Rotating machine fault source positioning method based on optimal acquisition point of mobile robot

Abstract

The invention relates to the technical field of sound source localization, in particular to a rotating machine fault source localization method based on an optimal acquisition point of a mobile robot, and the method comprises the steps: obtaining an initial sound signal; an optimal acquisition angle is determined by comparing signal qualities at different positions; determining an optimal acquisition distance; determining an optimal acquisition height; performing fault feature analysis to obtain a target frequency to locate a fixed frequency band sound source; the sound field imaging visually determines the position of a fault source. According to the invention, the robot flexibly moves to realize sound signal acquisition, gradually adjust acquisition parameters, and carry out signal quality comparison on the sound array at different angles, distances and heights until an optimal acquisition position is determined, so that adaptive detection on the object to be identified is ensured, the acquired sound signals can reflect fault source characteristics to the greatest extent, and the fault detection accuracy is improved. And sound field imaging is carried out in combination with the video image and the sound field image to visually display the position of the fault source, so that the reliability and intuition of positioning are enhanced.

Description

Technical Field

[0001] The invention relates to the technical field of sound source positioning, and in particular to a method for positioning a rotating machinery fault source based on an optimal acquisition point of a mobile robot. Background Art

[0002] Rotating machinery is widely used in key technical equipment such as processing and manufacturing, automobile production, mining and transportation, high-speed EMUs, and nuclear power units. However, these important mechanical equipment often need to serve for a long time under harsh and changeable working conditions, and are subjected to the continuous effects of factors such as alternating loads, impact loads, and speed fluctuations, resulting in frequent failures of rotating machinery systems. Once a rotating system failure occurs, it will lead to a series of chain reactions, which may affect the normal operation of the entire machine at the least, or even cause the machine to be destroyed and people to die, leading to catastrophic safety accidents. Therefore, it is crucial to carry out research on rotating machinery fault diagnosis and health status monitoring technology, detect potential early faults in the mechanical system as early as possible, and eliminate hidden fault hazards, in order to ensure the safe and reliable operation of the mechanical system and avoid major economic losses and casualties.

[0003] In modern factories, the application of automatic inspection robots is becoming popular. Automatic inspection robots can provide enterprises with equipment failure, environmental safety judgment and automatic alarm without human intervention, help enterprises better automate management, significantly enhance the ability of remote collaborative work, and improve work efficiency. It can also avoid inspection personnel's careless omissions, reduce labor costs, greatly reduce safety hazards and reduce the occurrence of safety accidents. Looking at the inspection robots currently used in various industries, most of them use machine vision technology to obtain various information about the equipment to judge the operating status of the equipment. However, the use of machine vision technology can only observe the surface information of the machine equipment. When the fault is not on the surface of the equipment, the use of this inspection robot cannot perform fault location and autonomous analysis well.

[0004] Traditional fault diagnosis systems mostly obtain local vibration signals through acceleration sensors. However, the vibration of mechanical equipment causes the installation position and accuracy of acceleration sensors to be poor. Some equipment cannot provide suitable conditions for the installation of vibration sensors, such as equipment in harsh environments such as high temperature, high corrosion, and high humidity.

[0005] The patent document with publication number CN112051063A discloses a method and system for locating the sound source of equipment faults. The patent uses the collected equipment sound signal as the signal to be analyzed, performs Fourier transform on the signal to be analyzed, calculates the fast spectral kurtosis index of the signal to be analyzed based on the transform result, and determines the fault feature frequency selection result based on the fast spectral kurtosis index; calculates the cross-spectral matrix and the delayed sum-guided response based on the fault feature frequency selection result and the signal to be analyzed, and obtains the delayed sum beamforming imaging result of the fault frequency selection; inputs the delayed sum beamforming imaging result into a trained dual-channel convolutional neural network to obtain the positioning information of the fault sound source point.

[0006] This shows the following problem: In the prior art, when the actual sound field environment is complex and changeable, it will lead to poor and unstable sound signal quality and inaccurate fault location results. Summary of the invention

[0007] To this end, the present invention provides a method for locating the fault source of rotating machinery based on the optimal collection point of a mobile robot, which is used to utilize the mobility characteristics of the robot to determine the optimal position for collecting sound signals to overcome the problems in the prior art of poor and unstable sound signal quality and inaccurate fault location results when the actual sound field environment on site is complex and changeable.

[0008] To achieve the above object, the present invention provides a method for locating a rotating machinery fault source based on an optimal collection point of a mobile robot, comprising:

[0009] Using an acoustic array to collect sound information of the object to be identified according to preset initial acquisition parameters to obtain an initial acoustic signal;

[0010] The acquisition parameters include an initial acquisition angle, an initial acquisition distance and an initial acquisition height;

[0011] The initial acquisition angle is gradually adjusted according to a preset angle adjustment step to obtain an adjusted acquisition angle;

[0012] Acquire a first acoustic signal of the object to be identified based on the adjusted acquisition angle, the initial acquisition distance, and the initial acquisition height, and determine an optimal acquisition angle based on the initial acoustic signal and the first acoustic signal;

[0013] The initial acquisition distance is gradually adjusted according to a preset distance adjustment step to obtain an adjusted acquisition distance;

