An automatic online monitoring method for mine air pressure unit

Abstract

The present application relates to the technical field of data processing, in particular to an automatic online monitoring method of a mine air pressure unit, comprising: obtaining an overall fault vibration frequency section of the mine air pressure unit; segmenting the overall fault vibration frequency section of the mine air pressure unit through an optimal decomposition result to obtain a plurality of frequency sections; in each frequency section, obtaining a fault prominence of a component in the frequency section according to fault difference; obtaining a local abnormality degree of a vibration signal of the component at a current time in the frequency section; then obtaining a comprehensive abnormality degree of each component at the current time; and monitoring the mine air pressure unit based on the comprehensive abnormality degree. The present application improves the accuracy of fault monitoring of the mine air pressure unit.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an automated online monitoring method for a mining wind compressor unit. Background Art

[0002] As key power or ventilation equipment in coal and metal mines, the stable operation of mining wind compressors is crucial to safe production in mines. Failures can lead to serious accidents such as interrupted underground air supply, insufficient air pressure, equipment shutdown, poisoning, or explosions. Therefore, developing fault monitoring technology has important safety and economic value. Because wind compressors contain multiple components, each generating different vibration signals when a component fails. Traditional fault detection relies on fixed frequency bands or overall weighted frequency domain features for fault monitoring. This results in frequency bands with significant fault variations being ignored or smoothed, leading to misidentification and untimely identification, which in turn reduces the accuracy of fault monitoring for mining wind compressors. Summary of the Invention

[0003] In order to solve the above problems, the present invention provides an automated online monitoring method for a mining wind turbine, the method comprising:

[0004] Collect the fault vibration signals and normal vibration signals of several components on the mining wind compressor unit, as well as the vibration signals at the current moment;

[0005] Decompose the fault vibration signal of the component into different numbers of components to obtain several decomposition results with different numbers of components; filter the decomposition results of all numbers of components based on the differences in the IMF components in the decomposition results to obtain the optimal decomposition result of the fault vibration signal of the component; obtain the overall fault vibration frequency segment of the mining wind compressor; and segment the overall fault vibration frequency segment of the mining wind compressor using the optimal decomposition result to obtain several frequency segments;

[0006] In each frequency band, based on the amplitude difference between the normal vibration signal and the fault vibration signal of the component, the fault difference of the component in the frequency band is obtained; based on the fault difference, the fault significance of the component in the frequency band is obtained; based on the amplitude difference between the vibration signal of the component at the current moment and the normal vibration signal and the fault vibration signal, the local abnormality degree of the vibration signal of the component at the current moment in the frequency band is obtained; based on the local abnormality degree and the fault significance degree, the comprehensive abnormality degree of each component at the current moment is obtained;

[0007] Monitor the mine wind compressor units based on the comprehensive abnormality level.

[0008] Preferably, the decomposition of the fault vibration signal of the component into different numbers of components to obtain decomposition results with different numbers of components includes the following specific methods:

[0009] Preset a starting quantity parameter and the number of terminal components parameter , the starting number of components parameter and the number of terminal components parameter All positive integers between are used as the number of components; the VMD decomposition algorithm is used to perform the first The decomposition of the number of components is obtained The decomposition results under the number of components.

[0010] Preferably, the decomposition results under all component numbers are screened according to the differences in the IMF components in the decomposition results to obtain the optimal decomposition result of the fault vibration signal of the component, including the specific method of:

[0011] According to the differences in the IMF components in the decomposition results, the modal decomposition scores of the decomposition results under each number of components are obtained;

[0012] Among the decomposition results of the fault vibration signal of the component under all component numbers, the decomposition result with the largest modal decomposition score is selected as the optimal decomposition result of the fault vibration signal of the component.

