Multi-sensor fusion vibration test data processing and analyzing system and method thereof

Abstract

The invention relates to the technical field of vibration test data processing and analysis systems, in particular to a multi-sensor fusion vibration test data processing and analysis system and method. A vibration data preprocessing unit; the vibration data fusion unit is connected with the vibration data preprocessing unit and is used for displaying the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in the same oscillogram in different colors and performing time alignment according to a set time window; a signal analysis and feature extraction unit; and the signal analysis information display unit is connected with the signal analysis and feature extraction unit and used for completing vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching and a time-frequency diagram function, and the accuracy and reliability of fault recognition are remarkably improved by combining multi-sensor data and an advanced feature extraction algorithm. The system not only can automatically detect and alarm abnormal vibration, but also can provide detailed fault positioning information.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration test data processing and analysis systems, in particular to a multi-sensor fusion vibration test data processing and analysis system and a method thereof. Background Art

[0002] With the continuous advancement of industrial technology, vibration analysis plays an increasingly important role in equipment condition monitoring and fault diagnosis. Traditional vibration analysis methods usually rely on a single type of sensor, such as an accelerometer or velocity sensor, to collect vibration data. This method can indeed provide useful information in some simple application scenarios, but when faced with complex industrial equipment, it is often difficult to fully capture multiple aspects of vibration characteristics.

[0003] In recent years, with the development of multi-sensor technology, some researchers have begun to try to combine multiple types of sensor data for analysis. This method can theoretically provide more comprehensive vibration information, but it faces many challenges in practical applications. First, different types of sensor data often have different characteristics and units, and how to effectively fuse these data has always been a difficult problem. Secondly, a large amount of multi-source data brings a huge computational burden to data processing and analysis. How to improve processing efficiency while ensuring analysis accuracy is also a problem that needs to be solved urgently.

[0004] In addition, existing vibration analysis systems also have deficiencies in data visualization and interactive analysis. Most systems can only provide basic time domain and frequency domain charts, lacking flexible data display and interactive functions, which makes it difficult for analysts to delve into complex vibration phenomena. At the same time, the limitations of automatic fault diagnosis functions also restrict the efficiency and accuracy of vibration analysis. Many systems can only identify some simple fault modes and often cannot provide accurate diagnostic results for complex or atypical fault conditions.

[0005] In view of the above problems, there is an urgent need for a vibration test data processing and analysis system that can effectively integrate multiple sensor data, provide flexible analysis tools, and have intelligent fault diagnosis capabilities. The present invention is proposed to meet this demand. Summary of the invention

[0006] The present invention proposes a multi-sensor fusion vibration test data processing and analysis system and method thereof, aiming to solve the problems existing in the existing vibration analysis system in terms of multi-sensor data fusion, data processing efficiency, analysis tool flexibility, and intelligent fault diagnosis. Through innovative system architecture design and advanced algorithm implementation, the present invention provides a comprehensive, efficient, and intelligent multi-sensor fusion vibration test data processing and analysis solution.

[0007] The present invention proposes a multi-sensor fusion vibration test data processing and analysis system, comprising:

[0008] A vibration data acquisition unit, used to measure the vibration information of the measured points of the product in the vibration test using an acceleration sensor, a velocity sensor and an eddy current sensor, and transmit the acquired data through a USB or LAN bus;

[0009] A vibration data preprocessing unit, connected to the vibration data acquisition unit, for preprocessing the received vibration data according to the measurement quantity, measurement conditions and data acquisition length to obtain original vibration data;

[0010] A vibration data fusion unit, connected to the vibration data preprocessing unit, is used to display the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in the same waveform diagram in different colors, and to perform time alignment according to a set time window;

[0011] a signal analysis and feature extraction unit, connected to the vibration data fusion unit, for performing automatic vibration analysis according to preset parameters, displaying an alarm message when a vibration fault is detected, and performing feature extraction on the current waveform signal; and

[0012] The signal analysis information display unit is connected to the signal analysis and feature extraction unit and is used to complete vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching and time-frequency diagram functions.

[0013] Preferably, the vibration data preprocessing unit is further used for:

[0014] When processing the acceleration sensor measurements, ensure that the effective measurement frequency range and resolution requirements are met, and calculate the data acquisition length based on the time interval length required for analysis;

[0015] When processing the speed sensor measurement value, ensure that the sampling frequency is not less than 1kHz;

[0016] When processing eddy current measurements, the data acquisition length is calculated based on the length of the time interval to be analyzed; and

[0017] According to the signal characteristics corresponding to different measured quantities, the maximum vibration amplitude of each measured point of the product in the vibration test is calculated, and based on this, the resolution of the acceleration amplitude is calculated.

