The varieties of energy structures that afford large amounts of energy to the global power economy has led to a reduced quality in the power distribution network (PDN). Most traditional methods use direct coupled distributed static compensator (DC-DSTATCOM) which is dependent on specific line parameters. Hence, it performs poor adaptability to the PDN. Aim at enhancing power quality (PQ); this study focuses on the deep belief learning network (DBLN) controlled inductively coupled distributed static compensator (IC-DSTATCOM). Considering the system variation and impact, an individual DBLN structure is accumulated for each and every phase. The weight obtained from the DBLN mechanism is caused for reference currents generation. The filtering performance of the IC-DSTATCOM is revealed better by combining the effect of the transformer and DSTATCOM impedance. So, the IC-DSTATCOM is augmented with better dynamic performance with PQ abilities as compared to DC-DSTATCOM in terms of shunt compensation such as, THD reduction, power factor (p.f.) improvement, better voltage regulation and load balancing etc. The IEEE – 2030 – 7 – 2017 and IEC – 61000 – 1 grid code are used to evaluate the effectiveness as per the benchmark value using MATLAB/Simulink