PSO优化灰色神经网络的珩磨油石磨损预测

Wear Prediction of Honing Oil Stone Based on PSO Optimize Grey Neural Network

为了预测油石的切削寿命,保证珩磨加工质量,引入灰色神经网络,通过将珩磨工艺加工参数作为模型输入来预测油石的磨损量,最终建立了珩磨油石磨损量预报模型。在油石磨损量预测过程中,针对神经网络存在收敛速度慢、容易陷入局部最优等缺陷,利用粒子群算法对灰色神经网络的参数进行优化。试验结果表明,基于粒子群算法改进的灰色神经网络具有更好的逼近能力和预测精度,便于合理更换油石。

In order to predict the cutting life of oil stone, ensure the quality of honing processing and facilitate reasonable replacement oil stone, the gray neural network has been introduced with the honing process parameters as model input to predict the honing oil stone wear, the wear prediction model of honing oil stone was established. In the process of oil stone wear prediction, the limitations like slow convergence speed of neural network and easily falling into local optimum are given. Particle Swarm Optimization（PSO） algorithm has been implemented to optimize the grey neural network parameters. The experience results show that the grey neural network based on improved particle swarm algorithm has better approximation performance and prediction precision.