CN116882713B - Data acquisition method and device of special equipment and electronic equipment - Google Patents
Data acquisition method and device of special equipment and electronic equipment Download PDFInfo
- Publication number
- CN116882713B CN116882713B CN202311146521.3A CN202311146521A CN116882713B CN 116882713 B CN116882713 B CN 116882713B CN 202311146521 A CN202311146521 A CN 202311146521A CN 116882713 B CN116882713 B CN 116882713B
- Authority
- CN
- China
- Prior art keywords
- maintenance
- special equipment
- data
- period
- training
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012423 maintenance Methods 0.000 claims abstract description 263
- 238000012549 training Methods 0.000 claims abstract description 113
- 238000007689 inspection Methods 0.000 claims abstract description 84
- 238000010801 machine learning Methods 0.000 claims abstract description 45
- 238000012360 testing method Methods 0.000 claims abstract description 43
- 238000004519 manufacturing process Methods 0.000 claims abstract description 20
- 238000012795 verification Methods 0.000 claims abstract description 17
- 239000013598 vector Substances 0.000 claims description 58
- 238000013480 data collection Methods 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000015654 memory Effects 0.000 claims description 9
- 238000012797 qualification Methods 0.000 claims description 8
- 238000011990 functional testing Methods 0.000 claims description 7
- 230000000737 periodic effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000009428 plumbing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06314—Calendaring for a resource
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Abstract
The application discloses a data acquisition method, a device and electronic equipment of special equipment, which relate to the technical field of special equipment inspection, and are characterized in that loss training data and availability training data are collected in a test environment, a machine learning model for evaluating the loss rationality of maintenance parts of the special equipment and a Bayesian network model for analyzing the availability probability of the special equipment are trained, a maintenance record data set is arranged in a production environment and comprises maintenance record data of all special equipment to be used, and in each inspection period of the production environment, the inspection data of the special equipment to be used are collected; generating a period update coefficient at the end of each test period; updating the verification period based on the period update coefficient; the effect of carrying out self-adaptive inspection on accident risks of faults of special equipment is achieved.
Description
Technical Field
The application relates to the technical field of special equipment inspection, in particular to a data acquisition method and device of special equipment and electronic equipment.
Background
Specialty equipment refers to machinery or devices for specific special purposes that are designed, manufactured, used, and managed to comply with specific safety specifications and regulations, such as hoisting machinery, elevators, boilers, pressure pipes, recreational facilities, farm vehicles, pressure vessels, and plumbing, etc.;
safety is an important concern for specialty devices due to their use and the specificity of the operating environment; in order to ensure safe operation of special equipment and prevent accidents, the special equipment is usually required to be maintained regularly; the maintenance personnel are personnel for maintaining and protecting equipment, and are generally affiliated to special equipment production, special equipment sales or third party institutions, so that the special equipment inspection specificity and enthusiasm are deficient; therefore, a professional inspection institution is also generally required to periodically inspect the special equipment;
at present, the inspection of special equipment by an inspection mechanism is often carried out in a fixed period or according to the needs, and detailed data of maintenance and repair of the special equipment by maintenance staff cannot be fully utilized, so that the use risk and the overall feasibility of the special equipment which is put into use cannot be comprehensively mastered, the inspection efficiency is lower when the risk is low, and accident potential cannot be found in time when the risk is high;
the Chinese patent with the application publication number of CN110909895A discloses an early warning method and system based on a special equipment history periodic inspection report, and the method comprises the following steps: step S1: collecting original record data in a historical inspection report of the special equipment through an inspection template; step S2: preprocessing the obtained original record data, constructing a training data set, a verification data set and a test data set width table, and obtaining analysis data; step S3: establishing a special equipment periodic inspection report fault early warning model by adopting a supervised decision tree-based classification algorithm; step S4: measuring the accuracy of the prediction result of the early warning model by utilizing various evaluation indexes; but the application fails to further adaptively adjust the inspection cycle based on the fault condition;
therefore, the application provides a data acquisition method and device for special equipment and electronic equipment.
Disclosure of Invention
The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a data acquisition method and device for special equipment and electronic equipment, and achieves the effect of carrying out self-adaptive detection on accident risk of special equipment faults.
