Method and system for detecting abnormal operation of manure scraper

Abstract

The invention provides a method and a system for detecting abnormal operation of a manure scraper, and relates to the field of electrical variable measurement, the system comprises a vibration detection module comprising a plurality of vibration sensing assemblies arranged on a scraper of the manure scraper; the motor detection module is used for acquiring current and voltage of a motor of the manure scraper; the abnormity judgment module is used for judging whether the operation of the manure scraper is abnormal or not according to the vibration data, the current and the voltage collected by the plurality of vibration sensing assemblies at the plurality of continuous time points; the anomaly detection module is used for judging whether a scraper plate clamps cattle hoof or not according to the vibration data collected by the plurality of vibration sensing assemblies at a plurality of continuous time points and the current and voltage of a motor of the manure scraper at the plurality of continuous time points after the operation of the manure scraper is judged to be abnormal; and the operation control module is used for controlling the motor to stop operating and generating warning information after judging that the cattle hoof is stuck by the scraper, and has the advantage of realizing automatic real-time monitoring of abnormal operation of the manure scraper.

Description

Technical Field

[0001] The invention relates to the field of measuring electrical variables, and in particular to a method and system for detecting abnormal operation of a manure scraper. Background Art

[0002] Manure scrapers are often used in cowsheds to clean cow dung. The principle of manure scrapers is to use a motor to drive a scraper through a chain to move in the manure channel, thereby scraping off the manure accumulated in the manure channel. During the process of the chain pulling the scraper, if the scraper is accidentally caught by the cow's hoof, it will cause injury to the cow's hoof, or it will directly pull the cow's hoof and trip the cow. Due to the small space in the cowshed and the heavy weight of the cow itself, once the cow trips, it is difficult to get up, and it is easy to be seriously injured or even die.

[0003] In the prior art, it is necessary to manually observe whether the chain is stuck or pulling the cow's hoof during the process of pulling the scraper. It is easy for the manure scraper to get stuck or pull the cow's hoof, and the operating state of the manure scraper cannot be adjusted in time, resulting in serious injury to the cow's hoof or death of the cow.

[0004] Therefore, it is necessary to provide a method and system for detecting abnormal operation of a manure scraper, so as to realize automatic real-time monitoring of abnormal operation of the manure scraper. Summary of the invention

[0005] The present invention provides a system for detecting abnormal operation of a manure scraper, comprising: a vibration detection module, comprising a plurality of vibration sensing components arranged at a plurality of positions of a scraper of the manure scraper; a motor detection module, for acquiring the current and voltage of the motor of the manure scraper; an abnormality judgment module, for judging whether an abnormality occurs in the operation of the manure scraper according to vibration data collected by the plurality of vibration sensing components at a plurality of consecutive time points and the current and voltage of the motor of the manure scraper at the plurality of consecutive time points; an abnormality detection module, for judging whether the scraper is stuck with a cow's hoof according to the vibration data collected by the plurality of vibration sensing components at a plurality of consecutive time points and the current and voltage of the motor of the manure scraper at the plurality of consecutive time points after the abnormality judgment module determines that an abnormality occurs in the operation of the manure scraper; and an operation control module, for controlling the motor to stop running and generating a warning message after the abnormality detection module determines that the scraper is stuck with a cow's hoof.

[0006] Furthermore, the vibration detection module is also used to: determine multiple vibration test positions on the scraper, and set a vibration sensing component at each vibration test position; obtain simulation test data of the vibration sensing component set at each vibration test position under multiple simulated scraper jamming scenarios, wherein the simulation test data includes vibration data at multiple continuous time points; screen the multiple vibration test positions based on the simulation test data collected by the vibration sensing component set at each vibration test position under multiple simulated scraper jamming scenarios to determine multiple vibration monitoring positions; and set the multiple vibration sensing components based on the multiple vibration monitoring positions.

[0007] Further, the vibration detection module screens the multiple vibration test positions according to the simulated test data collected by the vibration sensing component set at each vibration test position under the multiple simulated scraper stuck scenes, and determines multiple vibration monitoring positions, including: S11, for each vibration test position, according to the simulated test data collected by the vibration sensing component set at the vibration test position under the multiple simulated scraper stuck scenes, determining the vibration abnormality value of the vibration test position corresponding to each simulated scraper stuck scene; S12, for each vibration test position, according to the vibration abnormality value of the vibration test position corresponding to each simulated scraper stuck scene, judging whether the vibration test position is a target vibration test position, if so, executing S13, if not, screening out the vibration test position; S13, for any two target vibration test positions, according to the vibration abnormality value of the two target vibration test positions corresponding to each simulated scraper stuck scene, calculating the vibration abnormality similarity of the two target vibration test positions; S14, according to any two target vibration test positions According to the vibration anomaly similarity of any two target vibration test positions included in the target vibration test position group, multiple target vibration test positions included in the target vibration test position group are grouped to determine multiple target vibration test position groups; S15, for each target vibration test position group, according to the vibration anomaly similarity of any two target vibration test positions included in the target vibration test position group, multiple target vibration test positions included in the target vibration test position group are deduplicated to generate a deduplicated target vibration test position group; S16, for any two deduplicated target vibration test position groups, according to the vibration anomaly value of each simulated scraper stuck scene corresponding to each target vibration test position included in the two deduplicated target vibration test position groups, the correlation coefficient of the two deduplicated target vibration test position groups is calculated; S17, according to the correlation coefficient of any two deduplicated target vibration test position groups, multiple deduplicated target vibration test position groups are screened to determine multiple deduplicated target vibration test position groups after screening, and each target vibration test position included in the multiple deduplicated target vibration test position groups after screening is used as a vibration monitoring position.

