CN113919230A - Modeling evaluation method and system for operation of complex equipment - Google Patents
Modeling evaluation method and system for operation of complex equipment Download PDFInfo
- Publication number
- CN113919230A CN113919230A CN202111230046.9A CN202111230046A CN113919230A CN 113919230 A CN113919230 A CN 113919230A CN 202111230046 A CN202111230046 A CN 202111230046A CN 113919230 A CN113919230 A CN 113919230A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- equipment
- algorithm
- model
- algorithm model
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 92
- 238000000034 method Methods 0.000 claims abstract description 53
- 238000004519 manufacturing process Methods 0.000 claims abstract description 52
- 238000013209 evaluation strategy Methods 0.000 claims abstract description 37
- 238000007726 management method Methods 0.000 claims abstract description 31
- 238000004458 analytical method Methods 0.000 claims abstract description 30
- 238000010801 machine learning Methods 0.000 claims abstract description 11
- 238000005516 engineering process Methods 0.000 claims abstract description 8
- 238000005457 optimization Methods 0.000 claims description 56
- 238000012549 training Methods 0.000 claims description 33
- 230000008569 process Effects 0.000 claims description 13
- 238000003745 diagnosis Methods 0.000 claims description 10
- 238000013210 evaluation model Methods 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 2
- 238000011157 data evaluation Methods 0.000 claims description 2
- 230000008676 import Effects 0.000 claims description 2
- 238000004088 simulation Methods 0.000 claims description 2
- 230000007613 environmental effect Effects 0.000 abstract description 3
- 230000010354 integration Effects 0.000 abstract description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 230000004927 fusion Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013440 design planning Methods 0.000 description 1
- 238000005538 encapsulation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Manufacturing & Machinery (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
In order to solve the problems of low generalization, low accuracy and the like in the existing modeling method, the invention provides a modeling evaluation method for complex equipment operation based on machine learning. Aiming at the defects in the existing analysis model technology, the modeling evaluation method for the operation of the complex equipment is provided, multiple common algorithms of machine learning are subjected to universality and integration fusion to form a basic algorithm model base for the evaluation of the operation state of the equipment, and meanwhile, a management UI (user interface) for the algorithm model base, an evaluation strategy and the like is designed. The algorithm library takes the production parameters, the production mode and the environmental parameters of the complex equipment as input, provides data input for the model, integrates all algorithm results to vote on the analysis result, analyzes and evaluates the result for the running state of the complex equipment, and evaluates and optimizes the model of the algorithm model library. Enterprises can use the algorithm library to accurately evaluate the running state of the complex equipment, and effectively manage the equipment.
Description
Technical Field
The invention belongs to the technical field of intelligent manufacturing, and particularly relates to a modeling evaluation method and system for complex equipment operation.
Background
In the face of a new turn of intelligent manufacturing wave in the world, in order to improve product research and development efficiency, reduce research and development cost and quickly respond to markets, intelligent manufacturing and digital twin are used as the core of the change of manufacturing industry in high-end manufacturing fields such as aviation and aerospace in China, intelligent modeling analysis of equipment, workshops and factories is used as a hot spot in the digital twin, and the intelligent modeling analysis is used as the core in the whole life cycle of the production process to provide production strategies and intelligent management for enterprises. The method comprises the steps of establishing an operation state evaluation model of the power equipment by adopting a clustering-based equipment operation state evaluation method in the technical field of power automation, completing the establishment of an analysis model of the power equipment by taking historical data of the power equipment as a basis, and evaluating quantitative analysis, demonstration and comparison of the operation state of the power automation equipment through the model so as to realize the operation state evaluation of the power automation equipment; a method and a system for evaluating the state of a marine engine are established based on digital twins in the field of ship manufacturing, a corresponding digital twins model is established according to the parameter transfer relationship among a working process model, a fault diagnosis model and a structural strength finite element model, the input of the digital twins model is an operation parameter and an environment parameter, the model finally outputs a corresponding evaluation parameter, and the operation state of the marine engine is finally determined. A digital twin-driven intelligent line control chassis system and a fault diagnosis method thereof are provided in the aspect of automobile manufacturing, a digital twin model system of a line control chassis device is established through a digital twin technology, a virtual model and a physical entity are enabled to run consistently based on data driving, and real-time data interaction and fault diagnosis of the line control chassis device and an analysis model are realized.
