Intelligent mechanical motion state monitoring system driven by sensor bearing data

Abstract

The invention discloses a sensor bearing data-driven mechanical motion state intelligent monitoring system, which belongs to the technical field of monitoring equipment, and comprises an acquisition module, a processing module, a wireless communication module and an analysis module, the acquisition module and the processing module are arranged inside the bearing body, the wireless communication module and the analysis module are arranged outside the bearing body, and the acquisition module comprises a vibration acceleration sensor, a temperature sensor and a rotating speed sensor. The vibration acceleration sensor, the temperature sensor and the rotating speed sensor are used for collecting vibration acceleration data, temperature data and rotating speed data in the bearing operation process respectively, and the processing module comprises a microprocessor and a protocol converter. According to the method, the weighted root-mean-square acceleration is calculated, potential faults or abnormal conditions can be found in time, abnormal data are removed through the real abnormal degree, and the abnormal state of the bearing can be effectively monitored.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring devices, and particularly to an intelligent monitoring system for mechanical motion states driven by sensor bearing data. Background Art

[0002] Bearing monitoring and diagnosis is developing towards the direction of automation, intelligence, and integration. An intelligent bearing refers to integrating sensors with different uses on the basis of a traditional bearing to form a unique structural unit by combining them into one, and then performing information processing through a computer to achieve the purpose of real-time online monitoring.

[0003] A bearing is an important part for supporting a mechanical rotating body. A large number of facts have proved that many machine failures caused by bearing damage occur before the expected life of the bearing. Effectively monitoring the operating state of the bearing is an effective means to avoid catastrophic consequences caused by bearing failures. The selection of sensor measurement points is the primary problem to be solved in bearing state monitoring. Generally, the monitoring of the bearing is to obtain the working state signal of the bearing by installing a vibration sensor on the bearing housing or box. The signal collected by this method contains, in addition to the working information of the bearing itself, the noise signals generated by other moving parts in the equipment, which is very unfavorable for the monitoring of bearing failures; moreover, for the prediction of temperature data during the operation of the bearing, generally, the difference between the temperature data of a single bearing operation and the time series change trend is used to reflect the abnormality of the temperature data. Among them, the time series change trend often takes a local range with a fixed length as a unit. However, due to the obvious differences in the temperature data change trends at different positions of the bearing, only considering the temperature data within a local range with a fixed length is likely to cause information loss, thereby reducing the accuracy of a single temperature data and affecting the prediction and analysis of temperature data.

[0004] Therefore, there is an urgent need to provide an intelligent monitoring system for mechanical motion states driven by sensor bearing data to solve the above problems. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the disadvantages of the prior art that the monitoring of bearings is to obtain the working state signals of bearings by installing vibration sensors on the bearing housing or box body. The signals collected by this method contain not only the working information of the bearings themselves, but also the noise signals generated by other moving parts in the equipment, which is very unfavorable for the monitoring of bearing faults. Moreover, for the prediction of temperature data during the operation of bearings, generally, the difference between the temperature data of a single bearing operation and the time-series change trend is used to reflect the abnormality of the temperature data. Among them, the time-series change trend often uses a local range of a fixed length as a unit. However, due to the obvious differences in the temperature data change trends at different positions of the bearings, only considering the temperature data within a local range of a fixed length is likely to cause information loss, thereby reducing the accuracy of a single temperature data and affecting the prediction and analysis of temperature data. The present invention provides an intelligent monitoring system for mechanical motion states driven by sensor-bearing data.

[0006] To solve the above technical problems, a technical solution adopted by the present invention is: to provide an intelligent monitoring system for mechanical motion states driven by sensor-bearing data, including an acquisition module, a processing module, a wireless communication module, and an analysis module. The acquisition module and the processing module are arranged inside the bearing body, and the wireless communication module and the analysis module are arranged outside the bearing body;

[0007] The acquisition module, which includes a vibration acceleration sensor, a temperature sensor, and a rotational speed sensor. The vibration acceleration sensor, the temperature sensor, and the rotational speed sensor are respectively used to collect vibration acceleration data, temperature data, and rotational speed data during the operation of the bearing;

[0008] The processing module, which includes a microprocessor and a protocol converter. The processing module is used to receive the data collected by the vibration acceleration sensor, the temperature sensor, and the rotational speed sensor, and process the data;

[0009] The microprocessor analyzes, calculates, and arranges the collected data, compiles the valid data into data frames, and the protocol converter receives the data frames and outputs them to the wireless communication module;

[0010] The wireless communication module is used to transmit the data frames to a preset cloud data management and control platform, parse the data frames to generate real-time bearing operation state data, and store the real-time bearing operation state data;

[0011] The analysis module, which includes a retrieval unit and a calculation unit. The retrieval unit is used to retrieve the real-time bearing operation state data in real time, and the calculation unit is used to detect the bearing operation state and perform corresponding fault warnings.

