Equipment maintenance management method and device

Abstract

The invention discloses an equipment maintenance management method and device. The method comprises the following steps: collecting vibration data of equipment in different motion states; performing feature extraction and labeling on the vibration data; a training sample is generated according to the extracted feature values and corresponding labeling information, the labeling information is used for indicating equipment states corresponding to the vibration data, and the equipment states comprise a normal state and a specific fault state; training a fault diagnosis model by using the training sample; and performing fault diagnosis on the equipment by using the real-time monitoring data of the equipment and the fault diagnosis model, and determining the current state of the equipment. By means of the scheme, the real-time performance and intelligence of equipment vibration monitoring can be improved, and normal operation of equipment can be better guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment failure processing, and in particular to an equipment maintenance management method and device. Background Art

[0002] In modern industrial production, the stable operation of mechanical equipment is one of the key factors to ensure production efficiency and product quality. With the advancement of technology, more and more companies have begun to pay attention to preventive maintenance of equipment to reduce downtime and maintenance costs caused by sudden failures. Vibration monitoring, as a non-invasive detection method, can effectively reflect the working status of internal parts of mechanical equipment, and plays an important role in early detection of potential failures and extending the service life of equipment. At present, vibration measurement technology still faces some challenges, which are mainly reflected in the following aspects:

[0003] (1) Real-time problem: In some application scenarios, vibration data needs to be obtained and responded to quickly, but some existing solutions cannot meet this requirement.

[0004] (2) Adaptability to complex environments: Industrial sites often have a variety of adverse conditions such as high temperature, high humidity, and strong electromagnetic interference, which puts higher demands on the stability and reliability of vibration measurement equipment.

[0005] (3) Difficulty of data analysis: Although ISO 20816-3:2022 provides some basic analysis methods, it is still a difficult problem to effectively extract features and perform accurate fault diagnosis for complex vibration signals. Summary of the invention

[0006] The present invention provides an equipment maintenance management and device to improve the real-time and intelligence of equipment vibration monitoring and better ensure the normal operation of the equipment.

[0007] To this end, the present invention provides the following technical solutions:

[0008] The present invention provides an equipment maintenance management method, the method comprising:

[0009] Collect vibration data of the equipment in different motion states;

[0010] Extracting and labeling features of the vibration data;

[0011] Generate a training sample according to the extracted feature values ​​and corresponding annotation information, wherein the annotation information is used to indicate the device state corresponding to the vibration data, and the device state includes a normal state and a specific fault state;

[0012] Using the training samples to train a fault diagnosis model;

[0013] The real-time monitoring data of the equipment and the fault diagnosis model are used to perform fault diagnosis on the equipment to determine the current state of the equipment.

[0014] Optionally, the vibration data includes any one or more of the following: vibration velocity, vibration displacement, and vibration acceleration.

[0015] Optionally, the extracting features from the vibration data includes: extracting time domain features and frequency domain features from the vibration data to obtain time domain feature values ​​and frequency domain feature values.

[0016] Optionally, the time domain eigenvalues ​​include any one or more of the following: RMS value, peak value, kurtosis, skewness; the frequency domain eigenvalues ​​include any one or more of the following: main frequency, bandwidth.

[0017] Optionally, the extracting features from the vibration data further comprises: extracting time-frequency features from the vibration data to obtain time-frequency feature values.

[0018] Optionally, the labeling of the vibration data includes: labeling the vibration data according to the characteristic value and an industry standard threshold to obtain labeling information.

[0019] Optionally, generating training samples according to the extracted feature values ​​and corresponding annotation information includes:

[0020] Determine the best feature subset based on the extracted feature values;

[0021] A training sample is generated according to the optimal feature subset and the corresponding annotation information.

[0022] Optionally, the fault diagnosis model is any one of the following: a support vector machine, a neural network, or a decision tree.