[0018] Preferably, the vibration data fusion unit is further used for:

[0019] The waveform superposition method is used to integrate different measurement quantities and measurement conditions in the same time domain waveform diagram;

[0020] Supports setting different maximum coordinate values ​​on different coordinate axes of different charts in the time domain and frequency domain;

[0021] The time domain waveforms of acceleration and velocity vibration are displayed by maximum value and maximum flattening;

[0022] The waveform of eddy current vibration is displayed in different colors; and

[0023] When the same signal occurs, vibration data from multiple measurement points are aligned and displayed according to the request for synchronous display.

[0024] Preferably, the signal analysis and feature extraction unit is further used for:

[0025] Support fast and slow lens zoom display of waveform graph;

[0026] In the automatic analysis state, the faults that occur are recorded and the alarm information is displayed;

[0027] For physical quantities such as voltage, current, eddy current, acceleration and velocity, fault analysis is performed according to pre-set characteristic parameters. When the value of the physical quantity measured by the vibration waveform of a certain measuring point exceeds the set alarm threshold, the vibration waveform data is stored and the alarm information is displayed; and

[0028] For the measured physical quantity, frequency domain characteristic analysis is performed according to the preset characteristic parameters, and the relevant analysis results are displayed in the same waveform diagram.

[0029] Preferably, the signal analysis information display unit is further used for:

[0030] Supports time domain analysis waveform switching. By selecting a time domain waveform area in the waveform chart, the data of the area can be read from other waveform charts, and the waveform chart of the area can be quickly enlarged or slowly reduced.

[0031] Supports time-frequency diagram, spectrum, power spectrum, and frequency domain fast and slow shots;

[0032] Support switching the display of frequency domain power spectrum, time-frequency diagram and waterfall diagram by clicking the frequency domain analysis diagram;

[0033] Support switching to spectrum and power spectrum by clicking on the time-frequency graph and waterfall graph; and

[0034] Support switching to time domain fast and slow shots by clicking on the time domain waveform.

[0035] Preferably, the signal analysis information display unit is further used for:

[0036] Synchronously displaying the fault location information and the current display graphics; and

[0037] The fault location signal is displayed in a manner of intercepting in the time domain, zooming in with fast and slow lenses, and displaying the fault location information at the time when the fault occurs.

[0038] The multi-sensor fusion vibration test data processing and analysis method using the system includes the following steps:

[0039] The vibration data acquisition unit is used to measure the vibration information of the measured points of the product in the vibration test, and the acquired data is transmitted to the vibration data preprocessing unit through the USB or LAN bus;

[0040] Using a vibration data preprocessing unit to preprocess the received vibration data according to the measurement quantity, measurement conditions and data acquisition length to obtain original vibration data;

[0041] The vibration data fusion unit is used to display the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in different colors in the same waveform diagram, and time alignment is performed according to the set time window;

[0042] Performing automatic vibration analysis according to preset parameters using a signal analysis and feature extraction unit, displaying an alarm message when a vibration fault is detected, and extracting features from the current waveform signal; and

[0043] The signal analysis information display unit is used to complete vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching, and time-frequency diagram functions.

[0044] Preferably, the step of performing feature extraction using the signal analysis and feature extraction unit further comprises:

[0045] Support fast and slow lens zoom display of waveform graph;

[0046] In the automatic analysis state, the fault location information of the fault is recorded;

[0047] For physical quantities such as voltage, current, eddy current, acceleration and velocity, fault analysis is performed according to pre-set characteristic parameters. When the value of the physical quantity measured by the vibration waveform of a certain measuring point exceeds the set alarm threshold, the vibration waveform data is stored and the alarm information is displayed; and

[0048] For the measured physical quantity, frequency domain characteristic analysis is performed according to the preset characteristic parameters, and the relevant analysis results are displayed in the same waveform diagram. Through fast and slow shots and information recording and playback, analysts can easily locate and analyze vibration faults.

[0049] Preferably, the step of using the signal analysis information display unit to complete the analysis function also includes:

[0050] Supports time domain analysis waveform switching. By selecting a time domain waveform area in the waveform chart, the data of the area can be read from other waveform charts, and the waveform chart of the area can be quickly enlarged or slowly reduced.

[0051] Supports time-frequency diagram, spectrum, power spectrum, and frequency domain fast and slow shots;

[0052] Support switching the display of frequency domain power spectrum, time-frequency diagram and waterfall diagram by clicking the frequency domain analysis diagram;

[0053] Support switching to spectrum and power spectrum by clicking on the time-frequency graph and waterfall graph; and

[0054] Support switching to time domain fast and slow shots by clicking on the time domain waveform.

