Font Size: a A A

The Research Of Screw Pump Fault Diagnosis System Based On Information Fusion Technology

Posted on:2019-05-11Degree:MasterType:Thesis
Country:ChinaCandidate:H LiFull Text:PDF
GTID:2371330548976490Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
Emulsion explosive belongs to a new type of industrial explosives.It is widely used and has made great contributions to economic construction.In the production of emulsion explosives,due to various reasons,often lead to equipment failure,resulting in major accidents.The screw pump is one of the most important and troublesome devices in the production line.Therefore,it is of great significance to study a properly designed and accurate screw pump fault diagnosis system.In this paper,for the first time,the information fusion technology is applied to the fault diagnosis of the screw pump in the emulsion explosive production line,and the screw pump fault diagnosis system is designed.This article mainly carried on the following research:(1)On the basis of studying the production process of emulsion explosives,starting with the structure of screw pump,the mechanical failure and the plugging failure are mainly studied.The characteristics and mechanism of the two kinds of faults are deeply studied.The screw The overall implementation of the pump diagnostic system.(2)Focus on the information fusion technology,combined with the actual situation of the screw pump to determine the selection of feature-decision-making layer of information fusion structure model,based on information fusion technology fault diagnosis model.(3)The BP neural network is used to make up for the shortcoming of D-S evidence theory,and the combination of BP neural network and D-S evidence theory is selected to realize the fault diagnosis of the screw pump.Select modeling variables,the establishment of screw pump fault diagnosis system model.The state information of the screw pump is collected by the sensor and sent to the BP neural network for training.The output is normalized and then sent to the decision-making D-S evidence theory for evidence probability fusion to obtain the fusion fault support.The experimental results show that the fusion of multi-fault features can identify faults more accurately and lead to more reliable diagnosis conclusions.(4)In accordance with the requirements of the production of emulsion explosives,select the appropriate hardware and equipment to build a hardware platform for the diagnosis system;on this basis,put forward the system software design structure,and finally the software design of the diagnosis system,log in to the diagnosis System,the online monitoring of the operation of the screw pump and the operation of the production line,the screw pump failure to conduct a reliable analysis.Finally,the simulation test is carried out on the diagnosis system.The results show that the diagnosis system can effectively diagnose the faults,have better fault diagnosis ability,and can give reasonable suggestions.It can effectively guarantee the safe operation of the screw pump and reduce the accident rate.
Keywords/Search Tags:Screw pump, fault diagnosis, D-S evidence theory, BP neural network
PDF Full Text Request
Related items