An intelligent decision-making method and system based on industrial big data

By collecting and processing aluminum smelting energy consumption data in real time, combining edge computing and machine learning, building an energy consumption optimization model, the problem of insufficient energy efficiency optimization in the existing technology is solved, and intelligent decision-making and energy consumption reduction in the aluminum smelting process are achieved.

CN118568582BActive Publication Date: 2024-11-15NANJING XIANWEI INFORMATION TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411061087.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-15
Estimated Expiration
2044-08-05

AI Technical Summary

Technical Problem

The existing technology lacks real-time data processing and dynamic adjustment capabilities when processing industrial big data, cannot respond quickly to production changes, and lacks in-depth understanding of energy consumption and production processes, equipment status, and environmental factors, resulting in insufficient energy efficiency optimization.

Method used

Through sensors and the Internet of Things, collect energy consumption data on aluminum smelting in real time, use edge computing and machine learning for preliminary processing and enhancement, combine statistical analysis, time series analysis and graph neural network to build a multi-objective and multi-constraint energy consumption optimization model, and use Hybrid intelligent optimization algorithm solves, and introduces adaptive parameter adjustment and learning mechanisms to dynamically adjust the optimization solution.

Benefits of technology

It realizes intelligent decision-making and optimization of the aluminum smelting process, can quickly respond to production changes, accurately locate high energy consumption links, continuously improve the energy consumption optimization effect, and reduce production costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118568582B_ABST
    Figure CN118568582B_ABST
Patent Text Reader

Abstract

The present invention discloses an intelligent decision-making method and system based on industrial big data, which belongs to the technical field of intelligent decision-making systems, and specifically comprises: collecting energy consumption data in the aluminum smelting process in real time by means of sensors and the Internet of Things, and performing preliminary processing and real-time enhancement on the data by means of edge computing and machine learning methods; analyzing the links and steps with high energy consumption, and the correlation between energy consumption and production process parameters, equipment status, and ambient temperature by means of statistical analysis and data mining algorithms, combined with time series analysis and graph neural networks; constructing a multi-objective, multi-constrained energy consumption optimization model based on the analysis results, and solving it by means of a hybrid intelligent optimization algorithm to obtain an optimal energy consumption reduction plan; converting the plan into actual operation, and evaluating and adjusting the effect by means of real-time data monitoring and feedback mechanisms, thereby realizing intelligent decision-making and optimization of the aluminum smelting process.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent decision-making systems, and specifically relates to an intelligent decision-making method and system based on industrial big data. Background Art

[0002] Under the new situation of technological transformation in industrial production and manufacturing, with the further deepening of the application of big data analysis and mining technology, more and more manufacturing companies are demanding intelligent processing under the premise of stable and reliable production systems. In particular, with the development of the Internet, the volume of industrial big data in the production process has increased sharply. However, due to the limitations of data utilization methods and related systems, the utilization depth of industrial big data is insufficient and the utilization rate is not high. How to effectively integrate these data to make decisions quickly and accurately and improve work efficiency is an urgent problem that needs to be solved.

[0003] For example, the patent with publication number CN115496426A discloses an industrial big data intelligent analysis decision-making method and system, including: determining the target product and target tooling by tracking and monitoring industrial production; screening big data to obtain product backup data and tooling record data; analyzing product problems and making product problem judgments; analyzing tooling problems when there are no product problems; performing decision analysis, generating decision analysis results, and intervening in industrial production. When production problems occur, it is possible to determine the target product and target tooling, screen big data, obtain product backup data and tooling record data for problem analysis and troubleshooting, and first perform product problem analysis, then perform tooling problem analysis, and then perform decision analysis and industrial production intervention based on the analysis data, so as to quickly determine the cause of the processing problem, facilitate timely resolution of the corresponding problem, and reduce the impact on product production and processing.

