Fault prediction method based on ARMret model considering rare variables

Abstract

The invention discloses a fault prediction method based on an ARMret model considering rare variables. The fault prediction method comprises the steps of 1, mining rare elements and common elements and classifying the rare elements and the common elements; 2, mining HILP elements based on the rare elements, and reflecting the HILP elements in the form of a high-frequency variable set and a frequent association rule, the HILP referring to high risk and low probability; 3, for each environment feature in the training data set, sequentially repeating the steps 1-2; 4, solving the relative weight of each element; 5, on the basis of the relative weight of each element obtained in the step 4, calculating the predicted fault risk degree of the corresponding record according to the environment element contained in each fault record of the test data set and carrying out normalization; and 6, comparing the predicted fault risk degree with a real fault processing result correspondingly recorded in the test set, thereby evaluating the performance of the proposed prediction model. The fault prediction method is high in prediction accuracy and easy to implement.

Description

technical field [0001] The invention relates to an association rule mining fault distribution prediction model (Association Rule Mining with Rare Elements and Time series, ARMret) based on rare variables. Background technique [0002] In the external environmental characteristics of each transmission line system and transmission line system fault occurrence periods studied, there are often some environmental elements with low frequency and some fault occurrence periods. However, when faced with different environmental elements and different time periods in the environmental characteristics, the traditional ARM algorithm still uses the same and fixed importance diagnostic standard score calculation method and threshold setting method, which will result in rare environmental elements and rare fault time periods. Rare variables were discarded without any analysis. Considering that there is also a correlation between these rare variables and real faults, mining and analyzing th...

AI Technical Summary

Problems solved by technology

However, when faced with different environmental elements and different time periods in the environmental characteristics, the traditional ARM algorithm still uses the same and fixed importance diagnostic standard score calculation method and threshold setting method, which will result in rare environmental elements and rare fault time periods. of rare variables were discarded without any analysis

Images