Research On Online Monitoring And Fault Diagnosis System Of Metal Band Saw

With the increasing demand of domestic metal band sawing machines,whether the metal band sawing machines can run safely and directly determines the benefit of metal processing enterprises.Therefore,the construction of real-time monitoring and fault diagnosis system for metal band sawing machines is an urgent need for industrial production.A GZK4232 single-guided metal band saw machine is selected as the research object.The on-line monitoring and fault diagnosis system of metal band saw machine is studied by using sensor to monitor several parameters of saw machine.The main tasks are as follows:Analysis of the fault types of metal band saw machine: the principle of single guide column metal band saw machine is studied,the fault information collected in the field is summarized,and the common faults of metal band saw machine are divided into five categories: abnormal sawing,broken or damaged saw blade,no action of limit switch,fault of hydraulic system and motor,and the causes of the failure are analyzed.On-line monitoring and fault diagnosis system design of metal band saw machine: in order to improve the fault diagnosis efficiency of metal band saw machine,the on-line monitoring and fault diagnosis system of metal band saw machine is constructed,which is mainly divided into two parts: data acquisition module of lower computer and monitoring module of upper computer.Data acquisition module of the lower computer transmits the collected speed and temperature data to the upper computer through the RS485.when the upper computer receives the data that exceeds the threshold,the system sends out acousto-optic alarm,and gives the fault diagnosis result,and sends the fault message to the staff through the TC35 module at the same time.An improved particle swarm optimization algorithm was introduced to optimize the BP neural network for fault diagnosis: classification and coding of common faults in metal band sawing machines by fault tree and binary coding analysis.Aiming at the defect that the BP neural network can not converge to the global minimum point,the improved particle swarm optimization algorithm is used to optimize the fault diagnosis strategy of BP neural network.the BP neural network and the improved BP neural network are simulated respectively.the fault diagnosis accuracy of the BP neural network is 84.6%,and the fault diagnosis accuracy of the improved BP neural network is 96.1%.Physical system platform test: the field test takes the GZK4232 single guide column type metal band saw machine as the research object,through the saw cutting Q235?45# and so on different material steel,measures the saw machine speed and after the work starts the saw machine oil temperature and the water temperature change situation.Simulates the saw machine fault,the upper computer monitors the module acousto-optic alarm and outputs the fault type and the reason,simultaneously sends the fault information to the staff.Experimental results and analysis: after field test,the error of saw cutting steel speed is 1.8%,the error of saw cutting Q235 steel speed is 2.1%,the error of oil temperature is 2.6%,and the error of water temperature is 1.4%.The system can detect the fault of saw blade breaking,too high water temperature and too high hydraulic oil temperature,and the correct rate of fault diagnosis is 93.3%.The on-line monitoring and fault diagnosis function of metal band saw machine is realized.The application of the on-line monitoring system and intelligent fault diagnosis technology on the metal band sawing machine solves the difficulty for the staff to obtain effective data to analyze the cause of the fault after the sawing machine fails,and also improves the speed of troubleshooting and the degree of automation of the sawing machine.