An intelligent electromechanical monitoring system based on industrial Internet

Abstract

The present invention discloses an intelligent electromechanical monitoring system based on the industrial Internet, which relates to the field of intelligent electromechanical monitoring technology, including a sensing and acquisition module, a data transmission module, an intelligent analysis module, a user interface module and a control module; the existing intelligent electromechanical monitoring system based on the industrial Internet collects equipment data in real time through sensors, and combines big data analysis and artificial intelligence to realize equipment status monitoring, fault prediction and operation optimization. However, there are still problems such as insufficient analysis accuracy, weak prediction ability, weak regional equipment performance management and slow response speed; the present invention can timely detect anomalies, predict trends and generate maintenance suggestions through functions such as equipment data classification, standardized processing, correlation analysis, fault probability modeling and health scoring. By optimizing the operating status of the equipment, the equipment safety, reliability and management intelligence level are significantly improved, the maintenance cost and failure risk are reduced, and the continuous stability of production is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of intelligent electromechanical monitoring technology, and in particular to an intelligent electromechanical monitoring system based on industrial Internet. Background Art

[0002] The existing intelligent electromechanical monitoring system based on the Industrial Internet deploys high-precision sensors and edge computing devices to achieve real-time collection of electromechanical equipment operation data, including key parameters such as vibration amplitude, temperature, current and voltage. With the help of big data analysis and artificial intelligence technology, the system can accurately monitor the status of equipment, quickly identify potential faults and perform predictive maintenance, while optimizing equipment operating parameters to improve energy efficiency. Through the cloud platform and remote control functions, operation and maintenance personnel can achieve centralized management of equipment and cross-regional collaboration.

[0003] However, the system still lacks data standardization and analysis accuracy, and there is a lack of standardization in the integration and analysis of multi-device and multi-regional data, resulting in low accuracy of anomaly detection and processing, weak prediction ability, lack of causal analysis, and simple threshold judgment in fault prediction. It is difficult to accurately predict the probability of failure and analyze the correlation between devices, and the regional equipment performance management is weak. There is a lack of effective means for regional management of large-scale equipment, and it is difficult to identify regional problems in a timely manner. The response speed is slow and it is difficult to make dynamic adjustments. After an anomaly occurs, static analysis is mainly used, and it is impossible to respond quickly or adjust dynamically. Further optimization is urgently needed in terms of hardware reliability, algorithm accuracy and user-friendliness to meet the needs of complex industrial environments. Summary of the invention

[0004] The purpose of the present invention is to solve the problems mentioned in the above background technology and to propose an intelligent electromechanical monitoring system based on the industrial Internet.

[0005] The object of the present invention can be achieved by the following technical solutions: an intelligent electromechanical monitoring system based on industrial Internet, comprising an intelligent analysis module and a control module;

[0006] The intelligent analysis module is responsible for distinguishing the collected data to obtain the numbers DQAQSQ and analyzing the distinguished data, specifically:

[0007] Standardize the data to obtain data interpolation, and then interpolate the obtained data to supplement the corresponding missing data points; perform correlation analysis on the standardized data to obtain the correlation coefficient, and build a fault probability model to analyze whether a certain equipment abnormality is related to changes in other equipment parameters to obtain the probability of equipment failure; then calculate the health score for each electromechanical equipment to obtain a health score, and perform trend analysis on the health status of the corresponding electromechanical equipment. By predicting the future operating status of the equipment based on the time series model, the data is imported into the model to calculate the vibration amplitude; then combine the health score result with the trend prediction result, and when there is a potential risk in the health status or the health status needs attention and the trend prediction is abnormal, maintenance is performed in advance; the performance of the electromechanical equipment within the distinguishing mark is calculated to obtain the overall performance score of the equipment in the area, and the equipment operation data is grouped through cluster analysis to obtain the sum of the squares of the distances from all data points to the centers of their respective clusters; the overall performance score of the equipment in the area and the sum of the squares of the distances from all data points to the centers of their respective clusters are weighted to obtain the regional anomaly coefficient;

[0008] The control module is used to receive vibration amplitude abnormal instructions, regional performance abnormal instructions and related fault instructions, obtain the area where the distinguishing mark is issued through the regional performance abnormal instruction, and lock the electromechanical equipment in the corresponding marked area by receiving the vibration amplitude abnormal instruction, and obtain the unique identification code corresponding to the electromechanical equipment; then lock the unique identification code of the relevant electromechanical equipment in the area by the relevant fault instruction; draw a causal network diagram based on the electromechanical equipment with correlation according to the threshold value of abnormal electromechanical equipment.

