Research On Model Reference Adaptive Control Method And Its Application To Thermal Process Control

A class of direct model reference adaptive control algorithm is presented firstly under the influence of bounded disturbances and unmodeled dynamic for a practical plant. Considering the single-input single-output plant, the error model of the stable adaptive scheme proposed by Narendra with unmodeled dynamic is given. Based on Lyapunov stability theory, a new adaptive law is derived which is suitable for object with unmodeled dynamic using heuristic ideas by reselecting a new Lyapunov function.The proposed model reference adaptive control method is applied to the overall control of the common important equipment of the large scale continuous production process, industrial boiler and evaporator. In order to complete the calculation of the output amount within one sampling period, it can not apply too high sampling rate. Because the advanced control algorithm is quite complex and the computational cost is relatively large. Both of the industrial boiler and evaporator are complex systems which are multi-variable, strong coupling, nonlinear, long time-delay. But the parameters of the type of object change usually relatively slowly, so they are very suitable for verifying all kinds of new advanced algorithm.The control plant in this paper is the high precision semi-physical simulation experimental system SMPT-1000. The configuration of the control system is completed based on the Siemens PCS7system and S7-400PLC. The operation results under load variations show that the continuous-time model reference adaptive control can achieve the closed-loop control by applying the computer. The circulation of the heat quantity and working medium is normal and the combustion is stable. The experimental results show that the model reference adaptive control can meet the requirement of the process control.Nonlinear feedback exact linearization method and a-modification composite model reference adaptive control method are applied to the coordinated control system of the drum boiler-turbine units in the thermal power station. Considering the interference and system perturbation, the robustness of the static exact feedback linearization decoupling is analyzed. That controller is used to compensate the impact of the exact feedback linearization failure is proposed. The a-modification composite model reference adaptive controller for each loop is designed by deeming coupling to disturbance of the adaptive system. The method to design the system matrix of the model reference adaptive system according to the matching conditions is introduced. The simulation and comparison show that the a-modification composite model reference adaptive control method is superior to other existing programs the same as based on the premise of the feedback linearization.