Health monitoring method of transmission tower based on association rule algorithm

Abstract

In order to avoid the influence of environmental factors on the accuracy of abnormal monitoring of transmission tower (TT) and improve the applicability of health monitoring of TT, the health monitoring method of TT based on association rule algorithm is studied. The sensors of temperature, wind speed, wind direction and wind deflection angle are selected as the health monitoring nodes of TTs. The fusion values of abnormal monitoring points of TTs are obtained by batch estimation algorithm. The fused data of TTs are used as the candidate item set of association rule algorithm. According to the minimum support and confidence, the candidate item set of abnormal mode records of TTs is scanned, The frequent itemsets of abnormal patterns of TTs are found, and the association rules between abnormal patterns and abnormal causes in frequent itemsets are generated to realize the health monitoring of TTs. The test results show that the data fusion method can effectively avoid the impact of the instability of data acquisition of each monitoring node. When the minimum confidence and support are 0.6 and 0.7 respectively, the monitoring accuracy of this method is better, almost close to 1. In different environments, the monitoring accuracy of TT is higher than 0.9, which has strong applicability and high monitoring accuracy.