1Common fault diagnosis method Mechanical equipment fault diagnosis technology is a multi-disciplinary and comprehensive practical technology based on machine science and information science, system science, artificial intelligence and computer technology. It is based on computer, sensor, signal analysis and processing. Through equipment state detection and fault diagnosis, it can qualitatively and quantitatively grasp the equipment operating state parameters, predict equipment reliability, and identify and determine the cause, location and severity of the fault. And decision making.
1.1 Wear Residue Analysis Diagnostics Wear failure is one of the most common and major failure modes of gearbox gears. When tooth surface wear occurs during operation, the worn residue can be found in the lubricating oil. The so-called wear residue analysis method is to quickly obtain information about machine failure by measuring the residual content of mechanical parts wear residues in the lubricating oil. At present, there are three methods for measuring: one is to directly check the residue, and to measure the change of capacitance or inductance in the oil film gap, and the turbidity change of the lubricating oil to quickly obtain the information of the part failure. The second type is the collection of residues, such as magnetic probes, special filters, etc., collecting large pieces of exfoliated particles caused by fatigue of working surfaces such as gears and rolling bearings. The third method is oil sample analysis.
Oil analysis technology plays an important role in studying the location and process of mechanical wear, the type of wear failure, the mechanism of wear, and the evaluation of oil. It monitors and faults mechanical equipment without stopping or disassembling. An important means of diagnosis. The oil analysis technology is divided into two categories: (1) analysis of the physical and chemical properties of the oil itself, that is, the comprehensive analysis and inspection instrument for lubricating oil, which mainly affects the viscosity, acid value, moisture and insoluble matter of the lubricating oil in use. Physical and chemical indicators for analysis; (2) Analytical techniques for insoluble materials in oil, also known as wear debris monitoring technology.
The method of detecting debris is spectroscopic analysis and iron spectrum analysis. At present, the commonly used method is the iron spectrum analysis method. The iron spectrum analysis uses a high gradient strong magnetic field to separate the metal magnetic particles in the oil to make a spectrum. The composition, morphology and size of the abrasive chips are observed under a microscope, and the grinding is performed by a densitometer. The relative amount of shavings. It can judge the wear condition and predict the failure of the parts. This method can separate the wear debris of a wide range of lubricating oil, and has a wide application range. Due to the use of a special iron spectrometer, (direct reading type spectrometer, analytical type ferrography The instrument can qualitatively and quantitatively measure the degree of wear, and the reliability of the diagnosis is high. However, it has low monitoring ability for non-ferrous particles in the lubricating oil, and the required equipment is complicated, and needs to be completed by experienced professionals.
It is more effective to use the wear residue fault analysis technique to detect wear-type faults in the diagnostic transmission. Although the vibration diagnosis method can diagnose the wear-related fault to a certain extent, it has great limitations.
This method is therefore a powerful tool for determining the wear and tear of the transmission.
1.2 Vibration Monitoring Technology Diagnostics The most widely used diagnostic technology is the mechanical vibration signal. The reason is that the mechanical damage caused by vibration is high. According to statistics, mechanical failure caused by vibration accounts for more than 60%; secondly, the acquisition of vibration signals during mechanical operation is easy, and the vibration signal includes a large number of signals reflecting the state of mechanical equipment. Many mechanical faults can react abnormally in vibration state. come out. Vibration monitoring and diagnostic technology is a method of analyzing the state and fault of a device by detecting the vibration parameters of the device and its characteristics. Due to the extensiveness of vibration, the multi-dimensionality of parameters, the non-destructiveness of the vibration measurement method and the linearity, it is decided that the vibration monitoring and diagnosis of mechanical equipment is the preferred method for fault diagnosis of mechanical equipment.
