基于FPGA的RBF甲醇精馏塔温压解耦控制  

Temperature and Pressure Decoupling Control of Methanol Distillation Column Based on FPGA

作　　者：龚瑞昆 张文庆 GONG Rui-kun;ZHANG Wen-qing(College of Electrical Engineering,North China University of Science and Technology,Tangshan Hebei 063210,China)

机构地区：[1]华北理工大学电气工程学院,河北唐山063210

出　　处：《华北理工大学学报（自然科学版）》2021年第3期104-109,共6页Journal of North China University of Science and Technology：Natural Science Edition

基　　金：国家自然科学基金项目(61203343)。

摘　　要：针对甲醇精馏塔常压塔塔顶的精馏控制过程中塔顶温度和压强存在耦合,常用的PLC-PID控制存在不足,提出了一种基于FPGA的RBF神经网络解耦控制方法。RBF神经网络解耦算法选用高斯函数,并对神经网络的隐含层数据中心、连接权值和基宽进行了优化。训练结果显示进行了有效解耦,比常用的BP神经网络解耦算法响应时间短、逼近目标速度快。结合FPGA控制硬件,选用CycloneⅡ系列EP2C35F672C7型芯片,对硬件电路进行设计。通过在MATLAB上与传统的PLC-PID控制、基于FPGA的RBF未解耦控制和基于FPGA的BP解耦控制进行仿真对比研究,仿真结果显示,该种控制方法对现阶段甲醇精馏塔精馏过程的控制具有一定的改进,可提高控制的效率,进而提高精制甲醇产品的质量和企业的经济效益。In view of coupling between the top temperature and the top pressure in the distillation control process of the atmospheric tower of the methanol distillation column,and the common PLC-PID(programmable logic controller,proportion integral derivative)control has shortcomings,a RBF(radial basis function)neural network decoupling control method based on FPGA(field programmable gate array)was proposed.Gaussian functions RBF neural network decoupling algorithm was selected,and the hidden layer data center,connection weight and base width of the neural network were optimized.The training results show that the proposed method is effective and has shorter response time and faster approaching target speed than the commonly used BP neural network decoupling algorithm.Combined with FPGA control hardware,the EP2C35F672C7 chip of Cyclone II series was selected to design the hardware circuit.The simulation results show that this control method has a certain improvement on the control of methanol distillation column distillation process,and improves the control efficiency.Then the quality of refined methanol products and the economic benefits of the enterprise are improved.

关 键 词：控制 解耦 RBF FPGA 甲醇精馏塔 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]