A multi-source signal edge acquisition and intelligent operation and maintenance method for GMC400 dense bead shaft system

Abstract

The present invention discloses a multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type beaded shaft system. The multi-source sensor data of the GMC400 type beaded shaft system is synchronously acquired by using an intelligent edge device, and the multi-source sensor data is processed by edge computing technology such as redundant data dimension reduction and feature extraction, thereby alleviating the pressure of cloud computing. The edge computing results will be fed back to the remote operation and maintenance platform through the cloud database using the 4G network. The platform is based on the BS architecture and will perform real-time monitoring and alarm of the equipment, and visualize the multi-source sensor data, thereby realizing all-round intelligent operation and maintenance of multi-source sensor data acquisition, transmission, storage and analysis. The present invention can effectively acquire and operate the multi-source sensor data in the GMC400 type beaded shaft system, which plays an important role in improving the operating accuracy of the GMC400 type beaded shaft system.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent sensing and industrial big data technology, and specifically relates to a multi-source signal edge acquisition and intelligent operation and maintenance method for a GMC400 type dense bead shaft system. Background Art

[0002] The GMC400 type bead-type shaft system multi-source signal edge acquisition and intelligent operation and maintenance technology uses sensor data as the data source to collect sensor data from key parts of the measuring device, such as bearings, rotating shafts, etc., and uses the collected sensor data and the error data of the measuring device itself to extract information that affects the operating status and measurement accuracy of the shaft system, extract and predict information from the original data, and realize accurate processing of sensor data in the measuring device shaft system process.

[0003] During the operation of the GMC400 bead-type shaft system, a massive amount of data containing operating status information and measurement accuracy will be generated. Therefore, how to perform feature mining and information extraction on the data to improve the measurement accuracy can realize the prediction of future operating status, explore potential abnormal and sudden conditions of the measuring device shaft system, enhance the stability of the measuring device shaft system, and play a guiding role in the repair and maintenance of the measuring device shaft system, which is of great significance to the sustainable and stable measurement process.

[0004] To this end, a multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system was established, which plays an important role in improving the operation accuracy of the GMC400 type dense bead shaft system. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the deficiencies in the above-mentioned prior art and provide a multi-source signal edge acquisition and intelligent operation and maintenance method for a GMC400 type beaded shaft system. The multi-source sensor data of the GMC400 type beaded shaft system is synchronously acquired by an intelligent edge acquisition device, and the monitoring of the GMC400 type beaded shaft system is realized through a remote operation and maintenance platform, so as to solve the problem of insufficient accuracy retention of the GMC400 type beaded shaft system.

[0006] The present invention adopts the following technical solutions:

[0007] The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 dense bead shaft system includes the following steps:

[0008] The vibration, displacement and temperature signals of the lower bearing outer side, upper bearing outer side and encoder bracket of the GMC400 bead-type precision shaft system are collected; the PID parameters are collected through the network port; the variable speed and load working conditions are set to conduct the GMC400 bead-type precision shaft system status monitoring verification experiment; the redundant data dimensionality reduction and feature extraction methods are used to perform edge computing on the data signals obtained from the verification experiment, and effective information is selected from the original signal for data upload to the cloud; the multi-source sensor data behind the cloud is visualized on the remote operation and maintenance platform, and the dynamic accuracy of the bead-type precision shaft system is predicted using a multi-source fusion dynamic accuracy prediction model. The bead-type precision shaft system is monitored and alarmed in real time based on the prediction results, realizing remote intelligent operation and maintenance of the GMC400 bead-type precision shaft system.

[0009] Specifically, acceleration sensors, temperature sensors, and eddy current sensors are installed on the outside of the lower bearing, the outside of the upper bearing, and the encoder bracket of the GMC400 bead-type shaft system, and intelligent edge acquisition equipment is used to collect vibration, displacement, and temperature signals.

[0010] Furthermore, the intelligent edge acquisition device includes an acquisition module and a processing module. The acquisition module realizes the acquisition of multi-source sensor data, and the processing module realizes the storage, transmission and analysis process of multi-source sensor data.

[0011] Specifically, the variable speed and load working conditions are set as follows:

[0012] The axis turntable starts rotating at a constant speed of 10,000 pulses from 0s, starts accelerating at 3s, and the speed accelerates from 10,000 pulses to 100,000 pulses. The acceleration ends at 45s, at which time the speed reaches 100,000 pulses. Finally, it starts to decelerate at 46s and stops at 48s. During the acceleration period, the pulse changes are 10,000, 20,000, 30,000, ..., 90,000, 100,000, and continues to accelerate.

[0013] Specifically, the redundant data dimensionality reduction method uses a revolving door data compression method to reduce the dimensionality of redundant data, and establishes an upper and lower revolving door model to realize the judgment basis of data compression.

