CN118013396A - Industrial data management system and method based on edge calculation - Google Patents
Industrial data management system and method based on edge calculation Download PDFInfo
- Publication number
- CN118013396A CN118013396A CN202410185161.6A CN202410185161A CN118013396A CN 118013396 A CN118013396 A CN 118013396A CN 202410185161 A CN202410185161 A CN 202410185161A CN 118013396 A CN118013396 A CN 118013396A
- Authority
- CN
- China
- Prior art keywords
- equipment
- data
- risk
- state
- overhaul
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000013523 data management Methods 0.000 title claims abstract description 28
- 238000004364 calculation method Methods 0.000 title claims abstract description 6
- 230000002159 abnormal effect Effects 0.000 claims abstract description 75
- 238000012423 maintenance Methods 0.000 claims abstract description 46
- 238000012795 verification Methods 0.000 claims abstract description 39
- 238000007726 management method Methods 0.000 claims abstract description 23
- 238000012544 monitoring process Methods 0.000 claims abstract description 12
- 238000005065 mining Methods 0.000 claims abstract description 10
- 238000001514 detection method Methods 0.000 claims description 68
- 230000005856 abnormality Effects 0.000 claims description 47
- 238000007405 data analysis Methods 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 10
- 238000013075 data extraction Methods 0.000 claims description 6
- 230000008439 repair process Effects 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 17
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 239000012080 ambient air Substances 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000012098 association analyses Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an industrial data management system and method based on edge calculation, which relate to the technical field of industrial data management and comprise the following steps: acquiring historical overhaul related data of the target equipment through an equipment overhaul report, and acquiring current state data of the target equipment through a data importing or real-time monitoring mode; analyzing the acquired overhaul related data, and judging whether abnormal conditions exist in the state data of the current target equipment according to the overhaul related data; if the state data of the current target equipment has abnormal conditions, mining the abnormal data of the target equipment, and extracting risk items in related elements of the target equipment according to the abnormal data of the target equipment to generate a risk item set; and verifying the risk item set, judging whether maintenance requirements exist on each equipment related element in the risk item set, and feeding back a verification result to related equipment management staff. The discovery speed and the accuracy of the abnormal condition of the equipment are improved, and the fault risk of the equipment is reduced.
Description
Technical Field
The invention relates to the technical field of industrial data management, in particular to an industrial data management system and method based on edge calculation.
Background
The electric power unit equipment is equipment which converts other forms of energy into electric energy by using the equipment, in use, according to the use frequency and the operation condition of the equipment, related personnel can make a regular maintenance plan to carry out regular equipment inspection and maintenance work, detailed records are carried out on each maintenance, including maintenance time, content, problems found, measures taken and the like, and maintenance staff generates maintenance reports and carries out archiving management according to the contents.
However, since the overhaul report simply records the overhaul process and result, the overhaul report is not deeply analyzed and summarized, and effective reference cannot be provided for unit equipment management. Meanwhile, when the same equipment is overhauled, overhauling staff changes, and due to different operation skills and experiences of the overhauling staff, data and conclusions related to the same equipment in an overhauling report recorded by archiving can have deviation, and the deviation can have adverse effects on subsequent management and maintenance of the equipment, even cause potential safety hazards, and endanger personal safety of operators.
Accordingly, in order to solve the above-mentioned problems or some of the problems, the present invention provides an industrial data management system and method based on edge computing.
Disclosure of Invention
The present invention is directed to an industrial data management system and method based on edge computing, so as to solve the problems set forth in the background art.
In order to solve the technical problems, the invention provides the following technical scheme: an industrial data management method based on edge calculation comprises the following steps:
s1: determining a supervision target device, acquiring historical overhaul related data of the target device through an equipment overhaul report, and acquiring current state data of the target device through a data importing or real-time monitoring mode;
s2: analyzing the acquired overhaul related data, and judging whether abnormal conditions exist in the state data of the current target equipment according to the overhaul related data;
S3: if the state data of the current target equipment has abnormal conditions, mining the abnormal data of the target equipment, and extracting risk items in related elements of the target equipment according to the abnormal data of the target equipment to generate a risk item set;
s4: and (3) verifying the risk item set generated in the step (S3), judging whether maintenance requirements exist for each equipment related element in the risk item set, and feeding back a verification result to related equipment management staff.
Further, in the step S1, a user selects any device as a target device through an application program, and obtains historical maintenance related data of the target device through a device maintenance report, including device information, maintenance time, maintenance purpose, maintenance content, maintenance personnel and maintenance results;
acquiring current state data of target equipment by a data importing or real-time monitoring mode, wherein the current state data comprise equipment running state data, equipment fault information and equipment running logs;
wherein the device basic information comprises a device name, a model number, a manufacturer, a production date and the like; the device failure information includes failure type, failure code, failure time, etc.
If the electronic data management system exists in the target equipment, automatically acquiring data through the system, otherwise, manually recording and arranging related data through staff, analyzing the related data through subsequent steps after the data acquisition, and timely repairing and maintaining the target equipment so as to ensure the normal operation of the equipment and ensure the production efficiency and safety.