[0009] As a preferred embodiment of the present invention, it also includes a sensing and acquisition module, a data transmission module and a user interface module;

[0010] The sensing and acquisition module is used to collect electromechanical monitoring data connected to the industrial Internet, and the electromechanical monitoring data includes electrical data, mechanical data, temperature data, pressure fluid data, environmental data and status data;

[0011] The data transmission module is responsible for transmitting the data collected by the sensing and acquisition module to the industrial Internet platform and the intelligent analysis module; it is also used to obtain vibration amplitude abnormality instructions and regional performance abnormality instructions and related fault instructions and send them to the control module;

[0012] The user interface module is used to receive and display a causal network diagram.

[0013] As a preferred embodiment of the present invention, the specific process of distinguishing the collected data to obtain the number DQAQSQ is:

[0014] Step 1: Classify by device type. Perform a preliminary classification of the data based on the device category and mark the corresponding devices as: AQ;

[0015] Step 2: Classify data attributes and archive data of the same attributes, and divide them into electrical data, mechanical data, and environmental data; mark the corresponding data attributes as: SQ;

[0016] Step 3: Divide according to the electromechanical location area; bind the regional identification. The data collected by each device is bound to a specific geographical location or regional number during installation. The specific geographical location is distinguished and marked as: DQ; the distinguishing marks of steps 1, 2 and 3 are integrated to obtain the number DQAQSQ.

[0017] As a preferred implementation of the present invention, the specific process of performing standardization processing on the data to obtain data interpolation is:

[0018] Calculate the mean and standard deviation of historical data; if a certain electromechanical device collects X1, X2, X3, ..., X at n time points n The value of; then through the mean formula: Output historical mean , through the standard deviation formula: Output historical standard deviation , and then standardize each piece of data through the formula: Get the standardized data points , compare the standardized data with the preset normal range. If the standardized data exceeds the normal range, it will be further processed and marked as abnormal data; the abnormal data points are smoothed by the mean of the sliding window: Output the smoothed value at the current time point t , k is the window size; time alignment of data, time synchronization during analysis, specific calculation process: determine the time interval, set the time interval uniformly according to application requirements, if there is no data at a certain time point, use the data of adjacent time points for linear interpolation, through the formula: Output data interpolation , are the data values ​​at adjacent time points, are the timestamps of adjacent time points, is the time point that needs to be interpolated.

[0019] As a preferred embodiment of the present invention, the specific process of performing correlation analysis on the standardized data to obtain the correlation coefficient is:

[0020] Get the data of two electromechanical devices; if the two devices are Y and U, collect n1 data points of the same type, and the data sequence of device Y is: Y1, Y2, Y3, ..., Y n1 , the data sequence of device U: U1, U2, U3, ..., U n1 ; Calculate the data mean of the two devices: by formula: Output , is the data mean of device Y, is the data mean of device U; then calculate the covariance and variance: the covariance is calculated by the formula: Get; variance through the formula: Get; Substitute the calculated covariance and variance into the relevant calculation formula to get the correlation coefficient. The process is: Get the correlation coefficient .

[0021] As a preferred embodiment of the present invention, the specific calculation process of the probability of failure of the device is:

[0022] The fault probability model is calculated using logistic regression: If the fault U is affected by equipment Y1, Y2, Y3, ..., Y n1 Parameter influence, the model formula is: Probability of output device failure , Y1, Y2, Y3, ..., Y n1 are monitoring parameters that affect the fault, β0, β1, ..., β n The model parameters are obtained by using historical data for model training, and the historical fault state U and the corresponding equipment parameters Y1, Y2, Y3, ..., Y n1 Input the model, and the model outputs the optimized β0, β1, ..., β n ; The real-time data Y1, Y2, Y3, ..., Y n1 Input model to get failure probability .

[0023] As a preferred embodiment of the present invention, the specific process of calculating the health score of each electromechanical device is as follows: obtaining the key health indicators of each electromechanical device for calculation. If the health score of a device is calculated by three key indicators: temperature Z1, vibration amplitude Z2, and current Z3, the historical minimum and maximum values ​​of each indicator are obtained; by formula: Output health score H; Z j is the value of the jth key indicator, ω j is the weight of the jth key indicator, satisfying ∑ω j =1, is the historical minimum and maximum value of the corresponding j indicator.

