Research On Predictive Control And Its Application In Temperature Process

Temperature is an important control parameter in practical application.Due to the increasing complexity of the controlled process,there are many problems in controlling temperature.How to improve the control effect and meet the control requirements of complex systems is an important research direction in the field of temperature control.In this paper,the ETAL temperature control experimental device was used to simulate the process of calcining cement clinker in rotary kiln in industrial production,which was difficult to control with such characteristics as large time delay.The process mathematical models were established,the appropriate predictive control algorithms were selected,and the temperature process was controlled by programmable controller.According to the step response of three temperature zones,the mathematical models of different temperature zones were established by test method.The model includes time-delay part and time-delay free part.The time-delay free part was described by an inertial model,and the time-delay part was described by an inertial with pure hysteresis part.The PID control scheme was selected for the part of temperature process without delay,and the parameters of the controller are optimized by using the simplex method.Because the temperature process is a typical slow process,and the sampling speed in the actual experiment can reach the level of milliseconds,the high sampling frequency improves the response speed,but it also ignores the variation trend of the error,resulting in the difference between the two temperatures in the sampling period.It will make the differential in the PID control algorithm invalid;if the sampling time is increased,the system will oscillate strongly.In order to solve this problem,PID control scheme based on differential sampling was proposed.Different sampling time were set for the deviation of temperature and the variation of deviation,so that the value obtained from sampling is obvious.The experimental results showed that the control quality was improved.The delay part of the temperature process choosed the predictive control algorithm,which were the predictive PI and the dynamic matrix control(DMC)algorithm.The parameter of predictive PI controller was optimized and self-adjusted online.The simulation results show that the above two controllers can maintain good control performance with strong anti-interference ability and robustness under the condition of partial or complete mismatch of parameters.Discreting the predictive PI control algorithm,developed a new array to store and update the control amount at the previous N moments,and the parameters of the differential sampling predictive PI controller was designed to adjust online with the input of deviation signal.Based on the step response of the process,MATLAB was used to calculate the dynamic matrix,output expectation matrix,control matrix and error correction matrix offline,and the matrix operation module in Automation Studio was combined to realize real-time control.According to the flow chart of the control algorithms,ST language was applied to program the control scheme in PLC.The experimental results show that the three control schemes were effective,robust and anti-jamming.