Further, the step S2 includes:
Step S2-1: constructing a device state anomaly type set A= { a 1,a2,…,am }; wherein a 1、a2、…、am represents one of the device state exception types, respectively, and m represents the number of the device state exception types; acquiring equipment maintenance related factors, selecting factors exceeding a preset value g with equipment state abnormality related coefficients as training features of an equipment state abnormality prediction model, and comprising: equipment load, ambient temperature, ambient air pressure, equipment operation sensitivity, equipment power feedback, equipment material wear level; constructing a device state anomaly prediction model according to the following formula, and analyzing a device state anomaly prediction index:
Wherein z represents the device anomaly prediction index; u k represents the regression coefficient of the kth equipment state anomaly prediction model, k= [1, k '], k' represents the number of training features of the equipment state anomaly prediction model, μ is an error factor, and is preset by related personnel; s k represents the training feature of the kth device state anomaly prediction model;
Step S2-2: acquiring an equipment abnormality prediction index as a prediction value, acquiring the probability of an actual equipment abnormality as an actual value, acquiring delta data samples, wherein delta is a constant, and setting by related personnel according to the actual data quantity; data samples with a difference of less than a% between the predicted value and the actual value are defined as high samples, and data samples with a difference of greater than or equal to a% between the predicted value and the actual value are defined as low samples; the data samples comprise equipment overhaul related data and equipment state abnormality type related data;
Constructing an equipment state anomaly prediction model according to the data samples, wherein the method comprises the following steps of:
Step S2-2-1: constructing a Loss function Loss (y, z) by using a mean square error function, selecting a weak learner to minimize the Loss function, initializing the weak learner, further iteratively training the weak learner, and setting the maximum iteration times T; for any data sample i, i e [1, delta ], a negative gradient f ti for any data sample i is calculated according to the following formula:
Wherein, Is the sign of the partial derivative; z i represents the corresponding equipment anomaly prediction index of the data sample i, and y i is the loss function corresponding to z i; z t-1 represents the model of the previous iteration and t represents the number of iterations;
Step S2-2-2: adding the weak learner selected in the step S2-2-1 into the trained model to obtain a new prediction model: namely, a device state abnormality prediction model by which a prediction abnormality index of a current device state can be output;
Wherein z t represents a new equipment state anomaly prediction model obtained by the t-th round of iteration; j represents a leaf area; h is used for combining with H tj and represents the decision tree fitting function of the round; h tj represents the best fit value, calculated according to the following manner:
Fitting a regression tree through the negative gradient f ti in the step S2-2-1, and setting the maximum depth of the regression tree; the leaf node area corresponding to the negative gradient f ti is L tj, j epsilon [1, delta ]; the best fit value h tj is calculated according to the following formula:
wherein arg min () represents the value of z i that minimizes the summation result; c is a constant, added by the relevant personnel;
Step S2-3: according to the analysis result of the equipment state abnormality prediction model on the target equipment, a prediction abnormality index z ' of the current equipment state is obtained, whether the equipment has risk hidden danger is judged according to the prediction abnormality index z ' of the current equipment state, if the prediction abnormality index z ' of the current equipment state exceeds a preset threshold gamma in a database, the current equipment has abnormal conditions, relevant personnel are required to take measures in time to repair or replace the equipment, and supervision and management of the equipment are enhanced. If the predicted abnormality index z' of the current equipment state does not exceed the preset threshold value gamma in the database, the current equipment is indicated to have no abnormal condition.
According to the operation data of the equipment, whether the operation condition of the equipment is overused, the peak load is too high or overload operation exists or not, and whether the operation life is too long or not is judged; by analyzing the overhaul records, whether the equipment frequently fails or not and whether the maintenance cost exceeds the budget or not can be judged, so that whether the equipment has risk hidden trouble or not is judged; through analysis of fault records, the conditions of fault types, fault frequencies, fault influence ranges and the like of equipment can be found out, and the fault processing mode is evaluated, so that whether the equipment has risk hidden danger or not is judged;
Further, in the step S3, if the state data of the current target device has an abnormal condition, the abnormal data of the target device is mined, which includes the following steps:
Step S3-1: acquiring state data of target equipment and related elements of the target equipment, which are judged to have abnormal conditions, and preprocessing the acquired data, wherein the abnormal data comprises overhaul related data of the target equipment and operation state data of the target equipment; establishing an abnormal data set D= { D 1,d2,...,dn }, wherein D 1,d2,...,dn respectively represents the 1 st, 2 nd, n th and n th groups of abnormal data in the abnormal data set D of the target device;
Step S3-2: and (3) self-defining a threshold value of minimum support and confidence, wherein the minimum support is used for measuring the frequency of occurrence of one item set in all transactions, the confidence is used for measuring the reliability of the association rule, and the selection of the threshold value and the confidence is critical to finding the meaningful association rule. The method comprises the steps of mining frequent item sets by applying an Apriori algorithm to a preprocessed abnormal data set D of target equipment, including the steps of generating candidate item sets, calculating support degree, deleting item sets with low support degree and the like until the next frequent item set cannot be generated;
Mining to obtain an abnormal frequent item set P, wherein the P comprises { P 1,p2,...,px }, wherein P 1,p2,...,px respectively represents the 1 st, 2 nd, the third and the x th frequent items in the frequent item set P;
Step S3-3: generating an association rule based on the abnormal frequent item set P, determining equipment elements with risks according to the generated association rule, and generating a target equipment risk item set P '= { P' 1,p′2,...,p′u }; wherein P '1,p′2,...,p′u represents the 1 st, 2 nd, u th risk items in the target device risk item set P', respectively; the corresponding equipment elements of the risk items may have a certain relevance with the abnormal data, and can be used as potential risk factors for subsequent risk assessment and management.
