Sensorless cluster based neural-fuzzy control strategy for four quadrant operation of three phase BLDC motor with load variations

Brushless DC (BLDC) motors are, in fact, a type of permanent magnet synchronous motors with a very high level of efficiency. An attempt has been made in this research work to design and implement four quadrant control of a BLDC motor drive with regenerative braking in either direction with a Position sensor less neural-fuzzy controller. The complete closed loop system being speed-controlled, four quadrant operation has been obtained using step speed input while the suitability of the developed model has been tested under full load stress during steady state. The results obtained satisfy the four quadrant operation requirements of advanced drives where controlled starts and stops are essential in both forward and reverse directions. This is evident in the effectiveness of current and torque tracking and ease of speed transition from motoring to regeneration and vice versa. The design simulation of four quadrant control of the BLDC motor is carried out using MATLAB. A simulink model is developed to simulate and analyze the operation of the motor. A permanent magnet synchronous machine with trapezoidal back EMF is modeled as a BLDC machine. The developed model finds applications in advanced industrial drives as an energy-efficient and cost-effective alternative to eliminate the effects of supply voltage drops and mechanical load variations.