Neuro-Fuzzy Logic Based Control Scheme for PV-Battery Integrated Sensorless Induction Motor Drive for Water Pumping

This paper deals with a bidirectional power flow scheme for a PV array-battery interfaced system for an induction motor drive used for water pumping. The bidirectional power flow is achieved by bidirectional buck-boost converter. The motor control is achieved by field-oriented control (FOC), without mechanical sensor. The proposed system uses rotor flux based MRAS (Model Reference Adaptive System), with minimum parameters requirement, which is achieved by neuro-fuzzy based speed estimation technique. The DC link current is sufficient to estimate three phase motor winding currents. The modified space vector modulation technique is used for phase currents estimation. The developed prototype is tested in the laboratory and its suitability is judged through various test results under different conditions of insolation variations and also current sensor failure condition.