Predicting Failures of High Voltage Electric Power Facilities via Multidimensional Information Analysis

Efficient maintenance of the power grid is of great importance. This paper presents a multi-dimensional public information analysis based high voltage electric power facilities failures prediction mechanism, which aims at finding the critical facility. This approach helps finding failures, which will cause a lot of loses, in time and even in advance. We first summarize the factors that may cause failures on power facility. We then present our models for simulating the effects of these factors separately. A system that could highlight suspicious locations for power facilities routing inspection is demonstrated. Several experiments have been carried out to show the effectiveness of our approach. Our system as well as source codes are released at http://www.y-droid.com/FailurePreditionHV.