Operation status monitoring method of microparticle shot blasting equipment based on multi-source data fusion

Abstract

The present application relates to the field of data processing technology, and in particular to a method for monitoring the operating status of a microparticle shot blasting device based on multi-source data fusion; the method comprises the following steps: fusing the same frequency bands in multiple abnormal spectra under the same abnormal state to obtain a comprehensive spectrum, and connecting the comprehensive spectra of multiple frequency bands to form an abnormal fusion spectrum; real-time acquisition of multiple characteristic signals during the operation of the device and obtaining a real-time spectrum; fusing the frequency bands of multiple real-time spectra to obtain a real-time fusion spectrum; comparing the differences between each abnormal fusion spectrum and the real-time fusion spectrum to obtain frequency domain similarity; calculating an anomaly score based on the frequency domain similarity, and judging the operating status of the device based on the anomaly score. The present application has the effect of improving the accuracy of detecting the operating status of the device.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method for monitoring the operating status of a microparticle shot blasting device based on multi-source data fusion. Background Art

[0002] Microparticle shot peening is similar to traditional shot peening, except that smaller-diameter projectiles are used to impact the surface of a component at high speed. Using microparticle shot peening equipment to treat gears results in higher surface hardness and less plastic deformation. Microparticle shot peening can also reduce surface roughness. Microparticle shot peening is primarily used in the processing and manufacturing of high-precision workpieces, such as gears. The operational stability of microparticle shot peening equipment is crucial for successful microparticle peening. Monitoring the equipment's operating status is crucial to promptly detect equipment failures and prevent serious damage. Mechanical equipment emits regular sounds during operation, and when equipment fails, it generates noise. Therefore, the operating status of the equipment can be monitored based on the acoustic signals generated during operation. Acoustic testing involves collecting the sound signals of the equipment under test using an acoustic sensor, transmitting these signals via a network to a central database, and comparing them with pre-stored standard sound signals. The equipment's operating status is determined based on differences in characteristics such as frequency or amplitude.

[0003] However, there is no single cause of failure during the operation of microparticle shot blasting equipment. There may be many types of failures, such as loose screws, worn parts, cavitation explosions, and tiny cracks in storage components. Traditional acoustic detection technology is mainly based on a single acoustic signal, and the sensitive frequency band of the acoustic signal is usually between 50kHz and 2MHz. Although this type of method can capture high-frequency abnormal signals (such as cavitation effects) during actual detection, it has low sensitivity for detecting low-frequency fault signals (such as slight cracks and valve leaks). It may even ignore low-frequency fault signals, which accumulate over time and eventually lead to serious damage to the equipment. Therefore, the equipment detection methods in related technologies are prone to missed detection and misjudgment during application. Summary of the Invention

[0004] In order to solve the problem of false detection and missed judgment in the equipment operation detection process in the related art, the present application provides a method for monitoring the operation status of microparticle shot blasting equipment based on multi-source data fusion.

[0005] This application provides a method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion, which adopts the following technical solutions:

[0006] A method for monitoring the operating status of a microparticle shot blasting device based on multi-source data fusion includes the following steps: obtaining a plurality of characteristic signals of the device under normal operating conditions and under a plurality of abnormal conditions, wherein the characteristic signals include at least two of acoustic signals, pressure signals, vibration signals, and flow signals; obtaining the frequency spectrum of the characteristic signal under each abnormal condition to obtain an abnormal frequency spectrum, and obtaining the frequency spectrum of the characteristic signal under a normal condition to obtain a normal frequency spectrum; dividing the frequency bands, calculating the frequency band differences between the abnormal frequency spectrum and the normal frequency spectrum in different frequency bands, and determining the fusion weights of different characteristic signals under each abnormal condition in each frequency band based on the frequency band differences;

[0007] Based on the fusion weight, the same frequency bands in multiple abnormal spectra under the same abnormal state are fused to obtain a comprehensive spectrum, and the comprehensive spectrum of multiple frequency bands is connected to form an abnormal fused spectrum; multiple characteristic signals during the operation of the equipment are collected in real time and a real-time spectrum is obtained; the frequency bands of multiple real-time spectra are fused to obtain a real-time fused spectrum; the differences between each abnormal fused spectrum and the real-time fused spectrum are compared to obtain frequency domain similarity; the anomaly score is calculated based on the frequency domain similarity, and the equipment operation status is judged based on the anomaly score.

