Search

CN-121980216-A - Industrial interconnection intelligent system oriented to energy equipment industry and application method thereof

CN121980216ACN 121980216 ACN121980216 ACN 121980216ACN-121980216-A

Abstract

The invention relates to the technical field of industrial Internet platforms, and particularly discloses an industrial Internet intelligent system oriented to the energy equipment industry and an application method thereof. The industrial interconnection intelligent system is based on a technical base of an industrial internet platform and comprises an asset center module, a model construction center module and a model arrangement center module, wherein the asset center module is used for carrying out standardized packaging on accessed data, mechanisms and algorithms, the model construction center module is used for constructing an algorithm model based on a mechanism model and a mechanism model based on the algorithm model, and the model arrangement center module is used for fusing an algorithm-mechanism model parallel operation mode and an algorithm-mechanism model serial operation mode. The method can effectively combine the capture capacity of the algorithm model to nonlinear complex characteristics and the deep understanding of the mechanism model to the physical process, thereby effectively improving the efficiency and reliability of the industrial model construction and the industrial model decision stage.

Inventors

  • WANG JIACHEN
  • LI JINHANG
  • ZHOU YANG
  • LI YANG
  • LI LEI
  • LI SHANXUAN

Assignees

  • 东方电气集团数字科技有限公司
  • 东方电气股份有限公司

Dates

Publication Date
20260505
Application Date
20260116

Claims (10)

  1. 1. An industrial interconnection intelligent system for the energy equipment industry is characterized in that: The industrial interconnection intelligent system is based on a technical base of an industrial internet platform and comprises an asset center module, a model construction center module and a model arrangement center module; The asset center module is used for carrying out standardized encapsulation on accessed data, mechanisms and algorithms; The model construction center module is used for constructing an algorithm model based on the mechanism model and a mechanism model based on the algorithm model; the model programming center module is used for fusing an algorithm-mechanism model parallel operation mode and an algorithm-mechanism model serial operation mode.
  2. 2. The energy equipment industry oriented industrial interconnection intelligent system of claim 1, wherein: The industrial internet platform has LaaS layers for providing resources including computing, networking and storage; The asset center module, the model construction center module and the model arrangement center module are PaaS layers based on the LaaS layers, and the PaaS layers are further provided with a resource scheduling engine, an asset management engine, a development environment management engine, a model construction engine and a model arrangement engine.
  3. 3. The industrial interconnect intelligent system for energy equipment industry according to claim 1 or 2, characterized in that: the model building center module builds an algorithm model based on the mechanism model, and under the mode, the mechanism model is used for simulating the actual production process, constructing simulation data of a training algorithm model and verifying the application effect of the algorithm model.
  4. 4. The industrial interconnect intelligent system for energy equipment industry according to claim 1 or 2, characterized in that: the model construction center module is used for constructing a mechanism model based on the algorithm model, and the algorithm model is used for predicting key undetermined parameters of the mechanism model by combining historical data under the mode so as to complete construction of the mechanism model.
  5. 5. The industrial interconnect intelligent system for energy equipment industry according to claim 1 or 2, characterized in that: And the model arrangement center module fuses the arranged algorithm-mechanism model parallel operation mode, under the mode, the algorithm model and the mechanism model predict the same input in an integrated learning mode, and output values are fused and weighted and then output as a final result.
  6. 6. The industrial interconnect intelligent system for energy equipment industry according to claim 1 or 2, characterized in that: And the model arrangement center module fuses the arranged algorithm-mechanism model serial operation mode, under the mode, the algorithm model firstly carries out one-stage calculation according to input and then outputs a corresponding result, and then the mechanism model carries out two-stage further judgment according to the mechanism rule and outputs a final result.
  7. 7. An application method of the industrial interconnection intelligent system according to any one of claims 1 to 6, wherein the application method includes an asset entry creation process, a model fusion construction task creation process, and a model fusion orchestration task creation process after logging on a platform; Wherein the asset entry creation process generates an open asset reuse service; The model fusion construction task creation process selects an available mechanism model/algorithm model from the open asset multiplexing service of the asset entry creation process; The model fusion orchestration task creation process selects available mechanism models and algorithm models from the open asset multiplexing services of the asset entry creation process.
  8. 8. The method of claim 7, wherein the asset entry creation process comprises the steps of: Step 1, creating asset items; step2, under the created asset entry, selecting a data asset, a mechanism asset or an algorithm asset; Step 3, uploading corresponding asset files; Step 4, configuring description information of the asset file; step 5, completing creation of the industrial model asset; And 6, opening the asset multiplexing service.
  9. 9. The method for applying the industrial interconnection intelligent system according to claim 7, wherein the model fusion construction task creation process comprises the following steps: Step a, creating a model fusion construction task; Step b, under the established model fusion construction task, selecting an algorithm model construction mode based on a mechanism model or a mechanism model construction mode based on the algorithm model; If the algorithm model construction mode based on the mechanism model is selected, executing the following steps c-1 to f-1; if a mechanism model construction mode based on an algorithm model is selected, executing the following steps c-2 to f-2; C-1, selecting an available mechanism model from the opened asset multiplexing service to multiplex to the task created in the step a; D-1, simulating actual industrial process construction simulation data based on the mechanism model introduced in the step c-1; step e-1, training an algorithm model based on the simulation data constructed in the step d-1; f-1, simulating an actual industrial process by using the mechanism model introduced in the step c-1, and testing and verifying the application effect of the algorithm model trained in the step e-1; c-2, selecting available algorithm models from the opened asset multiplexing service to multiplex the algorithm models into the task created in the step a; d-2, accessing input and output historical data related to undetermined key parameters of the mechanism model to be built; step e-2, predicting undetermined key parameters of the mechanism model to be built based on the algorithm model introduced in the step c-2 and the history data accessed in the step d-2; f-2, constructing a mechanism model based on the predicted value of the undetermined key parameter obtained in the step e-2; And h, completing the fusion construction of the model.
  10. 10. The method of claim 7, wherein the model fusion orchestration task creation process comprises the steps of: step A, creating a model fusion scheduling task; Step B, under the established model fusion scheduling task, selecting an algorithm-mechanism model parallel operation mode or an algorithm-mechanism model serial operation mode; if the algorithm-mechanism model parallel operation mode is selected, executing the following steps C-1 to E-1; If the algorithm-mechanism model serial operation mode is selected, executing the following steps C-2 to E-2; c-1, selecting available algorithm models and mechanism models from the opened asset multiplexing service to multiplex the available algorithm models and mechanism models into the task created in the step A; d-1, acquiring an output a1 of an algorithm model based on the algorithm model introduced in the step C-1 according to the original input; C-1, acquiring output a2 of a mechanism model based on the mechanism model introduced in the step C-1 according to the original input; e-1, carrying out fusion weighting on the model outputs a1 and a2 obtained in the step D-1; C-2, selecting available algorithm models and mechanism models from the opened asset multiplexing service to multiplex the available algorithm models and mechanism models into the task created in the step A; d-2, acquiring a one-stage output b1 based on the algorithm model introduced in the step C-2 according to the original input; E-2, acquiring a two-stage output b2 based on the mechanism model introduced in the step C-2 according to the one-stage output b1 acquired in the step D-2; and F, outputting a final result of the model fusion and arrangement workflow.