[0055] Preferably, the method further comprises the following steps:

[0056] Synchronously displaying the fault location information and the current display graphics; and

[0057] The fault location signal is displayed in a manner of intercepting in the time domain, zooming in with fast and slow lenses, and displaying the fault location information at the time when the fault occurs.

[0058] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0059] From the perspective of the overall architecture, the system of the present invention includes multiple functional modules such as vibration data acquisition, preprocessing, fusion, analysis and feature extraction, and information display. These modules work closely together to form a complete data processing and analysis chain. This modular design not only improves the flexibility and scalability of the system, but also effectively solves the complexity of multi-source data processing.

[0060] In terms of data fusion, the present invention uses innovative multi-dimensional data visualization technology to display different types of sensor data in different colors in the same chart and align the time axis. This method cleverly solves the problem of large differences in the characteristics of different sensor data, allowing analysts to intuitively compare and correlate different types of vibration data. This not only improves the efficiency of data analysis, but also provides the possibility of discovering potential relationships in complex vibration phenomena.

[0061] In terms of data processing efficiency, the present invention greatly improves the processing speed of the system through intelligent preprocessing algorithms and adaptive analysis strategies. For example, for different types of sensor data, the system will automatically select appropriate preprocessing methods according to their characteristics, such as adjusting the effective frequency range and resolution for acceleration data, or optimizing the sampling frequency for velocity data. This intelligent processing method not only improves efficiency, but also ensures the consistency of data quality.

[0062] In terms of the flexibility of the analysis tools, the present invention provides a wealth of time domain and frequency domain analysis functions, such as fast and slow lens zooming, time-frequency diagram analysis, waterfall diagrams, etc. The flexible combination of these tools enables analysts to deeply study vibration characteristics from multiple angles. In particular, the system supports the function of quickly switching between different analysis views, which provides strong support for the diagnosis of complex vibration problems.

[0063] In terms of intelligent fault diagnosis, the present invention significantly improves the accuracy and reliability of fault identification by combining multi-sensor data and advanced feature extraction algorithms. The system can not only automatically detect and alarm abnormal vibrations, but also provide detailed fault location information. This intelligent diagnostic capability greatly reduces the workload of analysts and improves the timeliness and accuracy of fault diagnosis.

[0064] In general, this invention successfully solves many technical challenges faced by existing vibration analysis systems through innovative system design and advanced algorithm implementation. It not only improves the comprehensiveness and accuracy of vibration analysis, but also greatly improves analysis efficiency and user experience. This all-round improvement enables the system to better meet the complex vibration monitoring and fault diagnosis needs of modern industrial equipment, and provides strong technical support for equipment reliability management and preventive maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is the overall logic block diagram of the system of the present invention.

[0066] Figure 2 It is an overall logic block diagram of the data processing method of the present invention.

[0067] Figure 3 It is a detailed logic block diagram of the feature extraction step of the present invention.

[0068] Figure 4 It is a detailed logic block diagram of the information display step of the present invention. DETAILED DESCRIPTION

[0069] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects thereof are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0070] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0071] Example 1

[0072] See also Figure 1-4 The present invention provides a multi-sensor fusion vibration test data processing and analysis system and method thereof, which are used to solve the problem in the prior art that a single sensor cannot fully capture complex vibration characteristics.

[0073] First, the multi-sensor fusion vibration test data processing and analysis system of the present invention includes a vibration data acquisition unit 1, a vibration data preprocessing unit 2, a vibration data fusion unit 3, a signal analysis and feature extraction unit 4 and a signal analysis information display unit 5. These units are connected in sequence to form a complete data processing and analysis chain.

[0074] Specifically, the vibration data acquisition unit 1 uses an acceleration sensor, a velocity sensor, and an eddy current sensor to measure the vibration information of the product at the measured point in the vibration test. For example, when performing a vibration test on a certain industrial equipment, the acceleration sensor can capture high-frequency vibrations, the velocity sensor is suitable for medium-frequency vibrations, and the eddy current sensor can measure low-frequency displacements. These sensors transmit data to the vibration data preprocessing unit 2 via a USB or LAN bus, ensuring the high speed and reliability of data transmission.