In order to achieve the above object, in a first aspect, the present application provides a data acquisition method for a special device, including the following steps:
collecting loss training data and availability training data in a test environment, and training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment; presetting a test period;
setting an overhaul record data set in a production environment, wherein the overhaul record data set comprises overhaul record data of all special equipment to be used, and updating corresponding overhaul record data after maintenance personnel perform maintenance on the special equipment to be used each time;
collecting inspection data of the special equipment to be used in each inspection period of the production environment;
generating cycle update coefficients based on the service record data set, the machine learning model, the inspection data, and the bayesian network model at the end of each inspection cycle; updating the verification period based on the period update coefficient;
in the test environment, a tester collects a plurality of special devices with different use durations and use strengths in advance, the maintenance times of the maintenance parts, the replacement times of the maintenance parts, the use durations of the maintenance parts and the use strengths of the maintenance parts in each special device are different, and the functional test, the maintenance parts, the replacement of the maintenance parts and the usability evaluation are carried out on each special device for a plurality of times; the usability assessment is an assessment process that a professional tester performs a functional test on the special equipment to judge whether the special equipment can be used continuously; the result of the availability assessment is one of available and unavailable;
the maintenance part is a mechanical part which needs maintenance every time in special equipment;
the statistical modes of the service time of the maintenance part and the service strength of the maintenance part are as follows:
when the maintenance part is installed on the corresponding special equipment, the service time length of the special equipment is marked as t0, and the service strength is marked as s0;
at the time a after the maintenance part is installed, marking the service time length of the special equipment as ta and marking the service strength as sa;
the service time of the a-th time after the maintenance part is installed is ta-t0, and the service strength is sa-s0;
the loss training data comprises a feature vector set of each maintenance part and a loss rationality set of the maintenance part;
the feature vectors of the maintenance parts comprise the use time and the use strength of the maintenance parts;
the reasonable loss degree of the maintenance parts is the maintenance times of the maintenance parts;
the availability training data comprises available probability feature vectors and available probability labels of each special device;
for each special device, the available probability feature vectors are vectors which are arranged according to the serial numbers of the maintenance parts and are composed of the maintenance times and the replacement times of each maintenance part;
the available probability label is a quantized value of an evaluation result of usability evaluation of the special equipment, and the quantized value is one of 0 or 1;
the machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment is trained by the following steps:
for each maintenance part, taking the feature vector of the maintenance part in the loss training data as the input of a machine learning model, wherein the machine learning model takes the predicted loss rationality of the feature vector of each group of maintenance parts as the output, takes the loss rationality corresponding to the feature vector of the maintenance part in the loss training data as a prediction target, and takes the sum of prediction errors of all the loss rationality as a training target; training the machine learning model until the sum of the prediction errors reaches convergence, and stopping training to obtain a machine learning model for outputting predicted loss rationality according to the feature vectors of the maintenance parts;
the method for training the Bayesian network model for analyzing the available probability of the special equipment is as follows:
constructing a Bayesian network model structure;
the method comprises the steps that in the available rate training data, available probability feature vectors are used as input of a Bayesian network model, the Bayesian network model takes available probability labels of prediction of each group of available probability feature vectors as output, available probability labels corresponding to the available probability feature vectors in the available rate training data are used as prediction targets, and the sum of prediction errors of all available probability labels is minimized to be used as a training target; training the Bayesian network model until the sum of the prediction errors reaches convergence, and training the Bayesian network model of the available probability labels which are predicted according to the available probability feature vectors; the predicted available probability label is the available probability of the analyzed special equipment;
the manner of constructing the bayesian network model structure is as follows:
constructing a two-layer Bayesian network, wherein the number of nodes of the first layer is 2N, the number of nodes of the second layer is 1; wherein N is the number of maintenance parts in the special equipment; the nodes of the first layer are maintenance times and replacement times of the maintenance parts ordered according to the serial numbers of the maintenance parts in sequence; the nodes of the second layer are available probability labels of special equipment;
each node in the first layer has a node with a directed edge pointing to the second layer;
the maintenance record data comprise equipment numbers of special equipment to be used, service time of each maintenance part, service strength of each maintenance part, maintenance times of each maintenance part and replacement times of each maintenance part;
the mode of collecting the inspection data of the special equipment to be used is as follows:
collecting the ratio of the unqualified special equipment to be used to all the special equipment to be used as test data in the test period when each test period is finished;
the mode of generating the periodic update coefficients is as follows:
the number of the inspection period is marked as i, and i is a positive integer;
marking the test data at the end of the ith test period as Ei;
the number of the special equipment to be used is marked as J, wherein j=1, 2,3, …, J is the number of all the special equipment to be used;
at the end of the ith verification period:
marking the overhaul record data of the j-th special equipment to be used in the overhaul record data set as Pij, and marking the number of the maintenance part as k; wherein k=1, 2,3, …, K is the number of maintenance parts in the j-th special equipment to be used;
for the kth maintenance part in the jth special equipment to be used, forming a maintenance part feature vector by using the use duration and the use strength corresponding to the kth maintenance part in the maintenance record data Pij, and inputting the maintenance part feature vector into a machine learning model corresponding to the kth maintenance part to obtain the predicted loss rationality output by the machine learning model; marking the predicted loss rationality of the kth maintenance part in the jth special equipment to be used as Hijk;
for the j-th special equipment to be used, the maintenance times and the replacement times of all maintenance parts in the maintenance record data Pij are formed into an available probability feature vector, and the available probability feature vector is input into a Bayesian network to obtain the output predicted available probability; marking the predicted available probability of the j-th special device to be used as Gij;
marking the maintenance times of the kth maintenance part of the jth special equipment to be used as Xijk;
calculate j-th to be usedThe qualification rate Lijk of the kth maintenance part of the special equipment; the qualification rate Lijk is calculated in the following way:；
after the ith inspection period is finished, marking the generated period updating coefficient as Ri, and determining the calculation formula of Ri asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein b1 and b2 are respectively preset proportionality coefficients;
the way to update the verification period is:
marking the inspection period at the end of the ith inspection period as Vi;
the calculation formula of the inspection period Vi is。
In a second aspect, the application provides a data acquisition device of special equipment, which comprises a training data collection module, a model training module, a maintenance record data collection module and a checking period updating module; wherein, each module is electrically connected;
the training data collection module is used for collecting loss training data and availability training data in a test environment and sending the loss training data and the availability training data to the model training module;
the model training module is used for training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment, and sending the machine learning model and the Bayesian network model to a test period updating module;
the maintenance record data collection module is used for setting a maintenance record data set in a production environment, wherein the maintenance record data set comprises maintenance record data of all special equipment to be used, and after maintenance is carried out on the special equipment to be used by maintenance personnel each time, the corresponding maintenance record data is updated and sent to the inspection period updating module;
the inspection period updating module is used for collecting inspection data of special equipment to be used in each inspection period of the production environment and generating a period updating coefficient based on the overhaul record data set, the machine learning model, the inspection data and the Bayesian network model at the end of each inspection period; the verification period is updated based on the period update coefficient.