[0014] Furthermore, when the constructed revolving door interval can cover the original data, the current coverage interval t0 is recorded. The next revolving door interval uses the coverage interval t0. If it can still cover the current interval, the coverage interval t is gradually increased. The increasing expression is:

[0015]

[0016] If the current interval cannot be covered, the penalty mode is used to reduce the coverage interval. The reduction expression is:

[0017] t(k上 ,k 下 )=t0(k 上 ,k 下 )×c2×nt(k 上 ,k 下 )>t0(k 上 ,k 下 )

[0018] Among them, t(k 上 ,k 下 ) represents the current revolving door coverage interval, t0(k 上 ,k 下 ) represents the coverage interval of the previous revolving door, c1 represents the reward factor, c2 represents the penalty factor, and n represents the cumulative number of coverages.

[0019] Furthermore, the mathematical model of the upper revolving door slope is:

[0020]

[0021] The mathematical model of the slope of the lower revolving door is:

[0022]

[0023] Among them, x represents the currently received data value, x0 represents the value of the first point in the stored data, and er represents the fault tolerance value.

[0024] Specifically, the feature extraction method is as follows:

[0025] First, the signal is subjected to EMD modal decomposition, and the empirical mode functions with concentrated frequency components are selected and then combined into a new signal. The time-frequency domain quantities reflecting energy changes, energy distribution and geometric shapes of the new signal are extracted and analyzed respectively to obtain the eigenvectors that can reflect the signal characteristics. The original signal is decomposed using the EMD algorithm; finally, a series of IMFs are obtained, the relative energy ratio of each IMF is calculated, and the first three IMFs are selected for combination as the new signal for feature extraction.

[0026] Specifically, the multi-source fusion dynamic precision prediction model includes three parts: multi-source heterogeneous signal preprocessing, multi-source data multi-dimensional fusion, and a prediction layer based on the LSTM algorithm; multi-source heterogeneous signal preprocessing, time-frequency domain feature information extraction, data dimension transformation, and data padding of multi-source heterogeneous data, and standardization of data length and format; use the LSTM algorithm model to mine sensitive fault information in the sequence to achieve deep fusion of multi-source information, and add the Attention mechanism to the training model to improve the balance of new and old information in the LSTM neural network, use the DTW mechanism to predict unknown faults, and output faults when new location faults occur to predict unknown faults.

[0027] Furthermore, the DTW mechanism predicts unknown faults, with two multi-source fusion sequences R n 、U m , calculate the Euclidean distance between the two sequences, build the Euclidean distance matrix between the two sequences, calculate the cumulative distance to obtain the DP matrix under the entire M(n,m); calculate the DTW value based on the cumulative distance value between the two sequences calculated after the DP matrix; use it as an indicator to measure the occurrence of unknown faults.

[0028] Compared with the prior art, the present invention has at least the following beneficial effects:

[0029] The GMC400 beaded shaft system uses multi-source signal edge acquisition and intelligent operation and maintenance methods to identify the main factors affecting dynamic accuracy, including assembly errors, vibration factors, and thermal deformation during operation, and to construct a fault tree affecting the dynamic accuracy of the GMC400 beaded shaft system. The intelligent edge acquisition equipment adopts a "dual module" design method of acquisition module and processing module. The acquisition module realizes the acquisition of multi-source sensor data, and the processing module realizes the storage, transmission, and analysis of multi-source sensor data. The data control and display module can control the acquisition process of the acquisition module and perform high-speed and low-speed visualization of multi-source sensor data. Before the acquisition begins, the acquisition parameters are preset through the data control and display module. When data acquisition begins, the sensor is connected to the acquisition module, and the data will adopt the "acquisition, reading, and storage" mode, which realizes real-time data display and local storage. A variable speed test was conducted. The shaft system turntable began rotating at a constant speed of 10,000 pulses at 0 seconds. Acceleration began at 3 seconds, with the speed increasing from 10,000 pulses to 100,000 pulses. Acceleration ended at 45 seconds, reaching 100,000 pulses. Finally, deceleration began at 46 seconds, and stopped at 48 seconds. During the acceleration period, the pulse rate varied from 10,000 to 20,000, 30,000, …, 90,000, and 100,000 pulses. Redundant data was reduced using a revolving door data compression method based on a "reward and punishment" model. Time-frequency domain quantities reflecting energy variation, energy distribution, and geometric shape were extracted from the original signal to generate feature vectors that reflect signal characteristics, thereby achieving data dimensionality reduction. The remote operation and maintenance platform was designed based on the "BS" framework, establishing a multi-layered management architecture to analyze and transmit operation and maintenance data streams. The operation and maintenance data streams ultimately function on the platform for real-time monitoring and alarming of equipment and visualization of multi-source sensor data.