[0004] The above existing technologies all have the following problems: 1) Lack of ability to process and dynamically adjust real-time data, resulting in a slow response to emergencies and changes, and an inability to optimize the production process in a timely manner; 2) Lack of in-depth understanding of the complex relationship between specific production processes, equipment status and environmental factors and energy consumption; 3) Lack of adaptive learning mechanism, unable to dynamically adjust decision analysis results and optimization plans based on real-time data and monitoring results, resulting in the inability to continuously improve energy efficiency during the production process. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes an intelligent decision-making method and system based on industrial big data, which collects energy consumption data in the aluminum smelting process in real time through sensors and the Internet of Things, and uses edge computing and machine learning methods to perform preliminary processing and real-time enhancement on the data; uses statistical analysis and data mining algorithms, combined with time series analysis and graph neural networks, to analyze the links and steps with high energy consumption, as well as the relationship between energy consumption and production process parameters, equipment status, and ambient temperature; based on the analysis results, constructs a multi-objective, multi-constrained energy consumption optimization model, and uses a hybrid intelligent optimization algorithm to solve it to obtain the optimal energy consumption reduction plan; converts the plan into actual operation, and evaluates and adjusts the effect through real-time data monitoring and feedback mechanism, thereby realizing intelligent decision-making and optimization of the aluminum smelting process.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An intelligent decision-making method based on industrial big data, comprising:

[0008] Step S1: Use sensors and the Internet of Things to collect energy consumption data in the aluminum smelting production process in real time, introduce edge computing methods to preliminarily process the collected energy consumption data, and use an adaptive data enhancement method based on machine learning to perform real-time enhancement preprocessing and integration on the preliminarily processed energy consumption data;

[0009] Step S2: Analyze the collected energy consumption data using statistical analysis and data mining algorithms, combine time series analysis and graph neural networks to find out the links and steps with high energy consumption, and use deep clustering analysis to find out the relationship between energy consumption and production process parameters, equipment status, and ambient temperature, and generate energy consumption reports and visualization charts based on the analysis results;

[0010] Step S3: Based on the results of data analysis and combined with the actual constraints of the aluminum smelting production process, a multi-objective, multi-constrained energy consumption optimization model is constructed, the energy consumption optimization model is solved using a hybrid intelligent optimization algorithm, and the solution process of the energy consumption optimization model is optimized using an adaptive parameter adjustment strategy to obtain the optimal energy consumption reduction solution;

[0011] Step S4: Convert the optimal energy consumption reduction plan into actual operation and formulate an implementation plan. During the implementation process, use real-time data monitoring and feedback mechanisms to track and evaluate the energy consumption reduction effect, introduce an adaptive learning mechanism, and dynamically adjust the optimal energy consumption reduction plan based on the monitoring results;

[0012] The specific steps of searching for the link with high energy consumption in step S2 include:

[0013] S2.1: Collect energy consumption data, pre-process the energy consumption data, and arrange them in chronological order to obtain the arranged energy consumption data ,in, represents the energy consumption data after the nth arrangement, and n represents the number of energy consumption data after the arrangement;

[0014] S2.2: Set the smoothing factor ,Will As the initial prediction value, the prediction value is updated through the improved exponential smoothing formula, and the energy consumption data is predicted according to the updated prediction value to obtain the energy consumption pattern. The improved exponential smoothing formula is:

[0015] ;

[0016] in, represents the predicted value of energy consumption data at time t, represents the energy consumption data at time t, , and Respectively indicate time , and The predicted value of energy consumption data at represents the trend smoothing coefficient, represents the trend coefficient, represents the trend value at time t, and b represents the energy consumption data division period;

[0017] S2.3: Based on , extract the energy consumption data features, build the XGBoost model based on the selected features, and identify the links or time periods with high energy consumption according to the output results of the XGBoost model;

[0018] The specific steps of step S3 include:

[0019] S3.1: According to the identified high energy consumption links, set numerical targets for energy consumption reduction, combine the actual constraints of the production process, and quantify the actual constraints of the production process into constraints of the mathematical model;