In a third aspect, the present application proposes an electronic device comprising: a processor and a memory, wherein the memory stores a computer program for the processor to call;
the processor executes the data acquisition method of the special equipment by calling the computer program stored in the memory.
Compared with the prior art, the application has the beneficial effects that:
according to the method, loss training data and availability training data are collected in a test environment, and a machine learning model for evaluating the loss rationality of the maintenance parts of special equipment is trained based on the loss training data; training a Bayesian network model for analyzing the availability probability of special equipment based on availability training data, presetting a test period, setting a maintenance record data set in a production environment, wherein the maintenance record data set comprises maintenance record data of all special equipment to be used, updating corresponding maintenance record data after maintenance is carried out on the special equipment to be used by maintenance personnel each time, collecting test data of the special equipment to be used in each test period of the production environment, generating a period update coefficient based on the maintenance record data set, a machine learning model, the test data and the Bayesian network model at the end of each test period, and updating the test period based on the period update coefficient; the failure rate of the inspection is analyzed in each inspection cycle of the inspection mechanism, and then the inspection cycle is automatically adjusted according to the potential safety hazard condition of the parts of the special equipment which are put into use at present and the overall feasibility probability of the special equipment, so that the effect of self-adaptive inspection on the accident risk of the special equipment failure is achieved.
Drawings
Fig. 1 is a flowchart of a data acquisition method of a special device in embodiment 1 of the present application;
fig. 2 is a module connection relationship diagram of a data acquisition device of a special device in embodiment 2 of the present application;
fig. 3 is a schematic structural diagram of an electronic device in embodiment 3 of the present application.
Detailed Description
The technical solutions of the present application will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Example 1
As shown in fig. 1, a data acquisition method of a special device includes the following steps:
step one: collecting loss training data and availability training data in a test environment, and training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment; presetting a test period;
step two: setting an overhaul record data set in a production environment, wherein the overhaul record data set comprises overhaul record data of all special equipment to be used, and updating corresponding overhaul record data after maintenance personnel perform maintenance on the special equipment to be used each time;
step three: collecting inspection data of the special equipment to be used in each inspection period of the production environment;
step four: generating cycle update coefficients based on the service record data set, the machine learning model, the inspection data, and the bayesian network model at the end of each inspection cycle; updating the verification period based on the period update coefficient; thus, the collection period of the inspection data of the newly produced special equipment by the special equipment inspection department is dynamically adjusted, and the balance between the inspection efficiency and the safety of the special equipment is achieved;
in the test environment, a tester collects a plurality of special devices with different use durations and use strengths in advance, the maintenance times of the maintenance parts, the replacement times of the maintenance parts, the use durations of the maintenance parts and the use strengths of the maintenance parts in each special device are different, and the functional test, maintenance of the maintenance parts, replacement of the maintenance parts and the usability evaluation are carried out on each special device for a plurality of times, so that sufficient and accurate training data are obtained; the usability assessment is an assessment process that a professional tester performs a functional test on the special equipment to judge whether the special equipment can be used continuously; the result of the availability assessment is one of available and unavailable; it can be understood that the service time and the service strength of the special equipment and the maintenance parts are synchronously updated when each special equipment is subjected to functional test, and the maintenance times and the replacement times of the maintenance parts are synchronously updated when maintenance parts are maintained and the maintenance parts are replaced;
it will be appreciated that specialty equipment, including but not limited to lifting machinery, elevators, boilers, pressure cookers, etc., each type of specialty equipment requires periodic maintenance to ensure the life safety of the user and the user's unit of property;
preferably, the time period of use of the special device may be counted from the time the special device is purchased;
preferably, the use intensity of the special equipment can be counted according to the specific functions of different special equipment; for example, a technique of automatically counting the running distance is added to the background program of the elevator; the hoisting machine may be evaluated using the mechanical energy consumed by the object per hoisting, which can be calculated by calculating the product of the weight of the object times the transport distance;
the maintenance part is a mechanical part which needs maintenance every time in special equipment; it can be understood that the types of the parts to be maintained need to be configured according to the specific functions of the special equipment;
the statistical modes of the service time of the maintenance part and the service strength of the maintenance part are as follows:
when the maintenance part is installed on the corresponding special equipment, the service time length of the special equipment is marked as t0, and the service strength is marked as s0;
at the time a after the maintenance part is installed, marking the service time length of the special equipment as ta and marking the service strength as sa;
the service time of the a-th time after the maintenance part is installed is ta-t0, and the service strength is sa-s0;
the loss training data comprises a feature vector set of each maintenance part and a loss rationality set of the maintenance part;
the feature vectors of the maintenance parts comprise the use time and the use strength of the maintenance parts;
the reasonable loss degree of the maintenance parts is the maintenance times of the maintenance parts; it can be understood that the loss rationality measures the loss of the maintenance part theoretically under the corresponding use duration and use strength of the maintenance part;