[0014] Acquire a second acoustic signal of the object to be identified based on the optimal acquisition angle, the adjusted acquisition distance, and the initial acquisition height, and determine an optimal acquisition distance based on the first acoustic signal and the second acoustic signal;

[0015] The initial acquisition height is gradually adjusted according to a preset height adjustment step length to obtain an adjusted acquisition height;

[0016] Acquire a third acoustic signal of the object to be identified based on the optimal acquisition angle, the optimal acquisition distance, and the adjusted acquisition height, and determine an optimal acquisition height based on the second acoustic signal and the third acoustic signal;

[0017] Perform spectrum analysis on the best quality sound signal obtained from the optimal collection point according to the best collection angle, the best collection distance, and the best collection height to obtain the target frequency;

[0018] Fixed bandwidth sound source localization;

[0019] Acquiring a sound field image of the target frequency;

[0020] Acquire a video image of the object to be identified;

[0021] The fault source location is determined based on the sound field image and the video image.

[0022] Further, the step of gradually adjusting the initial acquisition angle according to a preset angle adjustment step to obtain the adjusted acquisition angle includes:

[0023] Set the angle adjustment step size;

[0024] The initial acquisition angle is gradually increased according to the angle adjustment step size and an adjustment period to obtain the adjusted acquisition angle.

[0025] Further, determining the optimal acquisition angle based on the initial acoustic signal and the first acoustic signal includes:

[0026] Acquiring a first characteristic parameter of the first acoustic signal and an initial characteristic parameter of an initial acoustic signal;

[0027] Calculate the difference between each of the first characteristic parameters and the corresponding initial characteristic parameters to obtain a first difference value;

[0028] Calculating a ratio of the first difference value to the corresponding initial characteristic parameter to obtain a first difference ratio, and calculating a sum of the first difference ratios to obtain a total first difference ratio;

[0029] By comparing the plurality of total first difference ratios, the adjusted collection angle of the first acoustic signal when the total first difference ratio is the largest is determined as the optimal collection angle.

[0030] Further, the step of gradually adjusting the initial acquisition distance according to a preset distance adjustment step to obtain an adjusted acquisition distance includes:

[0031] Set the distance adjustment step size;

[0032] The initial acquisition distance is gradually increased according to the distance adjustment step size and an adjustment period to obtain the adjusted acquisition distance.

[0033] Further, determining the optimal collection distance based on the first acoustic signal and the second acoustic signal includes:

[0034] Acquiring a second characteristic parameter of the second acoustic signal;

[0035] Calculating the difference between each of the second characteristic parameters and each of the first characteristic parameters to obtain a second difference value;

[0036] Calculating a ratio of the second difference value to the corresponding first characteristic parameter to obtain a second difference ratio, and calculating a sum of the second difference ratios to obtain a total second difference ratio;

[0037] A plurality of the total second difference ratios are compared, and the adjusted collection distance of the second sound signal when the total second difference ratio is the largest is determined as the optimal collection distance.

[0038] Further, the step of gradually adjusting the initial acquisition height according to a preset height adjustment step to obtain the adjusted acquisition height includes:

[0039] Set the height adjustment step size;

[0040] The initial acquisition height is gradually increased according to the height adjustment step length and in accordance with an adjustment period to obtain the adjusted acquisition height.

[0041] Further, determining the optimal collection height based on the second acoustic signal and the third acoustic signal includes:

[0042] Acquiring a third characteristic parameter of the third acoustic signal;

[0043] Calculating the difference between each of the third characteristic parameters and each of the second characteristic parameters to obtain a third difference;

[0044] Calculating a ratio of the third difference ratio to the corresponding second characteristic parameter to obtain a third difference ratio, and calculating a sum of the third difference ratios to obtain a total third difference ratio;

[0045] By comparing a plurality of the total third difference ratios, the adjusted collection height of the third acoustic signal when the total third difference ratio is the largest is determined as the optimal collection height.

[0046] Further, the acquiring the sound field image of the target frequency includes:

[0047] A bandpass filter is used to filter out invalid frequencies in the target frequency to obtain a valid target frequency, and sound source positioning is performed according to the valid target frequency to obtain a sound field image of the fault source.

[0048] Further, the performing spectrum analysis on the best quality sound signal acquired according to the best acquisition angle, the best acquisition distance, and the best acquisition height to acquire the target frequency includes:

[0049] Acquiring an acoustic signal spectrum of the optimal acoustic signal;

[0050] The frequency domain analysis is performed on the spectrum of the acoustic signal to determine the frequency with the largest amplitude in the frequency domain as the target frequency.

[0051] Further, determining the fault source location based on the sound field image and the video image includes:

[0052] Fusing the video image with the sound field image to obtain a sound field image;

[0053] The fault source location is determined based on the sound field imaging.

[0054] Compared with the prior art, the beneficial effect of the present invention lies in that, through systematic sound information collection and gradual adjustment of collection parameters, the sound array collects sound signals at different angles, distances and heights, ensuring comprehensive coverage and high-sensitivity detection of the object to be identified, ensuring that the high-quality sound signal collected at the best collection point found can reflect the characteristics of the fault source to the greatest extent, thereby improving the positioning accuracy, and combining video images and sound field images to intuitively display the location of the fault source in space, enhancing the reliability and intuitiveness of positioning, and being suitable for automated monitoring in complex environments, so that staff can maintain and replace equipment as early as possible to avoid causing greater losses.