[0012] The present invention is further configured such that: the vibration acceleration sensor is mounted on the surface of a preset circuit board, and the circuit board is mounted inside the acquisition module;

[0013] The vibration acceleration sensor is a triaxial accelerometer, and the vibration acceleration data collected by the vibration acceleration sensor includes the acceleration data of the bearing body in the X, Y, and Z directions, the acceleration data of the bearing body at each frequency component, and the acceleration data signal.

[0014] The present invention is further configured such that: the steps of the vibration acceleration sensor in the acquisition module collecting vibration acceleration data and processing it through the processing module are as follows:

[0015] S1. Record the acceleration data of the bearing in the X, Y, and Z directions through the vibration acceleration sensor;

[0016] S2. Preprocess the collected acceleration data through the processing module to remove the noise signals collected during data acquisition, and obtain the acceleration data of each frequency component;

[0017] S3. Calculate the weighted root mean square acceleration value according to the acceleration data of each frequency component;

[0018] S4. Calculate the overall vibration acceleration data of the bearing body according to the weighted root mean square acceleration value.

[0019] The present invention is further configured such that: the removal of the noise signal from the acceleration data in step S2 includes the following steps:

[0020] S21. The microprocessor in the processing module receives the noise signal, sets different signal decomposition methods to decompose the noise signal, obtains a plurality of initial mode components, and transmits the plurality of initial mode components to the analysis module through the wireless communication module;

[0021] S22. The retrieval unit in the analysis module retrieves the initial mode components, calculates the sample entropy of all the initial mode components through the calculation unit, and performs normalization processing on it to obtain the normalized sample entropy of the initial mode components;

[0022] S23. By setting a threshold for the normalized sample entropy, perform singular value decomposition noise reduction processing on all initial mode components greater than the threshold, and do not process the initial mode components not greater than the threshold;

[0023] S24. Calculate the kurtosis coefficient of the initial mode components after singular value decomposition noise reduction processing through the calculation unit, and set a threshold for the kurtosis coefficient;

[0024] S25. Reject the initial mode components with kurtosis coefficients less than the threshold, retain the initial mode components not less than the threshold, and output the final mode components;

[0025] S26. Recombine all the processed mode components to obtain the acceleration data of each frequency component.

[0026] The present invention is further configured that the calculation formula of the weighted root mean square acceleration value in step S3 is as follows:

[0027]

[0028] Where, is the weighted root mean square acceleration value; is the measurement time length; is the th acceleration data of the frequency component; is the time differential element, obtained by integrating the time ; is the weight vector; is the time.

[0029] The present invention is further configured that the calculation formula of the overall vibration acceleration of the bearing body in step S4 is as follows:

[0030]

[0031] Where, is the overall vibration acceleration data of the bearing body; , , are the acceleration data in the X, Y, and Z axes respectively.

[0032] The present invention is further configured that the microprocessor in the processing module receives the temperature data collected by the temperature sensor, the retrieval module in the analysis module retrieves the temperature data, and processes the temperature data through the deep learning model set in the calculation unit to obtain the bearing sequential temperature data;

[0033] The calculation unit processes the bearing sequential temperature data to obtain the fitting residual of the bearing temperature data.

