景模拟数量,考虑到多目标模型的复杂性,利用Benders分解技术对模型进行降维,设计一种基于解集动态分析的多目标自适应优化算法对降维后的多目标主问题进行求解,并提出一种提高模型整体求解效率的预处理机制加速收敛。仿真结果表明所提方法能够有效求解含风电的多目标机组组合问题,并验证了所提多目标算法和预处理机制在求解模型中的优势。

The large-scale access of wind power brings new opportunities and challenges to the multi-objective energysaving and emission-reduction generation scheduling. The confidence interval is used to simplify the simulation quantity of wind power scenario due to the randomness of wind farm output. Taking into account the complexity of multiobjective model,the Benders decomposition technique is adopted to reduce the dimension of the model. A multi-objective adaptive optimization algorithm based on the dynamic analysis of solution set is proposed to solve the multiobjective master problem after dimension reduction,and a pretreatment mechanism for improving the overall solving efficiency of the model is proposed to speed up the convergence. The simulative results show that the proposed method can effectively solve the multi-objective unit commitment problem with wind power,and verify the advantages of the proposed multi-objective algorithm and pretreatment mechanism in solving the model.