[0008] Furthermore, the abnormality judgment module judges whether the operation of the manure scraper occurs abnormally according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points, including: calculating the real-time scraper abnormality value according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points; calculating the real-time motor operation abnormality value according to the current and voltage of the motor of the manure scraper at the multiple consecutive time points; judging to perform time difference check according to the real-time scraper abnormality value and the real-time motor operation abnormality value; if it is determined to perform time difference check, determining the starting point of vibration abnormality time according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points, determining the starting point of motor operation abnormality time according to the current and voltage of the motor of the manure scraper at the multiple consecutive time points, and calculating the time difference between the starting point of vibration abnormality time and the starting point of motor operation abnormality time; judging whether the time difference meets the preset conditions, and if so, judging that the operation of the manure scraper occurs abnormally.

[0009] Furthermore, the abnormality judgment module calculates the real-time scraper abnormality value based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points, including: determining the real-time vibration abnormality value of each vibration monitoring position based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points; determining the comprehensive vibration abnormality value corresponding to each deduplicated target vibration test position group based on the real-time vibration abnormality value of each vibration monitoring position; calculating the real-time scraper abnormality value based on the correlation coefficient of any two deduplicated target vibration test position groups and the comprehensive vibration abnormality value corresponding to each deduplicated target vibration test position group.

[0010] Furthermore, the abnormality judgment module calculates the real-time motor operation abnormality value according to the current and voltage of the motor of the manure scraper at the multiple consecutive time points, including: generating a real-time current waveform according to the current of the motor of the manure scraper at the multiple consecutive time points; generating a real-time active power waveform according to the current and voltage of the motor of the manure scraper at the multiple consecutive time points; extracting real-time current waveform features from the real-time current waveform; extracting real-time active power waveform features from the real-time active power waveform; calculating the real-time motor operation abnormality value according to the real-time current waveform features and the real-time active power waveform features.

[0011] Furthermore, the abnormality detection module determines whether the scraper is stuck to the cow's hoof based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points, including: determining the abnormal vibration monitoring position according to the real-time vibration abnormality value of each vibration monitoring position; determining the abnormal vibration characteristics of the abnormal vibration monitoring position according to the vibration data collected by the vibration sensing components at the abnormal vibration monitoring position at multiple consecutive time points; extracting the real-time current mutation characteristics from the real-time current waveform; extracting the real-time active power mutation characteristics from the real-time active power waveform; and judging whether the scraper is stuck to the cow's hoof based on the abnormal vibration characteristics of the abnormal vibration monitoring position, the real-time current mutation characteristics and the real-time active power mutation characteristics through a state judgment model.

[0012] Furthermore, the abnormal detection module determines the abnormal vibration characteristics of the abnormal vibration monitoring position based on the vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points, including: performing variational modal decomposition on the vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points to generate multiple inherent modal components; and determining the abnormal vibration characteristics of the abnormal vibration monitoring position from the multiple inherent modal components.

[0013] Furthermore, the real-time current mutation characteristics include at least a real-time current mutation slope, a real-time current mutation amplitude and a real-time current mutation duration; the real-time active power mutation characteristics include at least a real-time power mutation slope, a real-time power mutation amplitude and a real-time power mutation duration.

[0014] The present invention provides a method for detecting abnormal operation of a manure scraper, which is applied to the above-mentioned abnormal operation detection system of a manure scraper, comprising: judging whether an abnormal operation of the manure scraper occurs according to vibration data collected by multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points; after judging that the operation of the manure scraper occurs abnormally, judging whether a scraper is stuck to a cow's hoof according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points; after judging that the scraper is stuck to the cow's hoof, controlling the motor to stop running and generating a warning message.

[0015] Compared with the prior art, the method and system for detecting abnormal operation of a manure scraper provided by the present invention have at least the following beneficial effects:

[0016] 1. It can monitor the operation status of the manure scraper in real time without the need for continuous manual observation, which greatly reduces the burden of manual monitoring and improves the accuracy and timeliness of monitoring. Through the combination of vibration detection module and motor detection module, the system can quickly identify abnormalities when the manure scraper is stuck in the cow's hoof or may cause danger to the cow, and stop the motor operation in time through the operation control module, effectively avoiding the risk of hoof injury or cattle tripping and injury. It improves the overall safety of the cowshed, reduces safety accidents caused by abnormal operation of the manure scraper, and provides a safer living environment for cattle. By timely detecting and handling abnormal operation of the manure scraper, the system helps to reduce economic losses caused by injuries or deaths of cattle, and also avoids downtime and maintenance costs caused by equipment failure. It can automatically generate warning information, which is convenient for managers to quickly respond to and handle abnormal situations, and improves the efficiency and response speed of cowshed management.

