基于混合的GA-PSO神经网络算法

粒子群优化(PSO)算法是一类随机全局优化的技术,它通过粒子间的相互作用发现复杂搜索空间中的最优区域。提出了一种基于GA和PSO混合的算法(GA-PSO)用于神经网络训练。算法在产生下一代时,结合了交叉、变异算子和粒子群算法中的速度—位移公式,充分利用了遗传算法的全局寻优和粒子群算法收敛速度快的优点。经GA-PSO训练的神经网络应用于三元奇偶问题和IRIS模式分类问题,与BP、GA和PSO算法相比,该算法在提高训练误差精度的同时加快收敛速度,并能有效避免早熟收敛。仿真结果表明,GA-PSO算法是有效的神经网络训练算法。

Based on mix GA-PSO nerve network algorithm

Particle swarm optimization algorithm is a kind of stochastic global optimization technology.It finds optimal regions of complex search spaces through the interaction of individuals in a population of particles.This paper proposes a kind uses for training artificial neural network based on GA and the PSO mix algorithm.The algorithm when has next generation,unified crossover and mutation operator and the particle swarm optimization algorithm speed-displacement formula.The algorithm is successfully applied to 3-model and IRIS pattern classification problems,compared with BP,GA and PSO,this algorithm can improve the classification accuracy while speeding up the convergence process,and can avoid premature effective.Simulation results show the effectiveness of the proposed algorithm.