[0023] Optionally, the method further includes: if the current state of the device is an abnormal state, an alarm is issued.

[0024] Optionally, the method further comprises:

[0025] Collect vibration data and maintenance records of equipment over different time periods;

[0026] Performing feature extraction on the vibration data and the maintenance record respectively to obtain feature values;

[0027] Arrange the extracted feature values ​​in chronological order to form time series data;

[0028] Using the time series data to train a state prediction model;

[0029] The real-time monitoring data of the equipment and the state prediction model are used to predict the state of the equipment and determine the future state of the equipment.

[0030] Optionally, the state prediction model is any one of the following: ARIMA model, LSTM model.

[0031] Optionally, the method further includes: generating an equipment maintenance plan according to the future state of the equipment.

[0032] Optionally, the method further comprises: generating a monitoring report, wherein the monitoring report comprises: real-time monitoring data, analysis results and suggested measures.

[0033] The present invention also provides an equipment maintenance management device, the device comprising:

[0034] Data acquisition module, used to obtain real-time monitoring data of equipment;

[0035] The fault diagnosis module is used to perform fault diagnosis on the equipment using the real-time monitoring data and the pre-trained fault diagnosis model to determine the current state of the equipment.

[0036] Optionally, the fault diagnosis module is further used to trigger an alarm when it is determined that the current state of the device is an abnormal state.

[0037] Optionally, the device further includes: a state prediction module, configured to use the real-time monitoring data and the state prediction model to perform state prediction on the equipment and determine the future state of the equipment.

[0038] Optionally, the device further comprises: a planning module, configured to generate an equipment maintenance plan according to the future state of the equipment.

[0039] Optionally, the device further comprises: a report generation module, configured to generate a monitoring report, wherein the monitoring report comprises: real-time monitoring data, analysis results and suggested measures.

[0040] Optionally, the device further comprises: a display module, configured to display the real-time monitoring data in real time.

[0041] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the equipment maintenance management method are executed.

[0042] The equipment maintenance management and device provided by the present invention not only fully complies with the requirements of ISO 20816-3:2022, but also has the following beneficial effects:

[0043] Enhanced real-time processing capabilities: Through efficient signal processing algorithms, real-time online monitoring can be performed based on real-time collected vibration data to detect equipment abnormalities in a timely manner.

[0044] Introducing intelligent analysis technology: Combining machine learning and artificial intelligence technology to automatically identify and classify different vibration modes, improving the accuracy and reliability of fault diagnosis.

[0045] Early fault detection: Through real-time monitoring and intelligent analysis, faults can be discovered and warned in time at the early stages, avoiding sudden equipment failure and reducing production losses.

[0046] Extend equipment life: Regular vibration monitoring and maintenance can effectively extend the service life of equipment and reduce the cost of replacing new equipment.

[0047] Optimize maintenance plan: Based on the results of vibration data analysis, a maintenance plan can be scientifically formulated to avoid over-maintenance or under-maintenance and improve maintenance efficiency.

[0048] Furthermore, when abnormal vibration of the equipment is detected, an alarm can be immediately sounded to remind the operator to take measures to avoid further expansion of the fault.

[0049] Furthermore, by analyzing historical data, the future status of the equipment can be predicted, helping equipment maintenance personnel to develop maintenance plans and effectively reduce unexpected downtime.

[0050] Furthermore, detailed monitoring reports can be automatically generated, including vibration data, analysis results and recommended measures, to facilitate management personnel's decision-making.

