Design and FPGA realization of Incremental Fuzzy controller for Stirred Tank Heater

Abstract

In this paper, Field programmable gate array (FPGA) realization of fuzzy PI controller for Stirred Tank Heater system is presented. The commonly used PI controller in industry though it is simple for implementation lacks reliability when the dynamics of a system changes with time. Replacing PI controllers with AI systems can lead an effective control of a system even in the presence of model uncertainties. The digital implementation of AI systems in DSP or Microcontrollers have limitations of low computational accuracy and slow processing speed. Now adays, FPGAs are being largely used for digital implementation of algorithms because of their higher computational speed and accuracy. In this paper, fast and novel design approach for rapid prototyping of Fuzzy PI controller for Stirred Tank Heater is presented thoroughly. System Generator SIMULINK add on from VIVADO is used to generate Very High-Speed Integrated Circuit Hardware Description Language (VHDL) directly from MATLAB. The developed controller for stirred tank system is verified in Fixed point arithmetic using Xilinx Simulink blocks. The VHDL code is then generated, synthesized, implemented and made ready for physical implementation on Kintex-7 evaluation board.