Model predictive control with technique for order of preference by similarity to ideal solution method for four leg distribution static compensator to improve power quality and reduce switching frequency

In this paper, model predictive control (MPC) of four leg distribution static compensator (FL-DSTATCOM) is proposed to compensate the power quality issues and to reduce the higher neutral leg switching frequency of FL-DSTATCOM. Current control using conventional MPC, selects a switching state which will minimize the single constraint cost function, comprising the difference between reference and actual DSTATCOM currents. However, FL-DSTATCOM has a limitation of higher neutral leg switching frequency, and it can be reduced by adding an additional constraint to the cost function using a weighting factor. The reduction in neutral leg switching frequency also reduces the switching frequency of phase leg switches. But, the selection of weighting factor during multiple constraint case is a cumbersome task. Therefore, to simplify the weighting factor selection process, an optimization technique, namely, the technique for order of preference by similarity to ideal solution (TOPSIS) is applied, which further selects the optimal switching state. At the same time, compensation using FL-DSTATCOM mainly depends on reference current extraction and it is a challenging task during unbalance and distortions in supply voltages. To address this problem, an improved conductance factor-based approach is implemented in this paper, which makes reference current extraction simpler, effective and easy to interpret. The efficacy of the proposed method is evaluated using simulation and experimental studies.