Integration of Numerical Methods in a Hybrid Fuzzy Knowledge-based System for Multiobjective Optimization of Power Distribution System Operations

R.J. Sárfi and A.M.G. Solo (USA)


Distribution system operations, optimization, numerical methods, fuzzy logic, fuzzy knowledge base, knowledge engineering


Knowledge-based and numerical methods both have ad vantages in performance optimization of power distribu tion system operations. To overcome the shortcomings of the individual methods, a hybrid fuzzy knowledge-based system was developed that employs both knowledge-based and numerical methods for multiobjective optimization of power distribution system operations. While complying with network constraints, the power distribution network is optimized for system loss reduction, transformer load balancing, reduction of transformer aging, maintenance of a satisfactory voltage profile throughout the network, reac tive power compensation, and conservative voltage reduc tion (CVR) practice. This research paper focuses on the integration of numerical methods in the hybrid fuzzy knowledge-based system. A transformer aging routine is used by heuristics for transformer aging due to temporary transformer overloading and for daily transformer aging. A voltage update routine is used by heuristics for mini mum voltage requirements and CVR. Fuzzy sets are de fined for the fuzzy antecedents used by these rules and a standardized degree of desirability is defined using fuzzy sets to express the fuzzy consequent. Extensive simula tions found that the hybrid fuzzy knowledge-based system significantly enhanced performance, achieved all objec tives in a time-efficient manner, and produced significant monetary savings.

Important Links:

Go Back