Research On Modeling And Data-driven Control Of Semi-finish Grinding System

Cement grinding is the last link of cement production,It directly determines the final production and quality of cement.In recent years,the application of semi-finish grinding technology has somewhat increased the cement production and reduced energy consumption.The application of laser particle size analyzer in cement grinding explores the potentiality in quality improving,energy saving,consumption reduction and production increasing of grinding.In this paper,used the online laser particle size analyzer as detection method,combined with the semi-finish grinding process mechanism,encircled cement grinding mill load and cement particle size,carry out semi-finish grinding system modeling and data driven control research.This paper is supported by major project of Shandong Province“The Key Technology Research and Application Demonstration of Intelligent Factory(2015ZDXX010F01)” and internationaloperation project “The Research on Integrated Control System of Cement Production Process for Energy Saving and Emission Reduction(SQ2013ZOC600)”,The main research works of this paper are as follows:(1)Aiming at the two key parameters of mill load and cement particle size of cement semi-finishing grinding system,established its mathematical model based on the analysis of influencing factors.The speed of circulating fan and the mill current are chosen as the input and output parameters of the neural network,extreme learning machine neural network is used to establish the mathematical model of mill load;The method generates weights and thresholds randomly,and remains constant,the optimal solution can be obtained after determining the number of hidden layer neurons;The training speed of this method is fast,relatively insensitive to the learning rate,and it can achieve global optimal.The speed of the latter separator is chosen as the input of the model,and the less-than 45 ?m particle content is the output of the model;The mathematical model of cement particle size is established by least square method;The simulation results show that the model can reflect the dynamic change of cement particle size,and it lays the foundation for the following research on cement particle size control algorithm.(2)In order to improve the stability and robustness of particle size control,the adaptive PID control method for cement particle size system based on data-driven technology is designed and the dependence of control method on modelis solved.The compact form dynamic linearization(CFDL)method is used to deal with nonlinear cement particle size control system;The I/O data(the speed of the latter separator and the less-than 45 ?m particle content)of particle size control system is used to estimate its pseudo partial derivative(PPD),and the parameters of the PID controller are adjusted according to PPD;The simulation results show the effectiveness of the proposed control algorithm.(3)The particle size optimization control scheme of semi-finishing grinding is proposed,including the system’s hardware and software architecture,database design and control algorithm combining Bang-Bang control with data-driven adaptive PID control;Theparticle size optimal control system of semi-finishing grinding is developed and applied in engineering.