改进混沌搜索的AMPSO-BP激光铣削质量预测

为了更好地控制激光铣削的质量,建立了激光铣削质量和铣削层参数的神经网络模型。针对神经网络易陷入局部极小值的缺点,提出混沌搜索的自适应变异粒子群优化算法(AMPSO)获得神经网络最佳参数,建立了AMPSO-BP激光铣削质量预测模型。最后以某种材料的激光铣削质量预测为例,将文中所提算法与PSO-BP、BP神经网络预测结果相比,结果表明所提方法有很高预测精度且预测误差明显减小,在实际中有一定应用价值。

Laser Milling Quality Predictionof Improved Chaotic Search AMPSO-BP neural network

In order to better control the quality of laser milling,a neural network model of laser milling quality and milling layer parameters was established.Aiming at the shortcomings of neural network apt to fall into local minimum value,the adaptive mutation particle swarm optimization algorithm(AMPSO)of chaotic search is used to obtain the optimal parameters of neural network,and the AMPSO-BP laser milling quality prediction model is established.Finally,taking the laser milling quality prediction of a certain material as an example,the proposed algorithm is compared with the PSO-BP and BP neural network prediction results.The results show that the proposed method has high prediction accuracy and the prediction error is significantly reduced,it has certain application value in practice.