the availability training data comprises available probability feature vectors and available probability labels of each special device;
for each special device, the available probability feature vectors are vectors which are arranged according to the serial numbers of the maintenance parts and are composed of the maintenance times and the replacement times of each maintenance part; as one example: [2,0, 1], which is exemplified by the available probability feature vector of the special equipment composed of two maintenance parts, wherein the maintenance number and the replacement number of the maintenance part numbered 1 are 2 and 0, respectively, and the maintenance number and the replacement number of the maintenance part numbered 2 are 1 and 1, respectively;
the available probability label is a quantized value of an evaluation result of usability evaluation of the special equipment, and the quantized value is one of 0 or 1; specifically, when the usability evaluation result is unavailable, the usability probability label is 0/1, and when the usability evaluation result is available, the usability probability label is 1/0 correspondingly;
the machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment is trained by the following steps:
for each maintenance part, taking the feature vector of the maintenance part in the loss training data as the input of a machine learning model, wherein the machine learning model takes the predicted loss rationality of the feature vector of each group of maintenance parts as the output, takes the loss rationality corresponding to the feature vector of the maintenance part in the loss training data as a prediction target, and takes the sum of prediction errors of all the loss rationality as a training target; training the machine learning model until the sum of the prediction errors reaches convergence, and stopping training to obtain a machine learning model for outputting predicted loss rationality according to the feature vectors of the maintenance parts; the machine learning model is any one of a polynomial regression model or an SVR model;
the method for training the Bayesian network model for analyzing the available probability of the special equipment is as follows:
constructing a Bayesian network model structure;
the method comprises the steps that in the available rate training data, available probability feature vectors are used as input of a Bayesian network model, the Bayesian network model takes available probability labels of prediction of each group of available probability feature vectors as output, available probability labels corresponding to the available probability feature vectors in the available rate training data are used as prediction targets, and the sum of prediction errors of all available probability labels is minimized to be used as a training target; training the Bayesian network model until the sum of the prediction errors reaches convergence, and training the Bayesian network model of the available probability labels which are predicted according to the available probability feature vectors; it can be understood that the predicted available probability label is the available probability of the analyzed special equipment;
it should be noted that, the calculation formula of the prediction error is:wherein c is the number of the feature data, zc is the prediction error, ac is the predicted state value corresponding to the feature data of the c group, and wc is the actual state value corresponding to the training data of the c group; for example, in the machine learning model, feature data corresponds to feature vectors of the maintenance parts, and state values correspond to loss rationality; the feature data in the Bayesian network model corresponds to an available probability feature vector, and the state value corresponds to an available probability label;
the manner of constructing the bayesian network model structure is as follows:
constructing a two-layer Bayesian network, wherein the number of nodes of the first layer is 2N, the number of nodes of the second layer is 1; wherein N is the number of maintenance parts in the special equipment; the nodes of the first layer are maintenance times and replacement times of the maintenance parts ordered according to the serial numbers of the maintenance parts in sequence; the nodes of the second layer are available probability labels of special equipment;
each node in the first layer has a node with a directed edge pointing to the second layer;
further, the checking period is a period of checking the special equipment which is newly used to wait for use or needs to be used continuously by a special equipment checking mechanism, and the special equipment which is newly used to wait for use or needs to be used continuously is called special equipment to be used;
the maintenance record data comprise equipment numbers of special equipment to be used, service time of each maintenance part, service strength of each maintenance part, maintenance times of each maintenance part and replacement times of each maintenance part;
the mode of updating the corresponding overhaul record data is as follows:
the method comprises the steps that for equipment numbers of each special equipment to be used, after maintenance personnel carry out maintenance on the special equipment each time, the use duration of each maintenance part of the special equipment to be used and the use intensity of each maintenance part are updated according to the actual use duration and the use intensity of the special equipment respectively;
further, the maintenance times of the maintenance parts and the replacement times of the maintenance parts are updated according to the actual maintenance conditions and the replacement conditions of the maintenance parts; for example, when a maintenance part is maintained, the maintenance frequency corresponding to the maintenance part is increased by one on the basis of the original maintenance frequency;
it can be understood that the maintenance personnel are personnel for maintaining and protecting equipment, and are generally affiliated to special equipment production, special equipment sales or third party institutions, so that the special equipment inspection has a defect in the specialty and enthusiasm; due to the specificity of the special equipment, the safety performance of the special equipment also needs to be periodically functionally checked by a professional checking mechanism so as to screen out unqualified special equipment as much as possible, thereby avoiding possible accidents;
the mode of collecting the inspection data of the special equipment to be used is as follows:
collecting the ratio of the unqualified special equipment to be used to all the special equipment to be used as test data in the test period when each test period is finished;
the mode of generating the periodic update coefficients is as follows:
the number of the inspection period is marked as i, and i is a positive integer;
marking the test data at the end of the ith test period as Ei;
the number of the special equipment to be used is marked as J, wherein j=1, 2,3, …, J is the number of all the special equipment to be used;
