A Sigmoid Function Based Improved Observer for PV-Battery fed Sensorless PMaSyRM Driven Light Electric Vehicle

This paper presents a light electric vehicle (LEV) integrated with a solar photovoltaic (SPV) array. A bidirectional DC-DC (BDDC) converter is used to maintain the DC-link voltage and charging of the battery. The cascaded PI controllers are used to maintain the voltage and current of the LEV battery. The battery is charged via BDDC when solar insolation is available or through regeneration when brakes are applied. Otherwise, the battery supplies power to the permanent magnet assisted synchronous reluctance motor while the LEV is cruising or accelerating. For the estimation of speed and hence rotor position for vector control, the sigmoid function-based sliding mode observer (SMO) technique is used. In the traditional SMO technique, there is a chattering problem because of the discontinuous control function. In this paper, this problem is addressed with a sigmoid function. The speed and position are estimated with the SMO, which is used to generate pulses for the three-phase voltage source inverter (VSI) through hysteresis current control. The presented system is simulated in MATLAB Simulink-based platform. The Vehicle dynamics and driving cycle techniques are used to analyze the performance of the overall system control. Simulated results showing the satisfactory performance of the BDDC, sigmoid function-based SMO and sensorless vector control technique for driving cycle-based operation of LEV.