[0020] S3.2: Based on the identified high energy consumption links and the determined numerical targets for energy consumption reduction, combined with the constraints of the mathematical model, a multi-objective and multi-constraint energy consumption optimization model is constructed;

[0021] S3.3: Set the initial parameters of the genetic algorithm and simulated annealing algorithm, and set the decision variables in the energy consumption optimization model Encoding to form chromosomes of genetic algorithms or states of simulated annealing algorithms;

[0022] S3.4: Randomly generate an initial population that meets the constraints and calculate the objective function value of the individual in the energy consumption optimization model ,in, represents the operating parameters of the i-th device in the population;

[0023] The specific steps of step S3 also include:

[0024] S3.5: Set the maximum number of iterations to H, perform genetic operations such as selection, crossover, and mutation on the population to generate a new population, and update the objective function value of the individual in the energy consumption optimization model , calculate the energy difference ;

[0025] like , then accept the new solution as the current solution;

[0026] like , then with probability Accept the new interpretation, the new interpretation is , where W represents the current temperature;

[0027] S3.6: Based on the cooling rate Update the temperature. The updated temperature is: ;

[0028] S3.7: Repeat S3.4-S3.6 until the maximum number of iterations is reached;

[0029] S3.8: Monitor the state of the solution process in real time, dynamically adjust the parameters of the genetic algorithm and the simulated annealing algorithm using an adaptive parameter adjustment strategy according to the state of the solution process, evaluate the results obtained by the algorithm solution, determine and output the optimal energy consumption reduction plan;

[0030] Specifically, the calculation formula for the objective function value of the individual in the energy consumption optimization model in S3.4 is:

[0031] ;

[0032] Where m represents the number of devices. represents the energy consumption function of the ith device, represents the efficiency parameter of the ith device, represents the current temperature of the i-th device, Represents the load factor of the ith device.

[0033] Specifically, the constraints of the mathematical model in S3.1 include:

[0034] ;

[0035] in, It represents the processing volume or output of the i-th device in unit time. express The maximum value of represents the quality index of the jth raw material, express The minimum value of represents the temperature of the kth step in the aluminum smelting process, represents the pressure at the kth step in the aluminum smelting process, represents the reaction time of the kth step in the aluminum smelting process, , , Respectively , , The minimum value of , , Respectively , , The maximum value of E represents the total energy consumption of the aluminum smelting process. , represent the minimum and maximum values ​​of E respectively.

[0036] An intelligent decision-making system based on industrial big data, comprising: a data acquisition module, a data mining module, an intelligent decision-making module, and an evaluation module;

[0037] The data collection module is used to collect industrial data in real time, including production data, equipment status data, and environmental data, and to store, manage, and maintain the collected industrial data to ensure data security and accessibility;

[0038] The data mining module is used to discover valuable information, patterns and associations from the processed industrial data;

[0039] The intelligent decision-making module is used to generate an optimal energy consumption reduction plan using an optimization algorithm and a prediction model;

[0040] The evaluation module is used to monitor and evaluate the operating status of the intelligent decision-making system in real time and provide a friendly user interface and interaction method.

[0041] Specifically, the intelligent decision-making module includes: a decision model unit, an optimization algorithm unit, a parameter adjustment unit, a solution conversion unit, and a plan formulation unit;

[0042] The decision model unit is used to build and optimize the decision model, which includes a prediction model and an optimization model, simulates the production process, and predicts the results under different decision paths;

[0043] The optimization algorithm unit is used to use genetic algorithms and simulated annealing algorithms to solve decision-making problems under multi-objective and multi-constraint conditions and find the optimal energy consumption reduction solution;

[0044] The parameter adjustment unit is used to automatically adjust the energy consumption reduction scheme;

[0045] The scheme conversion unit is used to convert the energy consumption reduction scheme generated by the optimization algorithm into an operable and specific implementation plan;

[0046] The plan formulation unit is used to formulate a detailed implementation plan.

[0047] Specifically, an electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of an intelligent decision-making method based on industrial big data when executing the computer program.

