Modified augmented hopfield neural network for optimal thermal unit commitment

This paper develops a novel solution methodology of the thermal unit commitment problem (TUCP) using modified augmented Hopfield network (AHN) with enhanced performance. The modifications are mandatory to eliminate the error that conventional AHN structure is reported to suffer from. This error originates from the mapping process, the corner stone in using AHN as an optimization tool. A new solution algorithm is developed by combining the AHN with the proposed modifications. In order to verify the effectiveness of the new algorithm, it is applied and tested to some examples reported in literature and the solution is then compared with that obtained by counterpart artificial intelligence (AI) techniques. Unlike other AI techniques, the solution obtained using the modified AHN is more optimal and satisfying all the operating constraints.