[0030] Furthermore, the acquisition module collects the analog voltage signal of the sensor according to pre-set acquisition parameters. The signal passes through the digital anti-aliasing filter module and IEPE excitation module of the intelligent edge acquisition device to realize the acquisition and signal pre-conditioning of multi-source sensor data in the "FIFO" (first-in-first-out) mode. The data control and display module in the processing module reads and stores the multi-source sensor data in the acquisition module. The read data will enter the edge computing module for redundant data dimensionality reduction, feature extraction and other processing. The data processed by the edge computing becomes an operation and maintenance data stream and is uploaded to the remote operation and maintenance platform through the data transmission module. The remote operation and maintenance platform receives the operation and maintenance data stream from the processing module and transmits the operation and maintenance data stream to the cloud database for storage according to different data types. Finally, the operation and maintenance data stream is used on the operation and maintenance platform for real-time monitoring of equipment, issuance of collection work orders, visualization of multi-source sensor data, abnormal information alarms, and multi-role permission user management.

[0031] Furthermore, the data control and display module is divided into two parts: control and visualization. It can control the switch of the acquisition module, the acquisition parameter setting, and perform high-speed and low-speed visualization of multi-source sensor data. Before starting the acquisition, the acquisition parameters are preset through the data control and display module. The preset acquisition parameters include the number of acquisition channels, sampling frequency, measurement type, sensor sensitivity, IEPE excitation mode, signal input mode, filter switch, AD sampling range, trigger mode, input polarity, clock mode, and external clock input polarity. When starting data acquisition, the sensor is connected to the acquisition module, and the data will adopt the "acquisition, reading and storage" mode, which can realize real-time data display while storing it locally.

[0032] Furthermore, the redundant data dimensionality reduction method uses a revolving door data compression method to reduce the dimensionality of redundant data. This method establishes an upper and lower revolving door model to determine the basis for data compression. Furthermore, the compression process employs a "reward and penalty" model. When the constructed revolving door interval can cover the original data, the current coverage interval t_0 is recorded. The next revolving door interval uses this coverage interval t_0. If the current interval can still be covered, the coverage interval t is gradually increased. If the current interval cannot be covered, a penalty model is adopted to reduce the coverage interval.

[0033] Furthermore, the feature extraction method extracts time-frequency domain quantities reflecting energy variation, energy distribution, and geometric shape from the original signal to derive eigenvectors that reflect the signal's characteristics, thereby achieving data dimensionality reduction. First, the signal undergoes EMD modal decomposition. The empirical mode functions (IMFs) with the most concentrated frequency components are selected and then combined to form a new signal. Time-frequency domain quantities reflecting energy variation, energy distribution, and geometric shape from the new signal are extracted and analyzed to derive eigenvectors that reflect the signal's characteristics, thereby achieving data dimensionality reduction. The original signal is decomposed using the EMD algorithm. Finally, a series of IMFs are obtained. To select useful IMFs, the relative energy ratio of each IMF is calculated. The first three IMFs occupy the majority of the energy, indicating that they contain the most information. Therefore, these first three IMFs are selected for combination and used as the new signal for feature extraction.

[0034] Furthermore, the multi-source fusion dynamic accuracy prediction model consists of three parts: multi-source heterogeneous signal preprocessing, multi-source data multi-dimensional fusion, and a prediction layer based on the LSTM algorithm. Multi-source heterogeneous signal preprocessing is characterized by extracting time-frequency domain feature information, transforming data dimensions, and padding data from multi-source heterogeneous data, thereby standardizing data length and format. The LSTM algorithm model is used to mine sensitive fault information in the sequence, achieving deep fusion of multi-source information. The training model incorporates an Attention mechanism, improving the balance of new and old information in the LSTM neural network. The DTW mechanism is also used to predict unknown faults. This process is based on normal operating data. When a new location fault occurs, a fault output is generated to predict unknown faults. The calculated DTW value is used as an indicator to measure the occurrence of unknown faults. By setting a fault threshold, unknown faults can be predicted during shaft system operation.

[0035] Furthermore, the operation and maintenance platform has multiple management contents. Real-time monitoring displays the status of the equipment and the location of sensor layout. Project management is used to add sensor monitoring work order requirements. Work order management is used to view historical sensor and operation and maintenance data. Alarm management displays equipment abnormality information.

[0036] In summary, the present invention synchronously collects multi-source sensor data of the GMC400 type bead-type shaft system, uses edge computing technology to reduce the dimension of the multi-source sensor data, and uses a remote operation and maintenance platform to perform full-dimensional management of the multi-source sensor data, which plays an important role in improving the accuracy of the GMC400 type bead-type shaft system.

[0037] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1This is the internal architecture diagram of the acquisition module;

[0039] Figure 2 Fault tree analysis of factors affecting the accuracy of GMC400 beadlock shafting;

[0040] Figure 3 This is the sensor layout diagram;

[0041] Figure 4 To collect the physical picture of the module;

[0042] Figure 5 To process the physical picture of the module;

[0043] Figure 6 The diagram shows the processing module parameter table, where (a) is the high-speed acquisition and display of multi-source sensor data, (b) is the high-speed display of multi-source sensor data, (c) is the sensor parameter setting, and (d) is the acquisition module parameter setting.

