基于变异反向学习郊狼优化算法的光伏智能边缘终端优化配置方法

光伏智能边缘终端 (photovoltaic intelligent edge terminal,PVIET)是光伏电站安全稳定运行和智慧运营的重要设备之一,其功能强大但价格十分昂贵,容量较小的分布式光伏电站难以承受其高价格。为解决上述问题,文章提出了一种光伏智能边缘终端优化配置方法,从数量与位置2个方面建立了光伏智能边缘终端的优化布局模型。同时,针对郊狼优化算法(coyote optimization algorithm,COA)优化性能弱、可操作性低的问题提出了一种变异反向学习郊狼优化算法(mutation opposition-based learning COA,MOBL-COA)。根据典型案例的测试结果表明,所提的优化布局方法能够有效实现成本的降低,并且所提的MOBL-COA在解决优化布局问题时,在收敛速度、求解精度与算法稳定性上均表现出优异性能,验证了所提方法的有效性与可行性。

Application of Coyote Optimization Algorithm Based on Mutation and Opposition-Based Learning in Optimal Configuration of Photovoltaic Intelligent Edge Terminal

Photovoltaic intelligent edge terminal (PVIET) is one of the important equipment for safe and stable operation and intelligent operation of PV power station. It is powerful but expensive, and the distributed PV power station with small capacity cannot afford its high price. In order to solve the above problems, an optimal configuration method for PVIET is proposed, and the optimal layout model of PVIET is established from two aspects, i.e., quantity and location. At the same time, aiming at the weak optimization performance and low operability of coyote optimization algorithm (COA), a mutation opposition-based learning COA (MOBL-COA) is proposed. The test results of typical cases show that the proposed method can effectively reduce the cost, and the proposed MOBL-COA has excellent performance in convergence speed, solution accuracy and algorithm stability, which verifies the effectiveness and feasibility of the proposed method.