[0075] Then, the vibration data preprocessing unit 2 preprocesses the received vibration data according to the measurement quantity, measurement conditions and data acquisition length. Preferably, for the measurement value of the acceleration sensor, this unit ensures that the requirements of the effective measurement frequency range and resolution are met. For example, for common mechanical equipment vibration analysis, the effective frequency range may be 0-10kHz, and the resolution needs to reach 0.1g. For the measurement value of the velocity sensor, the sampling frequency is not less than 1kHz, so that the vibration characteristics of most industrial equipment can be accurately captured. For eddy current measurement quantities, the appropriate data acquisition length is calculated according to the analysis requirements, which may usually be between 1-10 seconds.

[0076] Secondly, the vibration data fusion unit 3 displays the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in different colors in the same waveform. For example, acceleration data may be represented in red, velocity data in blue, and eddy current data in green. This intuitive display method allows analysts to quickly identify different types of vibration data. At the same time, this unit also performs time alignment according to the set time window to ensure that the data from different sensors are synchronized in time.

[0077] Again, the signal analysis and feature extraction unit 4 performs automatic vibration analysis according to preset parameters. When a vibration fault is detected, this unit will display an alarm message. For example, if the vibration amplitude of a certain measuring point exceeds a preset threshold (such as the safe vibration limit of mechanical equipment), the system will immediately issue an alarm. In addition, this unit also extracts features from the current waveform signal, such as calculating characteristic parameters such as the root mean square value and peak factor of the vibration.

[0078] Finally, the signal analysis information display unit 5 completes the vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching and time-frequency diagram functions. For example, the time domain analysis can show the change trend of vibration over time, the frequency domain analysis can reveal the frequency composition of vibration, and the time-frequency diagram can show the change of vibration frequency over time.

[0079] In one embodiment of the present invention, the vibration data preprocessing unit 2 also has a more detailed processing function. For the acceleration sensor measurement value, in addition to ensuring that the requirements of the effective measurement frequency range and resolution are met, the data acquisition length is also calculated according to the length of the time interval to be analyzed. For example, if it is necessary to analyze transient vibrations within 0.1 seconds, the sampling rate may need to reach 100kHz or more. For the velocity sensor measurement value, in addition to ensuring that the sampling frequency is not less than 1kHz, low-pass filtering may also be required to remove high-frequency noise. For eddy current measurements, in addition to calculating the data acquisition length according to the length of the time interval to be analyzed, temperature compensation may also be required to improve measurement accuracy.

[0080] In addition, the unit also calculates the maximum vibration amplitude of each measured point of the product in the vibration test according to the signal characteristics corresponding to different measurement quantities, and uses this as a basis to calculate the resolution of the acceleration amplitude. For example, if the maximum vibration amplitude is 10g, then the resolution of the acceleration amplitude may need to be set to 0.01g to ensure that subtle vibration changes can be captured.

[0081] In another embodiment, the vibration data fusion unit 3 uses a waveform superposition method to fuse different measurement quantities and measurement conditions in the same time domain waveform diagram. This method allows analysts to intuitively compare different types of vibration data. At the same time, this unit also supports setting different coordinate maximum values ​​on different coordinate axes of different charts in the time domain and frequency domain. For example, the maximum value of acceleration may be set to 20g, while the maximum value of velocity may be set to 100mm / s.

[0082] For the time domain waveforms of acceleration and velocity vibration, this unit uses the maximum value flattened display. This display method can highlight the peak value of vibration and help identify potential faults. For the waveforms of eddy current vibration, different colors are used to distinguish different measurement channels. In addition, for the vibration data of multiple measurement points when the same signal occurs, this unit aligns the display according to the request for synchronous display, which helps to analyze the propagation characteristics of vibration at different locations.

[0083] Next, we will introduce the functions of the signal analysis and feature extraction unit 4 in detail. First, the unit supports fast and slow zoom display of the waveform. This means that analysts can quickly browse the overall vibration trend or slow down to carefully observe the vibration details of a specific time period. Second, in the automatic analysis state, this unit can record the occurrence of faults and display alarm information. For example, if the characteristic frequency of the bearing is detected to be abnormal, the system will immediately issue an alarm and record the relevant data.

[0084] Furthermore, for physical quantities such as voltage, current, eddy current, acceleration and velocity, this unit performs fault analysis according to pre-set characteristic parameters. When the value of the physical quantity measured by the vibration waveform of a certain measuring point exceeds the set alarm threshold, the system will store the vibration waveform data and display the alarm information. For example, for a certain model of motor, if the vibration speed exceeds 5mm / s, the system will issue an alarm. Finally, this unit can also perform frequency domain characteristic analysis according to the pre-set characteristic parameters for the measured physical quantity, and display the relevant analysis results in the same waveform diagram. This comprehensive analysis can help technicians quickly locate the cause of the fault.