Intelligent manufacturing and digital twin are taken as important directions of transformation development of the current manufacturing industry, enterprises expect to solve the intelligent management problem in the actual production process through modeling and analyzing of workshops, production lines and equipment, and assist managers to make decisions through mass data modeling and analyzing results in the production process. In the digital twin system, real equipment is mapped to virtual modeling for real-time driving based on data driving, and real-time interaction of data is realized. However, when model analysis is performed on equipment, a large amount of invalid data is often generated during initial construction, which causes extremely low accuracy of an actual application model trained by the model and also fails to accurately evaluate the real-time state of the equipment; secondly, at present, a fixed algorithm model is often established after feature mining for data analysis modeling of a device, or a model based on artificial experience knowledge, and the model established in the early production stage can influence the cost control of an enterprise along with the difficulty in accurately identifying the running state of the device after the later production task and mode adjustment of the enterprise; in addition, after the production mode of the equipment is adjusted, the fault type or the running state of the equipment is increased frequently, the traditional customized algorithm analysis model cannot be adjusted in a self-adaptive mode, and the risk of safe production of the equipment is increased.
Before actual production of equipment, modeling analysis is an important component for realizing digital twin of an enterprise intelligent production line and is a necessary link for providing a relatively ideal production scheme and increasing the core competitiveness of the enterprise. In recent years, industrial equipment such as industry, metallurgy and the like has a trend of large-scale and complex development, the complex equipment cannot be managed and monitored by establishing an accurate physical model completely by a traditional method, and in addition, a large amount of reaction operation mechanisms and real-time equipment state data are generated in the production process of the equipment. Based on factors such as production optimization and technology, for better utilization, an analysis method based on a statistical analysis model and qualitative experience knowledge is formed. However, these analysis methods have the following problems: 1) data characteristics of external factors are not combined into the model, so that the generalization of the model is low; 2) a large amount of historical data is needed for establishing an analysis model, and the requirement of the data size of model training cannot be met before equipment is produced and operated, so that the accuracy rate of the model is low; 3) the used algorithm model is fixed, and the situation that the device is not suitable for actual production exists after the device is operated on line.
Disclosure of Invention
In order to solve the problems of low generalization, low accuracy and the like in the existing modeling method, the invention provides a modeling evaluation method for complex equipment operation based on machine learning. Aiming at the defects in the existing analysis model technology, the modeling evaluation method for the operation of the complex equipment is provided, multiple common algorithms of machine learning are subjected to universality and integration fusion to form a basic algorithm model base for the evaluation of the operation state of the equipment, and meanwhile, a management UI (user interface) for the algorithm model base, an evaluation strategy and the like is designed. The algorithm library takes the production parameters, the production mode and the environmental parameters of the complex equipment as input, provides data input for the model, integrates all algorithm results to vote on the analysis result, analyzes and evaluates the result for the running state of the complex equipment, and evaluates and optimizes the model of the algorithm model library. Enterprises can use the algorithm library to accurately evaluate the running state of the complex equipment, and effectively manage the equipment.
The invention is realized by the following technical scheme: the modeling evaluation method for the operation of the complex equipment is provided, an object-oriented modeling mode is adopted for a general processing mode of a complex equipment multi-subsystem of a digital intelligent workshop production line, and the multi-modeling evaluation method is mainly based on four parts of basic equipment resources, evaluation strategies, algorithm model base optimization, evaluation and decision.
After the above four parts are constructed, the method is combined with a complex equipment production operation flow, and the method can be realized by summarizing the following specific process that basic equipment resources firstly read equipment production parameters and external environment parameter information, then a plurality of corresponding models are selected for each subsystem of the complex equipment from model library resources through data cleaning, algorithm evaluation strategy parameters are called from evaluation strategies, and accuracy and generalization evaluation are carried out on the algorithm models of each subsystem. When an algorithm model needs to be optimized and set by an administrator using a development model or an evaluation strategy needs to be reset, a management UI interface for an algorithm model base and the evaluation strategy is provided in the model, and by setting parameters such as the strategy on the management UI, model retraining optimization and self-adaptation can be rapidly realized through multithreading execution, the state evaluation algorithm model of each subsystem is optimally bound, and then real-time state evaluation service of the whole equipment is formed again.
Basic device resources: the device process parameters are based on material information, device process information and resource information in a complex system, a unit base model, an algorithm base evaluation strategy, an algorithm optimization strategy, an evaluation and analysis basic model unit, an external information interface and the like are combined to form a multi-subsystem, multi-algorithm model, a multi-evaluation strategy and other multi-level reconfigurable models, the interior of the multi-subsystem, multi-algorithm model, multi-evaluation strategy and other multi-level reconfigurable models is in a modularized packaging mode to form a new multi-level algorithm model base of the complex device, the input of the multi-subsystem, multi-algorithm model, multi-evaluation strategy and other multi-level reconfigurable models is reconstructed feature data, the multi-subsystem, multi-algorithm model, multi-evaluation strategy and other multi-level reconfigurable models is directly and quickly mapped into the basic model resources, and state evaluation is carried out by taking the new model parameters as the input of production modeling.
