In recent years for reactive power compensation an inverter based conditioners have been used due to their faster response. To improve the quality of power in a distribution system, Distribution Static Synchronous compensator (DSTATCOM) which is a inverter based device which is been used broadly. To control these kind of devices, Proportional-Integral (PI) controller is been used with certain prespecified fixed parameters. Nowadays these kind of controllers performance are not up to the expectation due to the nonlinearity of the system .In this paper, the (Artificial Neural Network) ANN controller is described with the use of Fuzzy C-Means clustering (FCM) algorithm to generate appropriate weight to control direct and quadrature axes currents of DSTATCOMs. Simulations on wide range of processes are carried out by using MATLAB/Simulink software and responses are observed by changing the reference reactive current. The results are compared between the controllers in terms of several performance measures, in which the DSTATCOM improves the damping of a power system by proposed schemes.
Radial Basis Function Neural Network , Fuzzy C-Means Clustering, Proportional-Integral,
Cite this paper
P.Kanirajan, M.Joly. (2019) Modelling of DSTATCOM for Improving Power Quality using ANN Controller by Fuzzy C-Means Clustering. International Journal of Power Systems, 4, 70-76