[0024] As a preferred embodiment of the present invention, the specific process of importing data into the model to calculate the vibration amplitude is:

[0025] Substitute the data into the ARIMA model formula: ,in, is the vibration amplitude at time t, is the coefficient of the autoregressive (AR) part, indicating the current data Z t With past data Z t−1 , Z t−2 , the linear relationship between is the coefficient of the moving average (MA) part, which represents the difference between the current data Zt and the past error term , , the relationship between... is the noise term at time t.

[0026] As a preferred embodiment of the present invention, the specific process of calculating the performance of the electromechanical equipment within the distinguishing mark to obtain the overall performance score of the equipment in the area is:

[0027] By formula: Output the overall performance score P of the equipment in the area 区域 , where W is the total number of electromechanical equipment in the corresponding area, and j1=1, 2, 3, ..., W; κ is the preset importance weight of the j1th equipment, satisfying , H j1 Provide a health score for the j1th device.

[0028] As a preferred implementation of the present invention, the specific process of grouping the operation data of the device to obtain the sum of the squares of the distances of all data points to the centers of their respective clusters is:

[0029] By formula: Output the sum of the squared distances C of all data points to the centers of their respective clusters, where g is the number of clusters and the number of predefined categories in the clustering. is a data point, the i3th data point in the j2th cluster represents the operating parameters of a single device; is the center value of the j2th cluster, which is the mean of all data points in the cluster, indicating the typical operating status of the equipment, and ; is the deviation between the i3th data point and the cluster center.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. The present invention integrates advanced sensing technology, data processing algorithms and intelligent control methods, and comprehensively covers the entire process from data acquisition to abnormal response. Through the sensing and acquisition module, the system acquires multi-dimensional data including electrical, mechanical, temperature, pressure, environment, etc. in real time, and uses the data transmission module to quickly transmit the information to the intelligent analysis module and the industrial Internet platform. The intelligent analysis module generates standardized data with unique identification through equipment type classification, data attribute archiving, and geographic area binding, and realizes accurate anomaly detection and processing through the calculation of historical mean and standard deviation. Time alignment and interpolation technology ensure the synchronization of data analysis, and correlation analysis further reveals the potential impact relationship between devices.

[0032] 2. The fault probability model based on logistic regression of the present invention can not only accurately predict the failure probability of equipment, but also intuitively display the relationship between equipment through a causal network diagram, providing data support for operation and maintenance decisions. The system can identify potential risks in advance and make maintenance recommendations through a partitioned evaluation of health scores, combined with trend prediction based on time series models (such as ARIMA), thereby significantly reducing the failure rate. At the same time, the system achieves overall performance optimization management of large-scale equipment through regional performance scoring, cluster analysis, and calculation of regional anomaly coefficients, and timely generates regional anomaly instructions and takes targeted measures.

[0033] 3. The present invention combines the abnormal vibration amplitude instruction, the abnormal regional performance instruction and the related fault instruction through the control module, locks the specific equipment and regional problems through multi-level data analysis, and dynamically adjusts the equipment operation status (such as continuing to run, stopping operation, reducing power). The user interface module provides clear and intuitive visualization results, including cause-and-effect network diagrams and health status trend diagrams, to support users to make quick decisions and precise control. Intelligent monitoring and management of the entire life cycle of electromechanical equipment is realized. Its main advantages include: strong real-time performance, accurate fault prediction, comprehensive analysis, and fast response. By optimizing the operating status of the equipment, the system can significantly improve the safety and stability of the equipment, reduce unplanned downtime and maintenance costs, and help enterprises achieve efficient, intelligent and refined management of the production process, while providing a reliable technical foundation for building an intelligent manufacturing environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0035] Figure 1 It is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. 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 are within the scope of protection of the present invention.