[0008] The beneficial effect is that there may be multiple abnormal states during the operation of the equipment, and the signals of different abnormal states usually show high-frequency or low-frequency characteristics. Therefore, in this application, multiple characteristic signals are obtained under different abnormal states. Different characteristic signals are collected by different sensors. By comparing the differences between different characteristic signals and normal characteristic signals in different frequency bands under the same abnormal state, the importance of different characteristic signals in different frequency bands to the abnormality detection of the equipment can be determined, and the fusion weights corresponding to each frequency band can be obtained. Based on the fusion weight, the abnormal spectra of multiple characteristic signals collected under the same abnormal state are fused to obtain an abnormal fusion spectrum. According to the frequency domain similarity between the real-time fusion spectrum of multiple characteristic signals collected in real time and the abnormal fusion spectrum, the operating status of the equipment can be detected.

[0009] In this method, the importance of different frequency bands under different abnormal conditions is analyzed. For example, when some mechanical failures occur in the low-frequency band, the greater the difference between the abnormal spectrum and the normal spectrum of a certain characteristic signal in the low-frequency band, the more important this characteristic signal and this frequency band are for equipment abnormality detection. It also shows that the characteristic signal is more sensitive to low-frequency changes, reducing the occurrence of low-frequency fault information omissions and improving the accuracy of equipment operation fault detection.

[0010] Optionally, the step of calculating the frequency band difference between the abnormal spectrum and the normal spectrum in different frequency bands includes: for each abnormal spectrum in each abnormal state, obtaining the absolute value of the difference between the elements of the same order in the same frequency band of the abnormal spectrum and the normal spectrum; and taking the sum of the absolute values of the differences corresponding to all elements in the frequency band as the frequency band difference of the frequency band.

[0011] The beneficial effect is that the difference between each element in the abnormal spectrum and the normal spectrum is obtained and accumulated, thereby obtaining the frequency band difference between a certain frequency band in the abnormal spectrum and the corresponding frequency band in the normal spectrum.

[0012] Optionally, the step of determining the fusion weights of different characteristic signals in each frequency band under each abnormal state based on the frequency band differences includes: taking the sum of the frequency band differences corresponding to different characteristic signals as the total difference, and for each abnormal state, taking the ratio of the frequency band difference of any characteristic signal in any frequency band to the total difference as the fusion weight of the characteristic signal in the frequency band.

[0013] The beneficial effect is: normalizing the frequency band differences to facilitate subsequent frequency band fusion.

[0014] Optionally, the step of fusing the same frequency bands in multiple abnormal spectra under the same abnormal state based on the fusion weight to obtain a comprehensive spectrum includes: taking the product of the fusion weight corresponding to each frequency band and the elements in each frequency band as a local element, and for each abnormal spectrum under each abnormal state, taking the sum of the local elements of the same position in the same frequency band in multiple abnormal spectra as a comprehensive element; multiple comprehensive elements constitute the comprehensive spectrum.

[0015] The beneficial effect is that each abnormal state corresponds to multiple abnormal spectra, each of which corresponds to a fusion weight and multiple frequency bands. The elements of multiple abnormal spectra under the same abnormal state are weighted and summed using the fusion weight, resulting in a comprehensive spectrum containing multiple characteristic information.

[0016] Optionally, in the step of dividing the frequency bands, the spectrum is divided into three segments.

[0017] The beneficial effect is that different sensors have high sensitivity to different frequency bands. Based on this characteristic, the spectrum is divided into multiple frequency bands, and a fusion weight is set for each frequency band, thereby improving the accuracy of the final device detection.

[0018] Optionally, the step of calculating the anomaly score based on the frequency domain similarity includes: obtaining the time domain similarity between the real-time feature signal and the abnormal feature signal; and taking the product of the time domain similarity and the frequency domain similarity as the anomaly score.

[0019] Optionally, the step of calculating the anomaly score based on frequency domain similarity includes: obtaining the time domain similarity between the real-time feature signal and the abnormal feature signal; taking the product of the time domain similarity and the frequency domain similarity as the first control value, and obtaining the integrated spectrum corresponding to the normal feature signal; calculating the second control value based on the similarity between the normal fusion spectrum of the normal feature signal and the normal fusion spectrum of the real-time feature signal, and taking the ratio of the first control value to the second control value as the anomaly score.

[0020] The beneficial effect is that this method not only compares the difference in frequency domain between abnormal characteristic signals and real-time characteristic signals, but also compares the difference between real-time characteristic signals and normal characteristic signals, thereby further improving the accuracy of equipment detection.