[0013] Preferably, the method of obtaining the modal decomposition score of the decomposition result under each number of components according to the difference of the IMF components in the decomposition result includes the following specific methods:

[0014] The first The decomposition results under the number of components The variance of all amplitudes within the IMF component is recorded as The first variance of the IMF component; The mean of the variances of all amplitudes in all IMF components in the decomposition results under the number of components is recorded as the mean variance; The absolute value of the difference between the first variance and the mean variance of the IMF component is recorded as The variance deviation value of the IMF component; The ratio of the variance deviation value to the variance mean of the IMF component is recorded as The decomposition factor of the first IMF component; The normalized value of the mean of the decomposition factors of all IMF components in the decomposition results under the number of components is used as the first The modal decomposition score of the decomposition results for each number of components.

[0015] Preferably, the method of segmenting the overall fault vibration frequency segment of the mining wind compressor unit by the optimal decomposition result to obtain several frequency segments includes:

[0016] The Hilbert transform is used to obtain the frequency of each IMF component at each moment in the optimal decomposition results of the fault vibration signals of all components; the mean of the frequencies of each IMF component at all moments is recorded as the frequency mean of each IMF component;

[0017] The mean of the frequency means of the first IMF component in the optimal decomposition results of the fault vibration signals of all components is recorded as the first center frequency; the mean of the frequency means of the second IMF component in the optimal decomposition results of the fault vibration signals of all components is recorded as the second center frequency; and so on, several center frequencies are obtained;

[0018] In the overall fault vibration frequency segment of the mining wind compressor set, all center frequencies are used as segmentation points, and the overall fault vibration frequency segment is divided into several frequency segments.

[0019] Preferably, the method of obtaining the fault difference of the component in the frequency band according to the amplitude difference between the normal vibration signal and the fault vibration signal of the component includes the following specific methods:

[0020] The first The normal vibration signal of each component is The first frequency band The amplitude corresponding to the first frequency is The fault vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as The difference factor of the frequency; The normal vibration signal of each component is The mean of the difference factors of all frequencies in the frequency band is taken as the The component in The fault diversity in each frequency band.

[0021] Preferably, the specific method of obtaining the fault significance of the component in the frequency band according to the fault difference includes:

[0022] The first The component in The fault differences on the first frequency band are The ratio of the sum of the fault differences of the components in all frequency bands is taken as the first The component in The fault significance level in each frequency band.

[0023] Preferably, the method of obtaining the local abnormality degree of the vibration signal of the component at the current moment in the frequency band according to the amplitude differences between the vibration signal of the component at the current moment and the normal vibration signal and the fault vibration signal includes the following specific methods:

[0024] The first The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the first frequency is The normal vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as Normal vibration difference value of the frequency; The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the first frequency is The fault vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as The fault vibration difference value of the frequency; The ratio between the normal vibration difference value and the fault vibration difference value of the frequency is recorded as The abnormal factor of the frequency; The cumulative sum of all abnormal factors in the frequency band is taken as the The vibration signal of the component at the current moment is The degree of local abnormality in a frequency band.

[0025] Preferably, the method of obtaining the comprehensive abnormality degree of each component at the current moment based on the local abnormality degree and the fault significance degree includes the following specific methods:

[0026] The first The vibration signal of the component at the current moment is The degree of local abnormality in the frequency band is The component in The product of the fault differences in the frequency bands is recorded as The component in The comprehensive abnormal factor on the frequency band; The normalized value of the cumulative sum of the comprehensive abnormal factors of each component in all frequency bands is taken as the first The comprehensive abnormality level of each component at the current moment.

[0027] Preferably, the mine air pressure unit is monitored based on the comprehensive abnormality degree, and the specific method comprises the following steps:

[0028] A threshold parameter is preset If the comprehensive abnormality degree of any component of the mine air pressure unit at the current time is greater than or equal to the threshold parameter The system immediately issues a fault alarm.

