Optimization And Application Of Multi-model Adaptive Contro Algorithm

In practical industrial manufacture,the accused plant often has problems such as large nonlinear characteristic,large time variation,external disturbance and parameter jump,which will fortify the hardness of the work of the control plant.Although the conventional adaptive control algorithm is extremely simple,it still can not control the complex control system effectively.Multi-model adaptive control mainly uses two or more models combined with adaptive control algorithm to obtain the desired steady state and transient performance indexes.It can solve the control problem of complex systems with large instability.It has good control characteristics and a wide range of applications.The study contents of this thesis are as follows:1.On the basis of studying the object mechanism of room temperature conditions,a mathematical model of fresh air system temperature is established,the PID algorithm is improved by population optimization technology,and a control scheme combining the model reference adaptive control and the improved PID algorithm is proposed,and finally the improved PID control is compared with the improved PID model reference adaptive control for simulation,and the results show that the control algorithm combined with the improved PID and the model reference adaptive control has small overshoot and fast response speed,excellent dynamic characteristics and excellent control performance without steady state error.2.Aiming at a series of nonlinear phenomena in the system,the multi-model switching adaptive control scheme of neural network optimized according to population optimization is designed.The use of population technology can improve the neural network weights,and solve the optimal weights;according to the population optimization of the neural network algorithm and multi-model adaptive algorithm to design adaptive control scheme,establish a suitable switching rule,so that the control system at any time can choose the optimal controller to control the system,the good approximation ability of the neural network enhances the controllable effectiveness of adaptive control,simulation analysis shows that the proposed optimization algorithm has the advantages of faster convergence speed and higher accuracy.3.In view of the problems such as poor system response caused by instability in the process of system operation,the PID multi-model adaptive control algorithm based on the genetic algorithm is designed,and three parameters of PID are optimized by genetic algorithm to obtain excellent parameters;then,based on the nonlinear control system,the control scheme combining the PID control mechanism and multi-model control improved by the genetic algorithm is designed to achieve real-time optimization of the controlled object and enable the plant to realize high-powered follow-up control.Simulation results show that compared with the PID multi-model adaptive control under the genetic algorithm,the response time of the system is shorter,the deviation after running to a stationary state is smaller,and the operating oscillation amplitude and frequency of the system are small.