[0017] 2. By screening and grouping, the most representative vibration monitoring locations can be determined, thereby improving the accuracy of anomaly detection. By simulating test data and calculating vibration anomaly values, it is possible to identify which vibration test locations are most sensitive to the detection of abnormal operation of the manure scraper. This helps to optimize the layout of vibration sensing components and reduce unnecessary equipment investment while maintaining efficient monitoring capabilities. By calculating the vibration anomaly similarity and correlation coefficient, multiple target vibration test locations are grouped and deduplicated, which helps to identify and exclude redundant or highly correlated test locations, thereby improving the robustness and stability of the entire system. By accurately selecting vibration monitoring locations, problems can be identified more quickly when the manure scraper is operating abnormally, thereby shortening response time, reducing potential losses and risks, and being able to more accurately identify true abnormal events, reduce the occurrence of false alarms, and improve reliability and user experience.

[0018] 3. By calculating the real-time scraper abnormality value and the real-time motor operation abnormality value, and combining the time difference check, it is possible to more accurately determine whether the manure scraper is operating abnormally. The comprehensive analysis of vibration data and motor current and voltage data provides multi-dimensional monitoring information, which helps to reduce false alarms and missed alarms. By extracting the real-time current waveform and the real-time active power waveform, and calculating the features, the abnormality of the motor operation state can be quickly identified. Through the state judgment model, it is possible to quickly determine whether the scraper is stuck with the cow's hoof, so that timely measures can be taken to avoid accidents. The variational mode decomposition of vibration data and the extraction of abnormal vibration features help to more accurately identify abnormal situations such as the scraper getting stuck with the cow's hoof. The introduction of real-time current mutation features and real-time active power mutation features provides additional monitoring dimensions and enhances reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0020] Figure 1 It is a module schematic diagram of a manure scraper operation abnormality detection system according to some embodiments of this specification;

[0021] Figure 2 is a schematic diagram of a process for determining multiple vibration monitoring positions according to some embodiments of this specification;

[0022] Figure 3 It is a flow chart of a method for detecting abnormal operation of a manure scraper according to some embodiments of this specification. DETAILED DESCRIPTION

[0023] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0024] Figure 1 is a schematic diagram of a module of a manure scraper operation abnormality detection system according to some embodiments of this specification, such as Figure 1 As shown, a manure scraper operation abnormality detection system may include a vibration detection module, a motor detection module, an abnormality judgment module, an abnormality detection module and an operation control module.

[0025] The vibration detection module may include a plurality of vibration sensing components arranged at a plurality of positions of the scraper of the manure scraper, wherein the vibration sensing components may include vibration sensors.

[0026] In some embodiments, the vibration detection module is further configured to:

[0027] Determine a plurality of vibration test positions on the scraper, and set a vibration sensing component at each vibration test position, for example, determine a plurality of uniform vibration test positions on the scraper at preset intervals;

[0028] Under multiple simulated scraper stuck scenes, simulated test data of the vibration sensing component set at each vibration test position is obtained, wherein the simulated test data includes vibration data (for example, vibration displacement, vibration velocity, vibration acceleration and amplitude, etc.) at multiple continuous time points. Under different simulated scraper stuck scenes, the positions where the simulated cow hooves are stuck on the scraper may be different and / or the body shapes of the cows corresponding to the simulated cow hooves stuck on the scraper may be different.

[0029] According to the simulated test data collected by the vibration sensing component set at each vibration test position under multiple simulated scraper stuck scenes, multiple vibration test positions are screened to determine multiple vibration monitoring positions;

[0030] According to the multiple vibration monitoring positions, multiple vibration sensing components are set.

[0031] Figure 2 is a schematic diagram of a process for determining multiple vibration monitoring positions according to some embodiments of this specification, such as Figure 2 As shown, in some embodiments, the vibration detection module screens multiple vibration test positions according to the simulated test data collected by the vibration sensing component set at each vibration test position in multiple simulated scraper stuck scenes, and determines multiple vibration monitoring positions, including:

[0032] S11, for each vibration test position, according to the simulated test data collected by the vibration sensing component set at the vibration test position under multiple simulated scraper stuck scenes, determine the vibration abnormality value of the vibration test position corresponding to each simulated scraper stuck scene;

[0033] S12, for each vibration test position, according to the vibration abnormality value corresponding to each simulated scraper stuck scene of the vibration test position, determine whether the vibration test position is a target vibration test position, if so, execute S13, if not, screen out the vibration test position;

[0034] S13. For any two target vibration test positions, according to the vibration anomaly values ​​of each simulated scraper stuck scene corresponding to the two target vibration test positions, calculate the vibration anomaly similarity of the two target vibration test positions;

[0035] S14, grouping a plurality of target vibration test positions according to the vibration anomaly similarity between any two target vibration test positions, and determining a plurality of target vibration test position groups;

[0036] S15. For each target vibration test position group, according to the vibration anomaly similarity between any two target vibration test positions included in the target vibration test position group, deduplication is performed on multiple target vibration test positions included in the target vibration test position group to generate a deduplication target vibration test position group;

[0037] S16. For any two deduplicated target vibration test position groups, calculate the correlation coefficient of the two deduplicated target vibration test position groups according to the vibration abnormality value of each target vibration test position included in the two deduplicated target vibration test position groups corresponding to each simulated scraper stuck scene;

[0038] S17. Screening multiple deduplicated target vibration test position groups according to the correlation coefficient of any two deduplicated target vibration test position groups, determining the screened multiple deduplicated target vibration test position groups, and using each target vibration test position included in the screened multiple deduplicated target vibration test position groups as a vibration monitoring position.