[0029] The beneficial effects of the technical scheme of the present application are as follows: the overall fault vibration frequency band of the mine air pressure unit is obtained; the overall fault vibration frequency band of the mine air pressure unit is segmented based on the optimal decomposition result to obtain a plurality of frequency bands; the fault prominence of the component on the frequency band is obtained according to the fault difference on each frequency band; the local abnormality degree of the vibration signal of the component at the current time on the frequency band is obtained; then the comprehensive abnormality degree of each component at the current time is obtained; and the mine air pressure unit is monitored based on the comprehensive abnormality degree. By dividing the plurality of frequency bands and independently analyzing the signal characteristics of each frequency band, the abnormal conditions in different frequency ranges of the vibration signal can be more efficiently analyzed, thereby improving the accuracy of fault monitoring of the mine air pressure unit. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0031] Figure 1 The step flow chart of the automatic online monitoring method of the mine air pressure unit of the present application. DETAILED DESCRIPTION

[0032] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific implementation, structure, features and effects of the automatic online monitoring method of the mine air pressure unit according to the present application are described in detail as follows by combining with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0034] The application provides a kind of mine wind pressure unit's automatic online monitoring method.

[0035] Please refer to Figure 1 It shows the step flow chart of a kind of mine wind pressure unit's automatic online monitoring method provided by the embodiment of the application, and the method comprises the following steps:

[0036] Step S001: collect the fault vibration signal and normal vibration signal of several components on the mine wind pressure unit, and the vibration signal at the current time.

[0037] Specifically, first, the fault vibration signal and normal vibration signal of several components on the mine wind pressure unit, and the vibration signal at the current time need to be collected, and the specific process is:

[0038] The several components on the mine wind pressure unit include bearings, motors, shaft couplings, impellers and housings. For any one component on the mine wind pressure unit, when the any one component on the mine wind pressure unit fails, the vibration signal of the any one component is collected by using an IEPE (Integrated Electronics Piezo-Electric) sensor, and is recorded as the fault vibration signal of the any one component. When the mine wind pressure unit is running normally, the vibration signal of the any one component is collected by using the IEPE sensor, and is recorded as the normal vibration signal of the any one component. At the current time, the vibration signal of the any one component is collected by using the IEPE sensor, and is recorded as the vibration signal of the any one component at the current time.

[0039] At this point, the fault vibration signal and normal vibration signal of several components on the mine wind pressure unit, and the vibration signal at the current time are obtained by the above method.

[0040] Step S002: decompose the fault vibration signal of the component in different component quantities to obtain several decomposition results under different component quantities; according to the difference of the IMF components in the decomposition results, the decomposition results under all component quantities are screened to obtain the optimal decomposition result of the fault vibration signal of the component; the overall fault vibration frequency band of the mine wind pressure unit is obtained; the optimal decomposition result is used to segment the overall fault vibration frequency band of the mine wind pressure unit to obtain several frequency bands.

[0041] It should be noted that, since mining wind compressors are usually composed of multiple components, when different components fail, the degree of abnormality of their vibration signals in different frequency bands is different; for example, when there is wear on the inner and outer rings of the bearings, the energy of their high frequencies increases significantly; when the motor rotor is eccentric and broken, the current bandwidth expands to produce low-frequency jitter; when the casing is loose or cracked, low-frequency large-amplitude vibration and intermittent amplitude changes are generated; the existing mining wind compressor fault monitoring ignores the differences in vibration signals in different frequency bands when different components fail, resulting in the anomalies of local frequency bands being smoothed out by the overall frequency domain, resulting in the phenomenon of misidentification or untimely identification of faults; therefore, this embodiment decomposes the fault vibration signals of the components, and screens them based on the comprehensive modal decomposition score to obtain the optimal number of components; multiple frequency bands are obtained through the decomposition results of the optimal number of components.

[0042] This embodiment is described by taking any component of a mining wind compressor as an example;

[0043] Preferably, in some implementations of the embodiments of the present invention, the specific method of decomposing the fault vibration signal of the component into different numbers of components to obtain decomposition results with different numbers of components is:

[0044] Preset a starting quantity parameter and the number of terminal components parameter , wherein this embodiment is based on and This example is described as an example, and this embodiment is not specifically limited. and Depends on the specific implementation situation;

[0045] The starting number of components parameter and the number of terminal components parameter All positive integers between are used as the number of components, including the starting number of components parameter and the number of terminal components parameter ;

[0046] The VMD (Variational Mode Decomposition) algorithm is used to decompose the fault vibration signal of the component. The decomposition of the number of components is obtained decomposition results under the number of components; the decomposition results include several IMF components;

[0047] Among them, the VMD decomposition algorithm is an existing technology, and this embodiment will not be described in detail here; if The number of components is , then the decomposition results include IMF components.

