Optimization Of PID Parameters For The Superheated Steam Temperature Control System Of Thermal Power Units Based On Data Driven

With the development of information science and technology,the production process and manufacturing technique of enterprises are becoming more and more complex.And the mechanism modeling or identification modeling of these processes faces enormous challenges.As a control method that does not require the precise model of the controlled object and only relies on the I/O data and other control information of the control system to complete the control tasks of the system,data-driven control has gradually become the focus of modern control field research.In this paper,the data-driven PID control algorithm is an online data-driven algorithm based on the basic ideas of Just In Time and Model On Demand.A Hammerstein nonlinear model system is used to simulate the algorithm.In the simulation process,the information selection criteria for k-Nearest Neighbors may fail to accurately select the data information from history data information with high correlation of the current query points,use the k-Vector Nearest Neighbors instead.the information selection criteria comprehensively examine the distance and angle information of the data vector to improve the accuracy of data selection.At the same time,the database selection rules that have a great impact on the data-driven control method are modified,including the size of the database capacity,the principle of data elimination and retention,so as to achieve efficient online update of data.The performance index of parameter updating was further optimized and the control quantity was increased to improve the accuracy of parameter updating.The simulation results show that the method is effective.The data-driven PID control method is extended to the thermal power unit superheated steam temperature control system.The main controller in the traditional cascade PID control is replaced with a data-driven PID controller,and a data-driven cascade PID control system of superheated steam temperature is designed.In the simulation,aiming at the characteristics of the superheated steam temperature system,such as strong nonlinear and high time delay,optimized the reference model of the data-driven PID control method to make the data control method more suitable for the superheated steam temperature system.In order to simultaneously illustrate the effectiveness of the data-driven control method,three groups of parameters with different control effects were selected to build the initial database of the system when the initial database was established.Experiments show that the data-driven control method can always optimize online to calculate the best control parameters and obtain the best control effect.By comparing the control effect of data-driven PID control with the traditional fixed-parameter PID control,it is found that the data-driven PID control can always complete the control task faster and more stably when the system changes control,better robustness and response speed.