[0039] Specifically, standard test data collected by the vibration sensing component set at the vibration test position under the normal working scenario of the scraper can be obtained in advance. For each vibration test position, the vibration abnormality value of the vibration test position corresponding to each simulated scraper jam scenario is determined based on the simulated test data collected by the vibration sensing component set at the vibration test position under multiple simulated scraper jam scenarios and the standard test data collected by the vibration sensing component set at the vibration test position under the normal working scenario of the scraper.

[0040] As an example only, the simulated test data collected by the vibration sensing component set at the vibration test position in multiple simulated scraper stuck scenes and the standard test data collected by the vibration sensing component set at the vibration test position in the scraper normal working scene can be cleaned first, outliers and noise are removed, and standardized to ensure that the data in the scraper normal working scene and the simulated scraper stuck scene have the same dimension and scale. From the simulated test data corresponding to the simulated scraper stuck scene after preprocessing, the vibration displacement characteristics (for example, maximum vibration displacement, mean vibration displacement, standard deviation of vibration displacement, etc.), vibration velocity characteristics (for example, maximum vibration velocity, mean vibration velocity, standard deviation of vibration velocity, etc.), vibration acceleration characteristics (for example, maximum vibration acceleration, mean vibration acceleration, standard deviation of vibration acceleration, etc.), amplitude characteristics (for example, maximum amplitude, mean amplitude, standard deviation of amplitude, etc.) corresponding to the simulated scraper stuck scene can be extracted, and the vibration displacement characteristics, vibration velocity characteristics, vibration acceleration characteristics and amplitude characteristics corresponding to the normal working scraper scene can be extracted from the standard test data corresponding to the preprocessed scraper normal working scene. Then, based on the vibration displacement characteristics, vibration velocity characteristics, vibration acceleration characteristics and amplitude characteristics corresponding to the simulated scraper stuck scenario and the vibration displacement characteristics, vibration velocity characteristics, vibration acceleration characteristics and amplitude characteristics corresponding to the scraper normal working scenario, the vibration abnormal value of the vibration test position corresponding to the simulated scraper stuck scenario is calculated.

[0041] For example, the vibration anomaly value of the vibration test location corresponding to the simulated scraper stuck scenario can be calculated according to the following formula:

[0042]

[0043] S (i,j,n) =cos(F (i,j,n) ,F (i,n) ),n=1、2、3、4

[0044] Among them, O i is the vibration anomaly value corresponding to the jth simulated scraper stuck scenario at the i-th vibration test position, P 1 is the preset parameter, P 1 Greater than 0, a 1 、a 2 、a 3 and a 4 is the preset weight, a 1 、a 2 、a 3 and a 4 Greater than 0, a 1 +a 2 +a 3 +a 4 =1, S (i,j,1) is the vibration displacement feature similarity between the vibration displacement feature corresponding to the jth simulated scraper stuck scenario at the ith vibration test position and the vibration displacement feature corresponding to the normal working scraper scenario at the ith vibration test position, S (i,j,2) is the vibration velocity feature similarity between the vibration velocity feature corresponding to the jth simulated scraper stuck scenario at the ith vibration test position and the vibration velocity feature corresponding to the normal working scraper scenario at the ith vibration test position, S (i,j,3) is the similarity between the vibration acceleration feature corresponding to the jth simulated scraper stuck scene at the ith vibration test position and the vibration acceleration feature corresponding to the normal working scraper scene at the ith vibration test position, S (i,j,4) S is the amplitude feature similarity between the amplitude feature corresponding to the jth simulated scraper stuck scene at the i-th vibration test position and the amplitude feature corresponding to the normal working scraper scene at the i-th vibration test position, (i,j,n) is the vibration displacement feature similarity, vibration velocity feature similarity, vibration acceleration feature similarity or amplitude feature similarity, F (i,j,n) is the vibration displacement feature, vibration velocity feature, vibration acceleration feature or amplitude feature corresponding to the jth simulated scraper stuck scene at the i-th vibration test position, F (i,n) is the vibration displacement characteristic, vibration velocity characteristic, vibration acceleration characteristic or amplitude characteristic corresponding to the i-th vibration test position in the normal working scene of the scraper, cos(F(i,j,n) ,F (i,n) ) is F (i,j,n) and F (i,n) The cosine similarity of .

[0045] The vibration anomaly values ​​corresponding to each simulated scraper stuck scenario at the vibration test position can be averaged to obtain the mean vibration anomaly value corresponding to the vibration test position. When the mean vibration anomaly value corresponding to the vibration test position is greater than the vibration anomaly value mean threshold, the vibration test position can be used as the target vibration test position.

