Real-time reconfigurable micro-system based on FPGA and CPLD for dual-mode PID control through backpropagation neural network

In this paper, a practically usable dual-mode control micro-system based on CPLD and reconfigurable FPGA is described. FPGA can be dynamically reconfigured under the control of CPLD to implement two models, backpropagation neural network model and its training model, both of which are respectively directed to two control modes for industrial produce. One mode is neural network performed automatic control, the other one as human-interfered traditional control. That is, only one single FPGA is reconfigured with multifunction. This technique can be widely applied into other control fields such as the adaptive control in different environments, space-ship control, measuring control in rough situation, and even production control.