[0085] Next, let's look at the specific functions of the signal analysis information display unit 5. First, the unit supports time domain analysis waveform switching. By selecting a time domain waveform area in the waveform graph and reading the data of the area from other waveform graphs, the system can simultaneously display the fast zoom-in or slow zoom-out waveform graph of the area. This function allows analysts to flexibly view vibration characteristics at different time scales.

[0086] Secondly, this unit supports time-frequency plots, frequency spectra, power spectra, and frequency domain fast and slow motion. For example, time-frequency plots can show the change of vibration frequency over time, which is particularly useful for analyzing non-steady-state vibrations. Spectral analysis can reveal the frequency composition of vibrations, which helps to identify specific fault frequencies. Power spectra can show the energy distribution of different frequency components, which helps to determine the main source of vibration.

[0087] Thirdly, this unit supports switching between the frequency domain power spectrum, time-frequency diagram and waterfall diagram by clicking on the frequency domain analysis diagram. This flexible switching function allows analysts to observe vibration characteristics from different angles. For example, the waterfall diagram can clearly show the trend of vibration spectrum over time, which is particularly useful for analyzing vibration characteristics during startup or shutdown.

[0088] Finally, this unit also supports switching to frequency spectrum and power spectrum by clicking on the time-frequency graph and waterfall graph, and switching to time domain fast and slow shots by clicking on the time domain waveform. This multi-dimensional analysis perspective can help technicians fully understand complex vibration phenomena.

[0089] In another embodiment of the present invention, the signal analysis information display unit 5 also has a more advanced display function. First, the unit can synchronously display the fault location information with the current display graph. For example, when the system detects that a bearing may have a fault, it will not only mark the corresponding time point on the waveform graph, but also mark the characteristic frequency of the bearing on the spectrum graph.

[0090] Secondly, this unit can display the fault location signal in the form of time domain interception, graphic fast and slow lens magnification, and display of fault location information at the time of fault occurrence. This means that analysts can quickly locate the specific moment when the fault occurs and view the vibration characteristics before and after that moment in detail. For example, if the system detects an abnormal impact vibration at a certain moment, the analyst can quickly zoom in on the waveform around that moment and view the corresponding spectrum analysis results at the same time, so as to better understand the nature and possible causes of the fault.

[0091] Next, we will introduce the multi-sensor fusion vibration test data processing and analysis method of the present invention in detail. First, the method uses the vibration data acquisition unit 1 to measure the vibration information of the measured point of the product in the vibration test. For example, when performing a vibration test on a certain industrial pump, acceleration sensors, velocity sensors and eddy current sensors may be installed at multiple locations such as the pump body, motor and pipeline. These sensors transmit the collected data to the vibration data preprocessing unit 2 via a USB or LAN bus.

[0092] Then, the vibration data preprocessing unit 2 preprocesses the received vibration data according to the measurement quantity, measurement conditions and data acquisition length to obtain the original vibration data. For example, for acceleration data, it may be necessary to perform an integral operation to obtain velocity and displacement information; for velocity data, it may be necessary to perform a high-pass filter to remove the DC bias; for eddy current data, it may be necessary to perform temperature compensation to improve the measurement accuracy.

[0093] Secondly, the vibration data fusion unit 3 displays the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in different colors in the same waveform diagram, and performs time alignment according to the set time window. This fusion display method allows analysts to intuitively compare different types of vibration data. For example, they can simultaneously observe the changing trends of acceleration, velocity and displacement, thereby more comprehensively understanding the vibration status of the equipment.

[0094] Again, the signal analysis and feature extraction unit 4 performs automatic vibration analysis according to preset parameters. When a vibration fault is detected, the system will display an alarm message. For example, if the vibration amplitude of a bearing exceeds a preset threshold (such as the warning value specified in the ISO 10816 standard), the system will immediately issue an alarm. At the same time, this unit will also extract features from the current waveform signal, such as calculating statistical features such as kurtosis and skewness values. These features can reflect the nonlinear characteristics of the vibration signal and help early fault diagnosis.

[0095] Finally, the signal analysis information display unit 5 completes the vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching and time-frequency diagram functions. For example, time domain analysis can show the trend of vibration changes over time, which is helpful for identifying periodic faults; frequency domain analysis can reveal the frequency composition of vibration, which is helpful for identifying specific fault frequencies; time-frequency diagram can show the change of vibration frequency over time, which is particularly suitable for analyzing non-steady-state vibration.

