粒子群优化神经网络的工程陶瓷电火花加工效果预测模型

Process effect prediction model for EDM of engineering ceramics with neural network based on particle swarm optimization

针对工程陶瓷电火花加工的工艺电参数与工件的加工效果之间具有高度的非线性关系,难以建立精确数学模型的问题,建立了BP神经网络模型,以预测工程陶瓷电火花加工工艺效果。采用自适应位置变异粒子群算法,优化了网络模型的阈值和连接权值,解决了BP神经网络算法迭代速度慢、易于陷入局部最优解的问题;以碳化硼为例完成了算法的实现,对该材料工件的加工效果进行了预测。研究结果表明:自适应位置变异粒子群神经网络算法可以较好地反映电参数与表面粗糙度之间的非线性关系,算法的迭代次数显著减少,并具有较高的预测精度,模型的可靠性和有效性得以证实。

Aiming at the problem that there is a highly non-linear relationship between the process electrical parameters and processing effect of the workpiece in electrical discharge machining ( EDM) of engineering ceramics, and it is difficult to establish an accurate mathematical model,BP neural network model was established to predict the process effect of EDM of engineering ceramics, and particle swarm optimization algorithm with adaptive position variation was used to optimize the threshold and connection weights of the network model, which solved the problem that the BP neural network algorithm has a slow iteration speed and is easy to fall into the local optimal solution. Boron carbide was taken as an example, the algorithm was realized and the processing effect of the workpiece was predicted. The results indicate that the neural network algorithm based on the optimization of particle swarm optimization with adaptive position mutation can better reflect the nonlinear relationship between electrical parameters and surface roughness. The iteration times of the algorithm are significantly reduced, and the prediction accuracy is high. The reliability and validity of the model are confirmed.