Performance evaluation of DSTATCOM with three-phase T-type three-level inverter using anti-Hebbian based LMS control

Abstract In this paper, a three-phase T-type Three Level (TTL) inverter with Distributed Static Compensator (DSTATCOM) using Anti-Hebbian based Least Mean Square (AHLMS) control approach is proposed to alleviate the power quality issues. The design and modeling of a controller for the DSTATCOM device is constrained by the evaluation of shunt compensation and associated power quality issues. In this work, the AHLMS control approach is presented to control the DSTATCOM and its performance is evaluated under various loading conditions. The AHLMS is an upgraded variant of the Hebbian Least Mean Square (HLMS) control approach that employs an appropriate pattern of learning mechanism. Each controller is constructed as an individual neural network structure based on the system’s flexibility and experimental use. In order to produce reference source currents, the control approach seeks to isolate necessary active and reactive components from load currents. The AHLMS control approach outperforms in terms of voltage balancing, enhancing power factor correction, harmonic reduction, and voltage regulation of TTL inverter when there are disturbances triggered by even and uneven loading situations. The simulation model for the three-phase TTL inverter based DSTATCOM configuration is verified by MATLAB/Simulink. The performance of three-phase TTL inverter based DSTATCOM using AHLMS approach is analyzed under both uniformity and diversity loading conditions. Furthermore, the TTL inverter results are contrasted with that of the 2-level inverter with respect to DC link voltage, oscillations, and current harmonics at both the source and load sides. The TTL inverter DC link capacitor oscillates from 530 to 532 V with minimal oscillations compared to 2-level inverters under even loading. Overall, by using AHLMS control approach, the %THD of TTL inverter is obtained as 3.28%, which is less than the allowed %THD of 5%.