计及调峰能力差异性的系统多源优化调度

针对电力系统调峰能力、负荷及新能源出力的显著性差异,分别建立新能源全额消纳和消纳量最大化两种优化调度模型。前者系统调峰能力强、负荷量大,新能源按预测值消纳,火电承担基荷,水电根据风电、光伏和负荷的波动优化调整;后者适用调峰能力较弱情形,优先以新能源弃电量最少为优化目标,并结合运行成本最优,建立多源互补优化模型。针对传统非支配遗传算法(NSGA-Ⅱ)求解优化模型存在计算速度较慢,种群多样性较差等缺陷,提出基于正态分布交叉算子和动态变异算子改进的NSGA-Ⅱ算法,改进后的算法的计算速度和收敛性能均显著提升。算例仿真表明两种模型应对不同的调峰情形时,同时兼顾运行经济性和新能源消纳最大化,可作为大规模新能源并网后日前调度方式安排的依据,具有理论意义和实践价值。

Multi-Source Optimization Scheduling of System Considering the Differences of Peak Shaving Capacity

In allusion to the significant difference in the peak-shaving capacity of the power system,the load and the output of new energy,the optimization models for full accommodation and maximum accommodation of new energy were established respectively.The optimization model for full accommodation has a strong peak-shaving capacity,and new energy is absorbed according to the predicted value.Meanwhile,thermal power is responsible for the base load,and hydropower is adjusted according to fluctuations in wind power,photovoltaics and load.The optimization model for maximum accommodation which is applied to the situation that the peak-shaving ability of the power system is weak,takes new energy minimum abandonment as the optimization objective and combines with the optimal operating cost.Aiming to solve the slow convergence rate and poor population diversity of the conventional NSGA-Ⅱ,an improved NSGA-Ⅱalgorithm is proposed based on normal distribution crossover and dynamic mutation operator,which improves significantly the computational speed and convergence.The simulations show that the two models can deal with the system with different peakshaving capacity,taking the economics of operation and the maximization accommodation of new energy into account at the same time,which has theoretical and practical value,and can be used as the basis for the day-ahead scheduling arrangement of the large-scale new energy integration.