at the end of the ith verification period:
marking the overhaul record data of the j-th special equipment to be used in the overhaul record data set as Pij, and marking the number of the maintenance part as k; wherein k=1, 2,3, …, K is the number of maintenance parts in the j-th special equipment to be used;
for the kth maintenance part in the jth special equipment to be used, forming a maintenance part feature vector by using the use duration and the use strength corresponding to the kth maintenance part in the maintenance record data Pij, and inputting the maintenance part feature vector into a machine learning model corresponding to the kth maintenance part to obtain the predicted loss rationality output by the machine learning model; marking the predicted loss rationality of the kth maintenance part in the jth special equipment to be used as Hijk;
for the j-th special equipment to be used, the maintenance times and the replacement times of all maintenance parts in the maintenance record data Pij are formed into an available probability feature vector, and the available probability feature vector is input into a Bayesian network to obtain the output predicted available probability; marking the predicted available probability of the j-th special device to be used as Gij;
marking the maintenance times of the kth maintenance part of the jth special equipment to be used as Xijk;
calculating the qualification rate Lijk of the kth maintenance part of the jth special equipment to be used; the qualification rate Lijk is calculated in the following way:the method comprises the steps of carrying out a first treatment on the surface of the It will be appreciated that Lijk>1, the real maintenance times are larger than the theoretical maintenance times, the possibility that the maintenance parts are unqualified is shown, and the higher the possibility that the special equipment has potential safety hazards, the more the inspection frequency needs to be improved;
after the ith inspection period is finished, marking the generated period updating coefficient as Ri, and determining the calculation formula of Ri asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein b1 and b2 are respectively preset proportionality coefficients;
the way to update the verification period is:
marking the inspection period at the end of the ith inspection period as Vi;
the calculation formula of the inspection period Vi isThe method comprises the steps of carrying out a first treatment on the surface of the By checking the cycle based on the result of the previous round of checkingAnd the automatic adjustment is carried out, so that the effect of self-adaptive detection of accident risk of the special equipment fault is achieved.
Example 2
As shown in fig. 2, a data acquisition device of a special device comprises a training data collection module, a model training module, a maintenance record data collection module and a test period updating module; wherein, each module is electrically connected;
the training data collection module is mainly used for collecting loss training data and availability training data in a test environment and sending the loss training data and the availability training data to the model training module;
the model training module is mainly used for training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment, and sending the machine learning model and the Bayesian network model to a test period updating module;
the maintenance record data collection module is mainly used for setting a maintenance record data set in a production environment, wherein the maintenance record data set comprises maintenance record data of all special equipment to be used, and after maintenance is carried out on the special equipment to be used by maintenance personnel each time, the corresponding maintenance record data are updated and sent to the inspection period updating module;
the inspection period updating module is mainly used for collecting inspection data of special equipment to be used in each inspection period of the production environment, and generating period updating coefficients based on an overhaul record data set, a machine learning model, the inspection data and a Bayesian network model when each inspection period is finished; the verification period is updated based on the period update coefficient.
Example 3
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 3, there is also provided an electronic device 100 according to yet another aspect of the present application. The electronic device 100 may include one or more processors and one or more memories. In which a memory has stored therein computer readable code which, when executed by one or more processors, can perform the data acquisition method of a specialty device as described above.
The method or apparatus according to embodiments of the present application may also be implemented by means of the architecture of the electronic device shown in fig. 3. As shown in fig. 3, the electronic device 100 may include a bus 101, one or more CPUs 102, a Read Only Memory (ROM) 103, a Random Access Memory (RAM) 104, a communication port 105 connected to a network, an input/output component 106, a hard disk 107, and the like. A storage device in the electronic device 100, such as the ROM103 or the hard disk 107, may store the data collection method of the special device provided by the present application. The data acquisition method of the special equipment can comprise the following steps: step one: collecting loss training data and availability training data in a test environment, and training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment; presetting a test period; step two: setting an overhaul record data set in a production environment, wherein the overhaul record data set comprises overhaul record data of all special equipment to be used, and updating corresponding overhaul record data after maintenance personnel perform maintenance on the special equipment to be used each time; step three: collecting inspection data of the special equipment to be used in each inspection period of the production environment; step four: generating cycle update coefficients based on the service record data set, the machine learning model, the inspection data, and the bayesian network model at the end of each inspection cycle; updating the verification period based on the period update coefficient;
further, the electronic device 100 may also include a user interface 108. Of course, the architecture shown in fig. 3 is merely exemplary, and one or more components of the electronic device shown in fig. 3 may be omitted as may be practical in implementing different devices.
The above preset parameters or preset thresholds are set by those skilled in the art according to actual conditions or are obtained by mass data simulation.
The above embodiments are only for illustrating the technical method of the present application and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present application may be modified or substituted without departing from the spirit and scope of the technical method of the present application.