Optimizing an algorithm model library: the method mainly comprises the steps of optimizing newly-added faults, working states and accuracy, wherein the strategies of newly-added faults and working states are optimized by retraining an original model or combining a plurality of algorithms, the method consists of four types of modules including fault label classification, historical training data cleaning, algorithm model training interfaces and algorithm model accuracy verification, each module is related to each basic unit control by calling a corresponding interface method inside and is packaged inside model base resources, and the modules cooperate with basic model resources to be used as the input of modeling analysis and evaluation; and optimizing the accuracy of the model, and improving the generalization by performing cross validation on the cleaned training data.
The algorithm evaluates the strategy: on the basis of an original system, basic equipment resource management and an algorithm model are combined to optimize a UI, an interface calling method is adopted, information such as strategy parameters and the like on the management UI is input into a multi-level algorithm model library of multiple subsystems, an integrated optimization system framework is configured, the integrated UI interface can be called according to optimization requirements of different users by performing optimization parameter setting, optimization strategy setting, strategy evaluation setting, optimal solution checking and original system detailed setting, and the optimal algorithm model and the evaluation strategy of each subsystem are rapidly calculated in a final equipment model through a multithreading technology.
State evaluation and decision: after the complex equipment is modeled, the comprehensive consideration is carried out on the aspects of process feasibility, system rationality, algorithm model applicability and the like, the comprehensive training of an algorithm model base of the complex equipment is carried out and is used as a model for evaluation and decision, after the training is finished, the accuracy of the analysis result of each algorithm model of each subsystem is verified by adopting a cross verification method through input data in basic model resources, the evaluation result of the operation state of each subsystem, such as the operation states of faults, idleness, operation, overload and the like, is provided for an integrated learning algorithm interface during equipment evaluation and is used for real-time operation state evaluation and fault diagnosis of the complex equipment, if the evaluation result does not meet the actual condition, the target algorithm model optimization is required until the accuracy of the evaluation result meets the production, automatic deployment can be carried out, and an evaluation report is provided for a user.
In summary, compared with the prior art, the invention has the following advantages and beneficial effects: the invention relates to a machine learning-based multi-level and multi-modeling integrated design method, which has the key points and points to be protected as follows: firstly, a machine learning basic algorithm model, an algorithm optimization strategy, an evaluation optimization strategy and the like are combined to form a multi-level multi-model algorithm model base, basic model resources and a control strategy are split into basic ideas, basic data resources are introduced, matched modeling objects are selected from the model base and the evaluation strategy respectively to realize multi-level one-to-one mapping association, and the generalization modification of an analysis model due to equipment change is completed efficiently. Secondly, the model is developed in a modular form, interacts with the integrated UI through an interface, is integrated through an evaluation strategy, and is combined with a multi-model optimization algorithm, so that multi-level multi-model optimization training can be quickly set and realized on the integrated UI framework.
According to the design planning requirements of different complex equipment, one-to-one mapping can be carried out based on a model base, an evaluation strategy and the like, meanwhile, an algorithm model has strong generalization and expansion capability, multi-level and multi-model analysis evaluation required by enterprise users is realized, and the flourishing property and high accuracy of modeling analysis are enhanced. The modularized form and the interface development of the integrated UI avoid the problems of errors and low efficiency in the original model library method, integrally solve the forward-looking problem before the equipment is put into production, effectively control the problem of model adjustment analysis brought by later-stage production, and greatly save the model adjustment time after the equipment is put into production.
In conclusion, the modeling analysis method effectively solves the problems of low generalization, low accuracy and the like in the existing modeling analysis process, reasonably, quickly and efficiently trains the model resource allocation of each subsystem model of the equipment, optimizes the operation through the multithreading technology, and has higher significance and practical use value.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention.
FIG. 1 is a flow chart of an embodiment modeling integration design method.
Fig. 2 is a schematic diagram of an embodiment of an underlying data resource principle.
FIG. 3 is a schematic diagram of an algorithm model optimization strategy principle according to an embodiment.
FIG. 4 is a schematic diagram of an embodiment device evaluation policy.
Fig. 5 is a first schematic diagram of the embodiment.
Fig. 6 is a second schematic diagram of the embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples and accompanying drawings, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not meant to limit the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be apparent to one of ordinary skill in the art that: it is not necessary to employ these specific details to practice the present invention. In other instances, well-known structures, circuits, materials, or methods have not been described in detail so as not to obscure the present invention.
Throughout the specification, reference to "one embodiment," "an embodiment," "one example," or "an example" means: the particular features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment of the invention. Thus, the appearances of the phrases "one embodiment," "an embodiment," "one example" or "an example" in various places throughout this specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, or characteristics may be combined in any suitable combination and/or sub-combination in one or more embodiments or examples. Further, those of ordinary skill in the art will appreciate that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
In the description of the present invention, it is to be understood that the terms "front", "rear", "left", "right", "upper", "lower", "vertical", "horizontal", "high", "low", "inner", "outer", etc. indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and therefore, are not to be construed as limiting the scope of the present invention.