Further, the step S4 includes:
step S4-1: verifying each risk item in the risk item set according to the target equipment risk item set generated in the step S3, and calculating the risk probability of any risk item p' 0 of the target equipment according to the following formula:
G0＝e1×(ω1/v1)+e2×(ω2-v1)；
Wherein G 0 represents the risk probability of any risk item p' 0 of the target device; e 1 > 0, representing the influence factor of the normal operation item; omega 1 represents the number of regular operational items for which there is a risk item p' 0 in the last service data record for the target device; v 1 denotes the number of necessary operations of the risk item p' 0; e 2 < 0, which represents the influence factors of other operation items; omega 2 represents the number of other operational items in the last service data record of the target device, excluding s 1;
Step S4-2: setting a probability threshold as G', if the risk probability G 0 of any risk item exceeds a preset probability threshold, indicating that the verification result of the risk item is true, wherein the equipment element corresponding to the risk item has a maintenance requirement; and marking the risk items of the target equipment and the corresponding equipment elements thereof, sequencing the risk items according to the value of the risk probability from large to small to generate a feedback data set, and outputting the feedback data set to a related equipment manager terminal to assist staff in managing and maintaining the target equipment so as to ensure the normal operation and safety performance of the equipment.
An industrial data management system based on edge computing, the system comprising: the device comprises a device detection data acquisition module, a detection data analysis module, a risk item detection module and a risk item verification module;
the equipment detection data acquisition module is used for acquiring historical overhaul data and current state data of the equipment according to an overhaul report of the target equipment;
The detection data analysis module is used for analyzing the current state of the target equipment according to the equipment detection data and judging whether the target equipment has risk factors or not;
The risk item detection module is used for detecting the risk elements according to the judgment result of the detection data analysis module to generate a risk item set;
the risk item verification module is used for verifying each risk in the risk item set and feeding back a verification result to related personnel;
The output end of the equipment detection data acquisition module is connected with the input end of the detection data analysis module, the output end of the detection data analysis module is connected with the input end of the risk item detection module, and the output end of the risk item detection module is connected with the input end of the risk item verification module.
Further, the equipment detection data acquisition module comprises a historical overhaul data acquisition unit and an equipment state data acquisition unit;
The historical overhaul data collection unit is used for extracting historical overhaul data from overhaul reports or maintenance records of the equipment, wherein the historical overhaul data comprises the maintenance history of the equipment, the maintenance records, the information of replacement parts, the operation records of maintenance personnel and the like, and has important significance for analyzing the health condition of the equipment, predicting the service life of the equipment, making a maintenance plan and the like. Information about historical status of the equipment and maintenance repairs is extracted from the structured or unstructured service reports.
The equipment state data acquisition unit is used for acquiring current state data of equipment in real time, including the current running state of the equipment, performance parameters and sensor data, and the equipment state data plays an important role in real-time monitoring of running conditions, fault early warning, performance optimization and the like of the equipment.
Further, the detection data analysis module comprises a detection data extraction unit and a data abnormality discrimination unit;
The detection data extraction unit is used for extracting relevant characteristics from the equipment detection data acquisition module, converting and dimension-reducing the extracted characteristic data so as to better characterize the equipment state and prepare for subsequent analysis modeling;
The data abnormality judging unit is used for carrying out abnormality detection and risk judgment on the extracted equipment characteristic data, timely finding possible risk factors and providing important data support and decision reference for equipment operation and maintenance.
Further, the risk item detection module comprises an abnormal data detection unit and a risk item set generation unit;
The abnormal data detection unit is used for further identifying and detecting the abnormal data judged by the detection data analysis module; and carrying out association analysis on the detected abnormal data and related equipment information, historical data and the like to determine the source and the influence range of the abnormal data.
The risk item set generating unit is used for judging the risk items contained in the current equipment according to the detection result of the abnormal data detecting unit, and integrating the detected risk items to generate a risk item set. So that the subsequent risk item verification module can verify and feed back more accurately and conveniently.
Further, the risk item verification module comprises a risk item set analysis unit and a verification result feedback unit;
The risk item set analysis unit is used for analyzing and verifying each risk item in the risk item set and confirming the authenticity and the influence degree of the risk item set;
the verification result feedback unit is used for feeding the verification result back to related personnel, and comprises: result summary and report generation, feedback notification, risk management advice, and decision support. By verifying the operation of the result feedback unit, the risk information can be timely transmitted and effectively responded, so that the response speed and decision efficiency of equipment management are improved, and the safety and reliability of equipment operation are ensured.
Compared with the prior art, the invention has the following beneficial effects:
1. According to the invention, the historical overhaul related data in the equipment overhaul report is utilized, and the past maintenance condition of the target equipment can be better known by analyzing the data, so that more sufficient background information and comparison basis are provided for the abnormal condition of the current state data.