[0048] Preferably, in some implementations of the embodiments of the present invention, based on the differences in the IMF components in the decomposition results, a specific method for obtaining the modal decomposition score of the decomposition results for each number of components is as follows:

[0049] The first The decomposition results under the number of components The variance of all amplitudes within the IMF component is denoted as The first variance of the IMF component; The mean of the variances of all amplitudes in all IMF components in the decomposition results under the number of components is recorded as the mean variance; The absolute value of the difference between the first variance and the mean variance of the IMF component is recorded as The variance deviation value of the IMF component; The ratio of the variance deviation value to the variance mean of the IMF component is recorded as The decomposition factor of the first IMF component; The normalized value of the mean of the decomposition factors of all IMF components in the decomposition results under the number of components is used as the first The modal decomposition score of the decomposition results under the number of components;

[0050] The specific formula is:

[0051]

[0052] Where, Indicates the The modal decomposition score of the decomposition results under the number of components; Indicates the The number of all IMF components in the decomposition results under the number of components; Indicates the The decomposition results under the number of components The variance of all amplitudes in an IMF component; Indicates the The mean of the variances of all amplitudes of all IMF components in the decomposition results with the number of components; Indicates taking the absolute value; represents the linear normalization function.

[0053] It should be noted that the variance of each IMF component reflects the degree of change in the amplitude within the component. By calculating the variance of each IMF and the mean variance of all IMF components, the overall fluctuation degree of the signal can be obtained; the calculation of variance deviation and decomposition factor can evaluate the representativeness and decomposition effect of each IMF component; if the decomposition factor of the decomposition result is large, it means that the component has high volatility or characteristics, which may be related to the fault signal; that is, the variance of each IMF component can characterize the degree of chaos of the component. The greater the difference in variance between different IMF components, the greater the difference in distribution and chaos between the IMF components, and the better the decomposition effect; then The vibration signal representing the fault of the component is VMD decomposition effect under the number of components.

[0054] Preferably, in some implementations of the embodiments of the present invention, since a lower number of components may ignore some subtle signal features, and an excessively high number of components may lead to overfitting and capture some noise components, selecting the optimal number of components is crucial for obtaining effective decomposition results. Therefore, the decomposition results under all numbers of components are screened based on the modal decomposition score. The specific method for obtaining the optimal decomposition result of the fault vibration signal of the component is as follows:

[0055] Among the decomposition results of the fault vibration signal of the component under all component numbers, the decomposition result with the largest modal decomposition score is selected as the optimal decomposition result of the fault vibration signal of the component.

[0056] Preferably, the fault vibration signals of all components on the mining wind compressor are converted into an overall fault vibration signal using Fourier transform to obtain a spectrum diagram of the overall fault vibration; all horizontal coordinates on the spectrum diagram of the overall fault vibration constitute a frequency sequence, which is recorded as the overall fault vibration frequency segment of the mining wind compressor.

[0057] Preferably, in some implementations of the embodiments of the present invention, the specific method of segmenting the overall fault vibration frequency segment of the mining wind compressor group according to the optimal decomposition result to obtain several frequency segments is:

[0058] The Hilbert transform is used to obtain the frequency of each IMF component at each moment in the optimal decomposition results of the fault vibration signals of all components; the mean of the frequencies of each IMF component at all moments is recorded as the frequency mean of each IMF component;

[0059] The mean of the frequency means of the first IMF component in the optimal decomposition results of the fault vibration signals of all components is recorded as the first center frequency; the mean of the frequency means of the second IMF component in the optimal decomposition results of the fault vibration signals of all components is recorded as the second center frequency; and so on, several center frequencies are obtained;

[0060] In the overall fault vibration frequency segment of the mining wind compressor group, all center frequencies are used as segmentation points, and the overall fault vibration frequency segment is divided into several frequency segments;

[0061] Among them, Fourier transform and Hilbert transform are existing technologies, and this embodiment will not be described in detail here; if The optimal decomposition result of the fault vibration signal of the component does not exist in the When the number of components is The optimal decomposition results of the fault vibration signals of all other components are The mean of the frequency means of the components.

