The Design And Implement Of Fuzzy Controller Based On FPGA

Fuzzy control is an important composition of intelligence control. It can provide effective controlling on the plant with imprecise mathematical model. So it can be used in various fields such as expert system, digital signal processing, process control, communication system. Traditionally, the fuzzy inference system is implemented by software or general microprocessor, but these methods have its limitation. Considering the character of fuzzy calculation, it is more similar to parallel data processing, so fuzzy control suit implement on exclusive circuit, which can raise not only processing speed but also system stability.The character of logic programmable provide Field Programmable Gate Array(FPGA) with the ability of implement of specific integrated circuit, which fit the demand of the design of system on chip, so it become the development trend of logic programmable device. Combining the FPGA technique and the theory of fuzzy control, a fuzzy controller based on FPGA is designed and implemented, which takes the advantage of the ability of parallel data processing of FPGA. It provides a hardware platform of the implement of fuzzy controller. Using subsection 2nd order polynomial to fit the fuzzy membership function, it's no need to design for each kind of membership function respectively. Using Active-Rule-Driven Inference to realize the fuzzy inference, it not only reduces the calculation, but also enhances the fuzzy output. Using logarithm subtraction to replace division, it rise the data processing and enhance the real-time performance. The fuzzy operation, which is disassembled properly, is implemented with 10 levels pipelines. The membership functions and the fuzzy inference rules can be configured depends on different plants.Use the fuzzy controller as a digital controller on DC/DC converter. Firstly analyze the math model of the converter, depend on which configure the parameter of the fuzzy controller. Under the environment of MATLAB/Simulink, use the PLECS toolbox to modeling the DC/DC converter and the simulation result improves the effect of the fuzzy controller. Use Xilinx XC3S500E FPGA chip to implement the fuzzy controller, and get a well experiment result.