[0048] Specifically, a computer-readable storage medium stores computer instructions, which, when executed, execute the steps of an intelligent decision-making method based on industrial big data.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The present invention proposes an intelligent decision-making system based on industrial big data, and optimizes and improves the architecture, operation steps and processes. The system has the advantages of simple process, low investment and operation costs, and low production work costs.

[0051] 2. The present invention proposes an intelligent decision-making method based on industrial big data, which can collect energy consumption data in the aluminum smelting production process in real time, and use edge computing and machine learning technology to process and enhance it in real time, ensuring the real-time and dynamic nature of the data, so that the system can quickly respond to changes in the production process and make corresponding optimization decisions; it specifically optimizes the energy consumption of the aluminum smelting production process, combines statistical analysis and data mining algorithms, and can deeply understand the relationship between energy consumption and production process parameters, equipment status, and ambient temperature, which helps to accurately locate links and steps with high energy consumption; by introducing adaptive parameter adjustment strategies and adaptive learning mechanisms, it can dynamically adjust the optimization plan according to real-time data monitoring and feedback results, and continuously improve the effect of energy consumption optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a schematic diagram of an intelligent decision-making method based on industrial big data of the present invention;

[0053] Figure 2 This is a workflow diagram of an intelligent decision-making method based on industrial big data in the present invention;

[0054] Figure 3 This is a high energy consumption search flow chart of an intelligent decision-making method based on industrial big data in the present invention;

[0055] Figure 4 This is a flowchart of an optimal energy consumption reduction solution for an intelligent decision-making method based on industrial big data in the present invention;

[0056] Figure 5 This is an architecture diagram of an intelligent decision-making system based on industrial big data in the present invention. DETAILED DESCRIPTION

[0057] Example 1

[0058] See also Figure 1-Figure 4 , an embodiment provided by the present invention: an intelligent decision-making method based on industrial big data, comprising the following steps:

[0059] Step S1: Use sensors and the Internet of Things to collect energy consumption data in the aluminum smelting production process in real time, introduce edge computing methods to preliminarily process the collected energy consumption data, and use an adaptive data enhancement method based on machine learning to perform real-time enhancement preprocessing and integration on the preliminarily processed energy consumption data;

[0060] Among them, energy consumption data includes electricity consumption, fuel consumption, and heat energy consumption.

[0061] Furthermore, the specific steps of step S1 include:

[0062] A1: Deploy sensor networks and edge devices on the aluminum smelting production line and perform necessary configuration and testing;

[0063] A2: Sensors collect energy consumption data in the production process in real time and transmit it to edge devices through IoT technology;

[0064] A3: Edge devices receive raw data from sensors and perform preliminary processing, such as data cleaning, outlier detection, and data compression;

[0065] A4: Introduce a machine learning model, and select a feature expansion data enhancement method to generate new features based on the characteristics of historical data and real-time data to perform real-time enhancement preprocessing on the preliminarily processed data to improve the quality and diversity of the data. The feature expansion data enhancement method is a prior art in this field and is not an inventive solution of the present application, so it will not be described in detail here.

[0066] A5: Integrate the data after preliminary processing and enhanced preprocessing into a unified dataset;

[0067] A6: The integrated data set is transferred to the data center.

[0068] Step S2: Analyze the collected energy consumption data using statistical analysis and data mining algorithms, combine time series analysis and graph neural networks to find out the links and steps with high energy consumption, and find out the relationship between energy consumption and production process parameters, equipment status, and ambient temperature through deep clustering analysis, and generate energy consumption reports and visualization charts based on the analysis results. The deep clustering analysis method is a prior art content in this field, and is not an inventive solution of this application, and will not be described in detail here;

[0069] Step S3: Based on the results of data analysis and combined with the actual constraints of the aluminum smelting production process, a multi-objective, multi-constrained energy consumption optimization model is constructed, the energy consumption optimization model is solved using a hybrid intelligent optimization algorithm, and the solution process of the energy consumption optimization model is optimized using an adaptive parameter adjustment strategy to obtain the optimal energy consumption reduction solution;

[0070] Among them, constraints include equipment capacity and raw material quality.