[0051] Furthermore, the solution of the present invention can also support remote access and monitoring functions, so that operators can view the status of equipment in real time at any location, thereby improving management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 is a flow chart of the equipment maintenance management method provided by an embodiment of the present invention;

[0054] Figure 2 It is a flow chart for predicting the state of a device in the method of the present invention;

[0055] Figure 3 It is a structural schematic diagram of a device failure handling apparatus provided by an embodiment of the present invention;

[0056] Figure 4It is another structural schematic diagram of the device failure handling apparatus provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0058] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0059] Mechanical vibration is a common phenomenon in the operation of mechanical equipment. The parts and complete machines of mechanical equipment have different degrees of vibration. The vibration of mechanical equipment often affects its working accuracy, aggravates the wear of the machine, and accelerates fatigue damage; and with the increase of wear and fatigue damage, the vibration of mechanical equipment will become more intense, eventually leading to equipment failure and damage. Vibration intensification is often a physical phenomenon that occurs with abnormal working conditions or even failure of machine parts. By measuring and analyzing mechanical vibration signals, we can understand the degree of deterioration and nature of failure of mechanical equipment.

[0060] In response to the preventive maintenance needs of equipment, the present invention provides an equipment maintenance management method and device, which pre-trains a fault diagnosis model, uses the real-time monitoring data of the equipment and the fault diagnosis model to perform fault diagnosis on the equipment, and determines the current state of the equipment. The current state of the equipment can be accurately determined, providing technical support for the fault diagnosis and normal operation of the equipment. Furthermore, the real-time monitoring data of the equipment and the pre-trained state prediction model can be used to predict the future state of the equipment. Furthermore, according to the state prediction results, an equipment maintenance plan can be generated to fully ensure the stability and reliability of the equipment operation.

[0061] like Figure 1 FIG. 1 is a flow chart of the equipment maintenance management method provided by the present invention, comprising the following steps:

[0062] Step 101, collecting vibration data of the device in different motion states.

[0063] The motion state includes data of a normal state and various fault states, and each piece of data may include multiple characteristic values. For example, the vibration data includes some data obtained by direct measurement, such as any one or more of the following: vibration velocity, vibration displacement, vibration acceleration, etc.

[0064] Furthermore, other characteristic values ​​of each piece of data may be obtained through feature extraction.

[0065] Step 102: extracting and labeling features of the vibration data.

[0066] For example, the vibration data is subjected to time domain feature and frequency domain feature extraction to obtain time domain feature values ​​and frequency domain feature values.

[0067] The time domain eigenvalues ​​may include but are not limited to any one or more of the following: RMS (root mean square) value, peak value, kurtosis, skewness, etc. The kurtosis is a statistic that describes the sharpness of the data distribution morphology, and is used to measure the thickness of the tail of the data distribution and the degree to which the data center tends to the center. The skewness is a measure of the direction and degree of skewness of the statistical data distribution. The calculation of each eigenvalue can adopt the existing calculation formula, which will not be described in detail.

[0068] The frequency domain eigenvalues ​​may include but are not limited to any one or more of the following: main frequency, bandwidth, etc. There are many ways to extract frequency domain eigenvalues, such as:

[0069] (1) Convert discrete time domain signals into frequency domain signals through Fast Fourier Transform (FFT) to obtain the frequency information of the signal;

[0070] (2) Calculate the power spectrum density, which describes the power distribution of information at different frequencies. By calculating the power spectrum density of a signal, we can understand the energy distribution of the signal in the frequency domain.

[0071] (3) Spectral feature extraction: for example, extracting from the frequency domain signal: the peak value of the spectrum, the intensity of the frequency component, the spectrum width, etc.

[0072] Of course, there may be other methods and frequency domain eigenvalues, which are not listed here one by one. In specific implementation, the required frequency domain eigenvalue may be selected as needed.

[0073] Furthermore, in some embodiments, the vibration data may be subjected to time-frequency feature extraction to obtain time-frequency feature values. For example, there may be but are not limited to the following methods:

[0074] (1) Apply wavelet transform to decompose the signal into time-frequency domain signals of different scales; wavelet transform can decompose the signal into a combination of wavelet functions of different scales and positions. By analyzing the wavelet coefficients, the characteristics of the signal at different time and frequency scales can be extracted.

