基于遗传算法的大容量输油泵模糊控制器的优化设计

The Optimal Design of Fuzzy Controller for High Power Petroleum Pump System Based on Genetic Algorithm

由于大容量输油泵系统存在严重的非线性和时变性 ,输油泵入口压力、出口压力及管道流量三个参数之间又存在耦合关系 ,采用传统的PID控制器难以对它们实现协调控制。针对大容量输油泵系统的特点 ,本文提出了一种由两个模糊控制器组成的大容量输油泵控制策略 ,并应用改进的遗传算法对模糊控制器的隶属函数和融合因子进行了优化。仿真和现场运行结果表明 ,该控制器能够对入口压力、出口压力及管道流量进行协调优化控制。

Owing to uncertainty and non-linearity of the highpower petroleum pump system and the couple relations of the entry pressure,the outlet pressure,and the flow of the system,it is difficult to achieve good control performance of the system by using conventional PID controller.Most of high power petroleum pump systems are still controlled manually.In this paper,a new fuzzy control strategy of the system which is composed of two fuzzy controllers is proposed.To avoid oscillation of the system caused by redundant amplitude of the controller output,the whole output of the controller is divided into benchmark variable and regulating variable.The benchmark variable of the controller output is obtained approximately from the prior operational data according to the settings of the flow.The regulating variable is the synthesis of the outputs of the two fuzzy controllers according to the deviation between the settings of the flow the inlet pressure the outlet pressure and the actual operational values of the system.Based on improved genetic algorithms,the membership functions and the parameters of the fuzzy controllers are optimized.The simulation results and operational results show that the good control performance of the high power petroleum pump system can be realized by the new control strategy.