Claims (12)
1. The data acquisition method of the special equipment is characterized by comprising the following steps of:
collecting loss training data and availability training data in a test environment, and training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment; presetting a test period;
setting an overhaul record data set in a production environment, wherein the overhaul record data set comprises overhaul record data of all special equipment to be used, and updating corresponding overhaul record data after maintenance personnel perform maintenance on the special equipment to be used each time;
collecting inspection data of the special equipment to be used in each inspection period of the production environment;
generating cycle update coefficients based on the service record data set, the machine learning model, the inspection data, and the bayesian network model at the end of each inspection cycle; updating the verification period based on the period update coefficient;
the mode of generating the periodic update coefficients is as follows:
the number of the inspection period is marked as i, and i is a positive integer;
marking the test data at the end of the ith test period as Ei;
the number of the special equipment to be used is marked as J, wherein j=1, 2,3, …, J is the number of all the special equipment to be used;
at the end of the ith verification period:
marking the overhaul record data of the j-th special equipment to be used in the overhaul record data set as Pij, and marking the number of the maintenance part as k; wherein k=1, 2,3, …, K is the number of maintenance parts in the j-th special equipment to be used;
for the kth maintenance part in the jth special equipment to be used, forming a maintenance part feature vector by using the use duration and the use strength corresponding to the kth maintenance part in the maintenance record data Pij, and inputting the maintenance part feature vector into a machine learning model corresponding to the kth maintenance part to obtain the predicted loss rationality output by the machine learning model; marking the predicted loss rationality of the kth maintenance part in the jth special equipment to be used as Hijk;
for the j-th special equipment to be used, the maintenance times and the replacement times of all maintenance parts in the maintenance record data Pij are formed into an available probability feature vector, and the available probability feature vector is input into a Bayesian network to obtain the output predicted available probability; marking the predicted available probability of the j-th special device to be used as Gij;
marking the maintenance times of the kth maintenance part of the jth special equipment to be used as Xijk;
calculating the qualification rate Lijk of the kth maintenance part of the jth special equipment to be used; the qualification rate Lijk is calculated in the following way:；
after the ith inspection period is finished, marking the generated period updating coefficient as Ri, and determining the calculation formula of Ri asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein b1 and b2 are respectively preset proportionality coefficients;
the way to update the verification period is:
marking the inspection period at the end of the ith inspection period as Vi;
the calculation formula of the inspection period Vi is。
2. The data acquisition method of special equipment according to claim 1, wherein in the test environment, a tester collects special equipment with different use durations and use strengths in advance, and the maintenance times, the replacement times, the use durations and the use strengths of the maintenance parts in the special equipment are all different, and the functional test, the maintenance part maintenance, the maintenance part replacement and the usability evaluation are performed on the special equipment for a plurality of times; the usability assessment is an assessment process that a professional tester performs a functional test on the special equipment to judge whether the special equipment can be used continuously; the result of the availability assessment is one of available and unavailable; the maintenance part is a mechanical part which needs maintenance every time in special equipment.
3. The data collection method of a special device according to claim 2, wherein the statistical manner of the usage duration of the maintenance component and the usage intensity of the maintenance component is:
when the maintenance part is installed on the corresponding special equipment, the service time length of the special equipment is marked as t0, and the service strength is marked as s0;
at the time a after the maintenance part is installed, marking the service time length of the special equipment as ta and marking the service strength as sa;
the service time of the a-th time after the maintenance part is installed is ta-t0, and the service strength is sa-s0.
4. A data collection method of special equipment according to claim 3, wherein the loss training data comprises a feature vector set of each maintenance part and a loss rational degree set of each maintenance part;
the feature vectors of the maintenance parts comprise the use time and the use strength of the maintenance parts;
the reasonable degree of loss of the maintenance part is the maintenance times of the maintenance part.
5. The method for collecting data of special equipment according to claim 4, wherein the availability training data comprises an available probability feature vector and an available probability label of each special equipment;
for each special device, the available probability feature vectors are vectors which are arranged according to the serial numbers of the maintenance parts and are composed of the maintenance times and the replacement times of each maintenance part;
the availability probability tag is a quantized value of an evaluation result of the availability evaluation of the special equipment, and the quantized value is one of 0 or 1.
6. The method for collecting data of a special device according to claim 5, wherein the training a machine learning model for evaluating the loss rationality of a maintenance component of the special device is as follows:
for each maintenance part, taking the feature vector of the maintenance part in the loss training data as the input of a machine learning model, wherein the machine learning model takes the predicted loss rationality of the feature vector of each group of maintenance parts as the output, takes the loss rationality corresponding to the feature vector of the maintenance part in the loss training data as a prediction target, and takes the sum of prediction errors of all the loss rationality as a training target; and training the machine learning model until the sum of the prediction errors reaches convergence, and stopping training to train the machine learning model which outputs predicted loss rationality according to the feature vectors of the maintenance parts for each maintenance part.
7. The method for collecting data of a specific device according to claim 6, wherein the manner of training a bayesian network model for analyzing the probability of availability of the specific device is:
constructing a Bayesian network model structure;
the method comprises the steps that in the available rate training data, available probability feature vectors are used as input of a Bayesian network model, the Bayesian network model takes available probability labels of prediction of each group of available probability feature vectors as output, available probability labels corresponding to the available probability feature vectors in the available rate training data are used as prediction targets, and the sum of prediction errors of all available probability labels is minimized to be used as a training target; training the Bayesian network model until the sum of the prediction errors reaches convergence, and training the Bayesian network model of the available probability labels which are predicted according to the available probability feature vectors; the predicted available probability label is the available probability of the analyzed special equipment.
