Research On Identification And Fusion Control Strategy Of Hydraulic Turbine Control System

At present,with the installed capacity of hydropower stations is increasing,the safe and stable operation of power plants has become a primary concern for technicians.Therefore,it is particularly important to effectively identify and control the turbine control system.For the conventional PID control,the intelligent control theory such as sliding mode control,neural network control,sliding mode control and genetic algorithm is introduced into the turbine control system.The main work of this article is as follows:(1)Aim to the complexity and time-variation of the turbine control system,the overall identification of the turbine control system by a nonlinear autoregressive neural network is studied.First of all,for a large number of missing,anomalous,incomplete data,through the introduction of MySQL database,using Python language to preprocess it,to improve the accuracy of the identification data;Secondly,because of dynamic neural network structure is complex,weight and other issues,using the improved L-M algorithm accelerates the speed of network training.Finally,it compares with the identification of static neural networks such as BP and RBF.The results show that the dynamic neural network has high identification precision and good generalization performance.(2)For the medium and high water head environment,the traditional PID control can not adjust the PID control parameters adaptively,so this paper introduced the fuzzy neural network adaptive PID control algorithm..Firstly,the neural PID is compared with the traditional PID control to illustrate the strong self-learning ability of the neural network.Secondly,Fuzzy and PID control and Fuzzy-PID composite control are used to verify that the fuzzy control has good reasoning ability,but this control accuracy is lowly.Combined with the advantages and disadvantages of several algorithms,the three control algorithms are combined to make the control system have the characteristics of fuzzy nonlinearity,neural self-learning ability and the accuracy of PID control.The results show that the fusion algorithm can adjust the control parameters well and has strong robustness.(3)For the high head environment,the speed of the turbine control system is easily overlooked,and the sliding mode control is used to dynamically track the speed change of the turbine.First,the identification problem was re-examined from the perspective of control,and the sliding mode variable structure control was introduced into the turbine control system.For the problem that the control system is under optimal control and the unit speed value cannot reach the given value,the system model is changed to three inputs,the steady-state error is eliminated,and the sliding mode variable structure controller is introduced into the turbine control system.Turbine rotation speed is an important parameter.Secondly,in order to solve the intrinsic chatter phenomenon in sliding mode variable structure control,fuzzy inference system and genetic algorithm are introduced to eliminate chattering.The simulation results show that in the high head environment,the sliding mode control has good robustness and effectively solves the chattering problem.