应用基于MPI的混合粒子群算法的三维水翼优化设计

提出一种三维水翼的优化设计方法.方法应用混合粒子群算法(HPSO)与边界元法相结合进行三维水翼的优化和性能计算工作、应用多级罚函数法解决水翼设计这一多约束、多变量的优化问题.基于免疫理论和惯性权值非线性递减策略的混合微粒群算法,能够有效抑制算法早熟收敛,平衡全局和局部搜索能力.优化设计过程中,水翼的剖面形状、攻角及展弦比作为设计变量,给定的压力分布形式、升阻力系数作为设计约束或设计目标.混合粒子群算法通过划分子种群、应用基于MPI通信机制的并行计算来实施,最大限度减小了计算时间.设计算例表明了文中提出的三维水翼优化设计方法收敛速度快、计算时间短、有效可行.

Optimal Design of Three-Dimensional Hydrofoils Using Hybrid Particle Swarm Optimization Based on MPI

A method for the optimal design of three-dimensional hydrofoils is presented. In this design method, the hybrid particle swarm optimization (HPSO) algorithm incorporating potential based on boundary element method (BEM) is utilized for the optimizations and performance computations. Multi-stage penalty function is adopted for this multi-constraint and multi-variable optimization problem. HPSO algorithm, based on immune theory and nonlinear decreasing inertia weight (NDIW) strategy, can effectively prevent premature convergence and keep balance between global and local searching abilities. During the optimal design, the foil geometry, angle of attack and aspect ratio are selected as design variables; a prescribed pressure distribution, lift coefficient and drag coefficient are set to be design constraints or design objective. HPSO algorithm is designed by dividing the whole particle swarm into several sub-swarms, and implemented by parallel computing based on message-passing interface (MPI) to minimize computation time. In this study, HPSO algorithm based on MPI is found to be a useful and effective tool for the optimal design of three-dimensional hydro-foils due to its faster convergence and less computation time.