2. The invention not only can acquire the equipment state information through the historical overhaul data, but also provides the method for acquiring the current state data of the target equipment through the data import or real-time monitoring, thereby realizing the real-time monitoring of the equipment state, ensuring that the abnormal situation can be discovered and processed more timely, and improving the discovery speed and accuracy of the equipment abnormal situation.
3. The method combines the abnormal data mining technology, can extract potential risk items from the abnormal data of the target equipment and generate the risk item set, and is helpful for more accurately identifying possible problems and potential safety hazards of the equipment; meanwhile, verification is carried out on the generated risk item set, and verification results are fed back to relevant equipment management staff, so that the equipment management staff can know the equipment state in time and take corresponding maintenance measures.
4. The invention is beneficial to timely taking targeted maintenance measures and reducing the equipment fault risk; the real-time monitoring and management of the state of the equipment are enhanced, and the safety and reliability of the operation of the equipment are improved; and more reliable data support and decision basis are provided for equipment management personnel, and management efficiency and decision scientificity are improved.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic block diagram of an industrial data management system based on edge computing according to the present invention;
FIG. 2 is a flow chart of an industrial data management method based on edge computing according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention is further described with reference to fig. 1,2 and embodiments.
Example 1: as shown in fig. 1, the present embodiment provides an industrial data management system based on edge computing, the system including: the device comprises a device detection data acquisition module, a detection data analysis module, a risk item detection module and a risk item verification module;
The equipment detection data acquisition module is used for acquiring historical overhaul data and current state data of the equipment according to an overhaul report of the target equipment;
Further, the equipment detection data acquisition module comprises a historical overhaul data acquisition unit and an equipment state data acquisition unit;
The historical overhaul data collection unit is used for extracting historical overhaul data from overhaul reports or maintenance records of the equipment, wherein the historical overhaul data comprises the maintenance history of the equipment, the maintenance records, the information of replacement parts, the operation records of maintenance personnel and the like, and has important significance for analyzing the health condition of the equipment, predicting the service life of the equipment, making a maintenance plan and the like. Information about equipment history status and maintenance repairs is extracted from structured or unstructured service reports, and the extracted data is cleaned, formatted, and integrated for subsequent data analysis and mining.
The equipment state data acquisition unit is used for acquiring current state data of the equipment in real time, including the current running state, performance parameters and sensor data of the equipment, and the equipment state data plays an important role in real-time monitoring of the running condition, fault early warning, performance optimization and the like of the equipment, and the extracted historical overhaul data is stored in a database or a data warehouse for subsequent access and analysis.
The detection data analysis module is used for analyzing the current state of the target equipment according to the equipment detection data and judging whether the target equipment has risk factors or not; the detection data analysis module comprises a detection data extraction unit and a data abnormality discrimination unit;
The detection data extraction unit is used for extracting relevant features from the equipment detection data acquisition module, converting and dimension-reducing the extracted feature data so as to better characterize the equipment state and prepare for subsequent analysis modeling;
The data abnormality judging unit is used for carrying out abnormality detection and risk judgment on the extracted equipment characteristic data, timely finding possible risk factors and providing important data support and decision reference for equipment operation and maintenance.
The risk item detection module is used for detecting the risk elements according to the judgment result of the detection data analysis module to generate a risk item set; the risk item detection module comprises an abnormal data detection unit and a risk item set generation unit;
The abnormal data detection unit is used for further identifying and detecting the abnormal data judged by the detection data analysis module; and carrying out association analysis on the detected abnormal data and related equipment information, historical data and the like to determine the source and the influence range of the abnormal data.
The risk item set generating unit is used for judging the risk items contained in the current equipment according to the detection result of the abnormal data detecting unit, and integrating the risk items obtained by detection to generate a risk item set. So that the subsequent risk item verification module can verify and feed back more accurately and conveniently.
The risk item verification module is used for verifying each risk in the risk item set and feeding back a verification result to related personnel; the risk item verification module comprises a risk item set analysis unit and a verification result feedback unit;
The risk item set analysis unit is used for analyzing and verifying each risk item in the risk item set and confirming the authenticity and the influence degree of the risk item set;
the verification result feedback unit is used for feeding back the verification result and corresponding advice to related personnel, and comprises the following steps: summarizing the verification results, generating a corresponding verification report, and clearly listing the verification result, the possible influence range and the suggested processing measures of each risk item; sending verification result notification to related personnel, including equipment operation staff, management staff or other related responsible persons, so that the equipment operation staff, management staff or other related responsible persons can know equipment risk conditions in time and take corresponding actions; providing corresponding risk management advice, including possible countermeasures, emergency treatment plans, or preventive maintenance measures, to reduce the likelihood of risk occurrence or mitigate the effects thereof, based on the verification result; decision support is provided for related personnel, so that the related personnel are helped to make reasonable equipment management and maintenance strategies, and safe and stable operation of equipment is ensured.