[0075] (2) Apply Short-Time Fourier Transform (STFT): decompose the signal into a series of frequency domain components in different time windows, so as to analyze the frequency characteristics of the signal at different time points.

[0076] (3) Extract characteristic values ​​from time-frequency domain signals, such as time-frequency spectrum, energy distribution, etc.

[0077] In the embodiment of the present invention, labeling the collected vibration data means labeling the device state corresponding to each vibration data, wherein the device state includes a normal state and a specific fault state (ie, a fault type). In other words, the labeling information is used to indicate the device state corresponding to the vibration data.

[0078] In an embodiment of the present invention, the vibration data can be labeled according to the characteristic value corresponding to each piece of data and the industry standard threshold value to obtain the labeling information. For example, the operating status of the equipment can be judged in combination with the evaluation criteria in the ISO 20816-3:2022 standard. The judgment process is as follows:

[0079] First, according to the ISO 20816-3:2022 standard, the vibration characteristic value thresholds of a specific equipment type are obtained. These thresholds include but are not limited to vibration velocity, vibration displacement, and vibration acceleration.

[0080] For example, the specific thresholds corresponding to a certain device are as follows:

[0081] Vibration speed: 7.1mm / s;

[0082] Vibration displacement: 100μm;

[0083] Vibration acceleration: 100m / s 2 ;

[0084] RMS value: 5.0;

[0085] Peak: 10.0;

[0086] Kurtosis: 3.0;

[0087] Main frequency: 5000Hz;

[0088] Bandwidth: 1000Hz;

[0089] Based on the above parameter thresholds, the extracted feature values ​​are compared with the corresponding thresholds. According to the comparison results, it is determined whether the device is in a normal state or a fault state, and the fault type can be determined according to different abnormal values. For example:

[0090] (1) Comparison of time domain features:

[0091] If the extracted RMS value is greater than a threshold value of 5.0, the RMS value is determined to be abnormal.

[0092] If the extracted peak value is greater than the threshold value 10.0, the peak value is determined to be abnormal.

[0093] If the extracted kurtosis is greater than the threshold value of 3.0, the kurtosis is determined to be abnormal.

[0094] (2) Frequency domain feature comparison:

[0095] If the extracted main frequency is greater than the threshold value of 5000 Hz, the main frequency is determined to be abnormal.

[0096] If the extracted bandwidth is greater than the threshold value of 1000 Hz, the bandwidth is determined to be abnormal.

[0097] (3) Comprehensive status assessment:

[0098] The evaluation results of all characteristic values ​​are summarized. If any characteristic value is judged to be abnormal, the device state is finally judged to be abnormal. The granularity and specific type of the fault type can be determined according to the abnormal characteristic value, which is not limited in this embodiment of the present invention; otherwise, the device state is judged to be normal.

[0099] Step 103: Generate training samples based on the extracted feature values ​​and corresponding annotation information.

[0100] It should be noted that, in specific implementation, the best feature subset can be determined based on the extracted feature values, and training samples can be generated based on the best feature subset and the corresponding annotation information. In other words, some key feature values ​​that best reflect the device status can be selected, and training samples can be generated based on these key feature values ​​and the corresponding annotation information, thereby simplifying the computational complexity of model training and improving training efficiency.

[0101] The method for selecting key eigenvalues ​​may be, for example, recursive feature elimination, model-based feature selection, etc., which is not limited in the embodiment of the present invention.

[0102] In some embodiments, vibration data obtained by direct measurement, such as the above-mentioned vibration velocity, vibration displacement, vibration acceleration, etc., may also be used as characteristic parameters, together with the extracted characteristic values, as input parameters of the model.

[0103] In addition, since the units of different parameter characteristics are different, each input parameter can be further normalized to obtain normalized parameters, such as minimum value, maximum value, mean value, standard deviation, etc.

[0104] Step 104: train a fault diagnosis model using the training samples.