[0021] Optionally, the frequency domain similarity calculation step includes: obtaining the absolute value of the difference between the same-order elements in the real-time fusion spectrum and the abnormal fusion spectrum, and taking the sum of the absolute values of the differences corresponding to multiple elements as the adjustment value; The result with the negative value of the adjustment value as the base and the exponent as the frequency domain similarity is taken as the frequency domain similarity.

[0022] Optionally, the time domain similarity between the real-time feature signal and the abnormal feature signal is the average of the Euclidean distances between multiple real-time feature signals and corresponding abnormal feature signals.

[0023] Optionally, the step of determining the operating status of the device based on the anomaly score includes setting an anomaly threshold, and issuing an alarm in response to the anomaly score being greater than the anomaly threshold.

[0024] This application has the following technical effects:

[0025] On the one hand, multiple sensors are used to collect characteristic signals of multiple dimensions. On the other hand, the fusion weights of the characteristic signal frequency bands are determined based on the differences between the characteristic signals and normal signals in different frequency bands under different abnormal conditions. The abnormal characteristic signals are then fused. By comparing the differences in the time and frequency domains between the real-time characteristic signals and the abnormal characteristic signals, it is determined whether the current equipment status is similar to the signal corresponding to a certain abnormal state, thereby realizing the detection of the equipment status. At the same time, multiple sensors are used to comprehensively analyze the low-frequency and high-frequency fault signals of the equipment, reducing the omission of low-frequency fault signals and improving the accuracy of equipment operation status monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of the method for monitoring the operating status of microparticle shot blasting equipment by multi-source data fusion in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The present invention discloses a method for monitoring the operating status of microparticle shot peening equipment based on multi-source data fusion. During equipment operation, the method collects data from multiple dimensions under various abnormal conditions and data from multiple dimensions under normal operating conditions. The method extracts abnormal spectra from the abnormal conditions and normal spectra from the normal operating conditions. The method compares the frequency band differences between the normal and abnormal spectra, and determines the impact of characteristic signals from different dimensions on the frequency band differences under each abnormal condition. This method then determines the fusion weights of the different characteristic signals in different frequency bands under each abnormal condition. Based on the fusion weights, the abnormal spectra of the characteristic signals from multiple dimensions under the abnormal conditions are weightedly fused to produce an abnormal fused spectrum. Similarly, equipment operating information is collected in real time to obtain a real-time fused spectrum. The abnormal fused spectrum is then compared with the real-time fused spectrum, and an anomaly score is calculated to determine the equipment operating status. Different characteristic signals under different abnormal conditions receive different fusion weights. This means that different data are given different attention when determining whether the equipment is in a certain abnormal condition. Therefore, different characteristic signals can be focused on based on the abnormal condition, thereby reducing the tendency to overlook low-frequency signals during anomaly detection and improving equipment detection accuracy.

[0028] Reference Figure 1 The method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion includes steps S1 to S5.

[0029] S1: Acquire multiple characteristic signals of the device in normal operating conditions and multiple abnormal conditions, where the characteristic signals include at least two of the following: acoustic signals, pressure signals, vibration signals, and flow signals.

[0030] Sensors essentially convert signals when the microparticle shot blasting equipment is in various abnormal states (such as bearing wear, tank microcracks, seal leaks, etc.) as well as in normal conditions. Therefore, in this embodiment, sensors are used to collect characteristic signals from multiple dimensions during the equipment's operation. In this embodiment, a flow sensor collects particle flow information, a dynamic pressure sensor collects pressure information at the spray gun, and an acoustic sensor collects sound and vibration information during equipment operation. In other embodiments, two or three of these characteristic signals may be collected.

[0031] After data collection is completed, multiple groups of abnormal characteristic signals corresponding to multiple abnormal states and one normal characteristic signal are obtained. Each group of abnormal characteristic signals includes multiple abnormal characteristic signals corresponding to multiple characteristic signals.

[0032] S2: Obtain the spectrum of the characteristic signal under each abnormal state to obtain the abnormal spectrum, and obtain the spectrum of the characteristic signal under the normal state to obtain the normal spectrum; divide the frequency bands, calculate the frequency band differences between the abnormal spectrum and the normal spectrum in different frequency bands, and determine the fusion weights of different characteristic signals under each abnormal state in each frequency band based on the frequency band differences.