[0044] Figure 7 Schematic diagram of relative energy ratio of IMF;

[0045] Figure 8 It is the accuracy prediction result of the multi-source fusion dynamic accuracy prediction model;

[0046] Figure 9 Schematic diagram of work order creation and query on the remote operation and maintenance platform, where (a) is the collection information setting, (b) is the sensor layout work order setting, and (c) is the cloud visualization of the collected data. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] In the description of the present invention, it is to be understood that the terms “include” and “comprise” indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0049] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0050] It should be further understood that the term "and / or" as used in the present specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0051] It should be understood that although the terms "first," "second," and "third" may be used to describe preset ranges in embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are merely used to distinguish one preset range from another. For example, without departing from the scope of embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0052] The word "if," as used herein, may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0053] The accompanying drawings illustrate various schematic diagrams of structures according to embodiments disclosed herein. These figures are not drawn to scale; for clarity, some details are exaggerated and some details may be omitted. The shapes of the various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in practice due to manufacturing tolerances or technical limitations. Those skilled in the art may design regions / layers with different shapes, sizes, and relative positions as needed.

[0054] The present invention provides a multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system. By using intelligent edge devices to synchronously acquire multi-source sensor data of the GMC400 type dense bead shaft system, and utilizing edge computing technology, redundant data dimension reduction, feature extraction and other processing are performed on the multi-source sensor data, thereby alleviating the pressure of cloud computing. The edge computing results will be fed back to the remote operation and maintenance platform through the cloud database using the 4G network. The platform is based on the BS architecture and will perform real-time monitoring and alarm of equipment, and visualize the multi-source sensor data, thereby realizing all-round intelligent operation and maintenance of multi-source sensor data acquisition, transmission, storage and analysis. The present invention can effectively acquire and operate the multi-source sensor data in the GMC400 type dense bead shaft system, which plays an important role in improving the operating accuracy of the GMC400 type dense bead shaft system.

[0055] The present invention provides a multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type beaded shaft system. The intelligent edge acquisition device adopts a "dual module" design method of acquisition module and processing module. The acquisition module realizes the acquisition of multi-source sensor data, and the processing module realizes the storage, transmission and analysis process of multi-source sensor data. The remote operation and maintenance platform design is based on the "BS" framework, establishing a five-layer management architecture to realize the analysis and transmission of operation and maintenance data streams. The operation and maintenance data streams are ultimately used in the operation and maintenance platform for real-time monitoring and alarming of equipment, issuance of collection work orders, visualization of multi-source sensor data, and multi-role permission user management; specifically, the following steps are included:

[0056] S1. Based on the indicators that affect the dynamic performance of the GMC400 bead-type shafting system, vibration, displacement, temperature, and PID parameters are determined as monitoring signals, and the input of the dynamic accuracy prediction model for the GMC400 bead-type shafting system is constructed;

[0057] The main factors affecting dynamic accuracy, including assembly error, vibration factors and thermal deformation during operation, are determined, and a fault tree affecting the dynamic accuracy of the GMC400 beadlock shafting is constructed.

[0058] S2. Install acceleration sensors, temperature sensors, and eddy current sensors on the outside of the lower and upper bearings of the GMC400 bead-type shaft system, as well as on the encoder bracket. Use an intelligent edge acquisition device to collect vibration, displacement, and temperature signals from these locations. PID parameters are input into the intelligent edge acquisition device through the network port for collection.

[0059] The intelligent edge acquisition device adopts a "dual-module" design method of acquisition module and processing module. The acquisition module realizes the collection of multi-source sensor data, and the processing module realizes the storage, transmission and analysis process of multi-source sensor data.

[0060] The data control and display module can control the acquisition process of the acquisition module and perform high-speed and low-speed visual display of multi-source sensor data. Before starting the acquisition, the acquisition parameters are preset through the data control and display module. When the data acquisition starts, the sensor is connected to the acquisition module, and the data will adopt the "acquisition, reading and storage" mode, which enables real-time data display and local storage.

[0061] The acquisition module collects the analog voltage signal of the sensor, and realizes the acquisition of multi-source sensor data and signal pre-conditioning in "FIFO" (first-in-first-out) mode through the digital anti-aliasing filter module and IEPE excitation module; the pre-set acquisition parameters include the number of acquisition channels, sampling frequency, measurement type, sensor sensitivity, IEPE excitation mode, signal input method, filter switch, AD sampling range, trigger mode, input polarity, clock mode and external clock input polarity.

[0062] See also Figure 1 The acquisition module collects the analog voltage signal of the sensor and performs a fault tree analysis on the factors that affect the dynamic accuracy of the GMC400 bead-type shaft system. Figure 2 , signal acquisition includes the following steps:

[0063] S201, acceleration sensor and temperature sensor are installed in three locations, namely on the outside of the lower bearing, the outside of the upper bearing and the encoder bracket. For the sensor layout diagram, please refer to Figure 3 .