[0037] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0038] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0039] See also Figure 1 As shown, an intelligent electromechanical monitoring system based on the industrial Internet includes a sensing and acquisition module, a data transmission module, an intelligent analysis module, a user interface module and a control module;

[0040] The sensing and acquisition module is used to collect electromechanical monitoring data connected to the industrial Internet. The electromechanical monitoring data includes electrical data, mechanical data, temperature data, pressure fluid data, environmental data and status data;

[0041] The data transmission module is responsible for transmitting the data collected by the sensing and acquisition module to the industrial Internet platform and intelligent analysis module;

[0042] The intelligent analysis module is responsible for distinguishing the collected data:

[0043] Step 1: Classify by equipment type. Perform a preliminary classification of the data according to the equipment category (such as motors, pumps, fans, and compressors, etc.), and mark the corresponding equipment as: AQ; Q = 1, 2, ..., ∞-1, ∞;

[0044] Step 2: Classify data by attributes and archive data of similar attributes, for example: electrical data: voltage, current, power; mechanical data: vibration amplitude, speed, load; environmental data: temperature, humidity, pressure; the corresponding data attributes are marked as: SQ;

[0045] Step 3: Divide by electromechanical location area; perform regional identification binding. The data collected by each device is bound to a specific geographical location or regional number (such as workshop area A, area B, area C) during installation. The specific geographical location is distinguished and marked as: DQ;

[0046] Then the distinguishing marks of step 1, step 2 and step 3 are integrated to obtain the number DQAQSQ;

[0047] To store the marked data, each electromechanical device is assigned a unique identification code (ID) by mapping the marked SQ with the device identification code, and the data is archived by device ID when stored, and a timestamp is set for each piece of data to record the specific time of data collection;

[0048] The intelligent analysis module is also used to analyze the differentiated data, specifically:

[0049] Standardize the data. The specific calculation process is as follows: calculate the mean and standard deviation of historical data; if a certain electromechanical device collects X1, X2, X3, ..., X at n time points n The value of; then through the mean formula: Output historical mean , through the standard deviation formula: Output historical standard deviation , i=1, 2, ..., n; then standardize each piece of data, through the formula: Get the standardized data points , compare the standardized data with the preset normal range (such as >3 or <-3). If the standardized data exceeds the normal range, further processing is required and marked as abnormal data; abnormal data points are smoothed by the mean of the sliding window: Output the smoothed value at the current time point t , k is the window size (usually 2-5), i1=2, 3, 4, 5;

[0050] It should be noted that: since data from different devices may be collected at different frequencies, the data needs to be time-aligned to ensure time synchronization during analysis. The specific calculation process is: determine the time interval, and uniformly set the time interval according to application requirements. If there is no data at a certain time point, use the data of adjacent time points for linear interpolation, through the formula: Output data interpolation , are the data values ​​at adjacent time points, are the timestamps of adjacent time points, is the time point that needs to be interpolated; and the control module interpolates the obtained data Supplement the missing data points;

[0051] Then, the correlation analysis is performed on the standardized data. The specific calculation process is as follows: obtain the data of two electromechanical devices; if the two devices obtained are Y and U, collect n1 data points of the same type respectively, and the data sequence of device Y is: Y1, Y2, Y3, ..., Y n1 , the data sequence of device U: U1, U2, U3, ..., U n1 ; Then calculate the data mean of the two devices: by formula: Output , is the data mean of device Y, is the data mean of device U; then calculate the covariance and variance: the covariance is calculated by the formula: Get; variance through the formula: Finally, substitute the calculated covariance and variance into the relevant calculation formula to get the correlation coefficient. The process is: Get the correlation coefficient ,and ∈[−1,1], if >0, then it is positively correlated. <0, then it is positively correlated. =0, there is no correlation;

[0052] Then build the fault probability model; it should be noted that the fault probability model analyzes whether the abnormality of a certain device is related to the parameter changes of other devices; the fault probability model is calculated using logistic regression: Assume that the fault U (1 indicates fault, 0 indicates normal) is affected by the devices Y1, Y2, Y3, ..., Y n1 Parameter influence, the model formula is: Probability of output device failure , Y1, Y2, Y3, ..., Y n1 are monitoring parameters that affect the fault, β0, β1, ..., β n is the model parameter (weight). The model parameter is trained by using historical data to transform the historical fault state U and the corresponding equipment parameters Y1, Y2, Y3, ..., Y n1 Input the model, and the model outputs the optimized β0, β1, ..., β n ;

[0053] The real-time data Y1, Y2, Y3, ..., Y n1 Input model to get failure probability ,like >Preset thresholds and generate relevant fault instructions; draw a causal network diagram based on the calculation results, marking the direct or indirect impact relationship between devices.

