P. Somasundaram, R. Manickavasagam, G. Sadasivam, K. Kuppusamy, and R.P. Kumidini Devi (India)
Voltage control, Artificial neural networks, coordinatedactivity, loss minimization.1.
The secondary voltage control for the generators implemented using Optimal Power Flow (OPF) program from a centralized point (eg. Energy Control Centre) presents many difficulties for real-time environment. This paper presents a highly decentralised secondary voltage controller using Artificial Neural Networks (ANN). The controller is built using feed forward networks with error back propagation learning and trained with the data obtained from off-line OPF. The only inputs to the controller are generator bus voltage and reactive power flows out of the lines connected to the generator. The output from the controller is the required reference voltage which is supplied to the Automatic Voltage Regulator (AVR). This method is highly attractive for on line applications as the controllers are fast, robust and act in a coordinated way. The test results of the proposed controller on two standard test systems namely Modified IEEE 30-bus system and IEEE 57-bus system were very encouraging and their performance was highly comparable to that of OPF.
Important Links:
Go Back