[0064] S202: The signal is input through the BNC port of the intelligent edge acquisition device and first enters the FPGA module, which has four 16-bit high-speed synchronous analog signal acquisition channels with a maximum synchronous sampling rate of 200 kSPS. The signal first passes through the IEPE excitation module, which enables the acquisition of multi-source sensor data and signal preconditioning in "FIFO" (first-in, first-out) mode. The IEPE excitation module integrates an independent IEPE excitation source for each channel, providing 4mA as a constant current source excitation, enabling signal conditioning for ICP / IEPE accelerometers and other related sensors.

[0065] S203, the analog signal will then pass through a digital anti-aliasing filter module, which uses a digital first-order sinc filter. The filter expression is:

[0066]

[0067] The digital anti-aliasing filter module also uses oversampling rate to filter high-frequency noise. The oversampling rate range is: 2x, 4x, 8x, 16x, 32x, and 64x. Figure 4 ,For the detailed parameters of the acquisition module, please refer to Table 1.

[0068]

[0069] The data control and display module in the processing module reads and stores the multi-source sensor data in the acquisition module. The data control and display module is divided into two parts: control and visualization. It can control the switch of the acquisition module, set acquisition parameters, and perform high-speed and low-speed visualization of the multi-source sensor data. Figure 5 .

[0070] The processing module is equipped with Ubuntu 18.0 system, adopts ARM 64 architecture, and is equipped with 128G solid-state storage. For its specific parameter table, please refer to Figure 6 .

[0071] Before data acquisition begins, the data control and display module pre-sets acquisition parameters. These parameters include the number of acquisition channels, sampling frequency, measurement type, sensor sensitivity, IEPE excitation mode, signal input method, filter switch, AD sampling range, trigger mode, input polarity, clock mode, and external clock input polarity. When data acquisition begins, the sensor is connected to the acquisition module, and the data is stored locally in a "sampling, reading, and storing" mode, enabling real-time data display and simultaneous local storage. See Table 2 for the specific visualization interface of the data control and display module.

[0072]

[0073] S3. Conduct a GMC400 bead-type shaft system condition monitoring verification experiment. During the experiment, set the variable speed and load working conditions, and collect and analyze its vibration, displacement and temperature signals;

[0074] A GMC400 precision beaded shafting condition monitoring verification experiment was conducted. A variable speed test was conducted. The shafting turntable began rotating at a constant speed of 10,000 pulses at 0 seconds. Acceleration began at 3 seconds, with the speed increasing from 10,000 pulses to 100,000 pulses. Acceleration ended at 45 seconds, reaching 100,000 pulses. Finally, deceleration began at 46 seconds and continued until cessation at 48 seconds. During the acceleration period, the pulse rate varied from 10,000 to 20,000, 30,000, ..., 90,000, and finally 100,000. Real-time signal acquisition was performed during the experiment.

[0075] S4. Use data dimensionality reduction and feature extraction methods to perform edge computing on data signals, thereby selecting effective information from the original signal and uploading the data to the cloud;

[0076] The revolving door data compression method based on the "reward and punishment" mode is used to reduce the dimensionality of redundant data. The time-frequency domain quantities reflecting the energy change, energy distribution and geometric shape of the original signal are extracted respectively, and the feature vectors that can reflect the signal characteristics are obtained, thereby achieving data dimensionality reduction.

[0077] The stored multi-source sensor data will enter the edge computing module for redundant data dimensionality reduction, feature extraction and other processing. The redundant data dimensionality reduction method uses the revolving door data compression method to reduce the dimensionality of redundant data. The judgment basis for data compression is realized by establishing an upper and lower revolving door model. The mathematical model expression of the upper revolving door slope is:

[0078]

[0079] The mathematical model expression of the slope of the lower revolving door is:

[0080]

[0081] Among them, x represents the currently received data value, x0 represents the value of the first point in the stored data, and er represents the fault tolerance value.

[0082] The revolving door data compression method is used to reduce the dimension of redundant data. The judgment basis of data compression is realized by establishing an upper and lower revolving door model. The compression process adopts a "reward and punishment" mode. When the revolving door interval can cover the original data, the current coverage interval t0 is recorded. The next revolving door interval uses the coverage interval t0. If it can still cover the current interval, the coverage interval t is gradually increased. The increment expression is:

[0083]

[0084] If the current interval cannot be covered, the penalty mode is used to reduce the coverage interval. The reduction expression is:

[0085] t(k 上 ,k 下 )=t0(k 上 ,k 下 )×c2×nt(k 上 ,k 下 )>t0(k 上 ,k 下 )

[0086] Among them, t(k 上 ,k 下 ) represents the current revolving door coverage interval, t0(k 上 ,k 下 ) represents the coverage interval of the previous revolving door, c1 represents the reward factor, c2 represents the penalty factor, and n represents the cumulative number of coverages.

