Adaptive ANN rule-based controller for a chopper fed PMDC motor electric vehicles drive

The paper presents a new adaptive neural network based speed controller for the PMDC motor drive using an online tunable ANN control structure. The Proposed ANN regulator is tuned online using the backpropagation algorithm to minimize the actual motor speed error and current deviation. The performance of the ANN control algorithm is evaluated using a laboratory prototype drive system. The laboratory set-up comprises a PMDC motor fed from a DC supply or battery source via a pulse frequency modulation (PFM) MOSFET DC-DC chopper. The output from the ANN controller is used to regulate the duty cycle ratio (α_D) of the chopper converter. This scheme would be suitable for small urban type and factory forklift-EV drives using small size battery. The new online ANN rule-based assignment algorithm is used to update the ANN network weights and biases to ensure continuous effective dynamic response while keeping the motor inrush current under specified tolerable limits. The ANN controller can be assembled using available microchip controllers with EPROM software subroutines. The test results validate the robustness and flexibility of the proposed new ANN based speed controller under different load variations and excursions. Also, it shows good flexibility in dealing with any drive parametric uncertainties and sudden load variations. The paper presents digital simulation, laboratory testing and control validation results.