Study of Fault Detection Method based on Association Rules
LU Hui-gu · Journal of Chengdu University of Information Technology · 2014
To improve the Automatic Weather Station equipment operation ability and the observation data quality condition is the key to affect the accuracy of weather forecasting service and the development of meteorological service.However,the Automatic Weather Station equipment in our country is running daily without build in fault detection device and perfect self-checking system.But the changes of meteorological elements(temperature,humidity,pressure,wind and rain)are often interrelated.In order to improve the quality of Automatic Weather Station data,should judge the equipment failure in a timely and effective manner.This paper research the test results of a large number of observation data and then put forward a comprehensive fault diagnosis method based on association rules.By using the Apriori algorithm,this paper obtains the strong association rules between the various meteorological elements and its variation.Through actual observation data validate that at 8am of July 26,2012 Jiangxi NanChang Automatic Weather Station humidity sensor failure.