Example 2: as shown in fig. 2, the present embodiment provides an industrial data management method based on edge computing, which is implemented based on an industrial data management system based on edge computing in the embodiment, and specifically includes the following steps:
s1: determining a supervision target device, acquiring historical overhaul related data of the target device through an equipment overhaul report, and acquiring current state data of the target device through a data importing or real-time monitoring mode;
the method comprises the steps that a user selects any equipment as target equipment through an application program, and historical overhaul related data of the target equipment are acquired through an equipment overhaul report, wherein the historical overhaul related data comprise equipment information, overhaul time, overhaul purposes, overhaul contents, overhaul personnel and overhaul results;
acquiring current state data of target equipment by a data importing or real-time monitoring mode, wherein the current state data comprise equipment running state data, equipment fault information and equipment running logs;
wherein the device basic information comprises a device name, a model number, a manufacturer, a production date and the like; the device failure information includes failure type, failure code, failure time, etc.
If the electronic data management system exists in the target equipment, automatically acquiring data through the system, otherwise, manually recording and arranging related data through staff, analyzing the related data through subsequent steps after the data acquisition, and timely repairing and maintaining the target equipment so as to ensure the normal operation of the equipment and ensure the production efficiency and safety.
Because the condition that overhaul staff changes when overhaul work is carried out on the same equipment, the operation difference and the data recording mode of the overhaul staff are possibly different, and then the acquisition mode and the recording standard of the same data of the equipment are influenced by personal behaviors, deviation exists in an overhaul report of archiving records on overhaul data and overhaul conclusions of the same equipment, the deviation can mislead the judgment and overhaul management of the equipment state, and the accident occurrence risk of the equipment is increased.
S2: analyzing the acquired overhaul related data, and judging whether abnormal conditions exist in the state data of the current target equipment according to the overhaul related data;
Step S2-1: constructing a device state anomaly type set A= { a 1,a2,…,am }; wherein a 1、a2、…、am represents one of the device state exception types, respectively, and m represents the number of the device state exception types; acquiring equipment maintenance related factors, selecting factors exceeding a preset value g with equipment state abnormality related coefficients as training features of an equipment state abnormality prediction model, and comprising: equipment load, ambient temperature, ambient air pressure, equipment operation sensitivity, equipment power feedback, equipment material wear level; constructing a device state anomaly prediction model according to the following formula, and analyzing a device state anomaly prediction index:
z＝u1s1+u2s2+u3s3+u4s4+u5s5+u6s6+μ;
Wherein z represents the device anomaly prediction index; u 1、u2、u3、u4、u5、u6 is the regression coefficient of the equipment state abnormality prediction model, μ is an error factor, and is preset by related personnel; s 1、s2、s3、s4、s5、s6 respectively represents equipment load, ambient temperature, ambient air pressure, equipment operation sensitivity, equipment power consumption feedback and equipment material abrasion degree;
Step S2-2: acquiring an equipment abnormality prediction index as a prediction value, acquiring the probability of an actual equipment abnormality as an actual value, acquiring delta data samples, wherein delta is a constant, and setting by related personnel according to the actual data quantity; data samples with a difference of less than a% between the predicted value and the actual value are defined as high samples, and data samples with a difference of greater than or equal to a% between the predicted value and the actual value are defined as low samples; the data samples comprise equipment overhaul related data and equipment state abnormality type related data;
Constructing an equipment state anomaly prediction model according to the data samples, wherein the method comprises the following steps of:
Step S2-2-1: constructing a Loss function Loss (y, z) by using a mean square error function, selecting a weak learner to minimize the Loss function, initializing the weak learner, further iteratively training the weak learner, and setting the maximum iteration times T; for any data sample i, i e [1, delta ], a negative gradient f ti for any data sample i is calculated according to the following formula:
Wherein, Is the sign of the partial derivative; z i represents the corresponding equipment anomaly prediction index of the data sample i, and y i is the loss function corresponding to z i; z t-1 represents the model of the previous iteration and t represents the number of iterations;
Step S2-2-2: adding the weak learner selected in the step S2-2-1 into the trained model to obtain a new prediction model: namely, a device state abnormality prediction model by which a prediction abnormality index of a current device state can be output;
Wherein z t represents a new equipment state anomaly prediction model obtained by the t-th round of iteration; j represents a leaf area; h is used for combining with H tj and represents the decision tree fitting function of the round; h tj represents the best fit value, calculated according to the following manner:
fitting a regression tree according to the negative gradient f ti in the step S2-2-1, and setting the maximum depth of the regression tree; the leaf node area corresponding to the negative gradient f ti is L tj, j epsilon [1, delta ]; the best fit value h tj is calculated according to the following formula:
wherein arg min () represents the value of z i that minimizes the summation result; c is a constant, added by the relevant personnel;
Step S2-3: according to the analysis result of the equipment state abnormality prediction model on the target equipment, a prediction abnormality index z ' of the current equipment state is obtained, whether the equipment has risk hidden danger is judged according to the prediction abnormality index z ' of the current equipment state, if the prediction abnormality index z ' of the current equipment state exceeds a preset threshold gamma in a database, the existence of abnormal conditions or risk hidden danger of the current equipment is indicated, relevant personnel are required to take measures in time to repair or replace the equipment, and supervision and management of the equipment are enhanced. If the predicted abnormality index z' of the current equipment state does not exceed the preset threshold value gamma in the database, the current equipment is indicated to have no abnormal condition or risk hidden trouble.