[0071] Step S4: Convert the optimal energy consumption reduction plan into actual operation and formulate an implementation plan. During the implementation process, use real-time data monitoring and feedback mechanism to track and evaluate the energy consumption reduction effect, introduce adaptive learning mechanism, and dynamically adjust the optimal energy consumption reduction plan according to the monitoring results.

[0072] Practical operations specifically refer to: technology improvement, equipment update, and management optimization.

[0073] The specific steps of searching for the link with high energy consumption in step S2 include:

[0074] S2.1: Collect energy consumption data, pre-process the energy consumption data, and arrange them in chronological order to obtain the arranged energy consumption data ,in, represents the energy consumption data after the nth arrangement, and n represents the number of energy consumption data after the arrangement;

[0075] S2.2: Set the smoothing factor ,Will As the initial prediction value, the prediction value is updated through the improved exponential smoothing formula, and the energy consumption data is predicted according to the updated prediction value to obtain the energy consumption pattern. The improved exponential smoothing formula is:

[0076] ;

[0077] in, represents the predicted value of energy consumption data at time t, represents the energy consumption data at time t, , and Respectively indicate time , and The predicted value of energy consumption data at represents the trend smoothing coefficient, represents the trend coefficient, represents the trend value at time t, and b represents the energy consumption data division period;

[0078] S2.3: Based on , extract the energy consumption data features, build the XGBoost model based on the selected features, and identify the links or time periods with high energy consumption based on the output results of the XGBoost model.

[0079] The specific steps of step S3 include:

[0080] S3.1: According to the identified high energy consumption links, set numerical targets for energy consumption reduction, combine the actual constraints of the production process, and quantify the actual constraints of the production process into constraints of the mathematical model;

[0081] S3.2: Based on the identified high energy consumption links and the determined numerical targets for energy consumption reduction, combined with the constraints of the mathematical model, a multi-objective and multi-constraint energy consumption optimization model is constructed;

[0082] Furthermore, the specific steps of S3.2 include:

[0083] B1: According to the numerical goal of reducing energy consumption, multiple objective functions are set, including minimizing energy consumption, maximizing energy utilization efficiency, and maximizing output;

[0084] B2: Select key factors affecting energy consumption as decision variables, such as equipment operating speed and production process parameters;

[0085] B3: Integrate production process parameters, equipment status, and ambient temperature into the model to ensure the feasibility and practicality of the model;

[0086] B4: Use linear programming or integer programming to express the objective function and constraints to form a complete energy consumption optimization model.

[0087] S3.3: Set the initial parameters of the genetic algorithm and simulated annealing algorithm, and set the decision variables in the energy consumption optimization model Encoding to form chromosomes of genetic algorithms or states of simulated annealing algorithms;

[0088] S3.4: Randomly generate an initial population that meets the constraints and calculate the objective function value of the individual in the energy consumption optimization model ,in, represents the operating parameters of the ith device in the population.

[0089] The specific steps of step S3 also include:

[0090] S3.5: Set the maximum number of iterations to H, perform genetic operations such as selection, crossover, and mutation on the population to generate a new population, and update the objective function value of the individual in the energy consumption optimization model , calculate the energy difference, the formula is:

[0091] ;

[0092] like , then accept the new solution as the current solution;

[0093] like , then with probability Accept the new interpretation, the new interpretation is , where W represents the current temperature;

[0094] S3.6: Based on the cooling rate Update the temperature. The updated temperature is: ;

[0095] S3.7: Repeat S3.4-S3.6 until the maximum number of iterations is reached;

[0096] S3.8: Monitor the state of the solution process in real time, and dynamically adjust the parameters of the genetic algorithm and the simulated annealing algorithm using an adaptive parameter adjustment strategy based on the state of the solution process. Evaluate the results obtained by the algorithm solution, determine and output the optimal energy consumption reduction plan.

