Abstract:

In order to solve the impact of disordered charging of large-scale electric vehicles on the power grid stability, two ordered charging mathematical model with minimum gird load peak-valley difference and minimum user charging expenses is established. The Whale Optimization Algorithm（WOA） has been improved for efficient and fast solution, two nonlinear inertia weights are incorporated into the algorithm to improve both the local and the global search ability, and a Teaching-Learning Strategy（TLS） is proposed to improve the position quality of individual whales. And the diversity of the population is improved by means of variation in TLS, which can effectively prevent the premature stagnation of iteration. The variation disturbance is used to improve the population diversity and prevent the premature stagnation of iteration. IWOA, WOA and Particle Swarm Optimization algorithm（PSO） are used to solve the benchmark functions and the ordered charging optimization problems of electric vehicles respectively. Finally, the comparison of those algorithms verifies the efficiency and practicability of IWOA.

Key words: electric vehicle, ordered charging, Improved Whale Optimization Algorithm（IWOA）, nonlinear inertia weight, Teaching-Learning Strategy（TLS）

摘要：

为解决大规模电动汽车无序充电对电网稳定性造成的影响，建立了电网层负荷峰谷差最小和用户层充电费用最小的两方面有序充电目标函数。为实现高效且快速的求解，对鲸鱼算法（Whale Optimization Algorithm，WOA）进行了改进，在该算法中加入两种非线性惯性权重来平衡局部搜索能力和全局搜索能力，并提出了一种教学策略（Teaching-Learning Strategy，TLS）来提高鲸鱼个体的位置质量，教学策略中采用变异手段增加种群的多样性，能有效防止迭代过早停滞。算例中分别利用IWOA、标准WOA、粒子群算法（Particle Swarm Optimization，PSO）测试基准函数，并对电动汽车有序充电优化目标进行求解，最后通过比较验证了IWOA的高效性和实用性。

关键词: 电动汽车, 有序充电, 改进鲸鱼算法（IWOA）, 非线性惯性权重, 教学策略（TLS）