[0046] For each target vibration test position, the abnormal flag of the target vibration test position corresponding to each simulated scraper jam scenario is determined according to the vibration abnormal value of each simulated scraper jam scenario corresponding to the target vibration test position. When the vibration abnormal value of the target vibration test position corresponding to a simulated scraper jam scenario is greater than the vibration abnormal value threshold, the abnormal flag of the target vibration test position corresponding to the simulated scraper jam scenario is 1, otherwise it is 0.

[0047] The vibration anomaly similarity of two target vibration test locations can be calculated according to the following formula:

[0048]

[0049] Among them, S (e,f) is the vibration anomaly similarity between the e-th target vibration test position and the f-th target vibration test position, P 2 is the preset parameter, P 2 Greater than 0, sign (e,n) is the abnormal sign of the nth simulated scraper stuck scene corresponding to the eth target vibration test position, sign (f,n) is the abnormal identification of the nth simulated scraper stuck scene corresponding to the fth target vibration test position, and N is the total number of simulated scraper stuck scenes.

[0050] The K-means algorithm can be used to group multiple target vibration test positions according to the vibration anomaly similarity between any two target vibration test positions to determine multiple target vibration test position groups.

[0051] The multiple target vibration test positions included in the target vibration test position group may be deduplicated according to the following process:

[0052] S21, two target vibration test positions whose vibration anomaly similarity is greater than a vibration anomaly similarity threshold are used as target vibration test positions to be deduplicated;

[0053] S22, averaging the vibration anomaly similarities of the target vibration test position to be deduplicated and other target vibration test positions included in the target vibration test position group, calculating the mean vibration anomaly similarity corresponding to the target vibration test position to be deduplicated, and filtering out the one with the larger mean vibration anomaly similarity of the two target vibration test positions to be deduplicated;

[0054] S23, determining whether the vibration anomaly similarities of any two target vibration test positions included in the target vibration test position group are both less than or equal to the vibration anomaly similarity threshold; if so, performing deduplication; if not, executing S21.

[0055] For each deduplicated target vibration test position group, the mean of the vibration anomaly values ​​of each target vibration test position included in the deduplicated target vibration test position group corresponding to the simulated scraper stuck scene can be used as the vibration anomaly value of the deduplicated target vibration test position group corresponding to the simulated scraper stuck scene. The correlation coefficient of two deduplicated target vibration test position groups can be calculated based on the vibration anomaly values ​​of each simulated scraper stuck scene corresponding to any two deduplicated target vibration test position groups.

[0056] For example, the correlation coefficient of two target vibration test position groups after deduplication can be calculated according to the following formula:

[0057]

[0058] Among them, r (e,f) For the e-th target vibration test position and the f-th target vibration test position, O (e,n) is the vibration anomaly value of the nth simulated scraper stuck scene corresponding to the eth target vibration test position group after deduplication, O (f,n) It is the vibration anomaly value of the nth simulated scraper stuck scenario corresponding to the fth target vibration test position group after deduplication.

[0059] Multiple deduplicated target vibration test position groups can be screened according to the following process:

[0060] S31. For each target vibration test position group after deduplication, the correlation coefficients of the target vibration test position group after deduplication and each other target vibration test position group after deduplication may be averaged to obtain the correlation coefficient mean of the target vibration test position group after deduplication;

[0061] S32, determining whether there is at least one target vibration test position group after deduplication whose correlation coefficient mean is less than the correlation coefficient mean threshold, if so, executing S33, if not, completing the screening;

[0062] S33, filter out the target vibration test position group after deduplication with the smallest mean value of correlation coefficient, and execute S31.

[0063] The motor detection module can be used to obtain the current and voltage of the motor of the manure scraper.

[0064] Specifically, the motor detection module may include a current sensor and a voltage sensor for obtaining the current and voltage of the motor of the manure scraper.

[0065] The abnormality judgment module can be used to judge whether the operation of the manure scraper is abnormal based on the vibration data collected by multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at multiple consecutive time points.

[0066] Specifically include:

[0067] Calculate real-time scraper anomaly values ​​based on vibration data collected by multiple vibration sensing components at multiple consecutive time points;

[0068] According to the current and voltage of the motor of the manure scraper at multiple consecutive time points, the real-time motor operation abnormal value is calculated;

[0069] According to the real-time abnormal scraper value and the real-time abnormal motor operation value, the time difference calibration is performed;

[0070] If it is determined to perform time difference calibration, the vibration abnormality time starting point is determined based on the vibration data collected by multiple vibration sensing components at multiple consecutive time points, the motor operation abnormality time starting point is determined based on the current and voltage of the manure scraper motor at multiple consecutive time points, and the time difference between the vibration abnormality time starting point and the motor operation abnormality time starting point is calculated;

[0071] Determine whether the time difference meets the preset conditions. If so, determine that the manure scraper is operating abnormally.

[0072] In some embodiments, the abnormality determination module calculates the real-time scraper abnormality value based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points, including:

[0073] Determine a real-time vibration abnormality value at each vibration monitoring location based on vibration data collected by multiple vibration sensing components at multiple consecutive time points;

[0074] According to the real-time vibration abnormality value of each vibration monitoring position, determine the comprehensive vibration abnormality value corresponding to each target vibration test position group after deduplication;

[0075] The real-time scraper anomaly value is calculated based on the correlation coefficient of any two deduplicated target vibration test position groups and the comprehensive vibration anomaly value corresponding to each deduplicated target vibration test position group.