[0033] Performing Fourier transform on the abnormal characteristic signal and the normal characteristic signal to obtain their corresponding frequency spectra, thereby obtaining an abnormal frequency spectrum corresponding to the abnormal characteristic signal and a normal frequency spectrum corresponding to the normal characteristic signal.

[0034] The abnormal spectrum and the normal spectrum are divided based on the frequency to obtain multiple frequency bands. In this embodiment, for any abnormal spectrum or normal spectrum, it is divided into three frequency bands, namely low frequency band, medium frequency band and high frequency band. The frequency corresponding to the low frequency band is: 1Hz-1kHz; the frequency corresponding to the medium frequency band is: 1kHz-50kHz; the frequency corresponding to the high frequency band is: 50kHz-2MHz. The specific number of divided frequency bands and the corresponding frequencies are determined by technical personnel. Taking the division of the spectrum into three frequency bands in this embodiment as an example, the various characteristic signals collected should have different sensitivities in different frequency bands. For example, one of the characteristic signals can produce a large change when a low-frequency fault occurs in the equipment, and another set of characteristic signals can produce a large change when a high-frequency fault occurs in the equipment, etc.

[0035] For different characteristic signals collected under different abnormal conditions, there are certain differences between their corresponding abnormal and normal spectra. Specifically, these can manifest as differences in the low-frequency band, the mid-frequency band, or the high-frequency band. Different sensors have different sensitive frequency bands, so the changes in the same frequency band for different characteristic signals are not consistent. Therefore, the fusion weights of different characteristic signals in different frequency bands can be determined based on the differences in the frequency bands. For example, if the anomaly under a certain abnormal state is mainly manifested in differences in the low-frequency band, the low-frequency band should be paid attention to during the detection process. For example, if the characteristic signal that is sensitive to changes in the low-frequency band is a pressure signal, then when determining whether the device operating status is in this abnormal state, more attention should be paid to the changes in the pressure signal in this frequency band.

[0036] First, the frequency band differences between the abnormal spectrum and the normal spectrum in different frequency bands are calculated; the steps of calculating the frequency band differences include: for each abnormal spectrum under each abnormal state, obtaining the absolute value of the difference between the elements of the same order in the same frequency band of the abnormal spectrum and the normal spectrum; and taking the sum of the absolute values of the differences corresponding to all elements in the frequency band as the frequency band difference of the frequency band.

[0037] Specifically, the calculation formula for the frequency band difference under each abnormal state can be expressed as:

[0038] Where, Indicates that the The abnormal spectrum The frequency band is the same as the normal spectrum Frequency band differences of the frequency bands; Indicates the The abnormal spectrum In the frequency band elements; The normal spectrum In the frequency band elements; Indicates the The amount of data in a frequency band can also be understood as the The length of the frequency band.

[0039] Indicates the difference between abnormal spectrum and normal spectrum The difference between certain elements in a frequency band is calculated by traversing all elements in the frequency band to obtain the overall difference of the frequency band. The larger the difference, the greater the change in the frequency band of the characteristic signal when the corresponding abnormality occurs during device operation. This indicates that more attention should be paid to this frequency band of the characteristic signal when determining whether the device is in an abnormal state.

[0040] Normalize the frequency band differences to obtain the fusion weight of the frequency band; specifically, the calculation formula of the fusion weight is: Where, Indicates the first The abnormal spectrum The fusion weight of each frequency band; Indicates the first The abnormal spectrum Frequency band differences of the frequency bands; Indicates the number of characteristic signal types.

[0041] S3: Based on the fusion weight, the same frequency bands in multiple abnormal spectra under the same abnormal state are fused to obtain a comprehensive spectrum; the comprehensive spectra of multiple frequency bands are connected to form an abnormal fusion spectrum.

[0042] The product of the fusion weight corresponding to each frequency band and the elements in each frequency band is used as the local element. For each abnormal spectrum under each abnormal state, the sum of the local elements of the same position in the same frequency band in multiple abnormal spectra is used as the comprehensive element; multiple comprehensive elements constitute the comprehensive spectrum.

[0043] For any abnormal state, it corresponds to multiple abnormal spectra, which are fused. Each abnormal spectrum corresponds to multiple frequency bands, so the same frequency bands in multiple abnormal spectra are fused first.

[0044] Specifically, the calculation formula of the comprehensive element can be expressed as: Where, Indicates the The integrated spectrum corresponding to the frequency band elements; Indicates the number of abnormal spectra under the same abnormal state; Indicates the first The abnormal spectrum The fusion weight of each frequency band; Indicates the first The abnormal spectrum The first frequency band elements.

