基于BP神经网络技术的化工材料生产工艺模型优化  

Optimization of chemical material production process model based on BP neural network technology

作　　者：胡轶然 郑奋 HU Yiran;ZHENG Fen(PLA Naval Military Medical University,Shanghai 200433,China)

机构地区：[1]中国人民解放军海军军医大学,上海200433

出　　处：《粘接》2023年第11期84-86,113,共4页Adhesion

摘　　要：针对传统化工材料生产工艺无法实现最优工艺参数求解及精度要求的问题,结合人工智能技术,提出一种基于神经网络与遗传算法相结合的非线性函数极值寻优方法,将其应用于羟基磷灰石复合材料制备的感应热沉积工艺中,进行遗传算法优化神经网络模型的最优涂层总量求解。实验结果表明,通过遗传算法结合BP神经网络,在以电流频率、沉积温度和沉积时间作为输入时,可准确预测涂层沉积质量。由此得出,构建的感应热沉积工艺的模型的实际值与预测值误差较小,可满足该材料制备工艺的精度需求。In view of the problems that the traditional chemical material production process can not achieve the optimal process parameters and accuracy requirements,in combination with artificial intelligence technology,a nonlinear function extreme optimization method based on combination of neural network and genetic algorithm was proposed.T method was applied to the induced thermal deposition process of the hydroxyapatite composite material preparation,and the neural network model optimized by genetic algorithm was used to solve the optimal totalcoating.The experimental results showed that the genetic algorithm combined with BP neural network could accurately predict the coating deposition quality when current frequency,deposition temperature and deposition time were taken as inputs.It has been demostrated that the error between the actual and predicted values of the induced thermal deposition model is small,which can meet the accuracy requirements of the material preparation process.

关 键 词：化工材料制备 工艺优化 BP神经网络 遗传算法 感应热沉积 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] TQ015.9[自动化与计算机技术—计算机科学与技术]