融合Q学习和梯度搜索JAYA算法的水电机组优化控制研究  

Research on the Integration of Q-learning and Gradient Search JAYA Algorithm in Hydroelectric Unit Optimization Control

作　　者：马元江 邹屹东 庞小淞 王忠全[1] 肖志怀[2] 陈飞 MA Yuan-jiang;ZOU Yi-dong;PANG Xiao-song;WANG Zhong-quan;XIAO Zhi-huai;CHEN Fei(Sichuan Power Company Yingxiuwan Hydropower General Plant,China State Grid,Chengdu 611830,Sichuan Province,China;School of Power and Mechanical Engineering,Wuhan University,Wuhan 430072,Hubei Province,China)

机构地区：[1]国网四川省电力公司映秀湾水力发电总厂,四川成都611830 [2]武汉大学动力与机械学院,湖北武汉430072

出　　处：《中国农村水利水电》2024年第10期157-163,共7页China Rural Water and Hydropower

基　　金：国网四川省电力公司科技项目(521901230001)。

摘　　要：为提高水电机组调节系统的动态性能,提出一种融合了Q学习和梯度搜索JAYA算法(QJAYA)的机组调节系统参数优化方法。通过引入电路等效理论推导有压管道电路等效模型,利用BP神经网络建立充分反映水轮机流量和力矩非线性特性的水轮机神经网络模型。结合调速器和发电机模型,构建了水电机组调节系统整体非线性精细化模型。利用QJAYA算法对水电机组调节系统的PID控制参数进行优化,并通过仿真试验验证了所提方法能够有效提高水电机组调节系统的动态性能。To improve the dynamic performance of the hydroelectric unit regulation system(HURS),a method for optimizing the parameters of the HURS is proposed,which integrates Q-learning and gradient search JAYA algorithm(QJAYA).The equivalent circuit theory is introduced to derive the equivalent model of the pressure pipeline circuit.A hydroelectric turbine neural network model,reflecting the nonlinear characteristics of turbine flow and torque adequately,is established using BP neural networks.Subsequently,combining with the governor and generator models,a refined nonlinear model of the HURS as a whole is constructed.The QJAYA algorithm is utilized to optimize the PID control parameters of the HURS,and through simulation experiments,it is verified that the proposed method can effectively improve the dynamic performance of the HURS.

关 键 词：电路等效 水电机组 调节系统 优化 

分 类 号：TK731[交通运输工程—轮机工程]