[0062] So far, several frequency bands have been obtained through the above method.

[0063] Step S003: In each frequency band, based on the amplitude difference between the normal vibration signal and the fault vibration signal of the component, obtain the fault difference of the component in the frequency band; based on the fault difference, obtain the fault significance of the component in the frequency band; based on the amplitude difference between the vibration signal of the component at the current moment and the normal vibration signal and the fault vibration signal, obtain the local abnormality degree of the vibration signal of the component at the current moment in the frequency band; based on the local abnormality degree and the fault significance degree, obtain the comprehensive abnormality degree of each component at the current moment.

[0064] It should be noted that each frequency band represents the energy distribution of different frequency components in the vibration signal. Usually, mechanical failures will be more significant in certain frequency ranges. For example, rolling bearing failures, gear failures, etc. usually produce stronger vibration signals in a specific frequency range. Therefore, based on the frequency domain feature differences between the fault vibration signal and the normal vibration signal in each frequency band, the fault significance of each component in each frequency band is obtained. If the abnormality level of the signal at the current moment and the fault significance level of the frequency band are both high, it means that the component has a greater failure risk. The comprehensive abnormality level is obtained by combining the local abnormality level and the fault significance level for a weighted assessment.

[0065] Preferably, in some implementations of the embodiments of the present invention, based on the amplitude difference between the normal vibration signal and the fault vibration signal of the component in each frequency band, the specific method for obtaining the fault difference of the component in each frequency band is:

[0066] The first The normal vibration signal of each component is The first frequency band The amplitude corresponding to the first frequency is The fault vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as The difference factor of the frequency; The normal vibration signal of each component is The mean of the difference factors of all frequencies in the frequency band is taken as the The component in Fault differences in each frequency band;

[0067] The specific formula is:

[0068]

[0069] Where, Indicates the The component in Fault differences in each frequency band; Indicates the The number of all frequencies in a frequency band; Indicates the The normal vibration signal of each component is The first frequency band The amplitude corresponding to the frequency; Indicates the The fault vibration signal of each component is The first frequency band The amplitude corresponding to the frequency; Indicates taking the absolute value.

[0070] Preferably, in some implementations of the embodiments of the present invention, the specific method for obtaining the fault significance of the component in each frequency band according to the fault difference is:

[0071] The first The component in The fault differences on the first frequency band are The ratio of the sum of the fault differences of the components in all frequency bands is taken as the first The component in The fault significance level in each frequency band.

[0072] Preferably, in some implementations of the embodiments of the present invention, in each frequency band, the greater the difference between the vibration signal at the current moment and the normal vibration signal, the greater the probability of a fault; the smaller the difference between the vibration signal at the current moment and the fault vibration signal, the greater the probability of a fault; therefore, in each frequency band, based on the amplitude differences between the vibration signal of the component at the current moment and the normal vibration signal and the fault vibration signal, the specific method for obtaining the local abnormality degree of the vibration signal of the component at the current moment in each frequency band is:

[0073] The first The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the first frequency is The normal vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as Normal vibration difference value of the frequency; The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the first frequency is The fault vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as The fault vibration difference value of the frequency; The ratio between the normal vibration difference value and the fault vibration difference value of the frequency is recorded as The abnormal factor of the frequency; The cumulative sum of all abnormal factors in the frequency band is taken as the The vibration signal of the component at the current moment is The degree of local abnormality in each frequency band;

[0074] The specific formula is:

[0075]

[0076] Where, Indicates the The vibration signal of the component at the current moment is The degree of local abnormality in each frequency band; Indicates the The number of all frequencies in a frequency band; Indicates the The normal vibration signal of each component is The first frequency band The amplitude corresponding to the frequency; Indicates the The fault vibration signal of each component is The first frequency band The amplitude corresponding to the frequency; Indicates the The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the frequency; denotes taking absolute value.