After the failure of the surface of the gearbox of the automobile, different vibrations and noises are generated during the operation of meshing with the normal gear. Using vibration detection technology, by measuring the vibration signal during the operation of the car, the measurement parameters of the vibration have speed, acceleration and displacement, and the measured parameters and sensors can be selected according to the gearbox frequency. In order to detect a sufficient number of signals that truly reflect the state of the gearbox gear, the vibration measurement points are properly selected. Generally, vibration sensitive points that can fully reflect the vibration state of the gearbox are selected, and the key points closest to the core of the diagnosis and the vulnerable points prone to deterioration are ensured to ensure the effectiveness of the vibration signal measurement. Because the vibration monitoring and diagnosis technology can characterize the mechanical dynamic characteristics and its changing process in real time, intuitively and accurately, the monitoring and diagnosis method is simple and practical, and is widely used. For example, Zhao Fengqiang and others used the vibration signal to calibrate the sound pressure of the gearbox noise of Changchun FAW Group's gearbox factory, and diagnosed the possible faults, and achieved satisfactory results. Zheng Dianwang and others studied the mechanism of cepstrum analysis based on vibration monitoring technology. Through the analysis of the simulation of the failure of the car transmission gantry, the enthalpy value of the gear shift of the car transmission was analyzed by cepstrum analysis. The theoretical analysis is consistent with the experimental results.
1.3 Acoustic Emission Technique Diagnostics The so-called Acoustic Emission (AE) refers to the physical phenomenon of transient stress waves generated by the rapid release of elastic energy when an object is subjected to deformation or external influence. The elastic wave can reflect some properties of the material, so the acoustic emission technology is a method for detecting and analyzing the acoustic emission signal and using the acoustic emission signal to diagnose the fault. The automobile gearbox is a high-speed rotating machine. Due to unbalance, asymmetry, thermal bending, etc. during operation, the rotor will be rubbed. At this time, the internal crystal lattice of the metal will slip or rearrange. In this process, the energy changes. The form of the elastic wave is released, that is, an acoustic emission signal is generated. Then, the data acquisition module completes the collection, conversion and storage of the acoustic emission detection data, and uses the computer to analyze and process the data in real time to monitor the gearbox gear online, which can capture the weak fault signal in time and accurately, greatly improving the data. The safety of car operation. However, it should be pointed out that due to the influence of the shape of the acoustic emission source, the propagation path of the wave and the waveform change, it is necessary to use the acoustic emission signal to judge the change of the internal defects of the structure. One of the key technologies is to eliminate it. Interference from background noise.
Acoustic emission monitoring is a dynamic non-destructive testing method, but it is different from other non-destructive testing methods. Acoustic emission signals are generated under external conditions and are extremely sensitive to changes in defects. Micro-cracks on the order of micrometers can be detected. The information about the occurrence and expansion has high detection sensitivity. But there is no information about the occurrence of stripping, which is caused by the source mechanism. In addition, since most materials have acoustic emission characteristics, acoustic emission detection is not limited by materials, and it is possible to continuously monitor defect safety and over-limit alarms for a long period of time, which is where acoustic emission detection is superior to other non-destructive testing. But most of the acoustic emission signals are very weak, people can't hear them directly, and they need to be detected by sensitive electronic instruments. Therefore, with regard to the application of the AE measurement method, in addition to the advantages of using the AE measurement method to the maximum extent, it should be used in combination with other measurement methods to collect the necessary information. The research and application of acoustic emission technology has made considerable progress in China. Such as the development of new acoustic emission instruments, artificial neural network analysis of acoustic emission signals, fuzzy comprehensive evaluation of acoustic emission source severity, research on inverse source of acoustic emission signals, acoustic emission research of composite deformation damage process, and state detection of dynamic equipment With the formulation of standards, China's acoustic emission technology is maturing.