[0096] In a specific embodiment of the present invention, the feature extraction step of the signal analysis and feature extraction unit 4 further includes the following: First, fast and slow lens zoom display of the waveform is supported. This means that analysts can switch freely between macro and micro scales, and can observe the overall vibration trend and carefully study the vibration details of a specific time period.

[0097] Secondly, in the automatic analysis state, the system can record the fault location information of the fault. For example, if an abnormal impact vibration is detected at a certain moment, the system will automatically record the vibration parameters at that moment, including time, amplitude, frequency and other information, to facilitate subsequent in-depth analysis.

[0098] Again, for physical quantities such as voltage, current, eddy current, acceleration and speed, the system performs fault analysis according to pre-set characteristic parameters. When the value of the physical quantity measured by the vibration waveform of a certain measuring point exceeds the set alarm threshold, the system will store the vibration waveform data and display the alarm information. For example, for a certain model of centrifugal pump, if the vibration speed of the bearing seat exceeds 7.1mm / s (the upper limit of Zone C for medium-sized machinery in the ISO 10816 standard), the system will issue an alarm and automatically store the relevant vibration data. This automatic storage function is particularly useful for capturing transient faults, because these faults may occur and disappear in a very short time.

[0099] Finally, for the measured physical quantity, the system performs frequency domain feature analysis according to the preset characteristic parameters and displays the relevant analysis results in the same waveform. For example, the system may automatically calculate and display the spectrum envelope of the vibration signal, which is particularly effective for detecting early failures of bearings or gears. By fast and slow lenses and information recording and playback, analysts can easily locate and analyze vibration failures. For example, they can quickly browse the vibration trend during the entire test process, then slow down and carefully observe the suspicious time period, and combine the frequency domain analysis results to determine the specific type and location of the fault.

[0100] In another embodiment of the present invention, the functions of the signal analysis information display unit 5 are richer and more flexible. First, the unit supports time domain analysis waveform switching. By selecting a time domain waveform area in the waveform graph and reading the data of the area from other waveform graphs, the system can simultaneously display a fast zoom-in or slow zoom-out waveform graph of the area. This function is particularly suitable for analyzing transient vibration events. For example, when analyzing the startup process of a centrifugal compressor, the analyst can first view the vibration trend of the entire startup process, and then quickly zoom in on the key transition stages (such as the moment of passing the critical speed) to observe the vibration characteristics of these stages in detail.

[0101] Secondly, the unit supports functions such as time-frequency diagram, spectrum, power spectrum and frequency domain fast and slow shots. The time-frequency diagram can intuitively display the change of vibration frequency over time, which is particularly useful for analyzing non-steady-state vibration. For example, when analyzing the vibration characteristics of a wind turbine, the time-frequency diagram can clearly show the change of blade vibration frequency under different wind speeds. Spectral analysis can reveal the frequency composition of vibration and help identify specific fault frequencies. For example, by observing the sideband frequency in the spectrum, it can be determined whether the gear has eccentricity or misalignment problems. The power spectrum can show the energy distribution of different frequency components, which helps to determine the main source of vibration.

[0102] Thirdly, the unit supports switching between the frequency domain power spectrum, time-frequency diagram, and waterfall diagram by clicking on the frequency domain analysis diagram. This flexible switching function allows analysts to observe vibration characteristics from different angles. For example, when analyzing the vibration characteristics of a reciprocating compressor, analysts can first view the spectrum diagram to identify the main vibration frequencies, then switch to the waterfall diagram to observe the change trend of these frequencies over time, and finally use the time-frequency diagram to analyze the time-varying characteristics of a specific frequency component in detail.

[0103] Finally, the unit also supports switching to the spectrum and power spectrum by clicking on the time-frequency graph and waterfall graph, and switching to the time domain fast and slow shots by clicking on the time domain waveform. This multi-dimensional analysis perspective can help technicians fully understand complex vibration phenomena. For example, when analyzing the vibration characteristics of a large steam turbine generator set, analysts can first identify abnormal vibration events in the time domain waveform, then quickly switch to the spectrum graph at that moment to view the frequency components, and then observe the changing trends of these frequency components throughout the test process through the waterfall graph.

[0104] The multi-sensor fusion vibration test data processing and analysis method of the present invention also includes an important step, which is to synchronously display the fault location information with the current display graphics. This means that when the system detects a potential fault, it will not only mark it on the corresponding graphical interface, but also provide detailed fault information. For example, if the system detects that the bearing may have a fault at a certain moment, it will mark the moment on the time domain waveform graph, mark the characteristic frequency of the bearing on the spectrum graph, and display relevant fault diagnosis information on the interface, such as "suspected outer ring fault".