[0076] Specifically, for each vibration monitoring position, the real-time vibration displacement characteristics, vibration velocity characteristics, vibration acceleration characteristics and amplitude characteristics of the scraper are extracted from the vibration data collected by the vibration sensing component set at the vibration monitoring position at multiple consecutive time points. Then, based on the real-time vibration displacement characteristics, vibration velocity characteristics, vibration acceleration characteristics and amplitude characteristics of the scraper and the vibration displacement characteristics, vibration velocity characteristics, vibration acceleration characteristics and amplitude characteristics corresponding to the normal working scene of the scraper, the real-time vibration abnormality value of the vibration monitoring position is calculated. The method of calculating the real-time vibration abnormality value of the vibration monitoring position is similar to the method of calculating the vibration abnormality value of the vibration test position corresponding to the simulated scraper stuck scene, which will not be repeated here.

[0077] The real-time vibration abnormality value of each vibration monitoring position included in the target vibration test position group after deduplication may be averaged to serve as the comprehensive vibration abnormality value corresponding to the target vibration test position group after deduplication.

[0078] A multiple regression model may be established, wherein the independent variables of the multiple regression model include the comprehensive vibration anomaly value corresponding to each target vibration test position group after deduplication, the dependent variables of the multiple regression model include the real-time scraper anomaly value, and the parameters of the multiple regression model include the correlation coefficient of any two target vibration test position groups after deduplication. The real-time scraper anomaly value is calculated by the multiple regression model according to the correlation coefficient of any two target vibration test position groups after deduplication and the comprehensive vibration anomaly value corresponding to each target vibration test position group after deduplication.

[0079] In some embodiments, the abnormality judgment module calculates the real-time motor operation abnormality value according to the current and voltage of the motor of the manure scraper at multiple consecutive time points, including:

[0080] Generate a real-time current waveform according to the current of the motor of the manure scraper at multiple consecutive time points;

[0081] Generate a real-time active power waveform according to the current and voltage of the motor of the manure scraper at multiple consecutive time points;

[0082] Extracting real-time current waveform features from the real-time current waveform, wherein the real-time current waveform features may include at least a real-time current peak value, a real-time current average value, and a real-time current fluctuation rate;

[0083] Extracting real-time active power waveform features from the real-time active power waveform, wherein the real-time active power waveform features may at least include a real-time active power peak value, a real-time active power average value, and a real-time active power fluctuation rate;

[0084] The real-time motor operation abnormality value is calculated based on the real-time current waveform characteristics and the real-time active power waveform characteristics.

[0085] Specifically, the current and voltage of the manure scraper motor at multiple consecutive time points in the normal working scenario of the scraper can be obtained, and the standard current waveform characteristics and standard active power waveform characteristics corresponding to the normal working scenario of the scraper can be extracted from the current and voltage of the manure scraper motor at multiple consecutive time points in the normal working scenario of the scraper. The real-time motor operation abnormality value can be calculated based on the standard current waveform characteristics and standard active power waveform characteristics corresponding to the normal working scenario of the scraper as well as the real-time current waveform characteristics and real-time active power waveform characteristics.

[0086] For example, the real-time motor operation abnormality value can be calculated according to the following formula:

[0087]

[0088] Among them, O electrical For P 2 is the preset parameter, P 2 Greater than 0, b 1 and b 2 is the preset weight, b 1 and b 2 Greater than 0, b 1 +b 2 =1, F (current,1) is the standard current waveform characteristic, (F (current,2) is the real-time current waveform characteristic, cos(F (current,1) ,F (current,2) ) is the cosine similarity between the standard current waveform feature and the real-time current waveform feature, F (voltage,1) is the standard active power waveform characteristic, F (voltage,2) is the real-time active power waveform characteristic, cos(F (voltage,1) ,F (voltage,2) ) is the cosine similarity between the standard active power waveform characteristics and the real-time active power waveform characteristics.

[0089] The abnormality detection module can be used to determine whether the scraper is stuck with the cow's hoof based on the vibration data collected by multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the scraper at multiple consecutive time points after the abnormality judgment module determines that the manure scraper is operating abnormally.

[0090] Specifically include:

[0091] Determine an abnormal vibration monitoring position according to the real-time vibration abnormality value of each vibration monitoring position, for example, a vibration monitoring position whose real-time vibration abnormality value is greater than a real-time vibration abnormality value threshold is taken as an abnormal vibration monitoring position;

[0092] Determine the abnormal vibration characteristics of the abnormal vibration monitoring position according to the vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points;

[0093] Extracting real-time current mutation features from the real-time current waveform, wherein the real-time current mutation features at least include real-time current mutation slope, real-time current mutation amplitude and real-time current mutation duration;

[0094] Extracting a real-time active power mutation feature from the real-time active power waveform, wherein the real-time active power mutation feature at least includes a real-time power mutation slope, a real-time power mutation amplitude, and a real-time power mutation duration;

[0095] The state judgment model is used to judge whether the scraper is stuck with the cow's hoof based on the abnormal vibration characteristics, real-time current mutation characteristics and real-time active power mutation characteristics of the abnormal vibration monitoring position. The state judgment model can be a convolutional neural network model.

