Neurofuzzy system with GA-based algorithm for knowledge management in network planning

• Published in: TENCON 2004 : 2004 IEEE Region 10 Conference : proceedings : analog and digital techniques in electrical engineering : 21-24 November, 2004, Chiang Mai, Thailand Vol. D; pp. 641 - 644 Vol. 4
• Main Authors: Sangpetch, T., Lo, K.L.
• Format: Conference Proceeding
• Language: English
• Published: Piscataway NJ IEEE 2004
• Subjects: Applied sciences; Artificial intelligence; Computer science; control theory; systems; Connectionism. Neural networks; Electric, optical and optoelectronic circuits; Electrical engineering. Electrical power engineering; Electrical power engineering; Electricity supply industry; Electronics; Exact sciences and technology; Fuzzy systems; Knowledge management; Miscellaneous; Neural networks; Operation. Load control. Reliability; Power networks and lines; Power system dynamics; Power system economics; Power system modeling; Power system planning; Power system reliability; Power system security; Recurrent neural networks; Recurrent neural nets; Fuzzy system; Telecommunication network planning; Decision support system; Evolutionary algorithm; Power system economics; Optimization method; Inference; Knowledge base; Network management; Modeling; Management planning; Optimization; Electrical network; Genetic algorithm; Reinforcement; Electricity supply industry; Power markets; Power system reliability; Objective function; Safety; Dynamic model; Mathematical model
• ISBN: 0780385608; 9780780385603
• DOI: 10.1109/TENCON.2004.1415014

This paper presents the necessity of power system reinforcements in today's electricity supply industry. The principal outcome of the paper is two-fold. First, the paper offers a dynamic market modeling that primarily considers power system economics, reliability, security, and operation constraints. The network planning model is formulated by using stochastic modeling. Afterward neurofuzzy system with GA-based algorithm is proposed as a methodological approach. The fuzzy inference system is also presented. It mainly prioritizes all power system constraints throughout the model and evaluates a proper decision making towards the fuzzy if-then rules. Within the paper moreover, genetic algorithm (GA) is applied for optimizing the objective function whilst the recurrent neural network (RNN) technique is chosen for knowledge acquisition by training and learning all data under different power system control and operation scenarios. Eventually, some conclusions on whether or not the network reinforcements are necessarily required within a competitive marketplace are drawn.