The embodiment provides a modeling evaluation method for complex equipment operation, which aims at a general processing mode of a complex equipment multi-subsystem of a digital intelligent workshop production line and adopts an object-oriented modeling mode, wherein the multi-modeling evaluation method is mainly based on four parts, namely basic equipment resources, an evaluation strategy, algorithm model base optimization, evaluation and decision, and is shown in figure 1.
After the above four parts are constructed, the method is combined with a complex equipment production operation flow, and the method can be realized by summarizing the following specific process that basic equipment resources firstly read equipment production parameters and external environment parameter information, then a plurality of corresponding models are selected for each subsystem of the complex equipment from model library resources through data cleaning, algorithm evaluation strategy parameters are called from evaluation strategies, and accuracy and generalization evaluation are carried out on the algorithm models of each subsystem. When an algorithm model needs to be optimized and set by an administrator using a development model or an evaluation strategy needs to be reset, a management UI interface for an algorithm model base and the evaluation strategy is provided in the model, and by setting parameters such as the strategy on the management UI, model retraining optimization and self-adaptation can be rapidly realized through multithreading execution, the state evaluation algorithm model of each subsystem is optimally bound, and then real-time state evaluation service of the whole equipment is formed again.
According to the complex equipment real-time state identification and evaluation modeling method based on machine learning, based on data such as field equipment real-time production parameters, environmental temperature and humidity factors and workshop energy, modeling is respectively carried out on different subsystems of complex equipment from each subsystem module, multiple production modes and energy control key control points of the complex equipment, evaluation is carried out according to different production modes, fault information prediction and operation state identification of the equipment are achieved, equipment faults are reduced to the maximum extent, safe production of the equipment is guaranteed, and intelligent operation and maintenance management and control capacity is improved, and the method is shown in fig. 5-6.
Basic data resources: the basic data resources are used as important sources of the complex equipment algorithm model base data, data support is provided for the establishment of the whole analysis model, the basic data resources mainly comprise four parts, namely a basic data table, an evaluation strategy, an algorithm model optimization strategy and an external data interface quantity, and a basic data resource principle schematic diagram is shown in FIG. 2. Connecting a database through configuration of ODBC (open data base), remotely reading various information in the database by a program, reading operation parameters of all equipment subsystem algorithm models, reading historical production parameters and external environment parameters in the database, and performing model training by combining various machine learning algorithms; classifying various working states of the equipment and associating the various working states with multi-level and multi-model in the model library resources; editing equipment evaluation strategy configuration, configuring initialization, reset and simulation finished api methods, performing real-time data evaluation through optimal evaluation models of all subsystems in an algorithm model library before providing real-time state evaluation service, and performing initialization calling to realize import and binding of basic model resources; after the algorithm analysis model is trained, data collection and check of corresponding statistical results can be performed through a statistical table in the data table.
Algorithm model optimization strategy: the method mainly comprises the steps of optimizing the accuracy of a newly added working state mode, a newly added equipment fault type and an algorithm model. The newly added working state mode of the device can adopt the working state types which are not identified by the model and are reserved when the training mode is started or the original historical data and the new working state mode are combined and the data are re-labeled, the data in the new working state of the device are added into the training data to carry out optimization training on the existing model, for example, the new working mode is added on a device model library management UI interface, the data in the new mode are imported, and the characteristics of the mode are obtained again to carry out training. And secondly, aiming at the fault type of the newly-added equipment, the new fault type can be generally and directly positioned to which subsystem fails, the fault data and the historical data of the subsystem can be imported for optimal training after marking, and the accuracy of the subsystem in judging the fault type is improved. And finally, the accuracy optimization of the algorithm model can set accurate optimization configuration information through the equipment model library management UI, wherein the optimization training, the timing training and the quantitative training can be set when an equipment state recognition program is idle, so that the higher accuracy of the model on the recognition state of the equipment can be kept, and the accuracy of the algorithm model of a certain subsystem can be improved through a mode of model combined verification on the operation state and the fault diagnosis of the subsystem. After the UI setting is finished through the algorithm model library management, strategies are loaded to the background of the equipment state recognition program, and the model optimization tasks are automatically and quickly finished through the multithreading technology. The schematic diagram of the algorithm model optimization strategy is shown in fig. 3.
