谷物干燥过程模糊支持向量机控制器设计.docVIP

谷物干燥过程模糊支持向量机控制器的设计* 朱德泉1, 2 钱良存1 骆敏舟2 (1. 安徽农业大学工学院, 合肥 230036; 2. 中国科学院合肥智能机械研究所, 合肥 230031) 摘 要：为了准确控制谷物干燥过程的温度和湿度, 设计了一种基于改进遗传算法和最小二乘算法的干燥过程模糊支持向量机控制器。利用模糊算法解除温湿度的耦合作用, 采用支持向量机实现模糊逻辑控制的全过程和信号的非线线处理, 同时采用混合学习算法优化控制器参数, 即先采用最小二乘算法离线优化支持向量机性能参数, 再采用改进遗传算法在线优化支持向量机性能参数和模糊比例因子, 以使其控制性能适应对象的变化而达到最优。仿真结果表明, 设计的模糊支持向量机控制器比常规PID控制器和经典模糊控制器具有更好的控制性能, 能够满足谷物干燥工艺要求。 关键词：干燥；最小二乘支持向量机；遗传算法；温度；湿度 中图分类号：TP273.9 S625.5+1 　　　文献标识码：A　　　国家标准学科分类代码：270.4328 Design of fuzzy support vector machines controller for drying process of grain Zhu Dequan1,2 Qian Liangcun1 Luo Minzhou2 ( 1. College of Engineering, Anhui Agricultural University, Hefei 230036, China; 2. Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei 230031, China ) Abstract: To accurately control temperature and humidity for drying process of grain, fuzzy support vector machines controller for drying process was designed based on improved genetic algorithm and least square Algorithm. Fuzzy algorithm was used to decouple between temperature and humidity. Using support vector machines, fuzzy logical control of complete process and treatment of non-linear signal were realized. The controller parameters were optimized by hybrid learning algorithm. First, least square algorithm was used for off-line optimization to form support vector machines control system. Then the improved genetic algorithm was used for on-line optimization to get the optimal performance parameters of support vector machines and the optimal fuzzy proportional parameters. In the simulation, compared with the general PID controller and the traditional fuzzy controller, the results demonstrated that the fuzzy neural network controller designed gets better performance to meet the requirement of drying technology of grain. Keywords: drying; least square support vector machine; genetic algorithm; temperature; humidity 1 引 言 谷物干燥系统是一个多变量、强耦合、大滞后的复杂非线线系统。在谷物干燥过程中温度和湿度是影响干燥速率和干后品质主要因素而两者之间存在着强耦合关系传统控制方法难以精确控制温度和湿度。模糊控制不需要对控制对象进行精确的数学