S3: if the state data of the current target equipment has abnormal conditions, mining the abnormal data of the target equipment, and extracting risk items in related elements of the target equipment according to the abnormal data of the target equipment to generate a risk item set; the method specifically comprises the following steps:
Step S3-1: acquiring state data of target equipment and related elements of the target equipment, which are judged to have abnormal conditions, preprocessing the acquired data, sorting and preparing the abnormal data and the data of the related elements of the equipment, ensuring that the data format accords with the input requirements of an Apriori algorithm, and converting the data into a representation in the form of a transaction item set, wherein each transaction represents a record, and the item set contains the elements appearing in the transactions. The abnormal data comprise overhaul related data of the target equipment and running state data of the target equipment; establishing an abnormal data set D= { D 1,d2,...,dn }, wherein D 1,d2,...,dn respectively represents the 1 st, 2 nd, n th and n th groups of abnormal data in the abnormal data set D of the target device;
Step S3-2: and (3) self-defining a threshold value of minimum support and confidence, wherein the support is used for measuring the frequency of occurrence of one item set in all transactions, the confidence is used for measuring the reliability of the association rule, and the selection of the threshold value and the confidence is critical to finding the meaningful association rule. The method comprises the steps of performing frequent item set mining on an abnormal data set D of a preprocessed target device by using an existing Apriori algorithm library or an existing FP-growth algorithm, wherein the steps comprise generating candidate item sets, calculating support, deleting item sets with low support and the like until a next frequent item set cannot be generated;
The frequent item set is a set of elements frequently appearing in the data set and is used for mapping the relation between the abnormality of the target equipment and related elements causing the abnormality, and the association and the rule between the abnormality of the target equipment and the related elements causing the abnormality can be further represented by mining the frequent item set.
Mining to obtain an abnormal frequent item set P, wherein the P comprises { P 1,p2,...,px }, wherein P 1,p2,...,px respectively represents the 1 st, 2 nd, the third and the x th frequent items in the frequent item set P;
Step S3-3: generating an association rule based on the abnormal frequent item set P, determining equipment elements with risks according to the generated association rule, and generating a target equipment risk item set P '= { P' 1,p′2,...,p′u }; wherein P '1,p′2,...,p′u represents the 1 st, 2 nd, u th risk items in the target device risk item set P', respectively; the corresponding equipment elements of the risk items may have a certain relevance with the abnormal data, and can be used as potential risk factors for subsequent risk assessment and management.
S4: verifying the risk item set generated in the step S3, judging whether maintenance requirements exist on each equipment related element in the risk item set, and feeding back a verification result to related equipment management staff;
step S4-1: verifying each risk item in the risk item set according to the target equipment risk item set generated in the step S3, and calculating the risk probability of any risk item p' 0 of the target equipment according to the following formula:
G0＝e1×(ω1/v1)+e2×(ω2-v1)；
Wherein G 0 represents the risk probability of any risk item p' 0 of the target device; e 1 > 0, representing the influence factor of the normal operation item; omega 1 represents the number of regular operational items for which there is a risk item p' 0 in the last service data record for the target device; v 1 denotes the number of necessary operations of the risk item p' 0; e 2 < 0, which represents the influence factors of other operation items; omega 2 represents the number of other operational items in the last service data record of the target device, excluding s 1;
Step S4-2: setting a probability threshold as G ', if the risk probability G 0 of any risk item exceeds a preset probability threshold G', indicating that the verification result of the risk item is true, wherein the equipment element corresponding to the risk item has maintenance requirements; and marking the risk items of the target equipment and the corresponding equipment elements thereof, sequencing the risk items according to the value of the risk probability from large to small to generate a feedback data set, and outputting the feedback data set to a related equipment manager terminal to assist staff in managing and maintaining the target equipment so as to ensure the normal operation and safety performance of the equipment.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An industrial data management method based on edge calculation is characterized in that: the method comprises the following steps:
s1: determining a supervision target device, acquiring historical overhaul related data of the target device through an equipment overhaul report, and acquiring current state data of the target device through a data importing or real-time monitoring mode;
s2: analyzing the acquired overhaul related data, and judging whether abnormal conditions exist in the state data of the current target equipment according to the overhaul related data;
S3: if the state data of the current target equipment has abnormal conditions, mining the abnormal data of the target equipment, and extracting risk items in related elements of the target equipment according to the abnormal data of the target equipment to generate a risk item set;
s4: and (3) verifying the risk item set generated in the step (S3), judging whether maintenance requirements exist for each equipment related element in the risk item set, and feeding back a verification result to related equipment management staff.
2. An industrial data management method based on edge computation according to claim 1, wherein: in the step S1, a user selects any equipment as target equipment through an application program, and historical overhaul related data of the target equipment is acquired through an equipment overhaul report, wherein the historical overhaul related data comprises equipment information, overhaul time, overhaul purposes, overhaul contents, overhaul personnel and overhaul results;
acquiring current state data of target equipment by a data importing or real-time monitoring mode, wherein the current state data comprise equipment running state data, equipment fault information and equipment running logs;
if the target equipment has an electronic data management system, automatically acquiring data through the system, otherwise, manually recording and arranging related data through staff.