The algorithm evaluates the strategy: on the UI interface of the equipment library model management, a webapi interface method is adopted in combination with a basic data resource management interface, a complex equipment multi-model modeling evaluation program based on machine learning is compiled based on C # language, a type object is formed after encapsulation, a WPF is adopted to develop and configure a framework of the model library configuration management system, and the framework mainly comprises a UI module of the management interface and an algorithm library model optimization module, as shown in FIG. 4. The model library management UI module comprises a configuration process for initializing an equipment state identification and fault diagnosis algorithm library and the like, after initialization, the operation is carried out by loading an analysis model and setting model accuracy, and if the accuracy requirement in an evaluation strategy is met, the equipment state analysis result of an optimal model can be directly output; meanwhile, an evaluation is set as an integrated algorithm comprehensive evaluation strategy, and at the moment, the model base of the equipment integrates the evaluation results and accuracy rate information of all subsystems to carry out comprehensive evaluation on the state and faults of the equipment, so that an api interface of the evaluation results is formed and is provided for an evaluation and decision module to use.
The optimization and evaluation strategies can be packaged with the model library objects, can be embedded in the entity model as user-defined, but the modification and adjustment workload of multiple models of different complex equipment is increased, and the optimization and evaluation strategies do not have good flexibility.
State evaluation and decision: the method comprises the steps that a real-time equipment state recognition program starts to run through model pre-training, a user adopts a management control interface developed by WPF to set up an evaluation and decision integrated UI control frame, a control and a model optimization interface are bound with each other, the model optimization program is compiled in an interface method, when a viewing control is provided, a data source in a statistical tool is indirectly called to carry out graphic output display, the reasonability and manufacturability of statistical data such as equipment state recognition, fault diagnosis and safety production under the current model parameter setting are analyzed, a state recognition evaluation report of equipment model analysis is provided at the same time, the user can call an internal interface to view detailed attribute information of the report, namely, the equipment after the target optimization is safely produced by analyzing the recognition state information, the fault information and the like of the equipment after the model program runs, and an evaluation result is used as an important reference of a new multi-level multi-model, and if the actual state information of the equipment is not met, selecting a target model according to decision evaluation for re-optimization training.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A modeling evaluation method for complex equipment operation is characterized in that an equipment operation state evaluation model comprising basic equipment resources, an equipment algorithm model library, algorithm evaluation, algorithm model library optimization, state evaluation and decision making parts is constructed;
based on the evaluation model, basic equipment resources firstly read equipment production parameters and external environment parameter information and then are cleaned through data;
selecting a plurality of corresponding algorithm models for each subsystem of the complex equipment from an equipment algorithm model library;
calling algorithm evaluation parameters from the algorithm evaluation strategy, and carrying out accuracy and generalization evaluation on the algorithm model of each subsystem;
when algorithm model base optimization setting or algorithm evaluation parameter resetting is needed, the equipment algorithm model base provides a management UI for algorithm model base optimization and algorithm evaluation parameters, algorithm model retraining optimization and self-adaptation are realized through the management UI, and state evaluation algorithm models of all subsystems are optimally bound;
thereby constituting a real-time status evaluation service of the entire apparatus.
2. The modeling and evaluating method for the operation of the complex equipment according to claim 1, wherein in the basic equipment resources, the equipment production parameters comprise equipment process parameters and equipment control program parameters, and the equipment process parameters comprise material information, equipment process information and resource information in the complex system;
inputting and mapping the characteristic data after the basic equipment resource is reconstructed into an equipment algorithm model library; and if the algorithm model base is optimized, performing state evaluation according to the optimized algorithm model as input.
3. The modeling and evaluation method for the operation of the complex equipment according to claim 2, wherein the evaluation model basic data resources comprise basic equipment resources, algorithm evaluation strategies and algorithm model base optimization strategies, a data interface is connected with the equipment algorithm model base through a data interface, and the basic data resource database is connected through configuration of ODBC;
remotely reading the operation parameters of each subsystem algorithm model of the equipment, remotely reading historical production parameters and external environment parameters, and performing algorithm model training by combining various machine learning algorithms; classifying various working modes of the equipment, and performing multi-level and multi-model association with the algorithm models in the equipment algorithm model library;
before providing equipment real-time state evaluation service, editing algorithm evaluation strategy configuration, performing real-time data evaluation on an optimal evaluation model of each subsystem in an algorithm model library through an api method for configuring initialization, reset and simulation termination, and performing initialization calling to realize import and binding of basic algorithm model resources;
and after the algorithm model is trained, collecting and checking data of corresponding statistical results through a statistical table in the basic equipment resources.
4. The method of claim 3, wherein the optimization of the algorithmic model library comprises optimizing for newly added fault types, newly added operating modes, and accuracy;
the optimization strategy for the newly added fault type and the newly added working mode is realized by retraining the original algorithm model or combining a plurality of algorithms; fault label classification, historical training data cleaning, algorithm model training interface and algorithm model accuracy verification modules, calling corresponding interface methods through the inside, associating with basic unit controls, encapsulating inside algorithm model library resources, and performing cooperation with basic model resources as input of modeling analysis and evaluation;
and optimizing the accuracy of the algorithm model, and improving the generalization by performing cross validation on the cleaned training data.
