Temperature Control System Based On Fuzzy Control

With the development of science and technique, the control theories have been continuously renewing and varying. From the earlier period' s classic Control theories to the sixty-seventy years' modern control theories based on the state-space described, They were used to solve the control problem of linear or nonlinear, constant or time-change parameter multi-in and multi-out system and have been acquiring the extensive application. But whether adopting the classic control the theories or the modern control theories to design a control system, we should need to in advance know the accurate mathematics model of controlled object. Then according to the model and the given target of performance, we choose appropriate control regulation to proceed control system of designing. However, the accurate mathematics model of controlled object can hardly be established in the many circumstance. Therefore, from seventies to now, the intelligence control theory provided us the most valid method to solve the control problem of complicated system, of which mathematics model can hardly be established.The fuzzy control as a branch of intelligence control make a great progress in its theory study and application of engineering. Now, the many industry process were all adopt the normal or the digital PID controller, but the results of controller are not quite satisfied. Fuzzy controller is especially suited for handling the process of the indetermination model. Because the essence of fuzzy control is nonlinear control and adaptive control, fuzzy control could be well done for those complicated, nonlinear systems with the characteristic of the parameter drift and the inaccurate examining signalIn the complicated industry controlled process, the controlled objects usually own the nonlinear, time-change characteristic and exists the category of interferences. So, being used the traditional control method, the static and dynamic output is not very satisfied. This paper adopts a fuzzy rule self-adjust control method to the need of industrial constant temperature control. The controller is composed of as follow: regression fuzzif ication, self-adjust fuzzy rule, and intelligent integral, intelligent sample.In the regression fuzzification, the error and error diversification input values are transformed to correspond fuzzy values using the form of the segment function, Comparing to basic fuzzy controller using the mete-genes. The choice of mete-gene is very important to the performance of fuzzy controller. So the method of the regression fuzzification not only avoid the bed performance offuzzy controller resulting form the wrong mete-genes but also make the control process optimized and enhance the robustness of fuzzy controller.Intelligent integral own the apery intelligent character of nonlinear integral. It can simulate the Memory characteristic of human and strategy of apery intelligent control, it have the choice to remember useful information, but "forget " useless information. Therefore, it can overcome the weakness of the general integral controller and restrain the set over and reduce the error of stable state.Intelligent sample control is a novel control method of close loop inside open loop. When the controlled variables are depart from hopeful value, the controller sent out the single of sampling switch on. Then controlled variables arrived the balance of value through fuzzy controller. At the same time, sampling switch is off. Control system is on the state of open loop; the controller supplies the energy of object. Method of intelligent sampling control is very useful for controlling the time delay system.Finally, the results of MATLAB simulation of industrial constant temperature objects are satisfied. Especially to the time-delay and nonlinear systems it shows a very brilliant performance comparing to the PID controller.