Research On Fault Diagnosis Method Of Railway Signal Equipment Based On Association Rules

The railway has always been the backbone of the modern comprehensive transportation system.It is an important infrastructure and the main means of transportation for public travel in our country.It is also an important industrial sector that promotes the national economy and promotes social development.In recent years,with the rapid development of China’s railway,the train speed continues to increase,the train density continues to increase,and the safety requirements of railway transportation become increasingly strict.Railway signal equipment plays an important role in ensuring the safety of trains and improving transportation efficiency,therefore,it is particularly important to ensure its normal operation.In daily maintenance and repair,maintenance personnel mainly rely on manual experience and describe the processing flow of faulty equipment in the form of words,so a large amount of fault text data has been accumulated.These rich text data contain huge information,which is of great significance to the analysis of railway signal equipment failures.In order to improve the efficiency of fault diagnosis of railway signal equipment,based on the analysis of fault text information,a fault diagnosis method of railway signal equipment based on association rules is proposed.The corresponding relationship among fault characteristics,types and causes is established to realize intelligent diagnosis of railway signal equipment.The specific research content is as follows:(1)This thesis analyzes the characteristics of the maintenance records of railway signal equipment,designs the fault diagnosis scheme of railway signal equipment combined with intelligent decision support system,and determines to use the production representation method to express the fault maintenance knowledge.(2)Fault feature extraction of railway signal equipment based on improved TF-IDF(Term Frequency Inverse Document Frequency)algorithm.Aiming at the specialty of railway field,the professional lexicon of railway signal equipment is constructed.Considering the influence of synonyms on feature word extraction,the weight formula is modified according to synonym weight rules,and the weights are discretized to realize fault feature extraction.(3)Fault diagnosis based on improved FP-Growth(Frequent Pattern Growth)algorithm.The improved FP-Growth algorithm is used to dig out the regular information among the fault characteristics,types and causes,realize knowledge acquisition,and store the knowledge in the fault knowledge base to provide a basis for fault diagnosis.Firstly,according to the characteristics of the data,an adaptive strategy is adopted to set the minimum support number and confidence level to reduce the user’s subjectivity in the value of algorithm parameters.Secondly,divide the database sub-library of each item according to the frequent 1-item set,and then construct the conditional FP-Tree of each item,which reduces the memory footprint and improves the running speed.Finally,save the association rules as knowledge in the knowledge base,and perform fault diagnosis through the inference engine in the intelligent decision support system.Experiments show that the running time of this method is better than the traditional FP-Growth algorithm,and the average diagnosis accuracy is 13.02% and9.28% higher than the case-based reasoning algorithm and Bayesian networks algorithm.(4)Based on the above-mentioned research on the key technology of fault diagnosis,design and implement the railway signal equipment fault repair prototype system,analyze and implement the main functional modules according to the system requirements,and prove the feasibility and effectiveness of the method in this thesis.