[0055] Furthermore, through the preset step size and period, the angle adjustment process can be effectively controlled to ensure a comprehensive search in the angle dimension and avoid missing the optimal angle. The parameters can be adjusted according to the actual situation to adapt to different application scenarios.

[0056] Furthermore, the characteristic changes of the acoustic signal are quantified through characteristic parameters and differences, which is convenient for comparison and optimization. By comparing the differences and paying attention to the changing trend of the signal, the angle where the signal characteristic changes most obviously can be selected, and the optimal angle can be selected quickly and effectively, thereby improving the accuracy and reliability of fault source locating.

[0057] Furthermore, by setting the distance adjustment step, different collection distances can be accurately controlled, and dynamic adaptation can be made according to the sound field characteristics and environmental changes of the object to be identified, ensuring that the best sound signal can always be obtained under different working conditions, thereby improving flexibility.

[0058] Furthermore, by obtaining the second characteristic parameter of the second acoustic signal and comparing it with the characteristic parameter of the first acoustic signal, changes in the acoustic signal can be effectively identified. By maximizing the second difference, it is ensured that during the acoustic signal acquisition process, the selected distance can reflect the acoustic characteristics of the object to be identified to the greatest extent, thereby improving the effectiveness of signal acquisition and helping to better capture the acoustic signal characteristics related to the fault, thereby improving the accuracy of fault diagnosis.

[0059] Furthermore, by setting the height adjustment step, precise control of different collection heights can be achieved during the acoustic signal collection process, thereby improving the accuracy of signal collection. Through careful height adjustment, the acoustic characteristics of the object to be identified at different heights can be more comprehensively captured, thereby enhancing the ability to identify potential fault signals and improving the accuracy and reliability of fault source locating.

[0060] Furthermore, by accurately determining the optimal collection height, the acoustic signal characteristics related to the fault can be acquired more quickly. By optimizing the collection height, the quality of the collected acoustic signal is ensured, thereby significantly improving the accuracy, efficiency and degree of automation of fault source location, thereby accelerating the speed of fault diagnosis, which is beneficial to the maintenance and management of rotating machinery and equipment.

[0061] Furthermore, by using a bandpass filter to filter out invalid frequencies in the sound signal, only the signal part of the target frequency that is effective for fault location is retained, which effectively improves the frequency selectivity and enhances the signal clarity of the target frequency. This enables the system to more accurately focus on the frequency components related to the fault source, avoiding misidentification caused by invalid frequency interference, making subsequent fault diagnosis more intuitive and accurate, thereby improving work efficiency.

[0062] Furthermore, by performing frequency domain analysis on the acoustic signal spectrum at the optimal collection point, the frequency with the largest amplitude can be accurately extracted from the complex signal as the most representative target frequency. This process ensures that the system can automatically identify the frequency among many frequency components, thereby effectively locating the key signals related to the fault and avoiding interference from irrelevant frequencies, thereby improving the accuracy and reliability of fault source location.

[0063] Furthermore, by fusing the sound field image with the video image, the spatial distribution of the acoustic signal can be effectively combined with the physical position in the visual image, so that the location of the fault source can be quickly and accurately calibrated in the video image, thereby improving the spatial accuracy of fault source positioning. The operator can intuitively see the location of the fault source and make maintenance decisions quickly, greatly improving the efficiency and timeliness of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1This is a flow chart of a method for locating a rotating machinery fault source based on an optimal acquisition point of a mobile robot according to an embodiment of the present invention;

[0065] Figure 2 A decision logic diagram for determining the optimal acquisition point for an embodiment of the present invention;

[0066] Figure 3 A flow chart for determining the best acquisition angle for an embodiment of the present invention;

[0067] Figure 4 The present invention is a flowchart for determining the optimal acquisition distance according to an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0070] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0071] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0072] See also Figure 1 ,like Figure 1 As shown, it is a flow chart of a method for locating a rotating machinery fault source based on an optimal collection point of a mobile robot provided by an embodiment of the present invention;

[0073] Specifically, an embodiment of the present invention provides a method for locating a rotating machinery fault source based on an optimal collection point of a mobile robot, comprising:

[0074] Step S1, using an acoustic array to collect sound information of an object to be identified according to preset initial collection parameters to obtain an initial sound signal;

[0075] The acquisition parameters include an initial acquisition angle, an initial acquisition distance and an initial acquisition height;

[0076] Step S2, gradually adjusting the initial acquisition angle according to a preset angle adjustment step length to obtain an adjusted acquisition angle;

[0077] Acquire a first acoustic signal of the object to be identified based on the adjusted acquisition angle, the initial acquisition distance, and the initial acquisition height, and determine an optimal acquisition angle based on the initial acoustic signal and the first acoustic signal;

[0078] Step S3, gradually adjusting the initial acquisition distance according to a preset distance adjustment step to obtain an adjusted acquisition distance;

[0079] Acquire a second acoustic signal of the object to be identified based on the optimal acquisition angle, the adjusted acquisition distance, and the initial acquisition height, and determine an optimal acquisition distance based on the first acoustic signal and the second acoustic signal;

[0080] Step S5, gradually adjusting the initial acquisition height according to a preset height adjustment step length to obtain an adjusted acquisition height;

[0081] Acquire a third acoustic signal of the object to be identified based on the optimal acquisition angle, the optimal acquisition distance, and the adjusted acquisition height, and determine an optimal acquisition height based on the second acoustic signal and the third acoustic signal;

[0082] Perform spectrum analysis on the best quality sound signal obtained from the optimal collection point according to the best collection angle, the best collection distance, and the best collection height to obtain the target frequency;

[0083] Fixed bandwidth sound source localization;

[0084] Step S6, acquiring the sound field image of the target frequency;

[0085] Acquire a video image of the object to be identified;

[0086] The fault source location is determined based on the sound field image and the video image.

