基于模糊神经网络PID的真空吸铸控制系统设计  

Design of Vacuum Suction Casting Control System Based on Fuzzy Neural Network PID

作　　者：孙振保 许书娟 王敬泽[3] SUN Zhenbao;XU Shujuan;WANG Jingze(College of Automotive Engineering,Guangxi City Vocational University,Chongzuo 532200,China;College of Architecture and Engineering,Guangxi City Vocational University,Chongzuo 532200,China;School of Materials Science and Engineering,Harbin University of Science and Technology,Harbin 150000,China)

机构地区：[1]广西城市职业大学汽车工程学院,广西崇左532200 [2]广西城市职业大学建筑工程学院,广西崇左532200 [3]哈尔滨理工大学材料科学与工程学院,黑龙江哈尔滨150000

出　　处：《热加工工艺》2022年第7期75-78,共4页Hot Working Technology

基　　金：黑龙江省自然科学基金项目(E201447)。

摘　　要：针对真空吸铸设备压力控制系统存在的非线性、强耦合和多变量扰动等问题,对真空吸铸工艺进行分析,并设计了一种用于压力实时优化的模糊神经网络PID控制器。利用BP神经网络强化模糊算法的在线推理功能,建立铸件质量、最低薄壁厚度、材料密度、吸铸室压力等参数与PID修正量之间的非线性模型,通过模糊算法在线整定PID输出。利用Matlab软件进行仿真分析,并与PID和模糊PID进行对比。结果表明:模糊神经网络PID无超调,既能使压力满足工艺设定又能有效控制减压速度等参数,具有良好的鲁棒性,满足充型和凝固等过程压力控制需求。Aiming at the problems of nonlinear, strong coupling and multivariable disturbance in the pressure control system of vacuum suction casting equipment, the vacuum suction casting process was analyzed, and a fuzzy neural network PID controller for real-time pressure optimization was designed. Using BP neural network to strengthen the on-line reasoning function of fuzzy algorithm, a nonlinear model between casting quality, minimum thin-walled thickness, material density,pressure of suction chamber and PID correction was established, and then the PID output was adjusted on-line by fuzzy algorithm. The Matlab software was used for simulation analysis and compared with PID and fuzzy PID. The results show that the fuzzy neural network PID has no overshoot, which can not only make the pressure meet the process setting, but also effectively control the decompression speed and other parameters. It has good robustness and meets the requirements of pressure control in the process of filling and solidification.

关 键 词：真空吸铸 压力 控制系统 模糊神经网络PID 

分 类 号：TG249.9[金属学及工艺—铸造] TP183[自动化与计算机技术—控制理论与控制工程]