[0097] The constraints of the mathematical model in S3.1 include:

[0098] ;

[0099] in, It represents the processing volume or output of the i-th device in unit time. express The maximum value of represents the quality index of the jth raw material, express The minimum value of represents the temperature of the kth step in the aluminum smelting process, represents the pressure at the kth step in the aluminum smelting process, represents the reaction time of the kth step in the aluminum smelting process, , , Respectively , , The minimum value of , , Respectively , , The maximum value of E represents the total energy consumption of the aluminum smelting process. , represent the minimum and maximum values ​​of E respectively.

[0100] The calculation formula for the objective function value of the individual in the energy consumption optimization model in S3.4 is:

[0101] ;

[0102] Where m represents the number of devices. represents the energy consumption function of the ith device, represents the efficiency parameter of the ith device, represents the current temperature of the i-th device, Represents the load factor of the ith device.

[0103] Example 2

[0104] See also Figure 5 ,Another embodiment provided by the present invention: An intelligent decision-making system based on industrial big data, comprising: a data collection module, a data mining module, an intelligent decision-making module, and an evaluation module;

[0105] A data collection module is used to collect industrial data in real time, including production data, equipment status data, and environmental data, and to store, manage, and maintain the collected industrial data to ensure data security and accessibility;

[0106] Data mining module for discovering valuable information, patterns and associations from processed industrial data;

[0107] Intelligent decision-making module, used to generate the best energy consumption reduction plan using optimization algorithms and prediction models;

[0108] The evaluation module is used to monitor and evaluate the operating status of the intelligent decision-making system in real time, and provides a friendly user interface and interaction method.

[0109] The data collection module includes: data acquisition unit, edge computing unit, and data enhancement unit;

[0110] Data acquisition unit, used to collect industrial data in the production process in real time through sensors and IoT devices, such as energy consumption, equipment operating status, and production efficiency indicators;

[0111] The edge computing unit is located at the source of data and is used to perform preliminary processing and screening of the collected data, including data cleaning, anomaly detection and simple data analysis, to reduce the amount of data transmitted to the central server and improve the response speed;

[0112] The data enhancement unit is used to perform enhanced preprocessing on the data after preliminary processing, such as data normalization, missing value filling, and data smoothing, to improve data quality.

[0113] The data mining module includes: statistical analysis unit, data mining unit, deep clustering unit, and visualization unit;

[0114] Statistical analysis unit, used to apply statistical methods to conduct descriptive analysis of the collected data and discover the basic characteristics, trends and distribution of the data;

[0115] Data mining unit, used to deeply mine hidden information and potential relationships in data through pattern recognition, association rules, and regression analysis methods;

[0116] A deep clustering unit, which uses deep learning techniques to perform cluster analysis on complex data sets and identify natural groups or patterns in the data, such as correlating energy consumption with production parameters;

[0117] The visualization unit is used to display the analysis results in the form of charts and dashboards to help decision makers intuitively understand the information and trends behind the data.

[0118] The intelligent decision-making module includes: decision model unit, optimization algorithm unit, parameter adjustment unit, solution conversion unit, and plan formulation unit;

[0119] A decision model unit is used to construct and optimize a decision model, wherein the decision model includes a prediction model and an optimization model, simulates the production process, and predicts the results under different decision paths;

[0120] The optimization algorithm unit is used to use genetic algorithms and simulated annealing algorithms to solve decision-making problems under multi-objective and multi-constraint conditions and find the optimal solution;

[0121] Parameter adjustment unit, used to automatically adjust production parameters or suggest adjustment plans to achieve the goals of reducing energy consumption and improving efficiency;

[0122] A scheme conversion unit, used to convert the energy consumption reduction scheme generated by the optimization algorithm into an operational and specific implementation plan;

[0123] The planning unit is used to develop detailed implementation plans, including timetables and resource allocation, to ensure the effective execution of decision-making plans.