[0087] The object to be identified is a rotating mechanical device on site in a factory workshop, such as a water pump, a compressor, a motor and other driving devices, in this embodiment, it is a motor. The acoustic array is a device equipped by the robot for obtaining sound signals, which can be different types of microphone arrays, as long as it can collect real-time sound field data. In this embodiment, it is a 16-element MEMS digital silicon microphone rectangular array UMA-16v2. The video image of the object to be identified can be obtained by various cameras, as long as it can collect video images of the scene in real time. In this embodiment, it is an IMX294USB camera. The initial acoustic signal is the acoustic signal obtained by the acoustic array according to the initial acquisition parameters. The first acoustic signal is the acoustic signal collected according to the optimal acquisition angle. The best sound signal collected is the best sound signal collected according to the best collection angle and the best collection distance. The third sound signal is the best sound signal collected according to the best collection angle, the best collection distance and the best collection height. The initial collection angle is the initial angle between the sound collection direction of the sound array and the object to be identified, which is generally set between 0°-30°, and is set to 0° in this embodiment. The initial collection distance is the initial distance between the sound collection point of the sound array and the object to be identified, which is generally set between 0.5-1m, and is set to 0.5m in this embodiment. The initial collection height is the height difference between the sound collection point of the sound array and the object to be identified, which is generally set between 1-1.5m, and the initial collection height is 1m in this embodiment.

[0088] The optimal acquisition angle is the angle between the sound array and the object to be identified where the sound signal quality is the best. The optimal acquisition distance is the distance between the sound array and the object to be identified where the sound signal quality is the best. The optimal acquisition height is the height between the sound array and the object to be identified where the sound signal quality is the best. The adjusted acquisition angle is the amplitude of adjusting the angle between the sound array and the object to be identified each time, which is generally set between 0-60°, and is set to 45° in this embodiment. The adjusted acquisition distance is the amplitude of adjusting the distance between the sound array and the object to be identified each time, which is generally set between 0m-1m, and is set to 0.5m in this embodiment. The adjusted acquisition height is the amplitude of adjusting the height between the sound array and the object to be identified each time, which is generally set between 0m-0.5m, and is set to 0.25m in this embodiment. The first characteristic parameter, the second characteristic parameter, and the third characteristic parameter refer to parameters such as signal-to-noise ratio and fault characteristic amplitude ratio that can determine the quality of the sound signal. In this embodiment, they refer to four sound signal quality evaluation indicators, namely root mean square, sound pressure level, signal-to-noise ratio, and fault characteristic amplitude ratio.

[0089] The root mean square value, also called the effective value, represents the energy of the sound signal generated by the rotating machinery. When the bearing is not faulty, it runs smoothly, the sound is small, and the corresponding RMS value is relatively small. As the fault gradually worsens, the RMS will also increase. As shown in formula (1):

[0090]

[0091] Sound pressure level, measured as "effective sound pressure" in logarithm, describes the overall noise level of the sound source. The reference sound pressure Pref is the minimum sound pressure value audible to the human ear at 1000Hz, 20μPa. The greater the noise of the detected sound source, the greater the total sound pressure level. As shown in formula (2):

[0092]

[0093] The signal-to-noise ratio, which represents the ratio of the useful signal power to the background noise power, is an important parameter for measuring signal quality. Ps is the effective signal power and Pn is the noise signal power. A high SNR value means good signal quality and little interference from noise on the signal. A low SNR value indicates that there are more noise components in the signal and the signal quality is poor. As shown in formula (3):

[0094]

[0095] The fault characteristic amplitude ratio can quantitatively evaluate the fault information in the signal envelope spectrum. Fcf is the theoretically calculated fault characteristic frequency, i is the i-th harmonic, l is the number of harmonics to be analyzed, df is the frequency resolution, Fi indicates the deviation between the theoretical Fcf and the actual Fcf due to the unstable gear speed, and the deviation range is [Fcf-4×df, Fcf+4×df], Hes represents the Hilbert envelope spectrum; the higher the RFCA value, the more bearing fault diagnosis information is contained in the spectrum. As shown in formula (4):

[0096]

[0097] Specifically, the three determinations of the acquisition parameters of the best acquisition point, namely, determining the best acquisition angle, determining the best acquisition distance, and determining the best acquisition height, can be any combination according to actual needs, and the order can be changed. In the present embodiment, the determination order is to determine the best acquisition angle, determine the best acquisition distance, and determine the best acquisition height. First, the robot is controlled to move and gradually adjust the acquisition angle, acquisition distance, and acquisition height to determine the best acquisition angle, the best acquisition distance, and the best acquisition height to obtain a position that is best for acquiring the acoustic signal of the object to be identified. Then, high-quality acoustic signals of the object to be identified are acquired, and frequency domain analysis is performed to determine the target frequency with the most obvious difference, and the sound source of the target frequency is located. The sound source is further located, and the acquired video stream image is fused into sound field imaging to intuitively analyze the actual position of the fault source.