[0034] The present invention is further configured that the analysis module predicts the bearing temperature using the bearing sequential temperature data, including the following steps:

[0035] Q1. The retrieval unit in the analysis module obtains the bearing sequential temperature data in real time and obtains the bearing temperature data at each set sampling moment according to the bearing sequential temperature data;

[0036] Q2. Obtain the abnormality degree of the temperature data at each sampling moment in each time period interval according to the trend deviation of the bearing temperature data at each sampling moment within the set time period interval;

[0037] Q3. Obtain the abnormality stability of the bearing temperature data at each sampling moment in each time period interval according to the difference between the abnormality degree and the trend deviation of the bearing temperature data in different time period intervals;

[0038] Q4. Obtain the abnormality authenticity of the bearing temperature data at each sampling moment in each time period interval according to the abnormality stability and the distribution of the bearing temperature data within the time period interval;

[0039] Q5. Combine the abnormality authenticity and the abnormality degree to obtain the true abnormality degree of the bearing temperature data at each sampling moment, use the true abnormality degree to obtain the abnormal data in the bearing time-series temperature data, eliminate the abnormal data, and complete the temperature prediction according to the temperature data after eliminating the abnormal data;

[0040] Specifically, in step Q3, the abnormality stability of the bearing temperature data at each sampling moment in each time period interval is obtained according to the difference between the abnormality degree and the trend deviation of the bearing temperature data in different time period intervals, and its calculation formula is:

[0041]

[0042] where is the abnormality stability of the bearing temperature data at the th sampling moment in the th time period interval; is the abnormality degree of the bearing temperature data at the th sampling moment in the th time period interval; is the abnormality degree of the bearing temperature data at the th sampling moment in the th time period interval; represents the preset moving time length;

[0043] In step Q4, the abnormality authenticity of the bearing temperature data at each sampling moment in each local interval is obtained according to the abnormality stability and the distribution of the bearing temperature data within the time period interval, and its calculation formula is:

[0044] ;

[0045] where is the abnormality authenticity of the bearing temperature data at the th sampling moment in the th time period interval; is the th sampling moment of the bearing temperature data at the The abnormal stability of a time period interval; is a normalization function; is the th fitting residual of the bearing temperature data at the th sampling moment among all the bearing temperature data within the th th sampling moment of the bearing temperature data within the number of bearing temperature data included in the th th sampling moment of the bearing temperature data within the number of bearing temperature data included in the th th sampling moment of the bearing temperature data within the th time period interval;

[0046] The present invention is further configured as follows: The rotational speed sensor in the acquisition module is installed on the end face of the bearing body. The rotational speed sensor includes a Hall element and a magnetic encoder. The Hall element is arranged on the end face of the outer ring of the bearing body. The sensitive area of the Hall element is close to the end face of the inner ring of the bearing. The magnetic encoder is mounted on the end face of the inner ring of the bearing body.

[0047] The present invention is further configured as follows: The analysis module detects the operating state of the bearing according to the vibration acceleration data, the predicted bearing temperature data, and the rotational speed data, and converts the vibration acceleration data, the predicted bearing temperature data, and the rotational speed data into data frames through a digital protocol converter and outputs them to the wireless communication module. The wireless communication module transmits the data frames to a preset cloud data management and control platform. The cloud data management and control platform analyzes the data frames to generate real-time bearing operating state data, performs corresponding fault warnings, and stores the real-time bearing operating state data.

[0048] The beneficial effects of the present invention are as follows:

[0049] 1. By calculating the weighted root mean square acceleration and calculating the overall vibration acceleration of the bearing body based on the weighted root mean square acceleration, the present invention can evaluate the vibration level of the bearing, can more accurately reflect the vibration characteristics of the bearing at different frequencies, and helps to detect potential faults or abnormal conditions in a timely manner;

[0050] 2. The present invention analyzes the trend changes of temperature data in each time period interval, constructs the abnormal degree of temperature data at each sampling moment, comprehensively considers the difference between the bearing temperature data at each sampling moment and the overall trend change within the time period interval, uses the true abnormal degree to eliminate abnormal data, completes the optimized prediction analysis of temperature data, warns of the occurrence of bearing faults, and can effectively monitor the abnormal state of the bearing;

[0051] 3. The present invention removes the noise signal from the acceleration data to avoid the influence of the noise signal collected during data acquisition on the accuracy of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is the flow chart of the intelligent monitoring system for mechanical motion state driven by sensor bearing data of the present invention;

[0053] Figure 2 is the flow chart of the method for the acceleration sensor to collect vibration acceleration data of the present invention;

[0054] Figure 3 is the flow chart of the method for removing the noise signal from the acceleration data of the present invention;

[0055] Figure 4 is the flow chart of the method for the analysis module to predict the bearing temperature using the bearing time-series temperature data of the present invention;

[0056] In the figure: 1. Acquisition module; 2. Processing module; 3. Wireless communication module; 4. Analysis module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] The following elaborates on the preferred embodiments of the present invention in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making the protection scope of the present invention more clearly defined.