[0096] Specifically, the real-time current mutation feature can be extracted from the real-time current waveform and the real-time active power mutation feature can be extracted from the real-time active power waveform through a mutation feature extraction model, wherein the mutation feature extraction model can be a convolutional neural network model.

[0097] In some embodiments, the abnormality detection module determines the abnormal vibration characteristics of the abnormal vibration monitoring position according to the vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points, including:

[0098] Performing variational modal decomposition on vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points to generate multiple natural modal components;

[0099] The abnormal vibration characteristics of the abnormal vibration monitoring position are determined from multiple natural modal components, wherein the abnormal vibration characteristics of the abnormal vibration monitoring position may at least include the vibration frequency (e.g., center frequency, frequency bandwidth, etc.), time domain characteristics (e.g., maximum amplitude, average amplitude, amplitude fluctuation, etc.), energy characteristics (e.g., energy value, energy proportion, etc.), phase characteristics (e.g., phase information, phase stability, etc.) of each natural modal component. The center frequency of each natural modal component is the main vibration frequency represented by the natural modal component. The frequency range or bandwidth of the natural modal component reflects the degree of dispersion of the frequency component. The maximum amplitude value of the natural modal component in the time domain reflects the vibration intensity of the frequency component. The average amplitude value of the natural modal component in the time domain is used to evaluate the overall vibration level of the frequency component. The degree of change of the amplitude of the natural modal component over time can reflect the stability or periodicity of the vibration. The energy or power spectral density of the natural modal component represents the energy contribution of the frequency component in the vibration signal. The energy proportion represents the proportion of the energy of the natural modal component to the total energy of the entire signal, which is used to evaluate the importance of the frequency component. Phase information can characterize the change of the phase of the natural modal component over time, and can reflect the phase delay or phase difference of the vibration signal. Phase stability can characterize the stability or rate of change of the phase of the natural modal component, and is used to evaluate the phase consistency of vibration.

[0100] The operation control module can be used to control the motor to stop running and generate a warning message after the abnormal detection module determines that the scraper is stuck with the cow's hoof.

[0101] Figure 3 is a flow chart of a method for detecting abnormal operation of a manure scraper according to some embodiments of this specification, such as Figure 3 As shown, a method for detecting abnormal operation of a manure scraper may include the following steps:

[0102] Step 310, judging whether the operation of the manure scraper is abnormal based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at multiple consecutive time points;

[0103] Step 320, after determining that the manure scraper is operating abnormally, determine whether the scraper is stuck to the cow's hoof based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at multiple consecutive time points;

[0104] Step 330, after determining that the scraper is stuck on the cow's hoof, the motor is controlled to stop running and a warning message is generated.

[0105] A method for detecting abnormal operation of a manure scraper can be applied to a system for detecting abnormal operation of a manure scraper. For more description of the method for detecting abnormal operation of a manure scraper, reference can be made to the relevant description of a system for detecting abnormal operation of a manure scraper, which will not be repeated here.

[0106] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A manure scraper operation abnormality detection system, characterized in that: include: A vibration detection module includes a plurality of vibration sensing components arranged at a plurality of positions of a scraper of the manure scraper; A motor detection module, used to obtain the current and voltage of the motor of the manure scraper; An abnormality judgment module is used to judge whether an abnormality occurs in the operation of the manure scraper according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points; An abnormality detection module is used to determine whether the scraper is stuck with the cow's hoof according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the scraper at the multiple consecutive time points after the abnormality judgment module determines that the manure scraper is running abnormally; The operation control module is used to control the motor to stop running and generate a warning message after the abnormality detection module determines that the scraper is stuck with the cow's hoof.

2. The manure scraper operation abnormality detection system according to claim 1 is characterized in that: The vibration detection module is also used for: Determine a plurality of vibration test positions on the scraper, and set a vibration sensing component at each vibration test position; Under multiple simulated scraper stuck scenarios, obtaining simulated test data of the vibration sensing component set at each vibration test position, wherein the simulated test data includes vibration data at multiple consecutive time points; According to the simulated test data collected by the vibration sensing assembly provided at each vibration test position under the multiple simulated scraper stuck scenes, the multiple vibration test positions are screened to determine multiple vibration monitoring positions; The plurality of vibration sensing components are arranged according to the plurality of vibration monitoring positions.