[0098] Specifically, through systematic sound information collection and gradual adjustment of collection parameters, the acoustic array collects sound signals at different angles, distances and heights, ensuring comprehensive coverage and high-sensitivity detection of the objects to be identified, and ensuring that the collected sound signals can reflect the characteristics of the fault source to the greatest extent, thereby improving positioning accuracy. Combined with video images and sound field images, the location of the fault source can be intuitively displayed in space, enhancing the reliability and intuitiveness of positioning. It is suitable for automated monitoring in complex environments, and is convenient for staff to maintain and replace equipment as early as possible to avoid greater losses.

[0099] Please continue reading Figure 2 ,like Figure 2 As shown, it is a flow chart of determining the optimal acquisition angle according to an embodiment of the present invention;

[0100] Specifically, the step of gradually adjusting the initial acquisition angle according to a preset angle adjustment step to obtain an adjusted acquisition angle includes:

[0101] Set the angle adjustment step size;

[0102] The initial acquisition angle is gradually increased according to the angle adjustment step size and an adjustment period to obtain the adjusted acquisition angle.

[0103] Specifically, the adjustment period is the interval time for the robot to make movement adjustments, which is generally set at 0-3s. In this embodiment, the adjustment period is 2s.

[0104] In the specific implementation process, the fault source of the motor with a speed of 2400r / min and a load of 0% was located. The initial acquisition angle was 0°, and the angle adjustment step was 45°. The acoustic signal of the target to be identified was initially acquired at the acquisition angle of 0°. After that, the acquisition angle was increased by 45° every 2s to obtain a new adjusted acquisition angle. The number of adjustment steps was 3 times, and the acquisition time at the initial angle and each adjusted angle was 5s.

[0105] Specifically, through the preset step size and period, the angle adjustment process can be effectively controlled to ensure a comprehensive search in the angle dimension, avoid missing the optimal angle, and adjust the parameters according to actual conditions to adapt to different application scenarios.

[0106] Specifically, determining the optimal acquisition angle based on the initial acoustic signal and the first acoustic signal includes:

[0107] Acquiring a first characteristic parameter of the first acoustic signal and an initial characteristic parameter of an initial acoustic signal;

[0108] Calculate the difference between each of the first characteristic parameters and the corresponding initial characteristic parameters to obtain a first difference value;

[0109] Calculating a ratio of the first difference value to the corresponding initial characteristic parameter to obtain a first difference ratio, and calculating a sum of the first difference ratios to obtain a total first difference ratio;

[0110] By comparing the plurality of total first difference ratios, the adjusted collection angle of the first acoustic signal when the total first difference ratio is the largest is determined as the optimal collection angle.

[0111] Table 1 Characteristic parameter values ​​at different angles

[0112]

[0113]

[0114] Specifically, the first characteristic parameter (the four indicator values ​​of the characteristic parameters corresponding to 45°, 90° and 135°) is respectively subtracted from the initial characteristic parameter (the four indicator values ​​of the characteristic parameters corresponding to 0°) to obtain the respective differences, and then the respective difference ratios are added to the indicator values ​​corresponding to the initial characteristic parameters to obtain the respective difference ratios. The difference ratios of the four indicators are then summed to obtain the total first difference ratio of the three adjusted acquisition angles, i.e. the above-mentioned total first difference ratio. As shown in Table 1, the specific characteristic parameter values, differences and difference ratios for different angles. The SNR values ​​in the table are only linear ratios, and the logarithm is not taken. This is convenient for calculating the difference ratio.

[0115] Compare the three total first difference ratios, and determine the adjusted collection angle of the first acoustic signal of 0° when the total first difference ratio is the largest as the optimal collection angle. (When 0°, the first difference ratio is 0, and the other three angles are all negative values, so the signal quality is worse than the initial angle) In the specific implementation process, according to the initial collection angle of 0°, the initial collection distance of 0.5m, and the initial collection height of 1m, the initial characteristic parameters of the initial acoustic signal collected are RMS0=0.295Pa, SPL0=83.025dB, SNR0=0.672, and RFCA0=4.475. The specific index values ​​of the adjustment parameters for adjusting the collection angles to 45°, 90°, and 135° after adjustment are shown in Table 1. Then, the first difference ratios are determined to be -23.05%, -73.62%, and -118.08%, and the optimal collection angle is determined to be 0°.

[0116] Specifically, the characteristic changes of the acoustic signal are quantified through characteristic parameters and differences, which is convenient for comparison and optimization. By comparing the differences, we can pay attention to the changing trend of the signal and make it possible to quickly and effectively select the optimal angle at which the signal characteristic changes most obviously, thereby improving the accuracy and reliability of fault source locating.

[0117] Please continue reading Figure 3 ,like Figure 3As shown, it is a flow chart of determining the optimal acquisition distance according to an embodiment of the present invention;

[0118] Specifically, the step of gradually adjusting the initial acquisition distance according to a preset distance adjustment step to obtain an adjusted acquisition distance includes:

[0119] Set the distance adjustment step size;

[0120] The initial acquisition distance is gradually increased according to the distance adjustment step size and an adjustment period to obtain the adjusted acquisition distance.