[0105] In specific implementation, the fault diagnosis model may adopt but is not limited to any of the following models:

[0106] (1) Support Vector Machine (SVM): Classification is performed by optimizing the problem and the decision function. SVM achieves classification by finding an optimal hyperplane to maximize the interval between different categories.

[0107] (2) Neural Network (NN): Trained through forward propagation and back propagation. A neural network consists of multiple layers of neurons. The output is calculated through forward propagation, and the weights are adjusted through back propagation to minimize the loss function.

[0108] (3) Decision Tree (DT): Classification by splitting criterion. Decision tree divides the data set into different subsets by recursively selecting the best splitting attribute until the stopping condition is met.

[0109] The training process of the module may adopt some existing training methods, which are not limited in the embodiment of the present invention.

[0110] Furthermore, an independent validation dataset can be used to validate the performance of the fault diagnosis model. The validation dataset contains device data with known status to ensure the accuracy of the validation results. Specifically, the performance of the fault diagnosis model can be evaluated by calculating indicators such as the accuracy, recall, and F1 score of the model. If the performance of the fault diagnosis model does not meet the requirements, the fault diagnosis model parameters can be adjusted or the feature values ​​can be reselected, and then the model can be retrained.

[0111] Step 105, using the real-time monitoring data of the device and the fault diagnosis model to perform fault diagnosis on the device to determine the current state of the device.

[0112] The trained fault diagnosis model is applied to the actual monitoring data to perform fault diagnosis. By collecting the vibration data of the equipment in real time, extracting the characteristic values ​​and inputting them into the fault diagnosis model, the fault diagnosis model will output the current status of the equipment (normal or specific fault mode).

[0113] Furthermore, according to the diagnosis results of the fault diagnosis model, corresponding maintenance measures can be taken in time to prevent the occurrence of equipment failures and ensure production safety and efficiency.

[0114] Furthermore, if the current state of the device is abnormal, an alarm is issued to remind the operator to take measures to prevent the fault from further expanding. The specific alarm method can be set as needed, for example, one or more of the following methods can be used for alarm: sound and light alarm, SMS notification, email notification, etc.

[0115] In some embodiments, the equipment maintenance management method of the present invention can also use the vibration data and maintenance records of the equipment in different time periods to train a state prediction model in advance, and use the state prediction model and the real-time monitoring data of the equipment to predict the state of the equipment and determine the future state of the equipment. Then, based on the prediction results, equipment maintenance activities can be carried out in advance to effectively prevent the occurrence of equipment failures, ensure safe production operation, and improve production efficiency.

[0116] like Figure 2 FIG. 1 is a flow chart of the method for predicting the state of a device in the present invention, which includes the following steps:

[0117] Step 201, collecting vibration data and maintenance records of equipment in different time periods.

[0118] The vibration data includes but is not limited to: vibration velocity, vibration displacement, vibration acceleration, etc. The maintenance record includes but is not limited to: maintenance date, maintenance type, maintenance reason, and other information.

[0119] Step 202 : extracting features from the vibration data and the maintenance record respectively to obtain feature values.

[0120] The feature extraction of the vibration data can refer to the previous Figure 1 The description of the illustrated embodiment will not be repeated here. The characteristic values ​​of the extracted vibration data may include, for example, but are not limited to, any one or more of the following: root mean square value, peak value, kurtosis, main frequency, bandwidth, and the like.

[0121] The feature values ​​extracted from the maintenance records may include, but are not limited to: maintenance frequency, maintenance type, etc.

[0122] Step 203, arranging the extracted feature values ​​in chronological order to form time series data.

[0123] Step 204: training a state prediction model using the time series data.

[0124] The state prediction model may adopt any one of the following, for example: an ARIMA (Autoregressive Integrated Moving Average) model, or an LSTM (Long Short-Term Memory) model.