5. The modeling and evaluating method for the operation of the complex equipment according to claim 4, characterized in that, aiming at the algorithm model base optimization strategy of the newly added working mode, after the original historical data and the new working mode are combined and the data are re-labeled, the algorithm models of the equipment subsystem are retrained; or, reserving the work mode type which is not identified by the model when the algorithm model training mode starts, adding data under the new work mode of the equipment into the training data, and carrying out optimization training on the existing algorithm model; adding a new working mode on a device algorithm model library management UI interface, importing data in the new working mode, and acquiring the characteristics of the working mode again for training;
aiming at the newly added fault type of the equipment, directly positioning a subsystem generating the new fault type, importing fault data and historical data of the subsystem, and performing optimization training after marking;
aiming at the accuracy optimization of the algorithm model, the accurate optimization configuration information is set through the equipment model library management UI, the equipment state recognition program is set to carry out optimization training, timing training and quantitative training when idle, and the accuracy of the algorithm model of the subsystem is jointly verified through the algorithm model.
6. The modeling and evaluating method for the operation of the complex equipment according to claim 5, wherein the algorithm evaluating strategy optimizes the UI by combining basic equipment resources and the algorithm model on the basis of the original algorithm model and adopting a method of calling an interface to input information such as strategy parameters on the management UI into a multi-level algorithm model library of a multi-subsystem and configure an integrated optimization evaluating model framework;
according to optimization requirements of different users, optimization parameter setting, optimization strategy setting, evaluation strategy setting, optimal solution checking and detailed setting of an original evaluation model are carried out, an integrated UI interface is called, and an optimal algorithm model and an evaluation strategy of each subsystem are calculated in a final evaluation model through a multithreading technology.
7. The modeling assessment method of complex plant operation of claim 6,
an algorithm evaluation strategy is based on an equipment library model management UI, combines a basic data resource management interface, adopts a webapi interface method form, and writes a machine learning-based complex equipment multi-algorithm model modeling evaluation program based on C # language;
the method comprises the steps that a framework of an algorithm model base configuration management system is developed and configured by adopting WPF, the framework is composed of a UI module of a management interface and an algorithm model base optimization module, the UI module is managed in an initialization configuration process of an equipment state identification and fault diagnosis algorithm base and the like, an algorithm model is loaded after initialization, and operation is carried out after the algorithm model is set accurately;
if the accuracy requirement in the evaluation strategy is met, directly outputting the equipment state evaluation result of the optimal algorithm model; meanwhile, an evaluation mode is set to be an integrated multi-level algorithm comprehensive evaluation strategy, at the moment, an algorithm model base of the equipment integrates evaluation results and accuracy rate information of all subsystems, comprehensive evaluation is carried out on the state and the fault of the equipment, and an api interface with the evaluation results is provided for an evaluation and decision module to use.
8. The modeling and evaluating method for the operation of the complex equipment according to claim 7, characterized in that the state evaluation and decision are considered overall in all aspects of process feasibility, system rationality and algorithm model applicability, and the algorithm model library of the complex equipment is comprehensively trained to be used as a model for the state evaluation and decision;
after training is finished, verifying the accuracy rate of each algorithm model analysis result of each subsystem by adopting a cross verification method through input data of basic model resources;
the evaluation result of the running state of each subsystem, including fault, idle, running and overload running states, is provided for an integrated algorithm model interface during equipment evaluation and is used for real-time running state evaluation and fault diagnosis of complex equipment;
and if the actual situation is not met, the target algorithm model needs to be optimized, automatic deployment can be carried out until the accuracy of the evaluation result meets the production, and an evaluation report is provided for a user.
9. The modeling and evaluating method for the operation of the complex equipment according to claim 8, characterized in that the state evaluation and decision-making starts to operate the real-time equipment state recognition program of the evaluation model through algorithm model pre-training, an evaluation and decision-making integrated UI control frame is built by adopting a WPF developed management control interface, and a control and a model optimization interface are bound with each other;
an algorithm model optimization program is compiled in the interface method, and when a viewing control is provided, a data source in a statistical tool is indirectly called to carry out graphic output display;
analyzing the reasonability and manufacturability of statistical data of equipment state identification, fault diagnosis and safety production under the condition of the algorithm model parameter setting, and providing an equipment state identification evaluation report;
checking the attribute information of the report by calling an internal interface, namely evaluating the identification state information and fault information of the equipment after the operation of a model program, carrying out safety production on the equipment after benchmarking optimization, and taking an evaluation result as the reference of a new algorithm model;
and if the actual state information of the equipment is not met, selecting a target algorithm model according to decision evaluation for re-optimization training.