[0058] Please refer to Figure 1 - Figure 4 , an intelligent monitoring system for mechanical motion state driven by sensor bearing data, including an acquisition module 1, a processing module 2, a wireless communication module 3, and an analysis module 4. The acquisition module 1 and the processing module 2 are arranged inside the bearing body, and the wireless communication module 3 and the analysis module 4 are arranged outside the bearing body. The wireless communication module 3 is used to transmit the data frame to a preset cloud data management and control platform, parse the data frame to generate real-time bearing operation state data, and store the real-time bearing operation state data; the analysis module 4 includes a retrieval unit and a calculation unit. The retrieval unit is used to retrieve the real-time bearing operation state data in real time, and the bearing operation state is detected through the calculation unit, and corresponding fault warnings are issued;

[0059] The acquisition module 1, the acquisition module 1 includes a vibration acceleration sensor, a temperature sensor and a rotational speed sensor. The vibration acceleration sensor, the temperature sensor and the rotational speed sensor are respectively used to acquire vibration acceleration data, temperature data and rotational speed data during the operation of the bearing;

[0060] The processing module 2, the processing module 2 includes a microprocessor and a protocol converter. The processing module 2 is used to receive the data acquired by the vibration acceleration sensor, the temperature sensor and the rotational speed sensor, and process the data;

[0061] The microprocessor analyzes, calculates and arranges the acquired data, compiles the valid data into a data frame. The protocol converter receives the data frame and outputs it to the wireless communication module 3. Preferably, the microprocessor is an STM32L0 series ARM processor;

[0062] Among them, the vibration acceleration sensor is installed on the surface of a preset circuit board, and the circuit board is installed inside the acquisition module 1;

[0063] The vibration acceleration sensor is a triaxial accelerometer. The vibration acceleration acquired by the vibration acceleration sensor includes the acceleration data of the bearing body in the X, Y, and Z directions, the acceleration data of the bearing body at each frequency component, and the acceleration data signal.

[0064] Among them, the steps of the vibration acceleration sensor in the acquisition module 1 acquiring vibration acceleration data and processing it through the processing module 2 are as follows:

[0065] S1. Record the acceleration data of the bearing in the X, Y, and Z directions through the vibration acceleration sensor;

[0066] S2. Preprocess the acquired acceleration data through the processing module 2, remove the noise signal acquired during data acquisition, and obtain the acceleration data of each frequency component;

[0067] The steps of removing the noise signal from the acceleration data in step S2 include the following steps:

[0068] S21. The microprocessor in the processing module 2 receives the noise signal, sets different signal decomposition methods to decompose the noise signal, obtains multiple initial mode components, and transmits the multiple initial mode components to the analysis module 4 through the wireless communication module 3;

[0069] S22. The retrieval unit in the analysis module 4 retrieves the initial mode components, calculates the sample entropy of all the initial mode components through the calculation unit, and performs normalization processing on it to obtain the normalized sample entropy of the initial mode components;

[0070] S23. By setting the threshold of normalized sample entropy, perform singular value decomposition noise reduction processing on all initial mode components greater than the threshold, and do not process the initial mode components not greater than the threshold;

[0071] S24. Calculate the kurtosis coefficient of the initial mode components after singular value decomposition noise reduction processing through a calculation unit, and set the threshold of the kurtosis coefficient;

[0072] S25. Eliminate the initial mode components with kurtosis coefficients less than the threshold, retain the initial mode components not less than the threshold, and output the final mode components;

[0073] S26. Recombine all the processed mode components to obtain the acceleration data of each frequency component;

[0074] S3. Calculate the weighted root mean square acceleration value according to the acceleration data of each frequency component;

[0075] S4. Calculate the overall vibration acceleration data of the bearing body according to the weighted root mean square acceleration value.

[0076] Among them, the calculation formula of the weighted root mean square acceleration value in step S3 is as follows:

[0077]

[0078] Among them, is the weighted root mean square acceleration value; is the measurement time length; is the acceleration data of the th frequency component; is the time differential element, obtained by integrating the time ; is the weight vector;

[0079] Among them, the calculation formula of the overall vibration acceleration of the bearing body in step S4 is as follows:

[0080]

[0081] Among them, is the overall vibration acceleration data of the bearing body; 、 、 are the acceleration data in the X, Y, and Z axes respectively.

