一种基于混合遗传算法的有源噪声控制方法

Active noise control method based on hybrid genetic algorithms

提出了一种基于混合遗传算法的格型IIR滤波器结构的有源噪声控制方法 .混合遗传算法将遗传算法与随机搜索算法结合起来 ,可以改善基本遗传算法的局部搜索能力 ,克服基本遗传算法存在未成熟收敛问题 .本文选择UNDX交叉算子作为遗传算法的主要算子 ,在保留当前最佳个体的同时 ,再对该最佳个体用随机搜索法搜索优化个体 .这样既保证了算法的全局收敛性 ,又提高了收敛速度 .仿真结果表明 ,该算法可以有效地实现噪声控制 .

A hybrid genetic algorithm (HGA) is presented for an active noise control (ANC) method with a lattice IIR filter. The HGA approach combines the genetic algorithm with the stochastic search algorithm to improve the local-search properties and overcome the premature convergence of the standard genetic algorithm (SGA). A real-coded genetic algorithm is adopted using the unimodal normal distribution crossover (UNDX) and the elitist model. And then, the stochastic search algorithm is used to optimize the elitist. This approach maintains the globe-search properties of the GA and improves the convergence speed. Simulation experiments show that the results of the application in active noise control are satisfactory.