Research And Development Of Roots Fan Fault Diagnosis System Based On Case-based Reasoning

As one of the main power transmission equipment of the modern machinery industry,rotating machinery has been widely used in many important industrial fields.Roots blower is a positive displacement fan in rotating machinery.It has many advantages and is widely used.However,due to the harsh environment of the equipment and the interaction of various random factors,rotor unbalance,Shaft misalignment,bearing damage and other types of failures,its operating status not only affects the operation of the equipment itself,but also hinders the subsequent production on the industrial site.Therefore,the research on the fault diagnosis of Roots fans is particularly important,and it also forces people New requirements and technologies are constantly being introduced in the fault diagnosis of mechanical equipment.At present,most of the diagnosis of the operation process of Roots fan mechanical equipment still uses periodic inspections and portable instrument detection.It is difficult to achieve realtime accurate diagnosis.Therefore,the development of a fault diagnosis system for the operation process of Roots fan mechanical equipment has Certain practical significance.This article combines the historical operation data of a domestic aluminum plant to carry out an in-depth study of the Roots blower mechanical equipment fault diagnosis system.Its main work is as follows:(1)Establish a fault diagnosis model.Through field investigation,the failure categories of Roots fan mechanical equipment are summarized and the characteristic variables affecting the failure are analyzed respectively.On this basis,case reasoning and mutual information methods are combined and applied to the fault diagnosis modeling of Roots fan operation process In the first,the mutual information value between the characteristic variable and the operating state category is calculated by the mutual information method,so as to realize the selection of the characteristic variable and the reasonable distribution of the weights.Then follow the steps of case retrieval,reuse,correction and storage to implement the fault diagnosis algorithm based on case reasoning,and give the corresponding algorithm steps.Finally,it is compared with other classic algorithms through the UCI data set.The results show that the mutual information method can effectively delete redundant features and allocate feature weights reasonably,improving the overall performance of the diagnostic model;(2)Roots fan fault diagnosis system development.On the basis of the above fault diagnosis model and algorithm implementation,combined with the operation process of Roots fan mechanical equipment,using C language and Automation Studio configuration software to develop a set of fault diagnosis system for the operation process of Roots fan,mainly including related man-machine The development of the interface and the preparation of the corresponding algorithm program,so as to realize the functions of operation monitoring,status warning,parameter dynamic display,curve drawing,historical data query and fault diagnosis;(3)Experimental study.Using the historical operation data of a domestic aluminum plant,the fault diagnosis system was tested online and offline.The experimental results show that the fault diagnosis system’s accuracy rate,false alarm rate,false alarm rate and other related performance indicators are better than other diagnostic methods.The screen is intuitive and concise,and the operation is convenient and fast.It can monitor the running status of Roots blower mechanical equipment in real time and quickly and accurately give diagnostic results.It gives maintenance suggestions in the early stage of the failure,to a certain extent,it can avoid workers from discovering the failure caused by regular inspection Timely problems improve the work efficiency on site and have certain practical value.