Modeling And Optimal Control Of Temperature System For Electric Boiler Based On Particle Swarm Algorithm

Electric boiler is a typical controlled object in industrial process control and scientific experiment.Improving the accuracy of temperature control of electric boilers is very important for improving product quality and ensuring production safety.Due to the characteristics of large hysteresis,unidirectional heating and time-varying characteristics of the electric boiler temperature system,it is difficult to achieve the desired results using traditional control methods.Therefore,it is necessary to study its advanced control methods.The main content of this paper is to study the modeling and optimization control method of electric boiler temperature system with the help of particle swarm optimization.This paper first introduced a miniature electric boiler experimental device developed by the research group,elaborated on the design ideas,control system composition and working principle.It is driven and heated by DC voltage,with small volume and high safety factor.The control system adopts Quarc control platform produced by Canada Quanser company,which is seamlessly connected with MATLAB / Simulink and has high openness.It is an ideal R & D platform for scientific experiments and student training.Due to the large hysteresis,it takes a long time to implement the control scheme and parameter setting directly on the experimental device.We can first model and optimize the scheme and parameters on the simulation system and then implement it on the device.Through this method,it is possible to increase efficiency and reduce experiment costs.Based on this purpose,this paper studies the modeling method of the electric boiler temperature system.According to the experimental data of the system open-loop response,the model structure of two first-order inertial links in series and hysteresis links is determined.Then use particle swarm optimization to optimize the static gain constant and the time constant of the two inertial links in the model,so that the established model can more realistically simulate the motion characteristics of the system.Secondly,the PID controller is designed in the simulation system,and the parameters of the PID are optimized by the particle swarm,and the optimized parameters are directly used in the experimental device to obtain better experimental results,and the simulation results and the experimental results are well fitted.Then,on the basis of traditional PID,further study the method of PID controller parameters adaptively changing in real time with tracking error,and use particle swarm to optimize the adjustable parameters.Finally,in order to improve the stability and dynamic performance of the system,an electric boiler temperature control scheme of Smith predictor + fuzzy control is proposed.Smith predictor is used to compensate the pure delay characteristic,and the delay link is separated from the closed loop,and the inner loop adopts fuzzy control strategy.The thesis elaborated the controller design process in detail,and used particle swarm optimization to optimize the three quantization factors of fuzzy controller and the value range of some theories in the membership function,and finally compared and summarized several control schemes.The experimental process and results show that with the help of particle swarm optimization,the performance of the control system can be greatly improved,and the efficiency of controller design and implementation can be improved.