3. An industrial data management method based on edge computation according to claim 1, wherein: the step S2 comprises the following steps:
Step S2-1: constructing a device state anomaly type set A= { a 1,a2,…,am }; wherein a 1、a2、…、am represents one of the device state exception types, respectively, and m represents the number of the device state exception types; acquiring equipment maintenance related factors, and selecting factors exceeding a preset value g with the equipment state abnormality related factors as training features of an equipment state abnormality prediction model; constructing a device state anomaly prediction model according to the following formula, and analyzing a device state anomaly prediction index:
Wherein z represents the device anomaly prediction index; u k represents the regression coefficient of the kth equipment state anomaly prediction model, k= [1, k '], k' represents the number of training features of the equipment state anomaly prediction model, μ is an error factor, and is preset by related personnel; s k represents the training feature of the kth device state anomaly prediction model;
step S2-2: acquiring an equipment abnormality prediction index as a predicted value, acquiring the probability of an actual equipment abnormality as an actual value, and acquiring delta data samples; data samples with a difference of less than a% between the predicted value and the actual value are defined as high samples, and data samples with a difference of greater than or equal to a% between the predicted value and the actual value are defined as low samples;
Constructing an equipment state anomaly prediction model according to the data samples, wherein the method comprises the following steps of:
Step S2-2-1: constructing a Loss function Loss (y, z) by using a mean square error function, selecting a weak learner to minimize the Loss function, initializing the weak learner, and iteratively training the weak learner; for any data sample i, i e [1, delta ], a negative gradient f ti for any data sample i is calculated according to the following formula:
Wherein, Is the sign of the partial derivative; z i represents the corresponding equipment anomaly prediction index of the data sample i, and y i is the loss function corresponding to z i; z t-1 represents the model of the previous iteration and t represents the number of iterations;
step S2-2-2: adding the weak learner selected in the step S2-2-1 into the trained model to obtain a new prediction model:
wherein z t represents a new equipment state anomaly prediction model obtained by the t-th round of iteration; j represents a leaf area; h is used for combining with H tj and represents the decision tree fitting function of the round; h tj represents the best fit value;
step S2-3: according to the analysis result of the equipment state abnormality prediction model on the target equipment, a prediction abnormality index z 'of the current equipment state is obtained, whether the equipment has risk hidden danger is judged according to the prediction abnormality index z' of the current equipment state, if the prediction abnormality index z 'of the current equipment state exceeds a preset threshold value gamma in a database, the current equipment has an abnormal condition, and if the prediction abnormality index z' of the current equipment state does not exceed the preset threshold value gamma in the database, the current equipment has no abnormal condition.
4. An industrial data management method based on edge computation according to claim 1, wherein: in the step S3, if the state data of the current target device has an abnormal condition, the abnormal data of the target device is mined, which includes the following steps:
Step S3-1: acquiring state data of target equipment and related elements of the target equipment, which are judged to have abnormal conditions, and preprocessing the acquired data, wherein the abnormal data comprises overhaul related data of the target equipment and operation state data of the target equipment; establishing an abnormal data set D= { D 1,d2,...,dn }, wherein D 1,d2,...,dn respectively represents the 1 st, 2 nd, n th and n th groups of abnormal data in the abnormal data set D of the target device;
step S3-2: the threshold value of the minimum support degree and the confidence degree is set in a self-defined mode, and an Apriori algorithm is applied to the abnormal data set D of the target device after preprocessing to mine frequent item sets; mining to obtain an abnormal frequent item set P, wherein the P comprises { P 1,p2,...,px }, wherein P 1,p2,...,px respectively represents the 1 st, 2 nd, the third and the x th frequent items in the frequent item set P;
Step S3-3: generating an association rule based on the abnormal frequent item set P, determining equipment elements with risks according to the generated association rule, and generating a target equipment risk item set P '= { P' 1,p′2,...,p′u }; wherein P '1,p′2,...,p′u represents the 1 st, 2 nd, u th risk items in the target device risk item set P', respectively.
5. An industrial data management method based on edge computation according to claim 1, wherein: the step S4 comprises the following steps:
step S4-1: verifying each risk item in the risk item set according to the target equipment risk item set generated in the step S3, and calculating the risk probability of any risk item p' 0 of the target equipment according to the following formula:
G0＝e1×(ω1/v1)+e2×(ω2-v1)；
Wherein G 0 represents the risk probability of any risk item p' 0 of the target device; e 1 > 0, representing the influence factor of the normal operation item; omega 1 represents the number of regular operational items for which there is a risk item p' 0 in the last service data record for the target device; v 1 denotes the number of necessary operations of the risk item p' 0; e 2 < 0, which represents the influence factors of other operation items; omega 2 represents the number of other operational items in the last service data record of the target device, excluding s 1;
Step S4-2: setting a probability threshold as G', if the risk probability G 0 of any risk item exceeds a preset probability threshold, indicating that the verification result of the risk item is true, wherein the equipment element corresponding to the risk item has a maintenance requirement; and marking the risk items of the target equipment and the corresponding equipment elements thereof, sequencing the risk items and the corresponding equipment elements according to the value of the risk probability from large to small to generate a feedback data set, and outputting the feedback data set to the related equipment manager terminals.
