梯级水电站长期多目标模糊优化调度新模型

Long-term multi-objective fuzzy optimization scheduling model of cascaded hydroelectric stations

梯级水电站不仅要满足电力系统运行要求,还要考虑发电和用水之间的协调,才能使综合效益最大化。提出一种兼顾年发电量和运行成本的梯级水电站长期多目标优化调度新模型。通过分别求解各个单目标优化问题和定义各单项目标的隶属度函数,把多目标问题模糊化;采用对各单项目标优化的目标值在一定范围内伸缩的方法来体现决策者的主观意愿;利用模糊最大满意度方法把多目标优化问题转化为单目标非线性规划问题;并构建了一种动态调整惯性因子的自适应粒子群算法。仿真计算验证了模型的正确性和求解方法的可行性,多目标模型比单目标模型获得了更佳的综合效益,模糊优化处理方法避免了目标权重选取的人为任意性,同时自适应粒子群算法计算速度快、收敛精度高。

The cascaded hydroelectric stations should not only satisfy the demand of power systems but also consider the coordination of generation and water consumption to maximize the comprehensive profits. A novel long- term multi- objective optimization scheduling model of cascaded hydroelectric stations is proposed,which comprehensively takes the annual power generation and water consumption as optimal objectives. The multi-objective problem is fuzzed by defining the membership degree function of each objective and solving every single objective model. The decision maker may slightly adjust the objective value of each single objective. The multi- objective problem is changed into single objective nonlinear programming problem by using the fuzzy satisfaction- maximizing method. A SAPSO（Self-Adaptive Particle Swarm Optimization）algorithm which can dynamically adjust the inertia factor is built up. Simulation results show that the multi- objective model is correct and the proposal approach is feasible. The multi-objective model can achieve better comprehensive profits than the single- objective model,and the fuzzy optimal method avoids the contrived randomicity in objective weight selection. SAPSO has rapid computational speed and higher convergence accuracy.