基于混合遗传算法的航空发动机数学模型解法

Hybrid genetic algorithm in solving aeroengine nonlinear mathematical model

关键词:  涡轮风扇发动机  非线性数学模型  平衡方程  收敛性  牛顿-拉夫逊法  混合遗传算法  计算效率

Key words:  Turbofan engine  Nonlinear mathematical model  Balance equation  Convergence  Newton-Raphson  Hybrid genetic algorithm  Computing efficiency

摘要:

针对发动机模型现有平衡方程迭代解法存在不收敛或计算效率差的不足,以涡轮风扇发动机为对象,结合遗传算法与牛顿-拉夫逊法的优点,设计了在模型不收敛点采用遗传算法与牛顿-拉夫逊法交替计算的平衡方程求解混合算法。理论分析与数值仿真结果表明,对于整个模型而言,该算法既保持了牛顿-拉夫逊法的高计算效率,又吸收了遗传算法全局收敛的优点,可实现模型大范围收敛。

Abstract:

Current balance equation solutions for aeroengine nonlinear mathematical models are of unconvergence sometimes or computing speed inefficient.With regard to this shortcoming,combined GA(Genetic Algorithm) with N-R(Newton-Raphson),an improved hybrid algorithm is developped to be applied to a turbofan engine static and dynamic performance model.In the improved hybrid algorithm,N-R is also the main solutions and GA alternates with N-R when N-R diverges.compared with current model,theoretical analysis and performance simulation results show that the algorithm not only can maintain N-R computing efficiency,but also improve convergence obviously.