10. A modeling evaluation system for complex equipment operation is characterized by comprising a basic equipment resource module, an equipment algorithm model library module, an evaluation strategy module, an algorithm model library optimization module and a state evaluation and decision module;
the basic equipment resource module is used for reading the production parameters of the equipment and the external environment parameter information and cleaning data;
the equipment algorithm model library module is used for selecting a plurality of corresponding algorithm models for each subsystem of the complex equipment;
the evaluation strategy module is used for calling algorithm evaluation parameters and carrying out accuracy and generalization evaluation on the algorithm model of each subsystem;
the algorithm model base optimization module is used for providing a management UI (user interface) for algorithm model base optimization and algorithm evaluation parameters through the equipment algorithm model base when the algorithm model base optimization setting or the algorithm evaluation parameter resetting is needed, realizing algorithm model retraining optimization and self-adaptation and optimally binding the state evaluation algorithm models of all the subsystems;
and the state evaluation and decision module is used for providing evaluation service for the real-time state of the whole equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111230046.9A CN113919230A (en) | 2021-10-20 | 2021-10-20 | Modeling evaluation method and system for operation of complex equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111230046.9A CN113919230A (en) | 2021-10-20 | 2021-10-20 | Modeling evaluation method and system for operation of complex equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113919230A true CN113919230A (en) | 2022-01-11 |
Family
ID=79242332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111230046.9A Pending CN113919230A (en) | 2021-10-20 | 2021-10-20 | Modeling evaluation method and system for operation of complex equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113919230A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114754973A (en) * | 2022-05-23 | 2022-07-15 | 中国航空工业集团公司哈尔滨空气动力研究所 | Wind tunnel force measurement test data intelligent diagnosis and analysis method based on machine learning |
| CN114912371A (en) * | 2022-06-13 | 2022-08-16 | 江苏大学 | Digital twin body modeling system and method for running stability detection of tubular pump |
| CN115290286A (en) * | 2022-10-10 | 2022-11-04 | 中国空气动力研究与发展中心高速空气动力研究所 | Integrated learning evaluation system for flow field stability of sub-span supersonic wind tunnel |
| CN116522088A (en) * | 2023-04-27 | 2023-08-01 | 哈尔滨工程大学 | Nuclear power plant operation data analysis method and system based on machine learning |
| CN116738767A (en) * | 2023-08-14 | 2023-09-12 | 中国恩菲工程技术有限公司 | Modeling method of process unit and generating method of process unit |
| TWI824700B (en) * | 2022-09-06 | 2023-12-01 | 中華電信股份有限公司 | An automated machine learning system, method and computer readable medium thereof |
| CN117171551A (en) * | 2023-11-02 | 2023-12-05 | 山东港口科技集团烟台有限公司 | Large-scale industrial equipment data analysis and intelligent management method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509178A (en) * | 2011-11-25 | 2012-06-20 | 江苏省电力公司淮安供电公司 | Distribution network device status evaluating system |
| WO2015113450A1 (en) * | 2014-01-29 | 2015-08-06 | 华为技术有限公司 | Method and apparatus for visualized operation and maintenance of network |
| CN107545112A (en) * | 2017-09-07 | 2018-01-05 | 西安交通大学 | Complex equipment Performance Evaluation and Forecasting Methodology of the multi-source without label data machine learning |
| CN108173671A (en) * | 2016-12-07 | 2018-06-15 | 博彦科技股份有限公司 | O＆M method, apparatus and system |
| CN109508848A (en) * | 2018-08-08 | 2019-03-22 | 武汉理工光科股份有限公司 | Enterprise's production safety risk assessment and management system |
| CN111144025A (en) * | 2019-12-31 | 2020-05-12 | 中国兵器装备集团自动化研究所 | Simulation model parameterization integration system and method based on multi-level reconstruction |
| CN112784501A (en) * | 2021-03-23 | 2021-05-11 | 中国核电工程有限公司 | Modeling system and method for residual life prediction model of equipment and prediction system |
| US20210271582A1 (en) * | 2018-06-28 | 2021-09-02 | Zte Corporation | Operation and maintenance system and method |
-
2021
- 2021-10-20 CN CN202111230046.9A patent/CN113919230A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509178A (en) * | 2011-11-25 | 2012-06-20 | 江苏省电力公司淮安供电公司 | Distribution network device status evaluating system |
| WO2015113450A1 (en) * | 2014-01-29 | 2015-08-06 | 华为技术有限公司 | Method and apparatus for visualized operation and maintenance of network |
| CN108173671A (en) * | 2016-12-07 | 2018-06-15 | 博彦科技股份有限公司 | O＆M method, apparatus and system |