[0082] Through the above steps, the weighted root mean square acceleration can be effectively calculated, thereby evaluating the vibration level of the bearing. This method can more accurately reflect the vibration characteristics of the bearing at different frequencies, and helps to detect potential faults or abnormal conditions in a timely manner.

[0083] Among them, the microprocessor in the processing module 2 receives the temperature data collected by the temperature sensor. The retrieval module in the analysis module 4 retrieves the temperature data, and processes the temperature data through the deep learning model set in the calculation unit to obtain the bearing sequential temperature data;

[0084] After the calculation unit processes the bearing sequential temperature data, the fitting residual of the bearing temperature data is obtained.

[0085] Among them, the analysis module 4 uses the bearing sequential temperature data to predict the bearing temperature, including the following steps:

[0086] Q1. The retrieval unit in the analysis module 4 obtains the bearing sequential temperature data in real time, and obtains the bearing temperature data at each set sampling moment according to the bearing sequential temperature data;

[0087] For the bearing temperature data at each sampling moment, a plurality of search lengths are set, and all the bearing temperature data included within each search length of the bearing temperature data at each sampling moment are used as each time period interval of the temperature data at each sampling moment;

[0088] Q2. According to the trend deviation of the bearing temperature data at each sampling moment within the set time period interval, obtain the abnormal degree of the temperature data at each sampling moment within each time period interval;

[0089] Q3. According to the abnormal degree and the difference in the trend deviation of the bearing temperature data in different time period intervals, obtain the abnormal stability of the bearing temperature data at each sampling moment within each time period interval;

[0090] Q4. According to the abnormal stability and the distribution of the bearing temperature data within the time period interval, obtain the abnormal authenticity of the bearing temperature data at each sampling moment within each time period interval;

[0091] Q5. Combine the abnormal authenticity and the abnormal degree to obtain the true abnormal degree of the bearing temperature data at each sampling moment, and use the true abnormal degree to obtain the abnormal data in the bearing sequential temperature data, eliminate the abnormal data, and complete the temperature prediction according to the temperature data after eliminating the abnormal data;

[0092] Combining the abnormal authenticity and the abnormal degree to obtain the true abnormal degree of the bearing temperature data at each sampling moment includes the following calculation steps: Taking the sum of the products of the bearing temperature data at each sampling moment in all time period intervals as the true abnormal degree of the bearing temperature data at each sampling moment;

[0093] Specifically, in step Q3, according to the abnormal degree and the difference in the trend deviation of the bearing temperature data in different time period intervals, obtain the abnormal stability of the bearing temperature data at each sampling moment within each time period interval, and its calculation formula is:

[0094]

[0095] Among them, is the abnormal stability of the bearing temperature data at the -th sampling moment within the -th time period interval; is the degree of abnormality of the bearing temperature data at the -th sampling moment in the -th time period interval; is the degree of abnormality of the bearing temperature data at the -th sampling moment in the -th time period interval; represents the preset moving time length;

[0096] In step Q4, the abnormal authenticity of the bearing temperature data at each sampling moment in each local interval is obtained according to the abnormal stability and the distribution of the bearing temperature data within the time period interval. The calculation formula is:

[0097] ;

[0098] Among them, is the abnormal authenticity of the bearing temperature data at the -th sampling moment within the -th time period interval; is the abnormal stability of the bearing temperature data at the -th sampling moment in the -th time period interval; is the normalization function; is the fitting residual of all the bearing temperature data at the -th sampling moment within the -th time period interval; is the number of the bearing temperature data included in the -th sampling moment in the -th time period interval; is the number of the bearing temperature data included in the -th sampling moment in the -th time period interval; is the fitting residual of all the temperature data at the -th sampling moment within the -th time period interval.

[0099] By analyzing the trend changes of temperature data in each time period, the abnormal degree of temperature data at each sampling moment is constructed. The difference between the bearing temperature data at each sampling moment and the overall trend changes in the time period is comprehensively considered, reflecting the degree of noise interference of the bearing temperature data at each sampling moment. According to the abnormal degree, the abnormal stability of the temperature data at each sampling moment in each local interval is obtained, reflecting the stability of the abnormal degree of the bearing temperature data at each sampling moment with the change of the time period length. The trend changes of multiple time periods are considered, and the abnormal authenticity of the temperature data at each sampling moment in each time period is obtained according to the abnormal stability. The abnormal stability of the temperature data in each time period and the fitting residual of the temperature data in each time period are comprehensively considered, which improves the reliability of the abnormal degree of the temperature data, and then the real abnormal degree of the bearing temperature data at each sampling moment is obtained. The real abnormal degree is used to eliminate the abnormal data and complete the optimized prediction analysis of the temperature data.