[0121] In the specific implementation process, the initial acquisition distance is 0.5m, the distance adjustment step is 0.5m, and the acoustic signal of the target to be identified is initially acquired at a 0.5m acquisition distance. After that, the acquisition distance is increased by 0.5m every 2s to obtain a new adjusted acquisition distance. The number of adjustment steps is 3 times, and the acquisition time at the initial distance and each adjusted distance is 5s.

[0122] Specifically, by setting the distance adjustment step, the initial collection distance can be accurately controlled, and it can dynamically adapt according to the sound field characteristics and environmental changes of the object to be identified, ensuring that the best sound signal can always be obtained under different working conditions, thereby improving flexibility.

[0123] Specifically, determining the optimal collection distance based on the first acoustic signal and the second acoustic signal includes:

[0124] Acquiring a second characteristic parameter of the second acoustic signal;

[0125] Calculating the difference between each of the second characteristic parameters and each of the first characteristic parameters to obtain a second difference value;

[0126] Calculating a ratio of the second difference value to the corresponding first characteristic parameter to obtain a second difference ratio, and calculating a sum of the second difference ratios to obtain a total second difference ratio;

[0127] A plurality of the total second difference ratios are compared, and the adjusted collection distance of the second sound signal when the total second difference ratio is the largest is determined as the optimal collection distance.

[0128] Table 2 Characteristic parameter values ​​at different distances

[0129]

[0130] Specifically, the second characteristic parameter (the four indicator values ​​of the characteristic parameters corresponding to 1m, 1.5m and 2m) is respectively subtracted from the first characteristic parameter (the four indicator values ​​of the characteristic parameters corresponding to 0° and 0.5m, which are slightly different from the indicator values ​​of the first characteristic parameters in Table 1 due to the complexity of the actual sound field and the measurement error) to obtain their respective differences, and then the respective difference ratios are added to the indicator values ​​corresponding to the first characteristic parameters to obtain their respective difference ratios. The difference ratios of the four indicators are then summed to obtain the total second difference ratios of the three adjusted acquisition distances, i.e. the above-mentioned total second difference ratios. As shown in Table 2, the specific characteristic parameter values, differences and difference ratios for different distances are shown. The SNR values ​​in the table are only linear ratios, and no logarithm is taken. This facilitates the calculation of the difference ratio.

[0131] The three total second difference ratios are compared to determine the adjusted collection distance of the second sound signal of 1.5 m when the total second difference ratio is the largest as the optimal collection distance.

[0132] In the specific implementation process, according to the optimal collection angle of 0°, the initial collection distance of 0.5m, and the initial collection height of 1m, the first characteristic parameters of the sound signal collected are RMS1=0.27Pa, SPL1=82.825dB, SNR1=0.507, and RFCA1=4.425. The specific index values ​​of the adjustment parameters for the adjusted collection distances of 1m, 1.5m, and 2m are shown in Table 2. The second difference ratios are determined to be -89.11%, -10.91%, and -108.19%, and the optimal collection distance is determined to be 1.5m.

[0133] Specifically, by obtaining the second characteristic parameter of the second acoustic signal and comparing it with the characteristic parameter of the first acoustic signal, changes in the acoustic signal can be effectively identified. By maximizing the second difference ratio, it is ensured that during the acoustic signal acquisition process, the selected distance can reflect the acoustic characteristics of the object to be identified to the greatest extent, thereby improving the effectiveness of signal acquisition and helping to better capture the acoustic signal characteristics related to the fault, thereby improving the accuracy of fault diagnosis.

[0134] Please continue reading Figure 4 ,like Figure 4 As shown, it is a flow chart of determining the optimal acquisition height according to an embodiment of the present invention;

[0135] Specifically, the step of gradually adjusting the initial acquisition height according to a preset height adjustment step to obtain an adjusted acquisition height includes:

[0136] Set the height adjustment step size;

[0137] The initial acquisition height is gradually increased according to the height adjustment step length and in accordance with an adjustment period to obtain the adjusted acquisition height.

[0138] In the specific implementation process, the initial acquisition height is 1m, the height adjustment step is 0.25m, and the acoustic signal of the target to be identified is initially acquired at the acquisition height of 1m. After that, the acquisition height is increased by 0.25m every 2s to obtain a new adjusted acquisition height. The number of adjustment steps is 3 times, and the acquisition time at the initial height and each adjusted height is 5s.

[0139] Specifically, by setting the height adjustment step, the initial collection height can be accurately controlled during the acoustic signal collection process, thereby improving the accuracy of signal collection. Through meticulous height adjustment, the acoustic characteristics of the object to be identified at different heights can be more comprehensively captured, thereby enhancing the ability to identify potential fault signals and improving the accuracy and reliability of fault source locating.

[0140] Specifically, determining the optimal collection height based on the second acoustic signal and the third acoustic signal includes:

[0141] Acquiring a third characteristic parameter of the third acoustic signal;

[0142] Calculating the difference between each of the third characteristic parameters and each of the second characteristic parameters to obtain a third difference;

[0143] Calculating a ratio of the third difference ratio to the corresponding second characteristic parameter to obtain a third difference ratio, and calculating a sum of the third difference ratios to obtain a total third difference ratio;

[0144] By comparing a plurality of the total third difference ratios, the adjusted collection height of the third acoustic signal when the total third difference ratio is the largest is determined as the optimal collection height.