[0087] The specific feature extraction method is:

[0088] First, the signal is subjected to EMD modal decomposition. The empirical mode functions (IMFs) with the most frequency components are selected and then combined to form a new signal. Time-frequency domain variables reflecting energy variation, energy distribution, and geometric shape are extracted and analyzed from the new signal to obtain eigenvectors that reflect the signal's characteristics, thereby achieving data dimensionality reduction. The original signal is decomposed using the EMD algorithm. Finally, a series of IMFs are obtained. To select the most useful IMFs, the relative energy ratio of each IMF is calculated. The first three IMFs occupy the majority of the energy, indicating that they contain the most information. Therefore, these first three IMFs are selected for combination and used as the new signal for feature extraction.

[0089] The mathematical model of energy change is:

[0090]

[0091] The mathematical model of energy distribution is:

[0092]

[0093] The mathematical model of the geometric shape is:

[0094] First, obtain the power spectrum value of the signal,

[0095]

[0096] Where x is the signal sequence, n is the length of the calculated signal, and x(W) is the FFT transform of the original signal.

[0097]

[0098] Among them, S(f i ) are the frequency components f i The spectral energy of , pi is the corresponding probability density, N is the total number of frequency components in the FFT,

[0099]

[0100] In order to compare different operating conditions, the results are normalized by the factor logN,

[0101]

[0102] Where x is the signal sequence and n is the length of the calculated signal.

[0103] S5. After being uploaded to the cloud, the data will be visualized on the remote operation and maintenance platform through multi-source sensor data, and the dynamic accuracy of the dense bead shaft system will be predicted using a multi-source fusion dynamic accuracy prediction model. Based on the prediction results, the dense bead shaft system will be monitored and alarmed in real time to realize remote intelligent operation and maintenance of the GMC400 dense bead shaft system.

[0104] The remote operation and maintenance platform is designed based on the "BS" framework, establishing a multi-layer management architecture to realize the analysis and transmission of operation and maintenance data streams. The operation and maintenance data streams are ultimately used on the operation and maintenance platform for the visualization of multi-source sensor data of the equipment, accuracy prediction of the multi-source fusion dynamic accuracy prediction model, and real-time monitoring and alarm of the dense bead shaft system.

[0105] The multi-source fusion dynamic accuracy prediction model consists of three parts: multi-source heterogeneous signal preprocessing, multi-source data multi-dimensional fusion, and prediction layer based on LSTM algorithm.

[0106] Multi-source heterogeneous signal preprocessing involves extracting time-frequency domain feature information, transforming data dimensions, and padding data to standardize data length and format. An LSTM algorithm model is used to mine sensitive fault information in the sequence, achieving deep fusion of multi-source information. The training model incorporates an Attention mechanism to improve the balance of new and old information in the LSTM neural network. The DTW mechanism is also used to predict unknown faults. This process is based on normal operating data and generates a fault output when a new location fault occurs, thereby predicting unknown faults.

[0107] The DTW mechanism predicts unknown faults. There are two multi-source fusion sequences R in the input model. n 、U m :

[0108] R n ={r1,r2,r3,…,r n}

[0109] U m ={u1,u2,u 3 ,…,u m}

[0110] Among them, R n represents the multi-source fusion sequence after processing under normal operation, U n Represents the multi-source fusion sequence after processing under unknown conditions,

[0111] Calculate the Euclidean distance between the two sequences and construct the Euclidean distance matrix between the two sequences:

[0112] M(n,m)=||Rn-Um|| 2

[0113] Calculate the cumulative distance and calculate the DP matrix under the entire M(n,m):

[0114]

[0115] The DTW value is calculated based on the cumulative distance between the two sequences after the DP matrix calculation. The calculated DTW value is used as an indicator to measure the occurrence of unknown faults. By setting the fault threshold, the unknown faults during the operation of the shaft system can be predicted. Please refer to the final accuracy prediction results. Figure 8 .

[0116] The remote operation and maintenance platform receives and processes the module's operation and maintenance data streams, which are then transferred to a cloud database for storage based on different data types. The platform then uses this data stream to monitor equipment in real time, issue collection work orders, visualize multi-source sensor data, provide anomaly information alarms, and manage multi-role permissions. The operation and maintenance platform encompasses five management areas: real-time monitoring, project management, work order management, alarm management, and user management. Real-time monitoring displays equipment status and sensor placement. Project management is used to request sensor monitoring work orders. Work order management is used to view historical sensor and operation and maintenance data. Alarm management displays device anomaly information. User management manages passwords and usernames.

