Compressor is a kind of machinery that relies on compressed gas to power the system. It is widely used and widely used in metallurgy, mining, building materials, aerospace and other industries, especially in petroleum and chemical industries. Less critical equipment. In the event of a failure of the compressor, the entire production line will be shut down, causing great property damage and even personal safety. Therefore, it is necessary to conduct on-line monitoring and fault diagnosis of the operating conditions of the compressor equipment, and timely discover and eliminate faults to ensure smooth production. At present, most of the maintenance methods used for compressors are fault repair and regular maintenance. The former has hysteresis, and the latter has great blindness. In this paper, based on the working characteristics of the screw compressor, NI Labview software is used as the development platform. It adopts online real-time monitoring and over-limit alarm for each static parameter. It uses the new technology of modern signal analysis for bearing vibration signal and uses power for vibration signal. Spectral analysis, correlation analysis, trend analysis, signal envelope extraction analysis based on EMD and wavelet packet combination and intelligent diagnostic analysis technology based on vector machine, based on this, a virtual instrument based compressor fault diagnosis system is built. 2 Fault diagnosis system overall structure The system is divided into two parts: signal acquisition and signal analysis. The signal acquisition part uses the pressure, temperature and vibration sensors to collect the measured parameters. After the signal is conditioned, it is sent to the data acquisition card for collection. The data collected by the acquisition card is sent to the computer for analysis. The measured parameters have static and dynamic quantities, and the static parameters have : Stage intake air temperature, stage exhaust temperature, stage intake air temperature, stage exhaust temperature, cooling water inlet temperature, cooling water temperature, cooling water outlet temperature, lubricating oil temperature, tank cover temperature, stage intake pressure, The stage exhaust pressure, the stage intake pressure and the stage exhaust pressure, the dynamic quantity parameter is the vibration signal of the bearing; the signal analysis part is a Labview monitoring and diagnosis platform, real-time display and monitoring of each static quantity, and multi-method analysis of the vibration signal And processing. The overall structure of the system is as shown. 3 software structure and each functional module 3 1 software overall structure The entire monitoring system is divided into three layers: initial main interface layer, parameter setting layer and monitoring and diagnosis layer. The parameter setting layer is used to set the unit number and monitoring parameters, and the monitoring and diagnosis layer analyzes and diagnoses the collected signals. The overall structure is as shown. 3 2 Interfaces and Function Modules The programs developed in Labview are called VIs (V iew In strum ent). All VIs include the Front panel and the Block diagram. 3 2 1 Parameter setting subVI Click on the initial interface to enter the button, the system enters the monitoring parameter setting interface, and sets the monitoring unit number, temperature and pressure monitoring parameters. The program establishes a global variable to save and pass the settings. parameter. The front panel is as shown. Parameter setting interface 3 2 2 Monitoring diagnostic sub-VI is divided into four functional areas on the front panel: static quantity parameter monitoring area, bearing real-time vibration signal display area, power spectrum and related analysis display area and function menu area. The static quantity monitoring area can display and monitor each parameter in real time, and can alarm when the monitored object exceeds the set value; the real-time display area of ​​the bearing vibration signal can display the bearing vibration signal of the two measuring points in real time; the power spectrum and the related analysis display area The power spectrum and related analysis results of the bearing vibration signal can be displayed; the menu function area is various analysis methods and methods for bearing vibration signals, mainly including: power spectrum analysis, correlation analysis, trend analysis, EMD/HS based and wavelet packet Combined signal envelope extraction analysis and vector machine based intelligent fault diagnosis analysis of bearing faults. The dynamic quantity analysis of the compressor of the system is mainly for the vibration signal of the bearing, because according to the research, most of the failure of the compressor can be reflected by the vibration signal of the bearing. The trend analysis mainly uses the wavelet analysis method to analyze the bearing vibration signal, and extracts the trend item of the vibration signal, which is used to reflect the trend of the working state of the compressor and judge whether there is a fault. The signal envelope analysis based on the combination of EMD/HS and wavelet packet is considering that in the actual situation, the measured bearing signal may be a modulated signal. There is a big defect in the traditional Fourier analysis method. Here, EMD /HS is used. The method decomposes the vibration signal into several IMF components, and extracts the signal envelope by using the Hilbert transform. However, due to the end effect in the EMD decomposition process, it is easy to generate false components. Here, the wavelet packet is used to denoise the signal in advance. And then carry out EMD / HS analysis, experiments show that this method is effective. Support Sector Machine (SVM) is a learning method proposed by V apnik et al. based on statistical theory. It shows unique advantages and good application prospects in solving small sample problems of pattern recognition. , has been used in areas such as pattern recognition and feature extraction. The SVM can train the fault classifier with a small amount of time domain fault data samples, and can perform multi-fault identification and diagnosis without signal pre-processing to extract feature quantities. These several analysis methods are respectively programmed into function subVIs, which are dynamically called when the monitoring and diagnostic interface responds to menu operations. 4 The simulation test verifies the feasibility of the test system by setting the analog signal. The test results are shown in 5 and 6. The static amplitude and bearing real-time vibration signal monitoring results are displayed. It can be seen that when a certain static quantity parameter exceeds the set value, the value will flash continuously to achieve the alarm purpose; the display is the menu function in the bearing The vibration signal power spectrum and related analysis results; the results of the analysis of the bearing vibration signal vector machine in the menu function are displayed. It can be seen that the monitoring and diagnosis system can effectively perform real-time monitoring and fault diagnosis on the compressor state, and the design interface is friendly, and the diagnosis method is feasible. 5 Conclusion Labview is a graphical programming language and development environment. It is powerful and flexible to build data acquisition, data analysis and data display storage systems. It is flexible and intuitive, which can greatly reduce system design and development time. The monitoring system developed in this paper can monitor the working state of the compressor in real time, diagnose the state trend and bearing fault and judge the fault. By further improving the system, it can be applied to the condition monitoring and fault diagnosis of the actual equipment. Oil Pressure Sensor,Auto Sensors,Engine Oil Pressure Sensor,Oil Pressure Sensor For Car Wenzhou Yuanda Electronic Technology Co., Ltd , https://www.wzyddz.com