基于ALO-BP神经网络和机械零件加工特征的工艺方案决策

Process Method Decision-making of the ALO-BP Neural Network and the production feature of mechanical parts

针对传统制造企业工艺知识应用不充分,用于工艺方案决策过程的部分BP神经网络决策效率和准确率不高的问题,采用蚁狮优化(Ant Lion Optimizer,ALO)算法优化的BP神经网络构建了基于零件加工特征的工艺方案智能决策模型。首先对产品及工艺数据进行预处理,其次应用蚁狮优化算法对BP神经网络的初始化权值和阈值进行优化,最终基于样本数据集开展神经网络训练,进一步建立智能决策系统,并以柴油机零件为对象进行了工艺方案决策方法的应用验证。实例验证表明,采用优化的BP神经网络后,决策的速度和精度都有明显的提升;所构建决策系统是可行的,能够用于工艺方案的决策。

To solve the problem of Inadequate application of process knowledge and low efficiency and accuracy of partial BP neural network decision-making in the process method decision-making procedure for traditional companies,this paper uses the BP neural network optimized by the algorithm of the ant lion optimizer(ALO)to establish the intelligent decision-making model for process method with mechanical parts production features.First of all,make pretreatment of products and process data,and secondly,ALO algorithm is used to optimize the initial weights and thresholds of BP neural network.Finally,the neural network is trained based on the sample dataset,establishing intelligent decision-making system and verifying the process method decision-making application with the target of diesel engine parts.Through the example verification,it can be proved that the optimized BP neural network can significantly improve the speed and accuracy of decision-making,the established decision-making system is feasible and can be used for the decision-making of process method.