8. The method for data collection of a special device according to claim 7, wherein the bayesian network model structure is constructed by:
constructing a two-layer Bayesian network, wherein the number of nodes of the first layer isThe number of nodes of the second layer is 1; wherein N is the number of maintenance parts in the special equipment; the nodes of the first layer are maintenance times and replacement times of the maintenance parts ordered according to the serial numbers of the maintenance parts in sequence; the nodes of the second layer are available probability labels of special equipment;
each node in the first layer has a directed edge pointing to a node of the second layer.
9. The method for collecting data of a special device according to claim 8, wherein the service record data includes a device number of each special device to be used, a use period of each maintenance part, a use strength of each maintenance part, a maintenance number of each maintenance part, and a replacement number of each maintenance part.
10. The method for collecting data of a special device according to claim 9, wherein the method for collecting test data of the special device to be used is as follows:
at the end of each inspection cycle, the ratio of the unqualified special equipment to be used to all the inspected special equipment to be used, which is the inspection data, is collected in the inspection cycle.
11. A data acquisition device of special equipment, which is realized based on the data acquisition method of the special equipment according to any one of claims 1-10, and is characterized by comprising a training data collection module, a model training module, a maintenance record data collection module and a checking period updating module; wherein, each module is electrically connected;
the training data collection module is used for collecting loss training data and availability training data in a test environment and sending the loss training data and the availability training data to the model training module;
the model training module is used for training a machine learning model for evaluating the loss rationality of the maintenance parts of the special equipment based on the loss training data; based on the availability training data, training a Bayesian network model for analyzing the availability probability of the special equipment, and sending the machine learning model and the Bayesian network model to a test period updating module;
the maintenance record data collection module is used for setting a maintenance record data set in a production environment, wherein the maintenance record data set comprises maintenance record data of all special equipment to be used, and after maintenance is carried out on the special equipment to be used by maintenance personnel each time, the corresponding maintenance record data is updated and sent to the inspection period updating module;
the inspection period updating module is used for collecting inspection data of special equipment to be used in each inspection period of the production environment and generating a period updating coefficient based on the overhaul record data set, the machine learning model, the inspection data and the Bayesian network model at the end of each inspection period; updating the verification period based on the period update coefficient;
the mode of generating the periodic update coefficients is as follows:
the number of the inspection period is marked as i, and i is a positive integer;
marking the test data at the end of the ith test period as Ei;
the number of the special equipment to be used is marked as J, wherein j=1, 2,3, …, J is the number of all the special equipment to be used;
at the end of the ith verification period:
marking the overhaul record data of the j-th special equipment to be used in the overhaul record data set as Pij, and marking the number of the maintenance part as k; wherein k=1, 2,3, …, K is the number of maintenance parts in the j-th special equipment to be used;
for the kth maintenance part in the jth special equipment to be used, forming a maintenance part feature vector by using the use duration and the use strength corresponding to the kth maintenance part in the maintenance record data Pij, and inputting the maintenance part feature vector into a machine learning model corresponding to the kth maintenance part to obtain the predicted loss rationality output by the machine learning model; marking the predicted loss rationality of the kth maintenance part in the jth special equipment to be used as Hijk;
for the j-th special equipment to be used, the maintenance times and the replacement times of all maintenance parts in the maintenance record data Pij are formed into an available probability feature vector, and the available probability feature vector is input into a Bayesian network to obtain the output predicted available probability; marking the predicted available probability of the j-th special device to be used as Gij;
marking the maintenance times of the kth maintenance part of the jth special equipment to be used as Xijk;
calculating the qualification rate Lijk of the kth maintenance part of the jth special equipment to be used; the qualification rate Lijk is calculated in the following way:；
after the ith inspection period is finished, the generated period update coefficient is marked as Ri, and the calculation of Ri is commonIs of the typeThe method comprises the steps of carrying out a first treatment on the surface of the Wherein b1 and b2 are respectively preset proportionality coefficients;
the way to update the verification period is:
marking the inspection period at the end of the ith inspection period as Vi;
the calculation formula of the inspection period Vi is。
12. An electronic device, comprising: a processor and a memory, wherein:
the memory stores a computer program which can be called by the processor;