[0045] Each element in the frequency band is fused to obtain a comprehensive spectrum, and multiple comprehensive spectra corresponding to multiple frequency bands are connected to obtain an abnormal fused spectrum under a certain abnormal state.

[0046] S4: collecting multiple characteristic signals during the operation of the device in real time and obtaining real-time spectrum of the real-time characteristic signals; fusing frequency bands of multiple real-time spectrums to obtain a real-time fused spectrum.

[0047] During device operation, real-time device operating information is collected and multiple characteristic signals are acquired. The spectrum of the acquired characteristic signals is obtained to create a real-time spectrum, and the multiple real-time spectrums are fused to create a real-time fused spectrum. The acquisition of the real-time fused spectrum is similar to that of the abnormal fused spectrum and will not be further explained here.

[0048] S5: Compare the differences between each abnormal fusion spectrum and the real-time fusion spectrum to obtain frequency domain similarity; calculate the anomaly score based on the frequency domain similarity, and judge the device operation status based on the anomaly score.

[0049] The operating status of the device can be judged based on the difference between the abnormal fusion spectrum and the real-time fusion spectrum.

[0050] Specifically, the calculation formula of frequency domain similarity can be expressed as: Where, Indicates the The frequency domain similarity between the abnormal fusion spectrum of each abnormal state and the real-time fusion spectrum; Indicates the first elements; Indicates the The first in the abnormal fusion spectrum elements; Indicates the amount of data in the real-time fused spectrum, which can also be understood as the length of the real-time fused spectrum.

[0051] Indicates the difference between the elements in the real-time fusion spectrum and the abnormal fusion spectrum. The larger the difference, the less similar the real-time fusion spectrum is to the abnormal fusion spectrum, indicating a significant difference between the current device operating status and the status corresponding to the abnormal fusion spectrum.

[0052] The anomaly score is calculated based on the frequency domain similarity between the real-time fusion spectrum and different anomaly fusion spectrums:

[0053] In one embodiment, the step of calculating an anomaly score based on frequency domain similarity includes: obtaining time domain similarity between the real-time feature signal and the abnormal feature signal; and taking the product of the time domain similarity and the frequency domain similarity as the anomaly score. In this embodiment, the step of calculating time domain similarity includes: obtaining the Euclidean distance between the real-time feature signal and the abnormal feature signal; and taking the average of the Euclidean distances corresponding to multiple feature signals under the same abnormal state as the time domain similarity.

[0054] The frequency domain similarity and time domain similarity between the real-time spectrum and the abnormal spectrum are calculated to further improve the accuracy of judging the operating status of the equipment.

[0055] The time domain similarity and frequency domain similarity under the same abnormal state are multiplied to obtain an abnormality score. A threshold for the abnormality score is set. When the abnormality score is greater than the abnormality threshold, an alarm is issued, indicating that the current device operation is highly similar to the abnormal state and this type of abnormality exists.

[0056] In another embodiment: the step of calculating the anomaly score based on frequency domain similarity includes: obtaining the time domain similarity between the real-time feature signal and the abnormal feature signal; using the product of the time domain similarity and the frequency domain similarity as a first control value to obtain a normal fusion spectrum corresponding to the normal feature signal; using the similarity between the normal fusion spectrum of the normal feature signal and the real-time fusion spectrum of the real-time feature signal as a second control value, and using the ratio of the first control value to the second control value as the anomaly score.

[0057] The step of obtaining the second control value includes: obtaining the frequency domain similarity between the normal fusion spectrum and the real-time fusion spectrum; obtaining the time domain difference between the normal characteristic signal and the real-time characteristic signal under the same abnormal state, and taking the product of the frequency domain similarity and the time domain similarity as the second control value.

[0058] The greater the frequency domain similarity between the normal fusion spectrum and the real-time fusion spectrum, the more normal the currently acquired characteristic signal is in the frequency domain, and therefore, the greater the likelihood that the device is currently operating normally. When the first control value is greater than the second control value, the real-time acquired characteristic signal is more similar to the abnormal characteristic signal, and therefore, the current status is more likely to be abnormal.

[0059] Here, the acquisition of the normal fusion spectrum is similar to the acquisition of the abnormal fusion spectrum, and will not be repeated here.