3. A manure scraper operation abnormality detection system according to claim 2, characterized in that: The vibration detection module screens the multiple vibration test positions according to the simulated test data collected by the vibration sensing component set at each vibration test position under the multiple simulated scraper stuck scenes, and determines multiple vibration monitoring positions, including: S11, for each vibration test position, according to the simulated test data collected by the vibration sensing component arranged at the vibration test position under multiple simulated scraper stuck scenes, determine the vibration abnormality value of the vibration test position corresponding to each simulated scraper stuck scene; S12, for each vibration test position, judging whether the vibration test position is a target vibration test position according to the vibration abnormality value corresponding to each simulated scraper stuck scene of the vibration test position, if so, executing S13, if not, screening out the vibration test position; S13, for any two target vibration test positions, according to the vibration anomaly values ​​of the two target vibration test positions corresponding to each simulated scraper stuck scene, calculating the vibration anomaly similarity of the two target vibration test positions; S14, grouping a plurality of target vibration test positions according to the vibration anomaly similarity between any two target vibration test positions, and determining a plurality of target vibration test position groups; S15, for each target vibration test position group, according to the vibration anomaly similarity between any two target vibration test positions included in the target vibration test position group, deduplication is performed on a plurality of target vibration test positions included in the target vibration test position group to generate a deduplication target vibration test position group; S16. For any two deduplicated target vibration test position groups, calculate the correlation coefficient of the two deduplicated target vibration test position groups according to the vibration anomaly value of each target vibration test position included in the two deduplicated target vibration test position groups corresponding to each simulated scraper stuck scenario; S17. Screening multiple deduplicated target vibration test position groups according to the correlation coefficient of any two deduplicated target vibration test position groups, determining the screened multiple deduplicated target vibration test position groups, and using each target vibration test position included in the screened multiple deduplicated target vibration test position groups as a vibration monitoring position.

4. The manure scraper operation abnormality detection system according to claim 3 is characterized in that: The abnormality judgment module judges whether the operation of the manure scraper is abnormal according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points, including: Calculating a real-time scraper abnormality value based on the vibration data collected by the plurality of vibration sensing components at a plurality of consecutive time points; Calculate the real-time motor operation abnormality value according to the current and voltage of the motor of the manure scraper at the plurality of consecutive time points; According to the real-time scraper abnormal value and the real-time motor operation abnormal value, it is determined to perform time difference calibration; If it is determined to perform a time difference check, the vibration abnormality time starting point is determined based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points, the motor operation abnormality time starting point is determined based on the current and voltage of the motor of the manure scraper at the multiple consecutive time points, and the time difference between the vibration abnormality time starting point and the motor operation abnormality time starting point is calculated; Determine whether the time difference meets the preset conditions, and if so, determine that the manure scraper is operating abnormally.

5. The manure scraper operation abnormality detection system according to claim 4, characterized in that: The abnormality judgment module calculates the real-time scraper abnormality value according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points, including: Determining a real-time vibration abnormality value of each vibration monitoring position according to the vibration data collected by the plurality of vibration sensing components at a plurality of consecutive time points; According to the real-time vibration anomaly value of each vibration monitoring position, determine the comprehensive vibration anomaly value corresponding to each target vibration test position group after deduplication; The real-time scraper anomaly value is calculated based on the correlation coefficient of any two target vibration test position groups after deduplication and the comprehensive vibration anomaly value corresponding to each target vibration test position group after deduplication.

6. The manure scraper operation abnormality detection system according to claim 5, characterized in that: The abnormality judgment module calculates the real-time motor operation abnormality value according to the current and voltage of the motor of the manure scraper at the plurality of consecutive time points, including: generating a real-time current waveform according to the current of the motor of the manure scraper at the plurality of consecutive time points; generating a real-time active power waveform according to the current and voltage of the motor of the manure scraper at the plurality of consecutive time points; extracting real-time current waveform features from the real-time current waveform; Extracting real-time active power waveform features from the real-time active power waveform; The real-time motor operation abnormality value is calculated according to the real-time current waveform characteristics and the real-time active power waveform characteristics.

7. The manure scraper operation abnormality detection system according to claim 6, characterized in that: The abnormality detection module determines whether the scraper is stuck with the cow's hoof according to the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points, including: Determine the abnormal vibration monitoring position according to the real-time vibration abnormality value of each vibration monitoring position; Determine the abnormal vibration characteristics of the abnormal vibration monitoring position according to the vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points; Extracting a real-time current mutation feature from the real-time current waveform; Extracting a real-time active power mutation feature from the real-time active power waveform; The state determination model is used to determine whether the scraper is stuck with the cow's hoof based on the abnormal vibration characteristics of the abnormal vibration monitoring position, the real-time current mutation characteristics and the real-time active power mutation characteristics.

8. The manure scraper operation abnormality detection system according to claim 7, characterized in that: The abnormal detection module determines the abnormal vibration characteristics of the abnormal vibration monitoring position according to the vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points, including: Performing variational modal decomposition on vibration data collected by the vibration sensing component at the abnormal vibration monitoring position at multiple consecutive time points to generate multiple natural modal components; An abnormal vibration feature of an abnormal vibration monitoring position is determined from the plurality of natural mode components.

9. The manure scraper operation abnormality detection system according to claim 7, characterized in that: The real-time current mutation characteristics at least include the real-time current mutation slope, the real-time current mutation amplitude and the real-time current mutation duration; The real-time active power mutation characteristics at least include a real-time power mutation slope, a real-time power mutation amplitude and a real-time power mutation duration.

10. A method for detecting abnormal operation of a manure scraper, characterized in that: A manure scraper operation abnormality detection system applied to any one of claims 1 to 9, comprising: Determine whether the operation of the manure scraper is abnormal based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points; After determining that the operation of the manure scraper is abnormal, determine whether the scraper is stuck to the cow's hoof based on the vibration data collected by the multiple vibration sensing components at multiple consecutive time points and the current and voltage of the motor of the manure scraper at the multiple consecutive time points; after determining that the scraper is stuck to the cow's hoof, control the motor to stop running and generate a warning message.

Images