[0124] The evaluation module includes: a system monitoring unit, a performance evaluation unit, a feedback adjustment unit, a user interface unit, and an interactive control unit;

[0125] System monitoring unit, used to monitor the production process and the execution of decision-making plans in real time and collect feedback data;

[0126] Performance evaluation unit, used to evaluate the performance of the system, including efficiency improvement and decision-making accuracy, to provide a basis for system optimization;

[0127] Feedback adjustment unit, used to automatically or manually adjust decisions based on evaluation results and real-time monitoring data

[0128] Solution to achieve closed-loop control;

[0129] User interface unit, used to design intuitive and easy-to-use user interfaces to facilitate users to view data, analyze results and make decisions;

[0130] The interactive control unit is used to provide users with interactive methods for the system, including query, screening, and setting operations, so that users can customize decision results according to their needs, increasing the flexibility and practicality of the system.

[0131] Example 3

[0132] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of an intelligent decision-making method based on industrial big data when executing the computer program.

[0133] A computer-readable storage medium stores computer instructions, which, when executed, execute the steps of an intelligent decision-making method based on industrial big data.

[0134] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

Claims

1. An intelligent decision-making method based on industrial big data, characterized in that: include: Step S1: Use sensors and the Internet of Things to collect energy consumption data in the aluminum smelting production process in real time, introduce edge computing methods to preliminarily process the collected energy consumption data, and use an adaptive data enhancement method based on machine learning to perform real-time enhancement preprocessing and integration on the preliminarily processed energy consumption data; Step S2: Analyze the collected energy consumption data using statistical analysis and data mining algorithms, combine time series analysis and graph neural networks to find out the links and steps with high energy consumption, and use deep clustering analysis to find out the relationship between energy consumption and production process parameters, equipment status, and ambient temperature, and generate energy consumption reports and visualization charts based on the analysis results; Step S3: Based on the results of data analysis and combined with the actual constraints of the aluminum smelting production process, a multi-objective, multi-constrained energy consumption optimization model is constructed, the energy consumption optimization model is solved using a hybrid intelligent optimization algorithm, and the solution process of the energy consumption optimization model is optimized using an adaptive parameter adjustment strategy to obtain the optimal energy consumption reduction solution; Step S4: Convert the optimal energy consumption reduction plan into actual operation and formulate an implementation plan. During the implementation process, use real-time data monitoring and feedback mechanisms to track and evaluate the energy consumption reduction effect, introduce an adaptive learning mechanism, and dynamically adjust the optimal energy consumption reduction plan based on the monitoring results; The specific steps of searching for the link with high energy consumption in step S2 include: S2.1: Collect energy consumption data, pre-process the energy consumption data, and arrange them in chronological order to obtain the arranged energy consumption data ,in, represents the energy consumption data after the nth arrangement, and n represents the number of energy consumption data after the arrangement; S2.2: Set the smoothing factor ,Will As the initial prediction value, the prediction value is updated through the improved exponential smoothing formula, and the energy consumption data is predicted according to the updated prediction value to obtain the energy consumption pattern. The improved exponential smoothing formula is: ; in, represents the predicted value of energy consumption data at time t, represents the energy consumption data at time t, , and Respectively indicate time , and The predicted value of energy consumption data is represents the trend smoothing coefficient, represents the trend coefficient, represents the trend value at time t, and b represents the energy consumption data division period; S2.3: Based on , extract the energy consumption data features, build the XGBoost model based on the selected features, and identify the links or time periods with high energy consumption according to the output results of the XGBoost model; The specific steps of step S3 include: S3.1: According to the identified high energy consumption links, set numerical targets for energy consumption reduction, combine the actual constraints of the production process, and quantify the actual constraints of the production process into constraints of the mathematical model; S3.2: Based on the identified high energy consumption links and the determined numerical targets for energy consumption reduction, combined with the constraints of the mathematical model, a multi-objective and multi-constraint energy consumption optimization model is constructed; S3.3: Set the initial parameters of the genetic algorithm and simulated annealing algorithm, and set the decision variables in the energy consumption optimization model Encoding to form chromosomes of genetic algorithms or states of simulated annealing algorithms; S3.4: Randomly generate an initial population that meets the constraints and calculate the objective function value of the individual in the energy consumption optimization model ,in, represents the operating parameters of the i-th device in the population; The specific steps of step S3 also include: S3.5: Set the maximum number of iterations to H, perform genetic operations such as selection, crossover, and mutation on the population to generate a new population, and update the objective function value of the individual in the energy consumption optimization model. , calculate the energy difference ; like , then accept the new solution as the current solution; like , then with probability Accept the new interpretation, the new interpretation is , where W represents the current temperature; S3.6: Based on the cooling rate Update the temperature. The updated temperature is: ; S3.7: Repeat S3.4-S3.6 until the maximum number of iterations is reached; S3.8: Monitor the state of the solution process in real time, dynamically adjust the parameters of the genetic algorithm and the simulated annealing algorithm using an adaptive parameter adjustment strategy according to the state of the solution process, evaluate the results obtained by the algorithm solution, determine and output the optimal energy consumption reduction plan; The calculation formula for the objective function value of the individual in the energy consumption optimization model in S3.4 is: ; Where m represents the number of devices. represents the energy consumption function of the ith device, represents the efficiency parameter of the ith device, represents the current temperature of the i-th device, Represents the load factor of the ith device.