[0077] Preferably, in some implementations of the embodiments of the present application, the specific method for obtaining the comprehensive abnormality degree of each component at the current time according to the local abnormality degree and the fault distinctiveness is as follows:

[0078] the product of the local abnormality degree of the vibration signal of the i-th component at the current time in the j-th frequency band and the fault distinctiveness of the i-th component in the j-th frequency band is denoted as the comprehensive abnormality factor of the i-th component in the j-th frequency band; the normalized value of the cumulative sum of the comprehensive abnormality factors of the i-th component in all frequency bands is taken as the comprehensive abnormality degree of the i-th component at the current time.

[0079] Specifically, the formula is as follows:

[0080]

[0081] In the formula, denotes the comprehensive abnormality degree of the i-th component at the current time; denotes the local abnormality degree of the vibration signal of the i-th component at the current time in the j-th frequency band; denotes the fault distinctiveness of the i-th component in the j-th frequency band; denotes the local abnormality degree of the vibration signal of the i-th component at the current time in the j-th frequency band; denotes the fault distinctiveness of the i-th component in the j-th frequency band; denotes the number of all frequency bands; denotes a linear normalization function.

[0082] Thus, the comprehensive abnormality degree of each component at the current time is obtained by the above method.

[0083] Step S004: monitoring the mine air pressure unit based on the comprehensive abnormality degree.

[0084] Preferably, in some implementations of the embodiments of the present application, the specific method for monitoring the mine air pressure unit based on the comprehensive abnormality degree is as follows:

[0085] a threshold parameter is preset , wherein the present embodiment is described by taking as an example, and the present embodiment is not specifically limited, wherein is determined according to specific implementation conditions;

[0086] ​​​​​​​​​​​If the comprehensive abnormality degree of any component on the mining wind compressor at the current moment is greater than or equal to the threshold parameter , it means that there is a fault in the mine wind compressor unit. The system immediately issues a fault alarm and prompts the staff to carry out timely repairs.

[0087] At this point, this embodiment is completed.

[0088] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An automated online monitoring method for a mining wind turbine, characterized in that: The method comprises the following steps: Collect the fault vibration signals and normal vibration signals of several components on the mining wind compressor unit, as well as the vibration signals at the current moment; Decompose the fault vibration signal of the component with different numbers of components to obtain decomposition results with different numbers of components; screen the decomposition results with all numbers of components according to the differences in the IMF components in the decomposition results to obtain the optimal decomposition result of the fault vibration signal of the component; obtain the overall fault vibration frequency segment of the mining wind compressor; segment the overall fault vibration frequency segment of the mining wind compressor according to the optimal decomposition result to obtain several frequency segments, specifically: use Hilbert transform to obtain the frequency of each IMF component at each moment in the optimal decomposition result of the fault vibration signal of all components; record the mean of the frequencies of all moments on each IMF component as the frequency mean of each IMF component; record the mean of the frequency mean of the first IMF component in the optimal decomposition result of the fault vibration signal of all components as the first center frequency; record the mean of the frequency mean of the second IMF component in the optimal decomposition result of the fault vibration signal of all components as the second center frequency; and so on to obtain several center frequencies; on the overall fault vibration frequency segment of the mining wind compressor, use all center frequencies as segmentation points to divide the overall fault vibration frequency segment into several frequency segments; In each frequency band, according to the amplitude difference between the normal vibration signal and the fault vibration signal of the component, the fault difference of the component in the frequency band is obtained; according to the fault difference, the fault significance of the component in the frequency band is obtained; according to the amplitude difference between the vibration signal of the component at the current moment and the normal vibration signal and the fault vibration signal, the local abnormality of the vibration signal of the component at the current moment in the frequency band is obtained; according to the local abnormality and the fault significance, the comprehensive abnormality of each component at the current moment is obtained; the method for obtaining the fault significance is: The component in The fault differences on the first frequency band are The ratio of the sum of the fault differences of the components in all frequency bands is taken as the first The component in The significance of the fault in each frequency band; Monitor the mine wind compressor units based on the comprehensive abnormality level.

