Research On Implicit Generalized Predictive Control And Virtual Simulation Of Ball Mill

With the rapid development of renewable energy in China,new energy is gradually occupying an important position in the power structure,and traditional thermal power is gradually developing towards peak shaving power.However,However,the changing working conditions during peak shaving require thermal power units to be equipped with the capability of more flexible controlling and stabilizing on the frequency output.The direct-blown steel ball mill is an important auxiliary system of the coal-fired unit.The dynamic physical parameters of the steel ball mill can represent the running state of the equipment,among which the outlet temperature and material level of the ball mill have a direct influence on the fuel quality,flexible control of the load,and operation reliability of the unit.Therefore,modeling and control system optimization of the ball mill is a field to be explored urgently in the pulverizing system of coalfired units.In this thesis,the double-in-double-out(DIDO)steel ball mill was taken as the research object,the operation mechanism,main factors,and control requirements of the steel ball mill are analyzed.The mathematical model of the outlet temperature and fill level of the steel ball mill was established,and the Implicit Generalized Prediction Control(IGPC)system was designed.Finally,the virtual simulation technology to achieve three-dimensional visualization of the two-dimensional data was used.First,the input/output data of the actual operation on the site were collected and preprocessed,and the system model parameters were identified by RLS.The CARIMA mathematical model of the outlet temperature and material level of the ball mill was established,and the System model validation;Then,the implicit generalized predictive control(IGPC)algorithm was designed for the outlet temperature and material level of the ball mill,respectively,and the control algorithm was tracked and simulated in Matlab,The simulation results:the initial values of the outlet temperature and material level of the steel ball mill are 79℃ and 850pa,respectively,and the hot air door opening and coal feed volume disturbance are increased by 15%at 300s,respectively,the stabilization time and overshoot of the outlet temperature simulation curve are 355s and 5.6%,respectively,and the stabilization time and overshoot of the material level simulation curve are 368s and 9.52%,respectively,the set value tracking simulation of both can achieve the expected effect.The anti-interference ability of implicit generalized predictive control and traditional PID control is compared,and the antiinterference ability of implicit generalized predictive control is obviously better than that of traditional PID.In the end,this thesis simulated the process control of a steel ball mill with the help of virtual reality technology,using 3Ds MAX to establish a three-dimensional physical model of the DIDO steel ball mill,and bringing it into the virtual engine Unity 3D to incorporate corresponding functions.Moreover,it also completed the communication between Unity 3D and Matlab by calling out Matlab’s local program through C#,realizing the combination of simulation data and virtual simulation;it used the particle system to realize the change of the size of the air volume,the stress of the pulverized coal,the temperature of the air-powder mixture,and the trajectory of the pulverized coal movement making the visualization of the outlet temperature control,fill level control and data simulation of the steel ball mill come true with the interactive function of the UI interface.Establishing a model from the on-site data,this thesis optimized the steel ball mill control system based on the IGPC and achieved good control quality.In the meantime,the virtual simulation system facilitated the staff to observe the dynamic changes in the process control of the steel ball mill and also paved a way for the digitalization of the development of smart power plants.