[0054] Then calculate the health score of each electromechanical device; obtain the key health indicators of each electromechanical device for calculation. If the health score of a device is calculated by three key indicators Z1 (temperature), Z2 (vibration amplitude), and Z3 (current), obtain the historical minimum and maximum values ​​of each indicator; through the formula: Output health score H, Z j is the value of the jth key indicator, ω j is the weight of the jth key indicator, satisfying ∑ω j =1, is the historical minimum and maximum value of the corresponding j indicator; the calculated health score H is matched with the preset health score interval of the corresponding electromechanical equipment, and the preset health score intervals include health score interval 1, health score interval 2, health score interval 3, health score interval 4 and health score interval 5; if the health score H belongs to health score interval 1, it means that the health status of the corresponding electromechanical equipment is at an ideal level; if the health score H belongs to health score interval 2, it means that the health status of the corresponding electromechanical equipment is relatively stable; if the health score H belongs to health score interval 3, it means that the health status of the corresponding electromechanical equipment needs attention; if the health score H belongs to health score interval 4, it means that there is a potential risk in the health status of the corresponding electromechanical equipment; if the health score H belongs to health score interval 5, it means that the health status of the corresponding electromechanical equipment needs timely maintenance;

[0055] Obtain the health status of the corresponding electromechanical equipment for trend analysis, and predict the future operating status of the equipment based on a time series model (such as the ARIMA model); ARIMA model formula: ,in, is the monitoring data at time t (such as vibration amplitude), is the coefficient of the autoregressive (AR) part, indicating the current data Z t With past data Z t−1 , Z t−2 , the linear relationship between is the coefficient of the moving average (MA) part, which represents the difference between the current data Zt and the past error term , , the relationship between... is the noise term at time t;

[0056] If the calculated vibration amplitude Zt is within the normal range, it indicates that the vibration amplitude of the equipment will remain normal for a period of time in the future. Otherwise, it is judged as abnormal vibration amplitude and a vibration amplitude abnormal instruction is generated. If it is found after multiple calculations that Zt shows a continuous upward trend, it is judged as abnormal vibration amplitude and a vibration amplitude abnormal instruction is generated.

[0057] Combine the health score results and trend forecasts to find the following: If there is a potential risk in the health condition or the health condition requires attention and the trend forecast is abnormal (such as the vibration amplitude continues to increase), early maintenance is recommended; if the health condition is relatively stable or at an ideal level and the trend forecast is stable, the equipment does not need maintenance;

[0058] Then calculate the performance of the electromechanical equipment within the distinguishing mark: by formula: Output the overall performance score P of the equipment in the area 区域 , where W is the total number of electromechanical equipment in the corresponding area, and j1=1, 2, 3, ..., W; κ is the preset importance weight of the j1th equipment, satisfying , H j1 Score the health of the j1th device;

[0059] Then, cluster analysis is used to group the equipment's operating data, and the deviation (sum of squares) between the data point and the center of its cluster is calculated; the larger the deviation value, the less concentrated the data distribution is, and the more abnormal it is, specifically:

[0060] By formula: Output the sum of the squared distances C of all data points to the center of their respective clusters, where g is the number of clusters and the number of predefined categories in the clustering (such as normal and abnormal, or more). is a data point, the i3th data point in the j2th cluster, representing the operating parameters of a single device (such as temperature, vibration amplitude, etc.); is the center value of the j2th cluster, which is the mean of all data points in the cluster, indicating the typical operating status of the equipment, and ; is the deviation between the i3th data point and the cluster center;

[0061] The overall performance score of the equipment in the area is P 区域 The regional anomaly coefficient QYCZ is obtained by weighted calculation of the square sum of the distances C from all data points to the centers of their respective clusters: Output area anomaly coefficient QYCZ, DC is the preset correction factor;

[0062] The regional anomaly coefficient QYCZ is compared with the preset corresponding anomaly detection threshold. If the overall performance score P 区域 If the performance is lower than the preset corresponding performance threshold, it is judged as regional performance abnormality and a regional performance abnormality instruction is generated;

[0063] The data transmission module is also used to obtain vibration amplitude abnormality instructions and regional performance abnormality instructions and related fault instructions and send them to the control module;

[0064] The control module receives the vibration amplitude abnormality instruction, the regional performance abnormality instruction and the related fault instruction, and sends the regional performance abnormality instruction to distinguish the marked area DQ; then the received vibration amplitude abnormality instruction is used to lock the electromechanical equipment in the corresponding area, and obtain the unique identification code corresponding to the electromechanical equipment; then the related fault instruction is used to lock the unique identification code of the related electromechanical equipment in the area; then a causal network diagram is drawn for the electromechanical equipment with correlation according to the threshold of the abnormal electromechanical equipment, and the direct or indirect influence relationship between the equipment is marked; then the drawn causal network diagram is sent to the user interface module for display; the user controls the working status of the corresponding electromechanical equipment on the user interface module display; the working status includes continuing to run, stopping running and reducing the running power, etc.