2. The automated online monitoring method for a mining wind turbine according to claim 1, characterized in that: The method of decomposing the fault vibration signal of the component with different numbers of components to obtain decomposition results with different numbers of components includes the following specific methods: Preset a starting quantity parameter and the number of terminal components parameter , the starting number of components parameter and the number of terminal components parameter All positive integers between are used as component quantities; The VMD decomposition algorithm is used to analyze the fault vibration signal of the component. The decomposition of the number of components is obtained The decomposition results under the number of components.

3. The automated online monitoring method for a mining wind turbine according to claim 1, characterized in that: The method of screening the decomposition results under all component quantities based on the differences in the IMF components in the decomposition results to obtain the optimal decomposition result of the fault vibration signal of the component includes the following specific methods: According to the differences in the IMF components in the decomposition results, the modal decomposition scores of the decomposition results under each number of components are obtained; Among the decomposition results of the fault vibration signal of the component under all component numbers, the decomposition result with the largest modal decomposition score is selected as the optimal decomposition result of the fault vibration signal of the component.

4. The method for automatic online monitoring of a mining wind turbine according to claim 3, characterized in that: The modal decomposition score of the decomposition result under each number of components is obtained according to the difference of the IMF components in the decomposition result, including the specific method as follows: The first The decomposition results under the number of components The variance of all amplitudes within the IMF component is denoted as The first variance of the IMF component; The mean of the variances of all amplitudes in all IMF components in the decomposition results under the number of components is recorded as the mean variance; The absolute value of the difference between the first variance and the mean variance of the IMF component is recorded as The variance deviation value of the IMF component; The ratio of the variance deviation value to the variance mean of the IMF component is recorded as The decomposition factor of the first IMF component; The normalized value of the mean of the decomposition factors of all IMF components in the decomposition results under the number of components is used as the first The modal decomposition score of the decomposition results for each number of components.

5. The automated online monitoring method for a mining wind turbine according to claim 1, characterized in that: The method of obtaining the fault difference of the component in the frequency band according to the amplitude difference between the normal vibration signal and the fault vibration signal of the component includes the following specific methods: The first The normal vibration signal of each component is The first frequency band The amplitude corresponding to the first frequency is The fault vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as The difference factor of the frequency; The normal vibration signal of each component is The mean of the difference factors of all frequencies in the frequency band is taken as the The component in The fault diversity in each frequency band.

6. The method for automatic online monitoring of a mining wind turbine according to claim 1, characterized in that: The method of obtaining the local abnormality degree of the vibration signal of the component at the current moment in the frequency band according to the amplitude differences between the vibration signal of the component at the current moment and the normal vibration signal and the fault vibration signal includes the following specific methods: The first The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the first frequency is The normal vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as Normal vibration difference value of the frequency; The vibration signal of the component at the current moment is The first frequency band The amplitude corresponding to the first frequency is The fault vibration signal of each component is The first frequency band The absolute value of the difference between the amplitudes corresponding to the frequencies is recorded as The fault vibration difference value of the frequency; The ratio between the normal vibration difference value and the fault vibration difference value of the frequency is recorded as The abnormal factor of the frequency; The cumulative sum of all abnormal factors in the frequency band is taken as the The vibration signal of the component at the current moment is The degree of local abnormality in a frequency band.

7. The automated online monitoring method for a mining wind turbine according to claim 1, characterized in that: The method of obtaining the comprehensive abnormality degree of each component at the current moment based on the local abnormality degree and the fault significance degree includes the following specific methods: The first The vibration signal of the component at the current moment is The degree of local abnormality in the frequency band is The component in The product of the fault differences in the frequency bands is recorded as The component in The comprehensive abnormal factor on the frequency band; The normalized value of the cumulative sum of the comprehensive abnormal factors of each component in all frequency bands is taken as the first The comprehensive abnormality level of each component at the current moment.

8. The automated online monitoring method for a mining wind turbine according to claim 1, characterized in that: The specific method of monitoring the mining wind turbine based on the comprehensive abnormality degree is as follows: Preset a threshold parameter If the comprehensive abnormality of any component on the mining wind compressor at the current moment is greater than or equal to the threshold parameter , the system will immediately issue a fault alarm.

Images