A practical neural network approach for power generation automation

Abstract

This paper presents a practical artificial neural network (ANN) based technique for the automation of power generation scheduling based on the consumer's load profile. A multi-layered neural network with backpropagation learning algorithm is used to predict the required power generation to fulfill the consumer's demands. The proposed technique has been applied to a typical co-generation power plant of 4/spl times/8 MW rating. Test results indicates that the ANN model can automatically perform generator scheduling accurately.