[0105] In addition, this method also supports the display of fault location signals in the form of time domain interception, graphic fast and slow lens magnification, and display of fault location information at the time of fault occurrence. This function allows analysts to quickly locate the specific moment when the fault occurs and view the vibration characteristics before and after that moment in detail. For example, if the system detects abnormal impact vibration at a certain moment, analysts can quickly zoom in on the waveform around that moment and view the corresponding spectrum analysis results and envelope spectrum analysis results at the same time, so as to better understand the nature and possible causes of the fault.

[0106] The multi-sensor fusion vibration test data processing and analysis system and method of the present invention has many technical effects. First, by fusing the data of multiple sensors, the system can fully capture complex vibration characteristics and improve the accuracy and reliability of vibration analysis. For example, when analyzing the vibration characteristics of a large hydro-turbine generator set, the acceleration sensor can capture high-frequency bearing fault signals, the speed sensor can accurately measure the vibration displacement of the rotor, and the eddy current sensor can accurately monitor the trajectory of the shaft. This multi-dimensional data acquisition and analysis greatly improves the accuracy of fault diagnosis.

[0107] Secondly, the system's automatic analysis and alarm functions can greatly improve the efficiency of vibration monitoring. For example, when continuously monitoring the operating status of a large compressor, the system can analyze vibration data 24 hours a day and immediately issue an alarm if an abnormality is found, which is more reliable and timely than manual monitoring.

[0108] Thirdly, the system's multi-dimensional visualization function makes complex vibration data more intuitive and easy to understand. For example, by displaying different types of vibration data in different colors in the same chart, analysts can quickly identify the relationship and difference between different measurement quantities, which is very helpful for understanding complex vibration phenomena.

[0109] Finally, the system's flexible analysis capabilities provide a powerful tool for in-depth research on vibration problems. For example, by flexibly switching between time domain and frequency domain analysis, analysts can observe vibration characteristics from multiple angles, which is particularly useful for diagnosing complex faults.

[0110] In general, the multi-sensor fusion vibration test data processing and analysis system and method of the present invention provides a comprehensive, efficient and intelligent solution for the field of vibration analysis and fault diagnosis. It can not only improve the accuracy and efficiency of vibration analysis, but also help engineers to understand complex vibration phenomena more deeply, thus providing strong support for equipment reliability management and preventive maintenance.

[0111] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Multi-sensor fusion vibration test data processing and analysis system, characterized by: include: A vibration data acquisition unit, used to measure the vibration information of the measured points of the product in the vibration test using an acceleration sensor, a velocity sensor and an eddy current sensor, and transmit the acquired data through a USB or LAN bus; A vibration data preprocessing unit, connected to the vibration data acquisition unit, for preprocessing the received vibration data according to the measurement quantity, measurement conditions and data acquisition length to obtain original vibration data; A vibration data fusion unit, connected to the vibration data preprocessing unit, is used to display the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in the same waveform diagram in different colors, and to perform time alignment according to a set time window; a signal analysis and feature extraction unit, connected to the vibration data fusion unit, for performing automatic vibration analysis according to preset parameters, displaying an alarm message when a vibration fault is detected, and performing feature extraction on the current waveform signal; and The signal analysis information display unit is connected to the signal analysis and feature extraction unit and is used to complete vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching and time-frequency diagram functions.

2. The multi-sensor fusion vibration test data processing and analysis system according to claim 1 is characterized in that: The vibration data preprocessing unit is also used for: When processing the acceleration sensor measurements, ensure that the effective measurement frequency range and resolution requirements are met, and calculate the data acquisition length based on the time interval length required for analysis; When processing the speed sensor measurement value, ensure that the sampling frequency is not less than 1kHz; When processing eddy current measurements, the data acquisition length is calculated based on the length of the time interval required for analysis; as well as According to the signal characteristics corresponding to different measured quantities, the maximum vibration amplitude of each measured point of the product in the vibration test is calculated, and based on this, the resolution of the acceleration amplitude is calculated.

3. The multi-sensor fusion vibration test data processing and analysis system according to claim 1 is characterized in that: The vibration data fusion unit is also used for: The waveform superposition method is used to integrate different measurement quantities and measurement conditions in the same time domain waveform diagram; Supports setting different maximum coordinate values ​​on different coordinate axes of different charts in the time domain and frequency domain; The time domain waveforms of acceleration and velocity vibration are displayed by maximum value and maximum flattening; The waveform of eddy current vibration is displayed in different colors; and When the same signal occurs, vibration data from multiple measurement points are aligned and displayed according to the request for synchronous display.