[0125] Among them, the ARIMA model extracts the time series patterns hidden behind the data through the autocorrelation and difference of the data, and then uses these patterns to predict future data. Specifically, time series analysis can be performed through the autoregressive (AR), moving average (MA) and difference (Integrated, I) parts.

[0126] Among them, the LSTM model can more effectively capture long-distance dependencies by introducing a gating mechanism, thus performing well in processing and predicting time series data. The basic structure of LSTM includes three main gating units: input gate, forget gate, and output gate. These gating units help LSTM remember important information and forget unimportant information by controlling the inflow and outflow of information.

[0127] It should be noted that when training the model, the extracted feature values ​​need to be sorted in chronological order to form time series data.

[0128] Similarly, an independent validation data set can be further used to validate the performance of the state prediction model. The validation data set contains device data with known states to ensure the accuracy of the validation results. Specifically, the performance of the state prediction model can be evaluated by calculating the prediction error of the model (such as mean square error (MSE), root mean square error (RMSE), etc.). If the state prediction model performance does not meet the requirements, adjust the state prediction model parameters or reselect the feature values, and then retrain the model.

[0129] Step 205: Use the real-time monitoring data of the device and the state prediction model to predict the state of the device and determine the future state of the device.

[0130] The real-time monitoring data of the equipment is the vibration data of the equipment over a period of time. The characteristic values ​​are extracted and input into the state prediction model. The future state of the equipment is determined according to the output of the state prediction model.

[0131] Furthermore, in some embodiments, an equipment maintenance plan may also be generated based on the future status of the equipment.

[0132] In some embodiments, a monitoring report may also be generated, the monitoring report including: real-time monitoring data, analysis results, and suggested measures, etc., to facilitate decision-making by management personnel.

[0133] The monitoring report can be automatically generated by the system, such as pre-designing a report template, which may include equipment information, monitoring time, vibration data, characteristic value analysis, fault diagnosis results and recommended measures, etc.; according to the equipment status prediction results, the corresponding monitoring report is generated using the report template. The monitoring report can be output in one or more formats, such as PDF, Excel, Word, etc.

[0134] Accordingly, an embodiment of the present invention also provides a device for equipment maintenance management, such as Figure 3 Shown is a structural schematic diagram of the device.

[0135] The equipment maintenance management device 300 includes:

[0136] Data acquisition module 301, used to obtain real-time monitoring data of the equipment;

[0137] The fault diagnosis module 302 is used to perform fault diagnosis on the device using the real-time monitoring data and the pre-trained fault diagnosis model to determine the current state of the device.

[0138] The fault diagnosis model can be pre-trained by a corresponding model training module. The training process can refer to the description in the above embodiment of the method of the present invention, which will not be repeated here.

[0139] It should be noted that the model training module can be a part of the equipment maintenance management device 300, or can be independent of the equipment maintenance management device 300, which is not limited in this embodiment of the present invention.

[0140] In some embodiments, the fault diagnosis module 302 is further configured to trigger an alarm when determining that the current state of the device is an abnormal state, for example, triggering an audible and visual alarm, a text message notification, an email notification, etc.

[0141] like Figure 4 , which is another structural schematic diagram of the equipment maintenance management device provided in an embodiment of the present invention.

[0142] and Figure 3 The difference between the illustrated embodiment is that, in this embodiment, the equipment maintenance management device 300 further includes:

[0143] The state prediction module 303 is used to use the real-time monitoring data and the state prediction model to predict the state of the device and determine the future state of the device.

[0144] Similarly, the state prediction model can be pre-trained by a corresponding model training module. The training process can refer to the description in the previous embodiment of the method of the present invention, and will not be repeated here.

[0145] In some embodiments, the equipment maintenance management apparatus 300 may further include: a planning module (not shown) for generating an equipment maintenance plan according to the future state of the equipment.

[0146] In some embodiments, the equipment maintenance management device 300 may further include: a report generation module (not shown) for generating a monitoring report, wherein the monitoring report includes: real-time monitoring data, analysis results and suggested measures.