[0145] Table 3 Characteristic parameter values ​​at different heights

[0146]

[0147] Specifically, the third characteristic parameter (the four indicator values ​​of the characteristic parameters corresponding to 1.25m, 1.5m and 1.75m) is respectively subtracted from the second characteristic parameter (the four indicator values ​​of the characteristic parameters corresponding to 1m, which are slightly different from the indicator values ​​of the second characteristic parameters in Table 2 due to the complexity of the actual sound field and the measurement error) to obtain their respective differences, and then the respective difference ratios are added to the indicator values ​​corresponding to the second characteristic parameters to obtain their respective difference ratios. The difference ratios of the four indicators are then summed to obtain the total third difference ratio of the three adjusted acquisition heights, i.e. the above-mentioned total third difference ratio. As shown in Table 3, these are the specific characteristic parameter values, differences and difference ratios for different heights. The SNR values ​​in the table are only linear ratios, and no logarithm is taken. This facilitates the calculation of the difference ratio.

[0148] By comparing several total third difference ratios, it is determined that the adjusted collection height of the third sound signal of 1.25 m when the total third difference ratio is the largest is the optimal collection height.

[0149] In the specific implementation process, according to the best acquisition angle of 0°, the best acquisition distance of 1.5m, and the initial acquisition height of 1m, the second characteristic parameters of the acoustic signal collected are RMS2=0.2575Pa, SPL2=82.425dB, SNR2=0.589, and RFCA2=4.9. The specific index values ​​of the adjustment parameters for the adjusted acquisition heights of 1.25m, 1.5m, and 1.75m are shown in Table 3. The third difference ratios are determined to be 68.47%, 22.11%, and -44.5%, and the best acquisition height is determined to be 1.25m.

[0150] Specifically, by accurately determining the optimal collection height, the acoustic signal characteristics related to the fault can be acquired more quickly. By optimizing the collection height, the quality of the collected acoustic signal is ensured, thereby significantly improving the accuracy, efficiency and degree of automation of fault source locating, thereby accelerating the speed of fault diagnosis, which is beneficial to the maintenance and management of rotating machinery and equipment.

[0151] Specifically, acquiring the sound field image of the target frequency includes:

[0152] A bandpass filter is used to filter out invalid frequencies in the target frequency to obtain a valid target frequency, and sound source positioning is performed according to the valid target frequency to obtain a sound field image of the fault source.

[0153] Specifically, the bandpass filter is a filter that can filter out invalid frequency signals except the effective target frequency. It can be an elliptical bandpass filter, a Butterworth bandpass filter, etc. The specific filter is not limited here. In this embodiment, it is a Butterworth bandpass filter, and the sound field information of the effective target frequency is imaged by the sound field imaging function to obtain a sound field image.

[0154] Specifically, by using a bandpass filter to filter out invalid frequencies in the sound signal, only the signal part of the target frequency that is effective for fault location is retained, which effectively improves the frequency selectivity and enhances the signal clarity of the target frequency. This enables the system to more accurately focus on the frequency components related to the fault source, avoiding misidentification caused by invalid frequency interference, making subsequent fault diagnosis more intuitive and accurate, thereby improving work efficiency.

[0155] Specifically, performing spectrum analysis on the best quality sound signal acquired according to the best acquisition angle, the best acquisition distance, and the best acquisition height to acquire the target frequency includes:

[0156] Acquiring an acoustic signal spectrum of the optimal acoustic signal;

[0157] The frequency domain analysis is performed on the spectrum of the acoustic signal to determine the frequency with the largest amplitude in the frequency domain as the target frequency.

[0158] In a specific implementation process, the amplitude of the 12-fold frequency among the frequencies of the acoustic signal of the best acoustic signal is the largest, and the 12-fold frequency of the object to be identified is used as the target frequency.

[0159] Specifically, by performing frequency domain analysis on the acoustic signal spectrum of the optimal acoustic signal, the frequency with the largest amplitude can be accurately extracted from the complex signal as the most representative target frequency. This process ensures that the system can automatically identify the frequency among many frequency components, thereby effectively locating the key signals related to the fault and avoiding interference from irrelevant frequencies, thereby improving the accuracy and reliability of fault source diagnosis.

[0160] Specifically, determining the fault source location based on the sound field image and the video image includes:

[0161] Fusing the video image with the sound field image to obtain a sound field image;

[0162] The fault source location is determined based on the sound field imaging.

[0163] Specifically, the sound field image of the target frequency is determined, the video image is fused with the sound field image, and the actual position of the sound source located by the sound field image in the video image is determined, which is the fault source position.

[0164] Specifically, by fusing the sound field image with the video image, the spatial distribution of the acoustic signal can be effectively combined with the physical position in the visual image, so that the location of the fault source can be quickly and accurately calibrated in the video image, thereby improving the spatial accuracy of fault source locating. Operators can intuitively see the location of the fault source and make maintenance decisions quickly, greatly improving the efficiency of fault diagnosis.