6. An industrial data management system based on edge computing, characterized in that: the system comprises: the device comprises a device detection data acquisition module, a detection data analysis module, a risk item detection module and a risk item verification module;
the equipment detection data acquisition module is used for acquiring historical overhaul data and current state data of the equipment according to an overhaul report of the target equipment;
The detection data analysis module is used for analyzing the current state of the target equipment according to the equipment detection data and judging whether the target equipment has risk factors or not;
The risk item detection module is used for detecting the risk elements according to the judgment result of the detection data analysis module to generate a risk item set;
the risk item verification module is used for verifying each risk in the risk item set and feeding back a verification result to related personnel;
The output end of the equipment detection data acquisition module is connected with the input end of the detection data analysis module, the output end of the detection data analysis module is connected with the input end of the risk item detection module, and the output end of the risk item detection module is connected with the input end of the risk item verification module.
7. An edge computing-based industrial data management system according to claim 6, wherein: the equipment detection data acquisition module comprises a historical overhaul data acquisition unit and an equipment state data acquisition unit;
The historical overhaul data acquisition unit is used for extracting historical overhaul data from overhaul reports or maintenance records of the equipment and extracting information about the historical state and maintenance repair of the equipment from the structured or unstructured overhaul reports;
The equipment state data acquisition unit is used for acquiring current state data of equipment in real time, including the current running state of the equipment, performance parameters and sensor data.
8. An edge computing-based industrial data management system according to claim 6, wherein: the detection data analysis module comprises a detection data extraction unit and a data abnormality discrimination unit;
The detection data extraction unit is used for extracting relevant features from the equipment detection data acquisition module, and converting and dimension-reducing the extracted feature data;
The data abnormality judging unit is used for carrying out abnormality detection and risk judgment on the extracted equipment characteristic data.
9. An edge computing-based industrial data management system according to claim 6, wherein: the risk item detection module comprises an abnormal data detection unit and a risk item set generation unit;
the abnormal data detection unit is used for further identifying and detecting the abnormal data judged by the detection data analysis module;
The risk item set generating unit is used for judging the risk items contained in the current equipment according to the detection result of the abnormal data detecting unit, and integrating the detected risk items to generate a risk item set.
10. An edge computing-based industrial data management system according to claim 6, wherein: the risk item verification module comprises a risk item set analysis unit and a verification result feedback unit;
The risk item set analysis unit is used for analyzing and verifying each risk item in the risk item set to confirm the authenticity of the risk item; the verification result feedback unit is used for feeding the verification result back to related personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410185161.6A CN118013396A (en) | 2024-02-19 | 2024-02-19 | Industrial data management system and method based on edge calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410185161.6A CN118013396A (en) | 2024-02-19 | 2024-02-19 | Industrial data management system and method based on edge calculation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118013396A true CN118013396A (en) | 2024-05-10 |
Family
ID=90959146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410185161.6A Pending CN118013396A (en) | 2024-02-19 | 2024-02-19 | Industrial data management system and method based on edge calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118013396A (en) |
-
2024
- 2024-02-19 CN CN202410185161.6A patent/CN118013396A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023071217A1 (en) | Multi-working-condition process industrial fault detection and diagnosis method based on deep transfer learning | |
| CN108460144B (en) | Coal equipment fault early warning system and method based on machine learning | |
| CN106888205B (en) | Non-invasive PLC anomaly detection method based on power consumption analysis | |
| US20190310979A1 (en) | Anomaly data priority assessment device and anomaly data priority assessment method | |
| CN110990393B (en) | Big data identification method for abnormal behaviors of industry enterprise data | |
| KR102618023B1 (en) | Failure prediction diagnosis system and method through pattern analysis according to failure type | |
| CN113687972A (en) | Method, device and equipment for processing abnormal data of business system and storage medium | |
| CN114416826A (en) | Equipment point inspection data statistical method, equipment point inspection data analysis method and computer storage medium | |
| CN115524002A (en) | Running state early warning method and system for power plant rotating equipment and storage medium | |
| CN117216801A (en) | Enterprise financial data safety management system and method based on artificial intelligence | |
| CN111831634A (en) | Cloud platform processing system and method based on big data | |
| CN117347869B (en) | Data analysis method and device for energy storage battery management system, electronic equipment and medium | |
| CN117591679A (en) | Intelligent analysis system and method for carbon footprint of building block type product based on knowledge graph | |
| CN117439916A (en) | Network security test evaluation system and method | |
| CN118013396A (en) | Industrial data management system and method based on edge calculation | |
| CN117074852A (en) | Power distribution network electric energy monitoring and early warning management method and system | |
| CN115600695A (en) | Fault diagnosis method of metering equipment | |
| US20230359191A1 (en) | System and method for determining a cause of an operating anomaly of a machine, computer program and electronically readable data storage device | |
| CN112256830B (en) | Equipment investigation information acquisition method and device and equipment fault investigation system | |
| CN117078024B (en) | Dangerous waste quantity change detection system and method based on video analysis | |
| CN112307090B (en) | Detection method and system based on tunnel security data | |
| CN118071156B (en) | Enterprise risk internal control automatic early warning system and method based on big data | |
| CN117217625B (en) | Enterprise safety supervision method and device based on data analysis | |
| CN117909113A (en) | Intelligent analysis method, device, equipment and storage medium for application faults | |
| CN117314386A (en) | Full life cycle management system and method for metering equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |