CN116934293A - Equipment simulation maintenance method and system based on artificial intelligence - Google Patents
Equipment simulation maintenance method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN116934293A CN116934293A CN202310673785.8A CN202310673785A CN116934293A CN 116934293 A CN116934293 A CN 116934293A CN 202310673785 A CN202310673785 A CN 202310673785A CN 116934293 A CN116934293 A CN 116934293A
- Authority
- CN
- China
- Prior art keywords
- data
- equipment
- maintenance
- fault
- artificial intelligence
- 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
- 238000012423 maintenance Methods 0.000 title claims abstract description 166
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004088 simulation Methods 0.000 title claims abstract description 34
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 24
- 238000013145 classification model Methods 0.000 claims abstract description 43
- 230000008569 process Effects 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 54
- 239000000463 material Substances 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 6
- 238000009434 installation Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 6
- 238000007667 floating Methods 0.000 description 6
- 238000004140 cleaning Methods 0.000 description 4
- 238000012706 support-vector machine Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000009877 rendering Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 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
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an artificial intelligence-based equipment simulation maintenance method and system, comprising the following steps: step S100, collecting data in the running process of equipment, and preprocessing the data to obtain an equipment data set, wherein the data are physical data of the equipment and data reflecting the running states of various modules of the equipment; step S200, carrying out fault class classification on the received equipment data set based on the SVM classification model to obtain fault data of a certain module, wherein the fault data reflects fault problems encountered by the certain module; and step S300, the SVM classification model generates a corresponding maintenance scheme based on the fault data, and the scheme enables equipment maintenance to be faster, more accurate, safer and more intelligent, reduces maintenance cost and risk, and improves reliability and service life of the equipment.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to an equipment simulation maintenance method and system based on artificial intelligence.
Background
In the field of equipment maintenance, an artificial intelligent equipment simulation maintenance system can predict and diagnose faults through virtual simulation of equipment and provide maintenance schemes and operation guidance, so that maintenance efficiency and quality are improved. This technique can be applied to various types of equipment such as mechanical devices, electronic devices, vehicles, aircraft, and the like.
Modern equipment is increasingly complex and maintenance difficulties and risks are correspondingly increased. The traditional maintenance mode generally needs to rely on manual or field operation, and has the problems of low maintenance efficiency, high operation risk and the like. Therefore, a more efficient and safe maintenance method is needed to cope with the actual demands.
Disclosure of Invention
In order to solve the technical problems, the application provides an artificial intelligence-based equipment simulation maintenance method, which solves the problems that the traditional maintenance mode usually needs to rely on manual or field operation, and has low maintenance efficiency, high operation risk and the like.
In a first aspect, an artificial intelligence based equipment simulation maintenance method provided by an embodiment of the present application includes:
step S100, collecting data in the running process of equipment, and preprocessing the data to obtain an equipment data set, wherein the data are physical data of the equipment and data reflecting the running state of each module of the equipment;
step S200, carrying out fault class classification on the received equipment data set based on an SVM classification model to obtain fault data of a certain module, wherein the fault data reflects fault problems encountered by the certain module;
and step S300, the SVM classification model generates a corresponding maintenance scheme based on the fault data.
As a possible implementation manner, step S200 further includes, before: establishing an SVM classification model; and training the SVM classification model.
As a possible implementation manner, the training the SVM classification model further includes:
acquiring pre-training data, wherein the pre-training data comprises data in the running process of equipment, fault records, maintenance records or experience and knowledge of experts in the field of maintenance systems, and second running data acquired when the equipment simulates different fault conditions in a virtual environment;
selecting data influencing fault class classification from the pre-training data as training data;
and training the SVM classification model by using the training data.
As a possible implementation manner, the method further includes step S400, wherein the corresponding maintenance scheme generated based on the fault data is manually adjusted, and the maintenance scheme includes fault reasons, detailed maintenance steps, maintenance step combination for maintaining parts, maintenance operation guidance flow and tool operation method required for maintenance.
As a possible implementation manner, step S100 further includes: and storing the data acquired in the running process of the equipment in a network module capable of remotely acquiring or sharing the data.
As one possible implementation, the fault data includes the number of a certain module of the equipment, a maintenance tool, maintenance content, a general maintenance procedure, maintenance materials, a set of parts involved in maintenance, and a maintenance status.
As a possible implementation manner, step S500 displays, through an interface, a corresponding maintenance scheme generated based on the fault data.
As a possible implementation manner, the second operation data collected when the equipment simulates different fault conditions in the virtual environment further comprises: the virtual environment is a simulation model environment established according to the characteristic information of the equipment, wherein the characteristic information comprises the physical structure, the kinematic characteristic, the dynamic characteristic, the control system, the sensor and the structural characteristics of each part of the equipment.
As a possible implementation, the equipment data set further includes: equipment model, installation location, maintenance record, operating condition, equipment weight, external dimensions, equipment material, equipment internal construction picture, equipment external picture, wherein the operating condition includes: operating time, operating speed, operating temperature, operating pressure, operating voltage, and operating current.
In a second aspect, the device simulation maintenance system based on artificial intelligence provided by the embodiment of the application includes a data acquisition module, configured to acquire data in a device operation process, and preprocess the data to obtain a device data set, where the data is physical data of the device and data reflecting an operation state of each module of the device;
the algorithm module is used for classifying the fault categories of the received equipment data set based on the SVM classification model to obtain fault data of a certain module, wherein the fault data reflects fault problems encountered by the certain module; the SVM classification model generates a corresponding repair scheme based on the fault data.
By adopting the scheme of the application, equipment maintenance is quicker, more accurate, safer and more intelligent, maintenance cost and risk are reduced, and reliability and service life of the equipment are improved.
Drawings
FIG. 1 is a flow chart of artificial intelligence based equipment simulation maintenance embodying the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the present application is a system method for quantitative assessment, and the specific embodiments described herein are for illustrative purposes only and are not intended to limit the present application. In addition, the technical features of the embodiments of the present application described below may be combined with each other as long as they do not collide with each other.
The artificial intelligence equipment simulation maintenance system is a maintenance auxiliary system based on an artificial intelligence technology and a virtual reality technology, can simulate the maintenance process of actual equipment, and assists maintenance personnel to carry out maintenance operation in an intelligent mode.
The virtual reality technology has been widely used in the fields of games, education, medical treatment, etc., and can provide an immersive virtual environment and interactive experience. The technology also provides foundation and support for the development of the virtual simulation maintenance software platform of the equipment.
Development of three-dimensional modeling and computer graphics technology: three-dimensional modeling and computer graphics technology are well established, and can perform high-precision modeling and presentation on real equipment, provide various visual effects such as illumination, shadow, texture and the like, and enhance the sense of reality and user experience of a virtual environment.
Based on the complexity and existing maintenance difficulty of modern devices in the prior art, as shown in fig. 1, a first embodiment of the present application provides an artificial intelligence-based equipment simulation maintenance method, which includes:
step S100, collecting data in the running process of the equipment, and preprocessing the data to obtain an equipment data set, wherein the data are physical data of the equipment and data reflecting the running states of various modules of the equipment.
The collected data also includes fault records, maintenance records, or experience and knowledge of the expert in the field of maintenance systems, which are the basis for performing fault diagnosis and maintenance. To improve the quality and accuracy of the data, the collected data may be preprocessed, for example, by data cleaning, denoising, interpolation, smoothing, etc., to obtain an equipment data set, where the equipment data set further includes: equipment model, installation location, maintenance record, operating condition, equipment weight, external dimensions, equipment material, equipment internal construction picture, equipment external picture, wherein the operating condition includes: operating time, operating speed, operating temperature, operating pressure, operating voltage, and operating current. The attribute data of the setup data set is shown in table 1 below.
In a specific application scenario, monitoring equipment such as a sensor needs to be installed to acquire various data in the operation process of equipment, including sensor data such as temperature, pressure, vibration, current, voltage, speed and the like, operation parameters, fault information and the like. And storing the data acquired in the running process of the equipment in a network module capable of remotely acquiring or sharing the data. For example, the equipment data set of the final equipment is uploaded to a network module, in particular to a server of the maintenance system for storage via a network (local area network or internal network). The equipment simulation maintenance system can realize the functions of remote collaboration and guidance, remote maintenance support, data sharing and analysis and the like through the remote access server, so that the use value and maintenance efficiency of the system are improved, and meanwhile, the network module can also realize the online updating and maintenance of the system so as to ensure the stability and safety of the system.
The attribute data of the equipment data set is as follows:
TABLE 1
| Sequence number | Attribute name | Attribute type | Description of the application |
| 1 | eqModel | Character string | Equipment model |
| 2 | insPosition | (Vector) | Mounting position |
| 3 | maintRecords | Character string | Maintenance record |
| 4 | runtime | Date | Run time |
| 5 | runSpeed | Floating point number | Speed of operation |
| 6 | runTemperature | Floating point number | Operating temperature |
| 7 | runPressure | Floating point number | Operating pressure |
| 8 | runVoltage | Floating point number | Operating voltage |
| 9 | runElectricity | Floating point number | Operating current |
| 10 | overSize | (Vector) | External dimension |
| 11 | weight | Floating point number | Weight of the equipment |
| 12 | materialMsg | Character string | Equipment material |
| 13 | eqInsideImg | Picture picture | Picture with internal structure of equipment |
| 14 | eqOutsideImg | Picture picture | Equipment appearance picture |
And step 200, carrying out fault class classification on the received equipment data set based on the SVM classification model to obtain fault data of a certain module, wherein the fault data reflects fault problems encountered by the certain module. The fault data includes the number of a module of the equipment, a service tool, service contents, general service procedures, service materials, a set of parts involved in the service, and a service status. The attribute data of the fault data is shown in table 2 below.
The attribute data of the failure data are as follows:
TABLE 2
The SVM classification model is a machine learning algorithm based on a support vector machine, and is a two-class model, the basic idea is to convert the equipment data into a high-dimensional space, and then find a hyperplane capable of dividing the equipment data.
In the equipment simulation maintenance system, a trained SVM classification model is utilized to classify equipment data sets into fault categories and different types of maintenance problems or equipment states so as to better identify the problems and take proper maintenance measures.
The step S200 further includes: an SVM classification model based on an SVM algorithm needs to be constructed, and then the SVM classification model is trained, and the specific operation is as follows: pre-training data is collected, the pre-training data including data during operation of the equipment, fault records, maintenance records, or experience and knowledge of an expert in the field of maintenance systems, and second operational data collected when the equipment simulates different fault conditions in a virtual environment. The equipment simulates second operation data acquired in different fault conditions in a virtual environment, and further comprises: the virtual environment is a simulation model environment established according to the characteristic information of the equipment, wherein the characteristic information comprises the physical structure, the kinematic characteristic, the dynamic characteristic, the control system, the sensor and the structural characteristics of each part of the equipment.
Selecting data influencing fault class classification from the pre-training data as training data; specifically, the characteristics of input data such as data before training are analyzed, and data influencing fault class classification is selected from the input data as training data.
And training the SVM classification model by using the training data, and adjusting parameters of the model to obtain better classification performance.
After training the SVM classification model, the classification accuracy and performance of some test set evaluation models can be selected; the trained SVM model is applied to new unknown data for classification prediction and maintenance decision.
The training method for training the SVM can adopt a pair of methods, and supposing that equipment maintenance is divided into A, B, C, D four classification maintenance samples, one class can be regarded as class 1, and the other classes are uniformly classified as class 2. Thus we can construct 4 SVM classification models, alternatively called SVM classifiers, respectively the following:
(1) The maintenance sample A is used as a positive set, and B, C and D are used as negative sets; obtaining a training result of A and a maintenance scheme of A; A. b, C, D has a certain commonality;
(2) The maintenance sample B is taken as a positive set, and A, C and D are taken as a negative set;
(3) The maintenance sample C is taken as a positive set, and A, B and D are taken as a negative set;
(4) Maintenance sample D was taken as the positive set and a, B, C as the negative set.
The maintenance sample A, B, C, D is equipment operation data corresponding to the device A, B, C, D, and for K classifications, the K classifiers need to be trained, and the classification speed is faster but the training speed is slower.
It should be noted that A, B, C, D devices have a certain commonality, for example A, B, C, D devices may be radar devices, such as lidar, millimeter wave radar, centimetre wave radar, etc
Another training method is a one-to-one method, and aims to be more flexible in training. We can construct an SVM between any two classes of repair samples, so that for samples of class K, there will be a class C (K, 2) classifier.
For example, we want to divide A, B, C into three classes, 3 classifiers can be constructed:
(1) Classifier 1: A. b, a step of preparing a composite material;
(2) Classifier 2: A. c, performing operation;
(3) Classifier 3: B. and C, performing the operation of the device.
When an unknown sample is classified, each classifier has a classification result, namely 1 ticket, and the class with the largest final ticket is the class of the whole unknown sample.
The advantage of using a one-to-one training method is that if a class is newly added, all SVMs do not need to be retrained, and only the classifier of the class of samples needs to be trained and newly added. Moreover, the training speed is high when a single SVM model is trained. The application can select any training method according to the requirement.
And step S300, the SVM classification model generates a corresponding maintenance scheme based on the fault data.
The SVM classification model is used for maintenance decision support, and the system can provide maintenance decision support for specific faults or problems based on classification results of the SVM model. The model can provide relevant suggestions, guidance and recommended maintenance measures for maintenance personnel according to the characteristics of the input data so as to solve the problems more efficiently and accurately.
The virtual environment is a simulation model environment established according to the characteristic information of the equipment, wherein the characteristic information comprises the physical structure, the kinematic characteristic, the dynamic characteristic, the control system, the sensor and the structural characteristics of each part of the equipment.
Utilizing the characteristic information of the equipment, utilizing modeling software to construct a three-dimensional model of the equipment, and importing the three-dimensional model of the equipment constructed by the modeling software into a unit platform, wherein the construction of the virtual environment specifically comprises the following steps:
the scene module or the scene module is used for creating a scene, setting an initialization view angle, adding or deleting objects in the scene, searching objects containing specified tags, starting or stopping the scene, rendering the scene, loading all models through json files, and displaying or hiding the scene. Through the entity object module, the module can add a component to an object, add or remove a tag of the object, acquire the object through the tag, load a model through json files, set the position and the rotation angle of the object in the world, change all material patterns of the object, judge whether the loading of the object is completed, clone and acquire the component of the object, and finally construct a virtual environment. Building a virtual environment using modeling software and a unit platform may refer to prior art content.
Step S400, manually adjusting the corresponding maintenance scheme generated based on the fault data, where the maintenance scheme includes fault cause, detailed maintenance step, maintenance step combination of maintenance related components, maintenance operation guidance flow, and tool operation method required for maintenance, and manually cleaning the maintenance scheme given by the system, for example, manually deleting or adding content and adjusting sequence of the generated maintenance scheme, and combining the actual situation to obtain an adjusted maintenance scheme, see table 3 below.
The maintenance scheme's attribute data is shown below:
TABLE 3 Table 3
And step S400, displaying the adjusted maintenance scheme through an interface.
The second embodiment of the application provides an artificial intelligence-based equipment simulation maintenance system. The artificial intelligence equipment simulation maintenance system comprises a data acquisition module, an algorithm module, a virtual simulation module, an interface module and a network module.
The data acquisition module is used for acquiring data in the running process of the equipment and preprocessing the data to obtain an equipment data set, wherein the data is physical data of the equipment and data reflecting the running state of each module of the equipment.
The data collected by the data collection module also comprises fault records, maintenance records or experience and knowledge of experts in the field of maintenance systems, and the data are the basis for fault diagnosis and maintenance. To improve the quality and accuracy of the data, the collected data may be preprocessed, for example, by data cleaning, denoising, interpolation, smoothing, etc., to obtain an equipment data set, where the equipment data set further includes: equipment model, installation location, maintenance record, operating condition, equipment weight, external dimensions, equipment material, equipment internal construction picture, equipment external picture, wherein the operating condition includes: operating time, operating speed, operating temperature, operating pressure, operating voltage, and operating current. The attribute data of the setup data set is shown in table 1 above.
In a specific application scenario, monitoring equipment such as a sensor needs to be installed to acquire various data in the operation process of equipment, including sensor data such as temperature, pressure, vibration, current, voltage, speed and the like, operation parameters, fault information and the like. And storing the data acquired in the running process of the equipment in a network module capable of remotely acquiring or sharing the data. For example, the equipment data set of the final equipment is uploaded to a network module, in particular to a server of the maintenance system for storage via a network (local area network or internal network). The equipment simulation maintenance system can realize the functions of remote collaboration and guidance, remote maintenance support, data sharing and analysis and the like through the remote access server, so that the use value and maintenance efficiency of the system are improved, and meanwhile, the network module can also realize the online updating and maintenance of the system so as to ensure the stability and safety of the system.
And the algorithm module is used for classifying the fault categories of the received equipment data set based on the SVM classification model to obtain fault data of a certain module, wherein the fault data reflects fault problems encountered by the certain module.
The SVM classification model generates a corresponding maintenance scheme based on the fault data, and can also manually clean the maintenance scheme given by the system, for example, manually adjust the generated maintenance scheme and combine the real situation.
The method comprises the steps of carrying out fault class classification on a received equipment data set based on an SVM classification model, training the SVM classification model before fault data of a certain module are obtained, wherein the SVM classification model is a machine learning algorithm based on a support vector machine and is a two-class model, and the basic idea is to convert the equipment data into a high-dimensional space and then find a hyperplane capable of dividing the equipment data.
In the equipment simulation maintenance system, a trained SVM classification model is utilized to classify equipment data sets into fault categories and different types of maintenance problems or equipment states so as to better identify the problems and take proper maintenance measures.
The application needs to construct an SVM classification model based on an SVM algorithm, and trains the SVM classification model, and the specific operation is as follows: acquiring pre-training data, wherein the pre-training data comprises data in the running process of equipment, fault records, maintenance records or experience and knowledge of experts in the field of maintenance systems, and second running data acquired when the equipment simulates different fault conditions in a virtual environment;
selecting data influencing fault class classification from the pre-training data as training data; specifically, the characteristics of input data such as data before training are analyzed, and data influencing fault class classification is selected from the input data as training data.
And training the SVM classification model by using the training data, and adjusting parameters of the model to obtain better classification performance.
After training the SVM classification model, the classification accuracy and performance of some test set evaluation models can be selected; the trained SVM model is applied to new unknown data for classification prediction and maintenance decision.
The training method for training the SVM can adopt a pair of methods, and supposing that equipment maintenance is divided into A, B, C, D four classification maintenance samples, one class can be regarded as class 1, and the other classes are uniformly classified as class 2. Thus we can construct 4 SVM classification models, alternatively called SVM classifiers, respectively the following:
(1) The maintenance sample A is used as a positive set, and B, C and D are used as negative sets; obtaining a training result of A and a maintenance scheme of A; A. b, C, D has a certain commonality;
(2) The maintenance sample B is taken as a positive set, and A, C and D are taken as a negative set;
(3) The maintenance sample C is taken as a positive set, and A, B and D are taken as a negative set;
(4) Maintenance sample D was taken as the positive set and a, B, C as the negative set.
The maintenance sample A, B, C, D is equipment operation data corresponding to the device A, B, C, D, and for K classifications, the K classifiers need to be trained, and the classification speed is faster but the training speed is slower.
It should be noted that A, B, C, D devices have a certain commonality, for example A, B, C, D devices may be radar devices, such as lidar, millimeter wave radar, centimetre wave radar, etc
Another training method is a one-to-one method, and aims to be more flexible in training. We can construct an SVM between any two classes of repair samples, so that for samples of class K, there will be a class C (K, 2) classifier.
For example, we want to divide A, B, C into three classes, 3 classifiers can be constructed:
(1) Classifier 1: A. b, a step of preparing a composite material;
(2) Classifier 2: A. c, performing operation;
(3) Classifier 3: B. and C, performing the operation of the device.
When an unknown sample is classified, each classifier has a classification result, namely 1 ticket, and the class with the largest final ticket is the class of the whole unknown sample.
The advantage of using a one-to-one training method is that if a class is newly added, all SVMs do not need to be retrained, and only the classifier of the class of samples needs to be trained and newly added. Moreover, the training speed is high when a single SVM model is trained. The application can select any training method according to the requirement.
The SVM classification model is used for classifying faults of equipment, classifying faults or problems in an equipment simulation maintenance system, and accurately classifying the faults or problems into corresponding fault types or problem categories by analyzing characteristics of input data. This helps identify and locate equipment failures and take corresponding maintenance measures.
The SVM classification model of the present application is used for fault prediction of equipment, and based on analysis of historical data and features, the SVM model can be used for predicting faults or problems possibly encountered by the equipment in the future. By training and predicting the input data, the system can be prepared in advance and preventive maintenance measures can be taken to avoid the impact of faults on the operation of the equipment.
The SVM classification model is used for maintenance decision support, and the system can provide maintenance decision support for specific faults or problems based on classification results of the SVM model. The model can provide relevant suggestions, guidance and recommended maintenance measures for maintenance personnel according to the characteristics of the input data so as to solve the problems more efficiently and accurately.
The virtual environment is a simulation model environment established according to the characteristic information of the equipment, wherein the characteristic information comprises the physical structure, the kinematic characteristic, the dynamic characteristic, the control system, the sensor and the structural characteristics of each part of the equipment.
Utilizing the characteristic information of the equipment, utilizing modeling software to construct a three-dimensional model of the equipment, and importing the three-dimensional model of the equipment constructed by the modeling software into a unit platform, wherein the construction of the virtual environment specifically comprises the following steps: the scene module or the scene module is used for creating a scene, setting an initialization view angle, adding or deleting objects in the scene, searching objects containing specified tags, starting or stopping the scene, rendering the scene, loading all models through json files, and displaying or hiding the scene. Through the entity object module, the module can add a component to an object, add or remove a tag of the object, acquire the object through the tag, load a model through json files, set the position and the rotation angle of the object in the world, change all material patterns of the object, judge whether the loading of the object is completed, clone and acquire the component of the object, and finally construct a virtual environment. Building a virtual environment using modeling software and a unit platform may refer to prior art content.
The SVM classification model generates a corresponding repair scheme based on the fault data. And (3) manually cleaning the maintenance scheme given by the system, such as manually deleting or adding the content of the generated maintenance scheme and adjusting the sequence, and combining the real situation to obtain the adjusted maintenance scheme.
The equipment simulation maintenance system is provided with an interface module and an operation flow, wherein the interface module is used for displaying a maintenance scheme.
As one possible implementation, the interface module has the main functions and operational portals of the presentation system.
As one possible implementation, the interface module has a three-dimensional model of the display equipment, supporting multiple interaction modes such as rotation, scaling, movement, etc., so that the user can view the various parts and components of the equipment.
As one possible implementation, the interface module may display the failure phenomenon and possible cause and give corresponding repair advice and solutions.
As one possible implementation, the interface module may demonstrate steps and operations during maintenance, including disassembly, replacement, debugging, and the like.
As a possible implementation manner, the interface module may display information such as an operation state of the equipment, a maintenance record, fault distribution, and the like, and support the form of charts and reports.
As one possible implementation, the interface module may provide system setup functionality, including language, font size, etc., so that a user can customize the appearance and style of the interface.
As one possible implementation, the interface module may provide system assistance and instructions for use, as well as a way to contact technical support.
The attribute data for the maintenance scenario presented by the interface is as follows in table 4:
TABLE 4 Table 4
The system is utilized for maintenance exercise, and the virtualized operation and maintenance exercise of equipment are carried out through virtual simulation, so that the maintenance efficiency and quality are improved. The maintenance personnel can use the virtual simulation model to carry out maintenance exercise and carry out maintenance operation according to feedback and guidance of the virtual simulation model.
By utilizing the system to verify the repair effect and update the repair record, the actual repair condition is compared with the predicted result, the repair effect is verified, the repair record and the data set are updated, and more accurate reference is provided for subsequent repair. For example, the repaired equipment may be monitored and validated using the sensor data to ensure that the repair effect is expected.
Those of ordinary skill in the art will appreciate that: the modules in the system in the embodiments may be distributed in the system in the embodiments according to the embodiment description, or may be located in one or more systems different from the present embodiment with corresponding changes. The modules of the above embodiments may be combined into one module, or may be further split into a plurality of sub-modules.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. An artificial intelligence-based equipment simulation maintenance method is characterized by comprising the following steps:
step S100, collecting data in the running process of equipment, and preprocessing the data to obtain an equipment data set, wherein the data are physical data of the equipment and data reflecting the running state of each module of the equipment;
step S200, carrying out fault class classification on the received equipment data set based on an SVM classification model to obtain fault data of a certain module, wherein the fault data reflects fault problems encountered by the certain module;
and step S300, the SVM classification model generates a corresponding maintenance scheme based on the fault data.
2. The artificial intelligence based equipment simulation maintenance method of claim 1, further comprising, prior to step S200:
establishing an SVM classification model;
and training the SVM classification model.
3. The artificial intelligence based equipment simulation maintenance method of claim 1, wherein the training the SVM classification model further comprises:
acquiring pre-training data, wherein the pre-training data comprises data in the running process of equipment, fault records, maintenance records or experience and knowledge of experts in the field of maintenance systems, and second running data acquired when the equipment simulates different fault conditions in a virtual environment;
selecting data influencing fault class classification from the pre-training data as training data;
and training the SVM classification model by using the training data.
4. The artificial intelligence based equipment simulation maintenance method according to claim 1, further comprising step S400 of manually adjusting a maintenance scheme corresponding to the generation of the fault data, wherein the maintenance scheme includes a fault cause, a detailed maintenance step, a maintenance step combination of maintenance related parts, a maintenance operation guidance flow, and a maintenance required tool operation method.
5. The artificial intelligence based equipment simulation maintenance method of claim 1, wherein step S100 further comprises: and storing the data acquired in the running process of the equipment in a network module capable of remotely acquiring or sharing the data.
6. The artificial intelligence based equipment simulation maintenance method of claim 1, wherein the fault data includes a number of a certain module of the equipment, a maintenance tool, maintenance contents, a general maintenance procedure, maintenance materials, a set of parts involved in maintenance, and a maintenance status.
7. The method and system for artificial intelligence based equipment simulation maintenance according to claim 1, wherein step S500 is performed by displaying a maintenance plan corresponding to the generated fault data through an interface.
8. The artificial intelligence based equipment simulation maintenance method of claim 3, wherein the equipment simulates second operational data collected when different fault conditions are simulated in a virtual environment, further comprising: the virtual environment is a simulation model environment established according to the characteristic information of the equipment, wherein the characteristic information comprises the physical structure, the kinematic characteristic, the dynamic characteristic, the control system, the sensor and the structural characteristics of each part of the equipment.
9. The artificial intelligence based equipment simulation maintenance method of claim 1, wherein the equipment data set further comprises: equipment model, installation location, maintenance record, operating condition, equipment weight, external dimensions, equipment material, equipment internal construction picture, equipment external picture, wherein the operating condition includes: operating time, operating speed, operating temperature, operating pressure, operating voltage, and operating current.
10. An artificial intelligence-based equipment simulation maintenance system is characterized in that,
the data acquisition module is used for acquiring data in the running process of the equipment and preprocessing the data to obtain an equipment data set, wherein the data are physical data of the equipment and data reflecting the running state of each module of the equipment;
the algorithm module is used for classifying the fault categories of the received equipment data set based on the SVM classification model to obtain fault data of a certain module, and the fault data reflects fault problems encountered by the certain module; and the SVM classification model generates a corresponding maintenance scheme based on the fault data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310673785.8A CN116934293A (en) | 2023-06-07 | 2023-06-07 | Equipment simulation maintenance method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310673785.8A CN116934293A (en) | 2023-06-07 | 2023-06-07 | Equipment simulation maintenance method and system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116934293A true CN116934293A (en) | 2023-10-24 |
Family
ID=88383409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310673785.8A Pending CN116934293A (en) | 2023-06-07 | 2023-06-07 | Equipment simulation maintenance method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116934293A (en) |
-
2023
- 2023-06-07 CN CN202310673785.8A patent/CN116934293A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113228100A (en) | Imaging modality intelligent discovery and maintenance system and method | |
| Henriksson et al. | Towards structured evaluation of deep neural network supervisors | |
| US20190304322A1 (en) | Performance metrics in an interactive computer simulation | |
| EP3547288A1 (en) | Assessing a training activity performed by a user in an interactive computer simulation | |
| US20220366244A1 (en) | Modeling Human Behavior in Work Environment Using Neural Networks | |
| Kuravsky et al. | Detecting abnormal activities of operators of complex technical systems and their causes basing on wavelet representations | |
| JP2020117209A (en) | Methods and systems for performing aircraft maintenance | |
| CN112420125A (en) | Molecular attribute prediction method and device, intelligent equipment and terminal | |
| CN113253709B (en) | Health diagnosis method and device suitable for rail transit vehicle | |
| Karaoğlu et al. | Applications of machine learning in aircraft maintenance | |
| CN117319223A (en) | Digital twinning technology-based pilot cockpit visualization method and system | |
| Ozbey et al. | The determination and analysis of factors affecting to student learning by artificial intelligence in higher education | |
| CN117132763A (en) | Power image anomaly detection method, device, computer equipment and storage medium | |
| Ulrich et al. | Operator timing of task level primitives for use in computation-based human reliability analysis | |
| CA3235482A1 (en) | System and method for predicting performance by clustering psychometric data using artificial intelligence | |
| Li et al. | Long short-term memory network-based user behavior analysis in virtual reality training system—a case study of the ship communication and navigation equipment training | |
| CN116934293A (en) | Equipment simulation maintenance method and system based on artificial intelligence | |
| US11715387B2 (en) | Standard operating procedures feedback during an interactive computer simulation | |
| Amico et al. | Critical issues in simulation | |
| Susarev et al. | Training simulators development technique for oil and gas industry automation control systems | |
| Groth | A framework for using SACADA to enhance the qualitative and quantitative basis of HRA | |
| Lyu et al. | Optimal burn-in strategy for high reliable products using convolutional neural network | |
| Wang | Application of computer images in virtual simulation technology-apparel as an example | |
| AU2023204030B2 (en) | Federated machine learning in adaptive training systems | |
| WO2023186317A1 (en) | System, apparatus and method for managing one or more assets |
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 |