基于差异演化算法的高速储能飞轮形状优化设计

Shape optimization of high speed energy storage flywheel based on differential evolution

采用差异演化算法对高速储能飞轮的形状进行优化,以实现储能飞轮具有最大的质量能量密度,并且以在高速旋转的状态下满足强度要求为设计目标。针对差异演化算法进化速度依赖种群个体差异的特点,提出权重因子随种群聚集度变化而变化的策略,并增加混沌变异操作,以提高算法跳出局部最优的能力。构建了仿真优化程序,通过Visual C++6.0调用通用有限元软件ANSYS来计算飞轮的应力分布。通过优化计算得到了满足强度要求的质量能量密度最大的飞轮形状,与同质量的圆盘飞轮相比,转动惯量提高了28.39%。

The shape of the high speed engrgy storage flywheel was optimized with the differential evolution algorithm to acquire the maximum engrgy density per mass and the necessary strength. Due to the evolution speed of the differential evolution depends on the difference of the species population individuals, a strategy to change the weighting factor according to the population aggregation degree was suggested, and a chaos mutation operation was introduced to avoid the local optimization. An optimization program was compiled, the general finite element analysis software ANSYS was called by the Visual C＋＋6.0 to calculate the stress distribution inside the flywheel. The shape of the flywheel with the maximum energy density per mass which satisfies the strength requirement was found by the optimization program. Compared with the original disk flywheel, the moment inertia of the optimized one was increased by 28.39%.