| CN107545112A (en) * | 2017-09-07 | 2018-01-05 | 西安交通大学 | Complex equipment Performance Evaluation and Forecasting Methodology of the multi-source without label data machine learning |
| US20210271582A1 (en) * | 2018-06-28 | 2021-09-02 | Zte Corporation | Operation and maintenance system and method |
| CN109508848A (en) * | 2018-08-08 | 2019-03-22 | 武汉理工光科股份有限公司 | Enterprise's production safety risk assessment and management system |
| CN111144025A (en) * | 2019-12-31 | 2020-05-12 | 中国兵器装备集团自动化研究所 | Simulation model parameterization integration system and method based on multi-level reconstruction |
| CN112784501A (en) * | 2021-03-23 | 2021-05-11 | 中国核电工程有限公司 | Modeling system and method for residual life prediction model of equipment and prediction system |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114754973A (en) * | 2022-05-23 | 2022-07-15 | 中国航空工业集团公司哈尔滨空气动力研究所 | Wind tunnel force measurement test data intelligent diagnosis and analysis method based on machine learning |
| CN114912371A (en) * | 2022-06-13 | 2022-08-16 | 江苏大学 | Digital twin body modeling system and method for running stability detection of tubular pump |
| TWI824700B (en) * | 2022-09-06 | 2023-12-01 | 中華電信股份有限公司 | An automated machine learning system, method and computer readable medium thereof |
| CN115290286A (en) * | 2022-10-10 | 2022-11-04 | 中国空气动力研究与发展中心高速空气动力研究所 | Integrated learning evaluation system for flow field stability of sub-span supersonic wind tunnel |
| CN116522088A (en) * | 2023-04-27 | 2023-08-01 | 哈尔滨工程大学 | Nuclear power plant operation data analysis method and system based on machine learning |
| CN116522088B (en) * | 2023-04-27 | 2024-01-26 | 哈尔滨工程大学 | Nuclear power plant operation data analysis method and system based on machine learning |
| CN116738767A (en) * | 2023-08-14 | 2023-09-12 | 中国恩菲工程技术有限公司 | Modeling method of process unit and generating method of process unit |
| CN117171551A (en) * | 2023-11-02 | 2023-12-05 | 山东港口科技集团烟台有限公司 | Large-scale industrial equipment data analysis and intelligent management method |
| CN117171551B (en) * | 2023-11-02 | 2024-01-30 | 山东港口科技集团烟台有限公司 | Large-scale industrial equipment data analysis and intelligent management method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113919230A (en) | Modeling evaluation method and system for operation of complex equipment | |
| Friederich et al. | A framework for data-driven digital twins of smart manufacturing systems | |
| CN111144025B (en) | Simulation model parameterized integration system and method based on multi-level reconstruction | |
| CN101976067A (en) | Flexible assembly line management control experiment platform and management control experiment method | |
| CN101539772B (en) | Product lifecycle management method and apparatus | |
| CN108052020A (en) | A kind of Chemical Processes Simulation device towards intelligence manufacture | |
| Figueiras et al. | Big data provision for digital twins in industry 4.0 logistics processes | |
| CN113128041A (en) | Digital factory MES system development method based on digital twin virtual technology | |
| Qiu et al. | A structured adaptive supervisory control methodology for modeling the control of a discrete event manufacturing system | |
| CN112068522A (en) | Reconfigurable production line manufacturing control system, method, and medium | |
| CN109213105A (en) | A kind of reconfigurable device realizes restructural method and dcs | |
| Kalla et al. | Integrating Iot, AI, And Big Data For Enhanced Operational Efficiency In Smart Factories | |
| Yang et al. | Verification of intelligent scheduling based on deep reinforcement learning for distributed workshops via discrete event simulation. | |
| Haskovic et al. | Role of the adviser module in the hybrid assembly subordinating control structure | |
| CN113960964B (en) | Flexible flow shop production scheduling system based on simulation optimization | |
| CN109992251A (en) | A kind of software development scheme generation method of service-oriented variable | |
| Azadeh et al. | Performance improvement of a multi product assembly shop by integrated fuzzy simulation approach | |
| US20210116900A1 (en) | Production Control System | |
| Poudel et al. | An integrated framework for dynamic manufacturing planning to obtain new line configurations | |
| Fang et al. | Research on automatic flow-shop planning problem based on data driven modelling simulation and optimization | |
| Niggemann et al. | On the applicability of model based software development to cyber physical production systems | |
| Majstorović et al. | Industry 4.0 in Serbia-state of development | |
| Abdoli et al. | A target-driven framework for assessing the effectiveness of product-service-system development approaches at System of Systems level | |
| Wang et al. | Digital Twin Model Construction and Management Method of Workshop Based on Cloud Platform | |
| Stetter et al. | Geometric and kinetic digital twin of a body-in-white assembly system for virtual commissioning |
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 |