[0117] The remote operation and maintenance platform displays sensor signals uploaded to the cloud in real time, and alarm management allows users to view abnormal information about the precision shafting system, thus enabling real-time monitoring and alarming of the precision shafting system. In the alarm management, basic information about the monitoring equipment, monitoring time, and alarm thresholds are set. When the monitored sensor signal exceeds the specified threshold, the alarm management displays abnormal information and improvement suggestions.

[0118] Users can set up work orders on the platform that include collection and sensor information. After the collection is completed, they can query the data by entering the device information, the corresponding sensor signal, and the collection time. For the specific query process, please refer to Figure 9 .

[0119] In another embodiment of the present invention, a multi-source signal edge acquisition and intelligent operation and maintenance system for a GMC400 type dense bead shaft system is provided. The system can be used to implement the above-mentioned multi-source signal edge acquisition and intelligent operation and maintenance method for a GMC400 type dense bead shaft system. Specifically, the multi-source signal edge acquisition and intelligent operation and maintenance system for a GMC400 type dense bead shaft system includes an acquisition module, an experimental module, a calculation module and an operation and maintenance module.

[0120] The acquisition module collects vibration, displacement, and temperature signals from the outer side of the lower bearing, the outer side of the upper bearing, and the encoder bracket of the GMC400 bead-type shaft system; and collects PID parameters through the network port.

[0121] The experimental module sets the variable speed and load working conditions to conduct the GMC400 type bead-type shaft system condition monitoring verification experiment;

[0122] The computing module uses redundant data dimensionality reduction and feature extraction methods to perform edge computing on the data signals obtained from the verification experiment, and selects valid information from the original signal for data upload to the cloud;

[0123] The operation and maintenance module visualizes multi-source sensor data of the cloud data on the remote operation and maintenance platform, and uses a multi-source fusion dynamic accuracy prediction model to predict the dynamic accuracy of the dense bead shaft system. Based on the prediction results, the dense bead shaft system is monitored and alarmed in real time to realize remote intelligent operation and maintenance of the GMC400 dense bead shaft system.

[0124] In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system, including:

[0125] The vibration, displacement and temperature signals of the lower bearing outer side, upper bearing outer side and encoder bracket of the GMC400 bead-type precision shaft system are collected; the PID parameters are collected through the network port; the variable speed and load working conditions are set to conduct the GMC400 bead-type precision shaft system status monitoring verification experiment; the redundant data dimensionality reduction and feature extraction methods are used to perform edge computing on the data signals obtained from the verification experiment, and effective information is selected from the original signal for data upload to the cloud; the multi-source sensor data behind the cloud is visualized on the remote operation and maintenance platform, and the dynamic accuracy of the bead-type precision shaft system is predicted using a multi-source fusion dynamic accuracy prediction model. The bead-type precision shaft system is monitored and alarmed in real time based on the prediction results, realizing remote intelligent operation and maintenance of the GMC400 bead-type precision shaft system.

[0126] In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory (Non-Volatile Memory), such as at least one disk memory.

[0127] The processor may load and execute one or more instructions stored in a computer-readable storage medium to implement the corresponding steps of the multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system in the above embodiment; the processor may load and execute the following steps:

[0128] The vibration, displacement and temperature signals of the lower bearing outer side, upper bearing outer side and encoder bracket of the GMC400 bead-type precision shaft system are collected; the PID parameters are collected through the network port; the variable speed and load working conditions are set to conduct the GMC400 bead-type precision shaft system status monitoring verification experiment; the redundant data dimensionality reduction and feature extraction methods are used to perform edge computing on the data signals obtained from the verification experiment, and effective information is selected from the original signal for data upload to the cloud; the multi-source sensor data behind the cloud is visualized on the remote operation and maintenance platform, and the dynamic accuracy of the bead-type precision shaft system is predicted using a multi-source fusion dynamic accuracy prediction model. The bead-type precision shaft system is monitored and alarmed in real time based on the prediction results, realizing remote intelligent operation and maintenance of the GMC400 bead-type precision shaft system.

[0129] In summary, the present invention provides a multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type beaded shaft system. Starting from the equipment mechanism, the error fault sources that affect the equipment accuracy are analyzed, and the equipment failure mechanism is established; in response to the sampling rate requirements in different scenarios, an intelligent edge acquisition device is designed. The intelligent edge acquisition device adopts a "dual-module" design method of an acquisition module and a processing module to meet the equipment information monitoring needs; in response to the problem of insufficient accuracy retention of the GMC400 type beaded shaft system, a multi-source fusion dynamic accuracy prediction model is used to monitor the status of the GMC400 type beaded shaft system, explore the potential abnormal sudden changes and accuracy reduction conditions of the GMC400 type beaded shaft system, avoid losses caused by failure of the production process and accuracy reduction, and guide the regular repair and maintenance of the equipment; a remote cloud platform based on big data operation and maintenance is designed for the operation and maintenance of the GMC400 type beaded shaft system, thereby solving the shortcomings of traditional operation and maintenance.

