A.T. Adediran (Nigeria) and A. Domijan (USA)
System restoration, system identification, FRIENDS, and Modeling.
Restoration schemes based on optimal switching of circuit breakers, relays and fuses are very often used in present day utility systems. A very pronounced problem often expected and experienced is how to predict which restoration plan is applicable at any given time, load condition, fault situation and system state. In contribution to novel ideas for solving the problem of efficient and fast restoration, an online expert system based optimized switching scheme for a flexible reliable and intelligent electric energy delivery systems (FRIENDS) network is proposed. System identification method is used in the state recognition phase to determine the new condition of the monitored system and expert system is used to automatically propose and affect a corrective switching scheme. A system identification method suitable for our FRIENDS network (the Auto Regressive Moving Average with Exogenous input – ARMAX) was selected to determine the system gain at a given point on the network for a given fault situation and an expert system was used to interpret this data and propose automatically an optimal switching scheme output command from the expert system. Generation of optimal switching schemes for many situations were carried out and stored and logic was utilized to determine optimal switching from the information in the output of the system identification given available expert system output commands. The proposed system reduces response time for restoration, increases system operation efficiency, requires less feedback from sensors and protective devices and reduces manpower requirements at the Quality Control Center (QCC).
Important Links:
Go Back
