Research And Implementation Of Equipment Fault Diagnosis Reasoning System Based On Cloud Platform

The fault diagnosis and maintenance of equipment is very important in today's rapid development of Internet technology.Equipment maintenance often requires experts to check and maintain the equipment according to past experience.The traditional method is through manual diagnosis,which has some problems such as regional limitation,high cost,waste of manpower and so on.Therefore,it is very necessary to research on remote fault diagnosis reasoning of equipment.The core idea of equipment fault diagnosis reasoning is to use fault description documents and fault knowledge bases to effectively carry out equipment fault diagnosis reasoning through the combination of text classification and knowledge graph.The dissertation mainly focuses on the key issues of equipment fault description information classification model,similarity comparison method,knowledge graph and so on.Aiming at the problem of low intelligence level of fault diagnosis,a reasoning method of equipment fault diagnosis based on cloud platform is proposed.Aiming at the difficult problem of complex unstructured fault information processing,an improved fault diagnosis classification model is proposed.After the fault information is segmented,the fault keyword formed by TF-IDF is fused with Word2 vec to construct the original keyword vector.After secondary vectorization,the fusion fault information model is used to construct the phrase word vector as the input of Text CNN classification model to obtain the fault classification information.Aiming at the distributed complex fault knowledge,the knowledge graph is introduced into the field of equipment fault diagnosis,and the classified entities and relationships are constructed into a graph database.The similarity measurement of the fault phenomenon uses the cosine similarity method.Finally,according to the similar phenomena obtained,the cause of the fault is found in the knowledge graph,and the final solution is obtained.The dissertation proposes a fault diagnosis and reasoning method based on text classification technology,and deploys the development system on the cloud platform.In order to verify whether the classification diagnosis method is reasonable and effective,a set of comparative experiments is designed for verification.It can be seen from the experimental results that in the same failure data set,the proposed method is higher than other classification methods in terms of Precision,Recall and F1 value.It shows that the equipment fault diagnosis and reasoning method in the dissertation can provide the diagnosis and solution method of the fault location in the equipment maintenance process,and improve the efficiency and accuracy of the reasoning result.