[0065] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent electromechanical monitoring system based on the industrial Internet, comprising an intelligent analysis module and a control module; characterized in that: The intelligent analysis module is responsible for distinguishing the collected data to obtain the numbers DQAQSQ and analyzing the distinguished data, specifically: Standardize the data to obtain data interpolation, and then interpolate the obtained data to supplement the corresponding missing data points; perform correlation analysis on the standardized data to obtain the correlation coefficient, and build a fault probability model to analyze whether a certain equipment abnormality is related to changes in other equipment parameters, and obtain the probability of equipment failure; Then, the health score of each electromechanical equipment is calculated to obtain a health score, and the health status of the corresponding electromechanical equipment is analyzed for trend. The future operation status of the equipment is predicted based on the time series model, and the data is imported into the model to calculate the vibration amplitude; Then combine the health score results with the trend forecast results. When there is a potential risk in the health condition or the health condition needs attention and the trend forecast is abnormal, maintenance is performed in advance. The performance of the electromechanical equipment within the distinguishing mark is calculated to obtain the overall performance score of the equipment in the area. The operation data of the equipment is grouped through cluster analysis to obtain the sum of the squares of the distances from all data points to the centers of their respective clusters. The overall performance score of the equipment in the area and the sum of the squares of the distances from all data points to the centers of their respective clusters are weighted to obtain the regional anomaly coefficient. The control module is used to receive vibration amplitude abnormal instructions, regional performance abnormal instructions and related fault instructions, obtain the area where the distinguishing mark is issued through the regional performance abnormal instruction, and lock the electromechanical equipment in the corresponding marked area by receiving the vibration amplitude abnormal instruction, and obtain the unique identification code corresponding to the electromechanical equipment; then lock the unique identification code of the relevant electromechanical equipment in the area by the relevant fault instruction; draw a causal network diagram based on the electromechanical equipment with correlation according to the threshold value of abnormal electromechanical equipment.

2. According to the intelligent electromechanical monitoring system based on industrial Internet according to claim 1, it is characterized in that: It also includes a sensing and acquisition module, a data transmission module and a user interface module; The sensing and acquisition module is used to collect electromechanical monitoring data connected to the industrial Internet, and the electromechanical monitoring data includes electrical data, mechanical data, temperature data, pressure fluid data, environmental data and status data; The data transmission module is responsible for transmitting the data collected by the sensing and acquisition module to the industrial Internet platform and the intelligent analysis module; it is also used to obtain vibration amplitude abnormality instructions and regional performance abnormality instructions and related fault instructions and send them to the control module; The user interface module is used to receive and display a causal network diagram.

3. According to the intelligent electromechanical monitoring system based on industrial Internet as claimed in claim 1, it is characterized in that: The specific process of distinguishing the collected data to obtain the number DQAQSQ is as follows: Step 1: Classify by device type. Perform a preliminary classification of the data based on the device category and mark the corresponding devices as: AQ; Step 2: Classify data attributes and archive data of the same attributes, and divide them into electrical data, mechanical data, and environmental data; mark the corresponding data attributes as: SQ; Step 3: Divide according to the electromechanical location area; bind the regional identification. The data collected by each device is bound to a specific geographical location or regional number during installation. The specific geographical location is distinguished and marked as: DQ; the distinguishing marks of steps 1, 2 and 3 are integrated to obtain the number DQAQSQ.

