The Design And Implementation Of Temperature Control Experiment Platform Based On Model-free Adaptive Control

After the development of conditional control theory and modern control theory, control theory has come to a new spring. Conditionally, in industry, we use control theory like this:firstly, we should build up the linear model of certain real plant which is usually nonlinear; then apply our control theory to the established model for controlling. So the controlling effect of this conditional method is determined by the precision of the established linear model. Although this method is often used in producing of industry, with the rapid developing of scientific and technology, this conditional method can not fulfill some practical needs for that the establishment of mathematic model is very difficult even imposable sometime.Thus we think that:if we can gain the same controlling effect without the establishing of mathematical model, then the problem will be solved. Data driven control is brought up in this situation. Only input and output information is used in data driven control, so it will be large scale used in the future and it is worthy to do some research work on data driven control.This paper builds up an experiment flat form for data driven control research. The boiler temperature control system is very useful in practical application, so we decide to choose this system as our object to be controlled. In this flat form, a software application system and an intelligent meter are built for the application of the data driven control algorithm-model-free adaptive control (MFAC).This paper is organized like this:the background of the research of data driven control and the research situation are given in section 1. Section 2 introduces the theory of model-free adaptive control algorithms. The simulation research of MFAC is given in section 3. After the software application system and the intelligent meter designed, one experiment is done in section 4 on the bases of the experiment flat form. After doing this, the dream, building an experiment flat form for data-driven control, has come true. Section 5 provides some final conclusions and directions for future work.