1.4 Fiber Sensing Technology Diagnostics Fiber optic sensing technology is a technology that uses optical fibers to be sensitive to certain physical quantities and converts external physical quantities into signals that can directly measure signals. Since the optical fiber can not only serve as a propagation medium for light waves, but also when the light wave propagates in the optical fiber, it characterizes the characteristic parameters (amplitude, phase, polarization state, wavelength, etc.) of the light wave due to external factors (such as temperature, pressure, strain, magnetic field, electric field, displacement, Indirect or direct changes occur by the action of rotation, etc., so that the optical fiber can be used as a sensing element to detect various physical quantities. A sensor is a device or device that senses a specified measurement and converts it into an output signal according to a certain law. Any sensor, especially those installed in the car's casing, must withstand harsh environments such as high temperatures, shocks, continuous vibrations, corrosive gases and electromagnetic fields. Optical fiber sensors (OFS, opticfibersensors) have been widely used in the automotive industry due to their strong anti-electromagnetic interference capability, high sensitivity, good electrical insulation, safety and reliability, corrosion resistance, and the ability to form optical fiber sensing networks. application. Fiber optic sensing technology has evolved toward integration, functionality and intelligence. The functionalization of the sensor means that the sensor itself not only has detection functions, but also signal processing and other functions. The combination of sensors and other functions creates new features. In recent years, a variety of sensors capable of detecting two or more different physical quantities have been developed abroad. A special ceramic is used to construct a multi-function sensor capable of detecting humidity and gas, and temperature and humidity, respectively.
1.5 Artificial neural network technology diagnosis method In the gearbox diagnosis method, the time domain and frequency domain characteristic parameters of the vibration acceleration signal of the box are often used to judge the presence or absence of the fault. Different types of gearboxes, sensor mounting positions, etc. have a great influence on these characteristic parameters. Coupled with the complexity of the structure of the car's gearbox, the mapping between feature parameters and faults is very complicated. These bring certain difficulties to the correct diagnosis. The main trend in the development of fault diagnosis technology is intelligent diagnosis. Artificial neural network is an edge discipline developed based on the latest achievements in neuroscience research. It is a nonlinear highly parallel processing network formed by a large number of simple neurons through interconnection. It is a mathematical abstraction of the human brain.
With its association, memory, self-organization and self-learning ability and strong nonlinear mapping ability, ANN has been widely used in the field of fault diagnosis. It uses domain experts to solve problems to train neural networks to acquire knowledge information, implicitly expresses neural network knowledge, and automatically generates knowledge expressed by network structure and weights. Several knowledge of the same problem is represented in the same network, which facilitates automatic knowledge acquisition and association reasoning, and is realized through interaction between neural sources. Improve diagnostic accuracy by continuously improving the diagnostic system through self-learning. Hu Kaicheng and Li Chuanqi discussed in detail how to better apply ANN to solve the problem of feature selection, feature extraction, diagnostic model and life prediction model in vehicle gearbox condition monitoring and fault diagnosis.
2 Outlook With the development of microelectronics technology, large-scale integrated circuit technology applications, computer technology and sensor technology, the increase of various signal analysis methods provides a good opportunity for the development of fault diagnosis technology. To make fault diagnosis more automatic, digital, intelligent and integrated; application software standardization, hardware specialization, standardization; diagnostic instruments and devices tend to develop engineering network systems. At present, people are working on research and improvement of sensors and monitoring instruments, research on fault diagnosis technology based on wavelet analysis, application research of fractal geometry in automobile fault diagnosis, and remote collaborative diagnosis technology based on Internet. The Internet's remote collaborative diagnosis technology can make full use of more technical experience and diagnostic data sharing to improve the accuracy of equipment diagnosis, and can dynamically monitor and diagnose equipment, enterprise, engineering and farm equipment in the diagnostic center to realize resources. Comprehensive goal of full utilization, high production efficiency, low cost and good construction quality.
Container Lock Shaft,Container Door Hing,Container Locking Bar,Container Steel Metal Ring
Wuxi Sixleaf Machinery Co. LTD , https://www.wuxisixleafmachine.com