4. According to the intelligent electromechanical monitoring system based on industrial Internet as claimed in claim 1, it is characterized in that: The specific process of performing standardization processing on the data to obtain data interpolation is as follows: Calculate the mean and standard deviation of historical data; if a certain electromechanical device collects X1, X2, X3, ..., X at n time points n The value of; then through the mean formula: Output historical mean , through the standard deviation formula: Output historical standard deviation , and then standardize each piece of data through the formula: Get the standardized data points , compare the standardized data with the preset normal range. If the standardized data exceeds the normal range, it will be further processed and marked as abnormal data; the abnormal data points are smoothed by the mean of the sliding window: Output the smoothed value at the current time point t , k is the window size; time alignment of data, time synchronization during analysis, specific calculation process: determine the time interval, set the time interval uniformly according to application requirements, if there is no data at a certain time point, use the data of adjacent time points for linear interpolation, through the formula: Output data interpolation , are the data values ​​at adjacent time points, are the timestamps of adjacent time points, is the time point that needs to be interpolated.

5. According to the industrial Internet-based intelligent electromechanical monitoring system of claim 1, it is characterized in that: The specific process of performing correlation analysis on the standardized data to obtain the correlation coefficient is as follows: Get the data of two electromechanical devices; if the two devices are Y and U, collect n1 data points of the same type respectively, the data sequence of device Y is: Y1, Y2, Y3, ..., Y n1 , the data sequence of device U: U1, U2, U3, ..., U n1 ; Calculate the data mean of the two devices: by formula: Output , is the data mean of device Y, is the data mean of device U; then calculate the covariance and variance: the covariance is calculated by the formula: Get; variance through the formula: Get; Substitute the calculated covariance and variance into the relevant calculation formula to get the correlation coefficient. The process is: Get the correlation coefficient .

6. The intelligent electromechanical monitoring system based on industrial Internet according to claim 1 is characterized in that: The specific calculation process of the probability of failure of the equipment is as follows: The fault probability model is calculated using logistic regression: If the fault U is affected by equipment Y1, Y2, Y3, ..., Y n1 Parameter influence, the model formula is: Probability of output device failure , Y1, Y2, Y3, ..., Y n1 are monitoring parameters that affect the fault, β0, β1, ..., β n The model parameters are obtained by using historical data for model training, and the historical fault state U and the corresponding equipment parameters Y1, Y2, Y3, ..., Y n1 Input the model, and the model outputs the optimized β0, β1, ..., β n ; The real-time data Y1, Y2, Y3, ..., Y n1 Input model to get failure probability .

7. The intelligent electromechanical monitoring system based on industrial Internet according to claim 1 is characterized in that: The specific process of calculating the health score of each electromechanical device is as follows: obtaining the key health indicators of each electromechanical device for calculation. If the health score of a device is calculated by three key indicators: temperature Z1, vibration amplitude Z2, and current Z3, then obtaining the historical minimum and maximum values ​​of each indicator; by formula: Output health score H; Z j is the value of the jth key indicator, ω j is the weight of the jth key indicator, satisfying ∑ω j =1, is the historical minimum and maximum value of the corresponding j indicator.

8. The intelligent electromechanical monitoring system based on industrial Internet according to claim 1 is characterized in that: The specific process of importing data into the model to calculate the vibration amplitude is: Substitute the data into the ARIMA model formula: ,in, is the vibration amplitude at time t, is the coefficient of the autoregressive (AR) part, indicating the current data Z t With past data Z t−1 , Z t−2 , the linear relationship between is the coefficient of the moving average (MA) part, which represents the difference between the current data Zt and the past error term , , the relationship between... is the noise term at time t.

9. The intelligent electromechanical monitoring system based on industrial Internet according to claim 1 is characterized in that: The specific process of calculating the performance of the electromechanical equipment within the distinguishing mark to obtain the overall performance score of the equipment in the area is: By formula: Output the overall performance score P of the equipment in the area 区域 , where W is the total number of electromechanical equipment in the corresponding area, and j1=1, 2, 3, ..., W; κ is the preset importance weight of the j1th device, satisfying , H j1 Provide a health score for the j1th device.

10. The intelligent electromechanical monitoring system based on industrial Internet according to claim 1 is characterized in that: The specific process of grouping the operation data of the device to obtain the sum of the squares of the distances from all data points to the centers of their respective clusters is as follows: By formula: Output the sum of the squared distances C of all data points to the centers of their respective clusters, where g is the number of clusters and the number of predefined categories in the clustering. is a data point, the i3th data point in the j2th cluster represents the operating parameters of a single device; is the center value of the j2th cluster, which is the mean of all data points in the cluster, indicating the typical operating status of the equipment, and ; is the deviation between the i3th data point and the cluster center.

Images