[0060] This method obtains the normal fusion spectrum of the equipment during normal operation. Taking into account the similarity between the real-time acquired characteristic signal and the abnormal characteristic signal, as well as the similarity between the real-time acquired characteristic signal and the normal characteristic signal, it further improves the accuracy of the equipment operation status judgment.

[0061] The operating status of the device is judged based on the abnormality score. For any abnormal status corresponding to the abnormality score, when the abnormality score is greater than 1, it is judged that the device operation has such abnormality.

[0062] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for monitoring the operating status of microparticle shot peening equipment based on multi-source data fusion, characterized in that: The method comprises the following steps: obtaining a plurality of characteristic signals of the device in a normal operating state and in a plurality of abnormal states, the characteristic signals including at least two of a sound signal, a pressure signal, a vibration signal, and a flow signal; obtaining a spectrum of the characteristic signal in each abnormal state to obtain an abnormal spectrum, and obtaining a spectrum of the characteristic signal in a normal state to obtain a normal spectrum; Divide frequency bands; Calculating the frequency band differences between the abnormal spectrum and the normal spectrum in different frequency bands, including: for each abnormal spectrum in each abnormal state, obtaining the absolute value of the difference between the elements of the same position sequence in the same frequency band of the abnormal spectrum and the normal spectrum; and taking the sum of the absolute values of the differences corresponding to all elements in the frequency band as the frequency band difference of the frequency band; Determining the fusion weights of different characteristic signals in each frequency band under each abnormal state based on the frequency band differences, including: taking the sum of the frequency band differences corresponding to different characteristic signals as the total difference, and for each abnormal state, taking the ratio of the frequency band difference of any characteristic signal in any frequency band to the total difference as the fusion weight of the characteristic signal in the frequency band; The same frequency bands in multiple abnormal spectra under the same abnormal state are fused based on the fusion weight to obtain a comprehensive spectrum, including: taking the product of the fusion weight corresponding to each frequency band and the element in each frequency band as a local element, and for each abnormal spectrum under each abnormal state, taking the sum of the local elements of the same position in the same frequency band in the multiple abnormal spectra as a comprehensive element; the multiple comprehensive elements constitute the comprehensive spectrum; The integrated spectrum of multiple frequency bands is connected to form an abnormal fusion spectrum; multiple characteristic signals during equipment operation are collected in real time to obtain a real-time spectrum; the frequency bands of multiple real-time spectrums are fused to obtain a real-time fusion spectrum; the differences between each abnormal fusion spectrum and the real-time fusion spectrum are compared to obtain frequency domain similarity; an anomaly score is calculated based on the frequency domain similarity, and the equipment operation status is judged based on the anomaly score.

2. The method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion according to claim 1 is characterized in that: In the step of dividing the frequency band, the spectrum is divided into three segments.

3. The method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion according to claim 1, characterized in that: The steps of calculating the anomaly score based on the frequency domain similarity include: obtaining the time domain similarity between the real-time feature signal and the abnormal feature signal; and taking the product of the time domain similarity and the frequency domain similarity as the anomaly score.

4. The method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion according to claim 1, characterized in that: The steps of calculating the anomaly score based on frequency domain similarity include: obtaining the time domain similarity between the real-time feature signal and the abnormal feature signal; taking the product of the time domain similarity and the frequency domain similarity as the first control value, and obtaining the integrated spectrum corresponding to the normal feature signal; calculating the second control value based on the similarity between the normal fusion spectrum of the normal feature signal and the normal fusion spectrum of the real-time feature signal, and taking the ratio of the first control value to the second control value as the anomaly score.

5. The method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion according to claim 3 is characterized in that: The calculation steps of frequency domain similarity include: obtaining the absolute value of the difference between the same-order elements in the real-time fusion spectrum and the abnormal fusion spectrum, and taking the sum of the absolute values of the differences corresponding to multiple elements as the adjustment value; The result with the negative value of the adjustment value as the base and the exponent as the frequency domain similarity is taken as the frequency domain similarity.

6. The method for monitoring the operating status of a microparticle shot blasting equipment based on multi-source data fusion according to claim 3 is characterized in that: The time domain similarity between the real-time feature signal and the abnormal feature signal is the average of the Euclidean distances between multiple real-time feature signals and the corresponding abnormal feature signals.

7. The method for monitoring the operating status of microparticle shot blasting equipment based on multi-source data fusion according to claim 1, characterized in that: include: The step of judging the operating status of the device based on the abnormality score includes: setting an abnormality threshold, and issuing an alarm in response to the abnormality score being greater than the abnormality threshold.

Images