4. The multi-sensor fusion vibration test data processing and analysis system according to claim 1, characterized in that: The signal analysis and feature extraction unit is also used for: Support fast and slow lens zoom display of waveform graph; In the automatic analysis state, the faults that occur are recorded and the alarm information is displayed; For physical quantities such as voltage, current, eddy current, acceleration and speed, fault analysis is performed according to pre-set characteristic parameters. When the value of the physical quantity measured by the vibration waveform of a certain measuring point exceeds the set alarm threshold, the vibration waveform data is stored and the alarm information is displayed; as well as For the measured physical quantity, frequency domain characteristic analysis is performed according to the preset characteristic parameters, and the relevant analysis results are displayed in the same waveform diagram.

5. The multi-sensor fusion vibration test data processing and analysis system according to claim 1, characterized in that: The signal analysis information display unit is also used for: Supports time domain analysis waveform switching. By selecting a time domain waveform area in the waveform chart, the data of the area can be read from other waveform charts, and the waveform chart of the area can be quickly enlarged or slowly reduced. Supports time-frequency diagram, spectrum, power spectrum, and frequency domain fast and slow shots; Support switching the display of frequency domain power spectrum, time-frequency diagram and waterfall diagram by clicking the frequency domain analysis diagram; Support switching to spectrum and power spectrum by clicking on the time-frequency graph and waterfall graph; and Support switching to time domain fast and slow shots by clicking on the time domain waveform.

6. The multi-sensor fusion vibration test data processing and analysis system according to claim 1, characterized in that: The signal analysis information display unit is also used for: Synchronously displaying the fault location information and the current display graphics; and The fault location signal is displayed in a manner of intercepting in the time domain, zooming in with fast and slow lenses, and displaying the fault location information at the time when the fault occurs.

7. A method for processing and analyzing multi-sensor fusion vibration test data using the system according to any one of claims 1 to 6, characterized in that: The following steps are involved: The vibration data acquisition unit is used to measure the vibration information of the measured points of the product in the vibration test, and the acquired data is transmitted to the vibration data preprocessing unit through the USB or LAN bus; Using a vibration data preprocessing unit to preprocess the received vibration data according to the measurement quantity, measurement conditions and data acquisition length to obtain original vibration data; The vibration data fusion unit is used to display the vibration data corresponding to different measurement quantities, measurement conditions and data acquisition lengths in different colors in the same waveform diagram, and time alignment is performed according to the set time window; The signal analysis and feature extraction unit is used to perform automatic vibration analysis according to preset parameters, display alarm information when a vibration fault is detected, and perform feature extraction on the current waveform signal; as well as The signal analysis information display unit is used to complete vibration time domain analysis, frequency domain analysis, time domain and frequency domain waveform switching, and time-frequency diagram functions.

8. The multi-sensor fusion vibration test data processing and analysis method according to claim 7 is characterized in that: The step of performing feature extraction using the signal analysis and feature extraction unit also includes: Support fast and slow lens zoom display of waveform graph; In the automatic analysis state, the fault location information of the fault is recorded; For physical quantities such as voltage, current, eddy current, acceleration and velocity, fault analysis is performed according to pre-set characteristic parameters. When the value of the physical quantity measured by the vibration waveform of a certain measuring point exceeds the set alarm threshold, the vibration waveform data is stored and the alarm information is displayed; and For the measured physical quantity, frequency domain characteristic analysis is performed according to the preset characteristic parameters, and the relevant analysis results are displayed in the same waveform diagram. Through fast and slow shots and information recording and playback, analysts can easily locate and analyze vibration faults.

9. The multi-sensor fusion vibration test data processing and analysis method according to claim 7, characterized in that: The step of using the signal analysis information display unit to complete the analysis function also includes: Supports time domain analysis waveform switching. By selecting a time domain waveform area in the waveform chart, the data of the area can be read from other waveform charts, and the waveform chart of the area can be quickly enlarged or slowly reduced. Supports time-frequency diagram, spectrum, power spectrum, and frequency domain fast and slow shots; Support switching the display of frequency domain power spectrum, time-frequency diagram and waterfall diagram by clicking the frequency domain analysis diagram; Support switching to spectrum and power spectrum by clicking on the time-frequency graph and waterfall graph; and Support switching to time domain fast and slow shots by clicking on the time domain waveform.

10. The multi-sensor fusion vibration test data processing and analysis method according to claim 7, characterized in that: The following steps are also included: Synchronize the display of fault location information and current display graphics; as well as The fault location signal is displayed in a manner of intercepting in the time domain, zooming in with fast and slow lenses, and displaying the fault location information at the time when the fault occurs.

Images