Power system computing with neural networks is one of the fastest growing fields in the history of power system engineering. Since 1988, a considerable amount of work has been done in investigating computing capabilities of neural networks and understanding their relevance to providing efficient solutions for outstanding complex problems of the electric power industry. A principal objective of a power utility is to provide electric energy to its customers in a secure, reliable and economic manner. Toward this aim, utility personnel are engaged in a variety of activities in areas of supervisory control and monitoring, evaluation of operating conditions, operation planning and scheduling, system development, equipment testing, etc. Over the past decades significant advances have been made in the development of new concepts, design of hardware and software systems, and implementation of solid-state devices which all contributed to the steadily improving power system performance that we are experiencing today. Advanced information processing technologies played an important role in these development efforts. Members of the Special Interest Group for Power Engineering of the INNS recognized the need for bringing together leading researchers in the field of neurocomputing with experts from power utilities and manufacturing companies to assess the current state of affairs and to explore the directions of further research and practice. This book is based on The Summer Workshop on Neural Network Computing for the Electric Power Industry which brought together approximately forty specialists with backgrounds in power engineering, system operation and planning, neural network theory and AI systems design. An informal and highly inspiring atmosphere of the workshop facilitated open discussion and exchange of expertise between the participants.Performance improvements over manual designs have been observed, the interplay between performance criteria ... Such important aspects of neural network applications as generalization, learning speed, connectivity and tolerance to network damage are strongly related to the choice of ... Goldberg, 1989) for synthesizing neural network architectures (Harp, Samad aamp; Guha, 1989, 1990; Harp aamp; Samad, anbsp;...
|Title||:||Neural Network Computing for the Electric Power Industry|
|Author||:||Dejan J. Sobajic|
|Publisher||:||Psychology Press - 2013-06-17|