This paper demonstrates the control and operation of dual voltage source inverter (DVSI)-based distribution static compensator (DSTATCOM) in a three-phase three-wire distribution system. In this paper, a neural network-based control algorithm known as the Sparse Least Mean Square (SLMS) technique is employed to enhance the power quality (PQ) of the system. The proposed topology is built using a two-level two-voltage source inverter (VSI), where one inverter is marked as VSI-1 supported by distributed energy resource (DER) DC link. The other one is represented as VSI-2 supplied by self-supported capacitor DC link. The proposed system automatically senses the DER power generation and performs multi-operation such as PQ as well as optimal active power flow. The observations of the DVSI and VSI are carried out using an FPGA SPARTAN-6 controller-based experimental setup where the solutions are taken into consideration of the permissible standard grid code.