[0165] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot, characterized in that: include: Using an acoustic array to collect sound information of the object to be identified according to preset initial acquisition parameters to obtain an initial acoustic signal; The acquisition parameters include an initial acquisition angle, an initial acquisition distance and an initial acquisition height; The initial acquisition angle is gradually adjusted according to a preset angle adjustment step to obtain an adjusted acquisition angle; Acquire a first acoustic signal of the object to be identified based on the adjusted acquisition angle, the initial acquisition distance, and the initial acquisition height, and determine an optimal acquisition angle based on the initial acoustic signal and the first acoustic signal; The initial acquisition distance is gradually adjusted according to a preset distance adjustment step to obtain an adjusted acquisition distance; Acquire a second acoustic signal of the object to be identified based on the optimal acquisition angle, the adjusted acquisition distance, and the initial acquisition height, and determine an optimal acquisition distance based on the first acoustic signal and the second acoustic signal; The initial acquisition height is gradually adjusted according to a preset height adjustment step length to obtain an adjusted acquisition height; Acquire a third acoustic signal of the object to be identified based on the optimal acquisition angle, the optimal acquisition distance, and the adjusted acquisition height, and determine an optimal acquisition height based on the second acoustic signal and the third acoustic signal; Performing spectrum analysis on the best quality sound signal acquired according to the best acquisition angle, the best acquisition distance, and the best acquisition height to obtain the target frequency; Fixed bandwidth sound source localization; Acquiring a sound field image of the target frequency; Acquire a video image of the object to be identified; The fault source location is determined based on the sound field image and the video image.

2. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 1 is characterized in that: The step of gradually adjusting the initial acquisition angle according to a preset angle adjustment step to obtain an adjusted acquisition angle includes: Set the angle adjustment step size; The initial acquisition angle is gradually increased according to the angle adjustment step size and an adjustment period to obtain the adjusted acquisition angle.

3. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 2 is characterized in that: The determining of the optimal acquisition angle based on the initial acoustic signal and the first acoustic signal comprises: Acquiring a first characteristic parameter of the first acoustic signal and an initial characteristic parameter of an initial acoustic signal; Calculate the difference between each of the first characteristic parameters and the corresponding initial characteristic parameters to obtain a first difference value; Calculating a ratio of the first difference value to the corresponding initial characteristic parameter to obtain a first difference ratio, and calculating a sum of the first difference ratios to obtain a total first difference ratio; By comparing the plurality of total first difference ratios, the adjusted collection angle of the first acoustic signal when the total first difference ratio is the largest is determined as the optimal collection angle.

4. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 3 is characterized in that: The step of gradually adjusting the initial acquisition distance according to a preset distance adjustment step to obtain an adjusted acquisition distance includes: Set the distance adjustment step size; The initial acquisition distance is gradually increased according to the distance adjustment step size and an adjustment period to obtain the adjusted acquisition distance.

5. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 4 is characterized in that: The determining the optimal acquisition distance based on the first acoustic signal and the second acoustic signal comprises: Acquiring a second characteristic parameter of the second acoustic signal; Calculating the difference between each of the second characteristic parameters and each of the first characteristic parameters to obtain a second difference value; Calculating a ratio of the second difference value to the corresponding first characteristic parameter to obtain a second difference ratio, and calculating a sum of the second difference ratios to obtain a total second difference ratio; A plurality of the total second difference ratios are compared, and the adjusted collection distance of the second sound signal when the total second difference ratio is the largest is determined as the optimal collection distance.

6. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 5, characterized in that: The step of gradually adjusting the initial acquisition height according to a preset height adjustment step to obtain an adjusted acquisition height comprises: Set the height adjustment step size; The initial acquisition height is gradually increased according to the height adjustment step length and in accordance with an adjustment period to obtain the adjusted acquisition height.

7. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 6, characterized in that: The determining the optimal acquisition height based on the second acoustic signal and the third acoustic signal comprises: Acquiring a third characteristic parameter of the third acoustic signal; Calculating the difference between each of the third characteristic parameters and each of the second characteristic parameters to obtain a third difference; Calculating a ratio of the third difference ratio to the corresponding second characteristic parameter to obtain a third difference ratio, and calculating a sum of the third difference ratios to obtain a total third difference ratio; By comparing a plurality of the total third difference ratios, the adjusted collection height of the third acoustic signal when the total third difference ratio is the largest is determined as the optimal collection height.

8. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 7, characterized in that: The acquiring the sound field image of the target frequency comprises: A bandpass filter is used to filter out invalid frequencies in the target frequency to obtain a valid target frequency, and sound source positioning is performed according to the valid target frequency to obtain a sound field image of the fault source.

9. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 8, characterized in that: The performing spectrum analysis on the best quality sound signal acquired according to the best acquisition angle, the best acquisition distance, and the best acquisition height to acquire the target frequency comprises: Acquiring an acoustic signal spectrum of the optimal acoustic signal; Perform frequency domain analysis on the spectrum of the acoustic signal, and determine the frequency with the largest amplitude in the frequency domain as the target frequency.

10. The method for locating the fault source of rotating machinery based on the optimal acquisition point of a mobile robot according to claim 9, characterized in that: Determining the fault source location based on the sound field image and the video image includes: Fusing the video image with the sound field image to obtain a sound field image; The fault source location is determined according to the sound field imaging.

Images