[0100] Among them, the speed sensor in the acquisition module 1 is installed on the end face of the bearing body. The speed sensor includes a Hall element and a magnetic encoder. The end face of the outer ring of the bearing body is provided with a Hall element. The sensitive area of ​​the Hall element is close to the end face of the inner ring of the bearing. The magnetic encoder is mounted on the end face of the inner ring of the bearing body. When the bearing rotates and the magnetized area in the magnetic encoder periodically passes through the sensitive area of ​​the Hall element, it causes the periodic change of the output voltage of the Hall element. By analyzing the period of the voltage, the current speed of the bearing is obtained.

[0101] Among them, the analysis module 4 detects the bearing operation status according to the vibration acceleration data, the predicted bearing temperature data and the speed data, and converts the vibration acceleration data, the predicted bearing temperature data and the speed data into data frames through the digital protocol converter and outputs it to the wireless communication module 3. The wireless communication module 3 transmits the data frames to the preset cloud data management and control platform. The cloud data management and control platform parses the data frames to generate real-time bearing operation status data, performs corresponding fault warnings, and stores the real-time bearing operation status data.

[0102] The above are only embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent monitoring system for mechanical motion state driven by sensor bearing data, characterized in that: It includes a collection module (1), a processing module (2), a wireless communication module (3) and an analysis module (4). The collection module (1) and the processing module (2) are arranged inside the bearing body, and the wireless communication module (3) and the analysis module (4) are arranged outside the bearing body; The collection module (1), which includes a vibration acceleration sensor, a temperature sensor and a rotational speed sensor. The vibration acceleration sensor, the temperature sensor and the rotational speed sensor are respectively used to collect vibration acceleration data, temperature data and rotational speed data during the operation of the bearing; The processing module (2), which includes a microprocessor and a protocol converter. The processing module (2) is used to receive the data collected by the vibration acceleration sensor, the temperature sensor and the rotational speed sensor, and process the data; The microprocessor analyzes, calculates and sorts out the collected data, compiles the valid data into a data frame, and the protocol converter receives the data frame and outputs it to the wireless communication module (3); The wireless communication module (3), which is used to transmit the data frame to a preset cloud data control platform, parse the data frame to generate real-time bearing operation state data, and store the real-time bearing operation state data; The analysis module (4), which includes a retrieval unit and a calculation unit. The retrieval unit is used to retrieve the real-time bearing operation state data in real time, and the bearing operation state is detected through the calculation unit, and corresponding fault warnings are given.

2. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 1, wherein: The vibration acceleration sensor is installed on the surface of a preset circuit board, and the circuit board is installed inside the collection module (1); The vibration acceleration sensor is a triaxial accelerometer. The vibration acceleration data collected by the vibration acceleration sensor includes the acceleration data of the bearing body in the X, Y, and Z directions, the acceleration data of the bearing body at each frequency component, and the acceleration data signal.

3. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 2, wherein: The steps for the vibration acceleration sensor in the collection module (1) to collect vibration acceleration data and process it through the processing module (2) are as follows: S1. Record the acceleration data of the bearing in the X, Y, and Z directions through the vibration acceleration sensor; S2. Preprocess the collected acceleration data through the processing module (2) to remove the noise signals collected during data collection, and obtain the acceleration data of each frequency component; S3. Calculate the weighted root mean square acceleration value according to the acceleration data of each frequency component; S4. Calculate the overall vibration acceleration data of the bearing body according to the weighted root mean square acceleration value.

4. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 3, characterized in that: The steps for removing the noise signal from the acceleration data in step S2 include the following steps: S21. The microprocessor in the processing module (2) receives the noise signal, sets different signal decomposition methods to decompose the noise signal, obtains a plurality of initial mode components, and transmits the plurality of initial mode components to the analysis module (4) through the wireless communication module (3); S22. The retrieval unit in the analysis module (4) retrieves the initial modal components, calculates the sample entropy of all the initial modal components through the calculation unit, and performs normalization processing on it to obtain the normalized sample entropy of the initial modal components; S23. By setting a threshold for the normalized sample entropy, perform singular value decomposition noise reduction processing on all initial modal components greater than the threshold, and do not process the initial modal components not greater than the threshold; S24. Calculate the kurtosis coefficient of the initial modal components after singular value decomposition noise reduction processing through the calculation unit, and set a threshold for the kurtosis coefficient; S25. Eliminate the initial modal components with kurtosis coefficients less than the threshold, retain the initial modal components not less than the threshold, and output the final modal components; S26. Recombine all the processed modal components to obtain the acceleration data of each frequency component.

5. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 4, characterized in that: The microprocessor in the processing module (2) receives the temperature data collected by the temperature sensor. The retrieval module in the analysis module (4) retrieves the temperature data, and processes the temperature data through the deep learning model set in the calculation unit to obtain the bearing time-series temperature data; The calculation unit obtains the fitting residual of the bearing temperature data after processing the bearing time-series temperature data.

6. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 5, wherein: The analysis module (4) predicts the bearing temperature using the bearing time-series temperature data, including the following steps: Q1. The retrieval unit in the analysis module (4) obtains the bearing time-series temperature data in real time, and obtains the bearing temperature data at the set sampling moments according to the bearing time-series temperature data; Q2. Obtain the abnormality degree of the temperature data at each sampling moment in each time period interval according to the trend deviation of the bearing temperature data at each sampling moment in the set time period interval; Q3. Obtain the abnormality stability of the bearing temperature data at each sampling moment in each time period interval according to the abnormality degree and the difference in the trend deviation of the bearing temperature data in different time period intervals; Q4. Obtain the abnormality authenticity of the bearing temperature data at each sampling moment in each time period interval according to the abnormality stability and the distribution of the bearing temperature data in the time period interval; Q5. Combine the abnormality authenticity and the abnormality degree to obtain the true abnormality degree of the bearing temperature data at each sampling moment, use the true abnormality degree to obtain the abnormal data in the bearing time-series temperature data, eliminate the abnormal data, and complete the temperature prediction according to the temperature data after eliminating the abnormal data; Specifically, in step Q3, the abnormality stability of the bearing temperature data at each sampling moment in each time period interval is obtained according to the abnormality degree and the difference in the trend deviation of the bearing temperature data in different time period intervals, and its calculation formula is: ； Among them, is the abnormal stability of the bearing temperature data at the -th sampling moment within the -th time period interval; is the abnormal degree of the bearing temperature data at the -th sampling moment in the -th time period interval; is the abnormal degree of the bearing temperature data at the -th sampling moment in the -th time period interval; represents the preset moving time length; In step Q4, the abnormality authenticity of the bearing temperature data at each sampling moment in each local interval is obtained according to the abnormality stability and the distribution of the bearing temperature data in the time period interval, and its calculation formula is: ； Among them, is the anomaly authenticity of the bearing temperature data at the -th sampling moment in the -th time period interval; is the anomaly stability of the bearing temperature data at the -th sampling moment in the -th time period interval; is the normalization function; is the fitting residual of all bearing temperature data at the -th sampling moment in the -th time period interval; is the number of bearing temperature data included in the -th sampling moment in the -th time period interval; is the number of bearing temperature data included in the -th sampling moment in the -th time period interval; is the fitting residual of all temperature data at the -th sampling moment in the -th time period interval.

7. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 6, characterized in that: The rotational speed sensor in the acquisition module (1) is installed on the end face of the bearing body. The rotational speed sensor includes a Hall element and a magnetic encoder. The Hall element is arranged on the end face of the outer ring of the bearing body, and the sensitive area of the Hall element is close to the end face of the inner ring of the bearing. The magnetic encoder is mounted on the end face of the inner ring of the bearing body.

8. The intelligent monitoring system for mechanical motion state driven by sensor bearing data according to claim 7, characterized in that: The analysis module (4) detects the operating state of the bearing according to the vibration acceleration data, the predicted bearing temperature data, and the rotational speed data, and converts the vibration acceleration data, the predicted bearing temperature data, and the rotational speed data into data frames through a digital protocol converter and outputs them to the wireless communication module (3). The wireless communication module (3) transmits the data frames to a preset cloud data management platform. After parsing the data frames, the cloud data management platform generates real-time bearing operating state data, conducts corresponding fault warnings, and stores the real-time bearing operating state data.

Images