Description

Industrial interconnection intelligent system oriented to energy equipment industry and application method thereof Technical Field The invention relates to the technical field of industrial Internet platforms, in particular to an industrial Internet intelligent system oriented to the energy equipment industry and an application method thereof. Background The energy equipment industry is strategic and basic industry for providing technical equipment for the energy industry, relates to a plurality of fields such as high-end manufacturing, engineering technology and the like, and covers a plurality of production scenes such as research and development design, production manufacturing, quality control, safety production and the like, so that the energy equipment industry has the characteristics of full data and rich knowledge. At present, china is greatly advancing the high-end, green and intelligent development of the energy equipment industry, and an industrial Internet platform is used as a core of the hub of industrial full-factor connection and the industrial resource allocation, and is a key for supporting ubiquitous connection, elastic supply and efficient allocation of manufacturing resources and assisting the transformation and upgrading of the traditional energy equipment industry. As an important component of an industrial Internet system, the industrial model is a key carrier for the fusion and innovation of OT knowledge and artificial intelligence and other new generation IT technologies in the industrial field, and is also an important way for fully mining and releasing the value of industrial data, knowledge and operation experience and promoting the construction of 'intelligent change and number change' of manufacturing industry. Industrial models can generally be classified into two types, an industrial mechanism model (white box model) and an industrial intelligent model (black box model) according to the degree of interpretability of the mechanism inside the model. The industrial mechanism model is constructed based on a system principle and an internal mechanism by abstracting and expressing key factors, variables and interactions involved in an industrial process, has the characteristics of clear principle and strong interpretability, but often has the problems of high complexity, difficult construction and upgrading and limited application range, while the industrial intelligent model is constructed based on data and an algorithm and based on a data driving idea, and is constructed by excavating potential relations among the data through intelligent algorithms such as statistics, machine learning, artificial intelligence and the like, has the characteristics of strong flexibility and iterative learning, and also has the problems of low interpretability and poor generalization. Currently, with popularization of industrial internet concepts, demands of enterprises for intelligent production and management by utilizing technologies such as big data and artificial intelligence are continuously increasing. The traditional mechanism model is difficult to cope with complex and changeable actual conditions, and the intelligent model is limited in application scene and poor in energizing effect in actual application because of insufficient generalization and interpretability. If the intelligent model and the mechanism model can be combined for fusion innovation, the advantages of the two models are combined for complementation, and the accuracy and reliability of model prediction and decision making are improved. Disclosure of Invention Aiming at the particularity of the industrial mechanism model and the industrial intelligent model and the technical requirements of the energy equipment industry on intelligent production and management by utilizing technologies such as big data, artificial intelligence and the like, the invention provides an industrial interconnection intelligent system based on an industrial Internet platform, which can effectively improve the efficiency and reliability of the industrial model construction and the industrial model decision stage, and an application method thereof. The technical aim of the invention is achieved by the following technical scheme, namely an industrial interconnection intelligent system for the energy equipment industry, which is based on a technical base of an industrial internet platform and comprises an asset center module, a model construction center module and a model arrangement center module; The asset center module is used for carrying out standardized encapsulation on accessed data, mechanisms and algorithms; The model construction center module is used for constructing an algorithm model based on the mechanism model and a mechanism model based on the algorithm model; the model programming center module is used for fusing an algorithm-mechanism model parallel operation mode and an algorithm-mechanism model serial operation mode. Further, the