2. The intelligent decision-making method based on industrial big data as claimed in claim 1, characterized in that: The constraints of the mathematical model in S3.1 include: ; in, It represents the processing volume or output of the i-th device in unit time. express The maximum value of represents the quality index of the jth raw material, express The minimum value of represents the temperature of the kth step in the aluminum smelting process, represents the pressure at the kth step in the aluminum smelting process, represents the reaction time of the kth step in the aluminum smelting process, , , Respectively , , The minimum value of , , Respectively , , The maximum value of E represents the total energy consumption of the aluminum smelting process. , represent the minimum and maximum values ​​of E respectively.

3. An intelligent decision-making system based on industrial big data, which is used to implement an intelligent decision-making method based on industrial big data as described in any one of claims 1-2, characterized in that: include: Data collection module, data mining module, intelligent decision-making module, and evaluation module; The data acquisition module is used to collect industrial data in real time, including production data, equipment status data, and environmental data, and to store, manage, and maintain the collected industrial data to ensure data security and accessibility; The data mining module is used to discover valuable information, patterns and associations from the processed industrial data; The intelligent decision-making module is used to generate an optimal energy consumption reduction plan using an optimization algorithm and a prediction model; The evaluation module is used to monitor and evaluate the operating status of the intelligent decision-making system in real time and provide a friendly user interface and interaction method.

4. The intelligent decision-making system based on industrial big data as described in claim 3 is characterized in that: The intelligent decision-making module includes: a decision model unit, an optimization algorithm unit, a parameter adjustment unit, a scheme conversion unit, and a plan formulation unit; The decision model unit is used to build and optimize the decision model, which includes a prediction model and an optimization model, simulates the production process, and predicts the results under different decision paths; The optimization algorithm unit is used to use genetic algorithms and simulated annealing algorithms to solve decision-making problems under multi-objective and multi-constraint conditions and find the optimal energy consumption reduction solution; The parameter adjustment unit is used to automatically adjust the energy consumption reduction scheme; The scheme conversion unit is used to convert the energy consumption reduction scheme generated by the optimization algorithm into an operable and specific implementation plan; The plan formulation unit is used to formulate a detailed implementation plan.

5. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of an intelligent decision-making method based on industrial big data described in any one of claims 1-2 are implemented.

6. A computer-readable storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, the steps of an intelligent decision-making method based on industrial big data as described in any one of claims 1-2 are executed.

Citation Information

Cited By

  • Industrial big data driven intelligent production decision support system and method

    CN121787925A