[0130] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0131] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0132] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0133] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, and can be electrical, mechanical, or other forms.

[0134] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0135] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0136] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0140] The above content is only for explaining the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A multi-source signal edge acquisition and intelligent operation and maintenance method for GMC400 type dense bead shaft system, characterized in that: The following steps are involved: The vibration, displacement and temperature signals of the lower bearing outer side, upper bearing outer side and encoder bracket of the GMC400 bead-type shaft system are collected; PID parameters are collected through the network port; variable speed and load working conditions are set to conduct the GMC400 bead-type shaft system status monitoring verification experiment; redundant data dimensionality reduction and feature extraction methods are used to perform edge computing on the data signals obtained from the verification experiment, and valid information is selected from the original signal for data upload to the cloud; multi-source sensor data is visualized on the remote operation and maintenance platform for the cloud data, and the dynamic accuracy of the bead-type shaft system is predicted using a multi-source fusion dynamic accuracy prediction model. The bead-type shaft system is monitored and alarmed in real time based on the prediction results, realizing remote intelligent operation and maintenance of the GMC400 bead-type shaft system. The redundant data dimensionality reduction method uses the revolving door data compression method to reduce the dimensionality of redundant data, and the judgment basis for data compression is realized by establishing the upper and lower revolving door models. When the revolving door interval can cover the original data, the current coverage interval is recorded. , the next turnstile interval will use this coverage interval , if the current interval can still be covered, then gradually increase the coverage interval t, and the increasing expression is: If the current interval cannot be covered, the penalty mode is used to reduce the coverage interval. The reduction expression is: Among them, t Represents the current revolving door coverage area, Represents the previous revolving door coverage interval, represents the reward factor, represents the penalty factor, n represents the number of cumulative coverages; the mathematical model of the upper revolving door slope is: The mathematical model of the slope of the lower revolving door is: in, Represents the currently received data value, Represents the value of the first point in the stored data, Represents the fault tolerance value.

2. The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system according to claim 1 is characterized in that: Acceleration sensors, temperature sensors, and eddy current sensors are installed on the outside of the lower bearing, the outside of the upper bearing, and the encoder bracket of the GMC400 bead-type shaft system, and intelligent edge acquisition equipment is used to collect vibration, displacement, and temperature signals.

3. The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system according to claim 2 is characterized in that: The intelligent edge acquisition device includes an acquisition module and a processing module. The acquisition module realizes the collection of multi-source sensor data, and the processing module realizes the storage, transmission and analysis of multi-source sensor data.

4. The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system according to claim 1 is characterized in that: The settings for variable speed and load conditions are as follows: The axis turntable starts rotating at a constant speed of 10,000 pulses from 0s, starts accelerating at 3s, and the speed accelerates from 10,000 pulses to 100,000 pulses. The acceleration ends at 45s, at which time the speed reaches 100,000 pulses. Finally, it starts to decelerate at 46s and stops at 48s. During the acceleration period, the pulse changes are 10,000, 20,000, 30,000, ..., 90,000, 100,000, and continues to accelerate.

5. The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system according to claim 1 is characterized in that: The specific feature extraction method is: First, the signal is subjected to EMD modal decomposition, and the empirical mode functions with concentrated frequency components are selected and then combined into a new signal. The time-frequency domain quantities reflecting energy changes, energy distribution and geometric shapes of the new signal are extracted and analyzed respectively to obtain the eigenvectors that can reflect the signal characteristics. The original signal is decomposed using the EMD algorithm; finally, a series of IMFs are obtained, the relative energy ratio of each IMF is calculated, and the first three IMFs are selected for combination as the new signal for feature extraction.

6. The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system according to claim 1 is characterized in that: The multi-source fusion dynamic precision prediction model consists of three parts: multi-source heterogeneous signal preprocessing, multi-source data multi-dimensional fusion, and a prediction layer based on the LSTM algorithm; multi-source heterogeneous signal preprocessing, time-frequency domain feature information extraction, data dimension transformation and data padding of multi-source heterogeneous data, and standardization of data length and format; the LSTM algorithm model is used to mine sensitive fault information in the sequence to achieve deep fusion of multi-source information, and the Attention mechanism is added to the training model to improve the balance of new and old information in the LSTM neural network. The DTW mechanism is used to predict unknown faults, and fault output is performed when a new location fault occurs to predict unknown faults.

7. The multi-source signal edge acquisition and intelligent operation and maintenance method for the GMC400 type dense bead shaft system according to claim 6 is characterized in that: The DTW mechanism predicts unknown faults. There are two multi-source fusion sequences R in the input model. n 、U m , calculate the Euclidean distance between the two sequences, build the Euclidean distance matrix between the two sequences, calculate the cumulative distance to get the whole The DP matrix under the DP matrix is used to calculate the DTW value based on the cumulative distance between the two sequences after the DP matrix is calculated; this is used as an indicator to measure the occurrence of unknown faults.

Images