With the increasingly severe global energy and environmental issues, the traditional economic load dispatching with the single goal of power generation cost has been unable to meet the national macro strategic requirements for energy conservation and emission reduction. This research mainly discusses the multi-objective optimization of spatial adaptive partitioning based on environmental economic dispatch problems. This study uses the method of calculating power generation costs and pollution control costs to select a compromise solution. When initializing the individual optimal position memory bank and the global optimal position memory bank, put the first-generation particle position into the individual optimal position memory bank, and put the optimal position obtained by comparing the first-generation particle into the global optimal position. In the location memory. Then calculate the fitness value corresponding to each objective function of each particle, compare the calculated result with the corresponding results of each Pareto solution in the internal memory bank and the external memory bank, and replace the particle position with the best. In the multi-objective processing strategy, in order to better obtain the Pareto optimal solution set and the Pareto optimal frontier, an external archive is set up to save and update the new elite Pareto optimal solution generated after each iteration. Since the quality of the particles in the random simulation particle swarm algorithm will directly affect the convergence and diversity of the final Pareto optimal solution set, this study selects the historical optimal particles according to the Pareto dominance relationship. When the multi-objective optimized power system economic dispatch research model with wind farms was adopted, the cost of power generation increased by 6.21 million yuan, the cost of polluting gas treatment was reduced by 8.1 million yuan, and the total cost was reduced by 1.89 million yuan. The research results show that the multi-objective optimization model proposed in this research has achieved a relatively satisfactory balance between power generation costs and environmental advantages.

随着全球能源和环境问题日益严峻，以发电成本为单一目标的传统经济负荷调度已无法满足国家节能减排的宏观战略要求。本研究主要讨论了基于环境经济调度问题的空间自适应分区的多目标优化。本研究使用计算发电成本和污染控制成本的方法来选择折衷方案。在初始化个体最优位置存储库和全局最优位置存储库时，将第一代粒子位置放入个体最优位置存储库，并将通过将第一代粒子进行比较而获得的最优位置置于全局最优位置。在位置存储器中。然后计算与每个粒子的每个目标函数相对应的适应度值，将计算结果与内部存储库和外部存储库中每个Pareto解的对应结果进行比较，并以最佳位置替换粒子位置。在多目标处理策略中，为了更好地获得Pareto最优解集和Pareto最优前沿，建立了一个外部档案库来保存和更新每次迭代后生成的新的精英Pareto最优解。由于随机模拟粒子群算法中粒子的质量将直接影响最终Pareto最优解集的收敛性和多样性，因此本研究根据Pareto优势关系选择历史最优粒子。采用风电场多目标优化电力系统经济调度研究模型，发电成本增加621万元，污染气体治理费用减少810万元，总成本减少1.89。万元。研究结果表明，该研究提出的多目标优化模型在发电成本和环境优势之间取得了较为令人满意的平衡。