METHOD AND APPARATUS FOR PREDICTING SYSTEM FAILURE AND GRADING STATUS OF POWER DISTRIBUTION PANEL SYSTEM

摘要

PURPOSE: An apparatus and method for estimating a quality index of a distribution system are provided to construct an optimum artificial neural network model by directly using data measured in the distribution system to improve the accuracy of an artificial neural network model. CONSTITUTION: A distribution system(110) provides electricity to a building. A measurement data input unit(130) provides the measurement data of the distribution system to the input of an artificial neural circuit network. A measurement data memory(140) stores the measurement data of a prior timing to calculate differential data. A differential data generating unit(150) subtracts the measurement data of the prior timing from the measurement data of a current timing and provides a subtracted result to the input of the artificial neural network. An estimation data generating unit(160) calculates an estimation value of an environmental element of the distribution system and provides the estimation value to the input of the artificial neural network.