ANN based control of Statcom for improving voltage profile in power system

This paper presents neural networks based approach for estimation of the control and operating parameters of Statcom used for improving voltage profile in a power system, which is emerging as a major problem in the day-to-day operation of stressed power systems. Statcom is an important voltage source converter FACTS device, which can be used in voltage control mode or reactive power injection mode. For stable operation and control of power systems it is essential to provide real time solution to the operator in energy control centers. Artificial neural networks are proposed here for this task as they have ability to synthesize complex mappings accurately and rapidly. Two multi layer feed-forward neural networks are developed to estimate the control/ operating parameters of statcom used for improving voltage profile at various loading condition of power system. To reduce the neural networks training time, the two ANNs have been developed simultaneously using parallel computing. The effectiveness of the proposed method is demonstrated on a benchmark 5-bus system. The results obtained clearly indicate the superiority of the proposed approach in terms of accuracy and speed.