[0147] In some embodiments, the equipment maintenance management device 300 may further include: a display module for displaying the real-time monitoring data in real time. In addition, characteristic values ​​and analysis results may be displayed to facilitate operators to monitor the equipment status. The displayed information may use charts (such as line charts, bar charts, scatter plots, etc.) to intuitively display vibration data and characteristic values. Furthermore, multiple view options may be provided, such as time domain view, frequency domain view, and time-frequency domain view, and users may switch between different views as needed.

[0148] The specific implementation of each of the above modules can be found in the description of the above method embodiments of the present invention, which will not be repeated here.

[0149] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0150] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0151] In several embodiments provided by the present invention, it should be understood that the disclosed device can be implemented in other ways.

[0152] The present invention also provides a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and the computer program can be executed when it is run. Figure 1 and / or Figure 2 The storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc. The storage medium may also include a non-volatile memory (non-volatile) or a non-transitory memory, etc.

[0153] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, a computer, a server or a data provider to another website site, a computer, a server or a data provider by wired or wireless means.

[0154] The embodiments of the present invention are described in detail above. The present invention is described in detail using specific implementation methods herein. The description of the above embodiments is only used to help understand the method and system of the present invention. It is only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should belong to the scope of protection of the present invention, and the content of this specification should not be understood as limiting the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for equipment maintenance management, characterized in that: The method comprises: Collect vibration data of the equipment in different motion states; Extracting and labeling features of the vibration data; Generate a training sample according to the extracted feature values ​​and corresponding annotation information, wherein the annotation information is used to indicate the device state corresponding to the vibration data, and the device state includes a normal state and a specific fault state; Using the training samples to train a fault diagnosis model; The real-time monitoring data of the equipment and the fault diagnosis model are used to perform fault diagnosis on the equipment to determine the current state of the equipment.

2. The equipment maintenance management method according to claim 1, characterized in that: The feature extraction of the vibration data comprises: The time domain features and frequency domain features are extracted from the vibration data to obtain time domain feature values ​​and frequency domain feature values.

3. The equipment maintenance management method according to claim 2, characterized in that: The feature extraction of the vibration data further comprises: The time-frequency feature extraction is performed on the vibration data to obtain the time-frequency feature value.

4. The equipment maintenance management method according to claim 1, characterized in that: The marking of the vibration data comprises: The vibration data is labeled according to the characteristic value and the industry standard threshold to obtain labeling information.

5. The equipment maintenance management method according to any one of claims 1 to 4, characterized in that: The method further comprises: Collect vibration data and maintenance records of equipment over different time periods; Performing feature extraction on the vibration data and the maintenance record respectively to obtain feature values; Arrange the extracted feature values ​​in chronological order to form time series data; Using the time series data to train a state prediction model; The real-time monitoring data of the equipment and the state prediction model are used to predict the state of the equipment and determine the future state of the equipment.

6. The equipment maintenance management method according to claim 5, characterized in that: The method further comprises: An equipment maintenance plan is generated according to the future state of the equipment.

7. An equipment maintenance management device, characterized in that: The device comprises: Data acquisition module, used to obtain real-time monitoring data of equipment; The fault diagnosis module is used to perform fault diagnosis on the equipment using the real-time monitoring data and the pre-trained fault diagnosis model to determine the current state of the equipment.

8. The equipment maintenance management device according to claim 7, characterized in that: The fault diagnosis module is also used to trigger an alarm when it is determined that the current state of the device is an abnormal state.

9. The equipment maintenance management device according to claim 7, characterized in that: The device also includes: The state prediction module is used to use the real-time monitoring data and the state prediction model to predict the state of the equipment and determine the future state of the equipment.

10. The equipment maintenance management device according to claim 9, characterized in that: The device also includes: A planning module is used to generate an equipment maintenance plan according to the future status of the equipment.

Images