Realization of Fuzzy PID Controller on FPGA for Source Measurement Unit

Abstract

The classic Source Measurement Unit's (SMU: Source Measurement Unit) shortcomings include a sluggish transient response and a significant overshoot. This paper proposed an optimization method that made use of fuzzy PID controller on FPGA for SMU. With the aid of the software Matlab, we first analyzed the SMU's circuit diagram and obtained the system's transfer function. Next, we constructed the system simulation model adjusted by a fuzzy PID controller in Simulink. The initial P, I, and D parameters were obtained by the PID tuner, and they were adjusted by the fuzzy controller until they can achieve the optimal performance. Finally, we designed a Verilog program with the initial parameters to implement the fuzzy PID controller, in which the parameters were modified dynamically according to error and its increment to make sure they were optimal. We tested the controller using the HDL simulation tool Modelsim to ensure the simulation results matched those from Simulink. The results of simulation showed that the proposed method outperformed the conventional approach. In detail, our fuzzy PID controller can achieve shorter rising time, shorter setting time, and smaller overshoot than the NI SourceAdapt technology does.