The Semantic Recognition And Auxiliary Positioning System Of Automobile Fault

With the continuous increase of China's car ownership,the number of complaints about car failures is rising.However,because of the poor timeliness,high cost,low efficiency and other factors,the existing automobile fault diagnosis technology can not meet the needs of users.Therefore,it is of great theoretical significance and application value to realize a low cost and high efficiency method for automobile fault diagnosis.The development of natural language processing technology and deep learning technology makes it possible for computers to understand the natural language in the maintenance data and explore its potential value.In this paper,based on the maintenance data of 10 fault categories diagnosed and recorded by auto repair experts,the correlation between the description of automobile fault surface phenomena and the location of automobile fault is analyzed by natural language processing technology and deep learning technology.The expert's experience is parameterized,and an auxiliary positioning method of automobile fault is proposed,which can effectively shorten the diagnosis time for automobile repair personnel,improve the diagnosis efficiency and is easy to promote.First of all,because there is no relevant corpus in the field of automobile fault,this paper establishes the corpus in the field of automobile fault,so that the natural language processing technology can be better applied in the field of automobile fault.Secondly,based on the deep learning technology,the correlation between the description language of automobile fault phenomena and the parts of automobile fault is analyzed,and the automobile fault data set and the parameters of convolutional neural network model are optimized to obtain the automobile fault auxiliary positioning model with higher accuracy.After that,the technology of automobile fault feature extraction is studied,and the Jieba word segmentation and TF-IDF algorithm are optimized to make it more suitable for the application requirements in the field of automobile fault,in order to obtain more accurate results of automobile fault positioning.Finally,based on the above research results,this paper designs and implements an auxiliary positioning system for automobiles fault.Users only need to input the description of the phenomena of automobile faults to get three most likely fault locations.In addition,this paper develops a web-based user interface for the system based on Springboot framework and Vue framework,which is convenient for system promoting and user operating.The accuracy of the auxiliary positioning model of automobile fault studied in this paper is 89.56% on the test set.The recall rate of tf-idf algorithm based on automobile fault domain corpus for fault feature extraction reaches 95%.The research of feature extraction technology in the field of automobile fault makes the diagnosis accuracy of automobile fault prediction model reach 94.44% in the control test.And the accuracy of fault description language prediction without feature extraction is only 88.89%.The automobile fault positioning system studied in this paper provides an efficient,low-cost method for automobile fault auxiliary positioning,which can shorten the diagnosis process effectively.It provides a new idea for the combination of traditional technology and artificial intelligence technology.