the processor performs a data acquisition method of a special device according to any one of claims 1-10 by calling a computer program stored in the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311146521.3A CN116882713B (en) | 2023-09-07 | 2023-09-07 | Data acquisition method and device of special equipment and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311146521.3A CN116882713B (en) | 2023-09-07 | 2023-09-07 | Data acquisition method and device of special equipment and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116882713A CN116882713A (en) | 2023-10-13 |
| CN116882713B true CN116882713B (en) | 2023-11-28 |
Family
ID=88272031
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311146521.3A Active CN116882713B (en) | 2023-09-07 | 2023-09-07 | Data acquisition method and device of special equipment and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116882713B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104850934A (en) * | 2015-04-14 | 2015-08-19 | 中国气象局气象探测中心 | Meteorological device maintenance and repair management system |
| CN109947088A (en) * | 2019-04-17 | 2019-06-28 | 北京天泽智云科技有限公司 | Equipment fault early-warning system based on model lifecycle management |
| CN110909895A (en) * | 2019-11-19 | 2020-03-24 | 科大国创软件股份有限公司 | Early warning method and system based on special equipment historical periodic inspection report |
| JP2020144569A (en) * | 2019-03-06 | 2020-09-10 | 日本電信電話株式会社 | Model learning device, label estimation device, method thereof, and program |
| CN112101662A (en) * | 2020-09-16 | 2020-12-18 | 珠海格力电器股份有限公司 | Equipment health condition and life cycle detection method, storage medium and electronic equipment |
| CN113777423A (en) * | 2021-08-19 | 2021-12-10 | 武汉铁路职业技术学院 | Test system based on electric automation equipment |
| CN114912635A (en) * | 2022-04-28 | 2022-08-16 | 广州铁路职业技术学院（广州铁路机械学校） | Highway maintenance decision-making method and system, storage medium and electronic equipment |
| CN115630716A (en) * | 2022-11-10 | 2023-01-20 | 南方电网数字平台科技(广东)有限公司 | Intelligent generation method and device for equipment maintenance plan |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11367053B2 (en) * | 2018-11-16 | 2022-06-21 | University Of Connecticut | System and method for damage assessment and restoration |
-
2023
- 2023-09-07 CN CN202311146521.3A patent/CN116882713B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104850934A (en) * | 2015-04-14 | 2015-08-19 | 中国气象局气象探测中心 | Meteorological device maintenance and repair management system |
| JP2020144569A (en) * | 2019-03-06 | 2020-09-10 | 日本電信電話株式会社 | Model learning device, label estimation device, method thereof, and program |
| CN109947088A (en) * | 2019-04-17 | 2019-06-28 | 北京天泽智云科技有限公司 | Equipment fault early-warning system based on model lifecycle management |
| CN110909895A (en) * | 2019-11-19 | 2020-03-24 | 科大国创软件股份有限公司 | Early warning method and system based on special equipment historical periodic inspection report |
| CN112101662A (en) * | 2020-09-16 | 2020-12-18 | 珠海格力电器股份有限公司 | Equipment health condition and life cycle detection method, storage medium and electronic equipment |
| CN113777423A (en) * | 2021-08-19 | 2021-12-10 | 武汉铁路职业技术学院 | Test system based on electric automation equipment |
| CN114912635A (en) * | 2022-04-28 | 2022-08-16 | 广州铁路职业技术学院（广州铁路机械学校） | Highway maintenance decision-making method and system, storage medium and electronic equipment |
| CN115630716A (en) * | 2022-11-10 | 2023-01-20 | 南方电网数字平台科技(广东)有限公司 | Intelligent generation method and device for equipment maintenance plan |
Non-Patent Citations (2)
| Title |
|---|
| 关于大数据运用在特种设备检验中的思考;檀昊等;《工程技术研究》(01);82-84+86 * |
| 黄魁.面向顾客个性化需求的预测性维修决策优化.《中国优秀博硕士学位论文全文数据库（硕士） 工程科技Ⅱ辑》.2022,(第6期),1-73. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116882713A (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106872657B (en) | A kind of multivariable water quality parameter time series data accident detection method | |
| CN103617110B (en) | Server device condition maintenance system | |
| CN103400310B (en) | Method for evaluating power distribution network electrical equipment state based on historical data trend prediction | |
| CN104218570A (en) | Method and system for online evaluating overall measuring errors of electric energy measuring device | |
| CN107392324A (en) | The specialized maintenance total management system of transformer | |
| EP2682836B1 (en) | Method for performing diagnostics of a structure subject to loads and system for implementing said method | |
| CN103606062A (en) | Relay protection state evaluation and aid decision-making maintenance method | |
| CN110363404A (en) | A kind of dry-type air-core reactor status data analysis method | |
| CN110276385A (en) | Mechanical part remaining life prediction technique based on similitude | |
| CN205621238U (en) | Damaged on -line monitoring alarm device of nuclear reactor fuel | |
| CN107832984A (en) | Consider the tractive power supply system methods of risk assessment of reliability of relay protection | |
| CN116882713B (en) | Data acquisition method and device of special equipment and electronic equipment | |
| CN105242129B (en) | A kind of transformer winding fault probability determination method | |
| CN109359742B (en) | Method for generating preventive maintenance period of subway subsystem | |
| CN106802599A (en) | A kind of diagnosing fault of power transformer system based on expert database | |
| CN105741184B (en) | Transformer state evaluation method and device | |
| CN114137302A (en) | Monitoring system for whole verification process of electric energy metering device | |
| CN107563608A (en) | A kind of power distribution communication net safety evaluation method | |
| CN112819262A (en) | Memory, process pipeline inspection and maintenance decision method, device and equipment | |
| CN115514099A (en) | Electric power utilization safety inspection system and method | |
| RU2667119C2 (en) | Products residual defectiveness monitoring method | |
| CN115758066A (en) | Method for counting carbon emission in whole life cycle of transformer | |
| CN108573119A (en) | In-service hoisting machinery overall structure remaining life real-time evaluation system | |
| CN105303315B (en) | A kind of power equipment reliability appraisal procedure counted and maintenance randomness influences | |
| CN113159411B (en) | Method and system for testing power grid meteorological risk early warning model |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |