Intelligent evaluation system and method for pollutant treatment technology in iron and steel industry

Abstract

The invention discloses an intelligent evaluation system and method for a pollutant treatment technology in the iron and steel industry, belongs to the technical field of pollutant treatment evaluation, and solves the problem that evaluation in the prior art only considers economic factors and ignores multiple factors such as technical feasibility and environmental influence through simple cost-benefit analysis, so that the evaluation efficiency is greatly improved. The system comprises an enterprise terminal, a data management module, an intelligent evaluation module and an association recommendation module. According to the technical scheme of the invention, the comprehensive evaluation model based on technology-economy-environment is constructed, the treatment technology is comprehensively evaluated from multiple dimensions such as technical feasibility, economic cost effectiveness and environmental influence, and the real-time performance, the accuracy and the pertinence of the data based on the evaluation are ensured. And personalized evaluation and recommendation services can be provided according to factors such as scales, production processes, pollutant emission characteristics and economic conditions of different enterprises.

Description

Technical Field

[0001] This invention belongs to the field of pollutant treatment assessment technology, specifically relating to an intelligent assessment system and method for pollutant treatment technologies in the steel industry. Background Technology

[0002] With increasingly stringent ultra-low emission standards in the steel industry, nitrogen oxides (NOx) are becoming more and more important. x The coordinated control of air pollutants such as particulate matter, SO2, and NO has become a key task for the industry. The steel production process involves multiple steps, including sintering, ironmaking, steelmaking, and rolling, generating pollutants including particulate matter, SO2, and NO. x The treatment of various pollutants, including heavy metals and high-salinity wastewater, varies significantly across different stages, making it difficult to conduct unified assessments and optimizations.

[0003] In recent years, with the rapid development of information technology, intelligent assessment systems have been gradually applied to various fields, providing new ideas and methods for solving complex problems. However, existing technology assessments generally rely on simple cost-benefit analyses, considering only economic factors and ignoring the feasibility and environmental impact of the technology. This fails to comprehensively reflect the overall performance of the treatment technology, and thus cannot provide enterprises / factories with recommendations for feasible pollution prevention technologies, fugitive control measures throughout the entire production process, and relevant case studies, covering the entire process from source reduction and process control to end-of-pipe treatment. To address these issues, we propose an intelligent assessment system and method for pollutant treatment technologies in the steel industry. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing an intelligent evaluation system and method for pollutant treatment technologies in the steel industry. This solves the problem that existing technology evaluations generally rely on simple cost-benefit analyses, which only consider economic factors and ignore factors such as the feasibility of the technology and its environmental impact, thus failing to comprehensively reflect the overall performance of the treatment technology.

[0005] To address the problem that existing technology assessments typically rely on simple cost-benefit analyses, considering only economic factors and neglecting feasibility and environmental impact, thus failing to comprehensively reflect the overall performance of pollution control technologies, we propose an intelligent assessment system and method for pollution control technologies in the steel industry. In short, the system consists of an enterprise-side module, a data management module, an intelligent assessment module, a recommendation module, and a system management module. During operation, the enterprise-side module first collects real-time data on pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs. Then, the data management module collects the pollution control technology parameters for the steel industry and uploads them to the corresponding database. The intelligent assessment module uses the enterprise's pollutant emissions, treatment technology parameters, and equipment operating parameters as input, executes a comprehensive evaluation model, and outputs the comprehensive assessment result for the corresponding treatment technology for the enterprise. Finally, the recommendation module obtains the comprehensive assessment result for the enterprise's treatment technology, matches it with application cases in a case library, and generates an intelligent recommendation queue of treatment technologies that meet the enterprise's needs. In this embodiment of the invention, a comprehensive evaluation model based on technology, economy, and environment is constructed. This model comprehensively assesses the treatment technology from multiple dimensions, including technical feasibility, economic cost-effectiveness, and environmental impact. By collecting real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs from the enterprise side, the model ensures the real-time nature, accuracy, and relevance of the data used in the assessment. Based on this rich data, more scientific and credible evaluation results are obtained. Compared with traditional evaluation methods based on experience and simple calculations, this system can more accurately grasp the actual effects and potential problems of treatment technologies. Furthermore, it can provide personalized evaluation and recommendation services based on factors such as the size of different enterprises, production processes, pollutant emission characteristics, and economic conditions, providing strong support for enterprises' technology selection and management decisions.

[0006] This invention is implemented as follows: an intelligent assessment system for pollutant treatment technologies in the steel industry, the system comprising:

[0007] On the enterprise side, it is used to collect real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise demand information, and to visualize and present a smart recommendation queue of treatment technologies.

[0008] The data management module is used to collect technical parameters for pollutant treatment in the steel industry, upload these parameters to the corresponding database, and store and update the technical parameters in the database.

[0009] The intelligent assessment module pre-builds a comprehensive evaluation model based on technology, economy, and environment, captures a modeling sample set, uses the modeling sample set to iteratively train the comprehensive evaluation model, and outputs a converged comprehensive evaluation model. Taking enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters as input, it executes the comprehensive evaluation model and outputs the comprehensive evaluation results of the enterprise's corresponding treatment technology.

[0010] The association recommendation module is used to obtain the comprehensive evaluation results of the enterprise's governance technology, match the comprehensive evaluation results with application cases in the case library, and generate a smart recommendation queue of governance technologies that meet the enterprise's needs.

[0011] The system management module communicates and connects with the enterprise terminal, data management module, and intelligent evaluation module, and is used to manage and control the enterprise terminal, data management module, and intelligent evaluation module.

[0012] Preferably, the system management module includes:

[0013] The system login unit is used to collect user login verification information and verify the user login verification information;

[0014] The system administrator unit is responsible for maintaining the basic database and updating, modifying, and deleting data from it.

[0015] The technician management unit is used to verify technician information and support technical evaluators in adjusting the weights and quantification methods of governance technical indicators.

[0016] The enterprise management unit interacts with the enterprise to obtain the governance technology parameters uploaded by the enterprise, and assesses the enterprise's control level based on the enterprise's pollutant emission data, governance technology parameters, and equipment operating parameters.

[0017] Preferably, the data management module includes:

[0018] The basic database is used to store basic enterprise information and enterprise management levels.

[0019] A technical database is used to store enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters;

[0020] The case study library stores application cases related to pollutant treatment technologies in the steel industry.

[0021] Preferably, the intelligent evaluation module includes:

[0022] The model building unit is used to pre-build a comprehensive evaluation model based on technology, economy, and environment, capture a modeling sample set, use the modeling sample set to iteratively train the comprehensive evaluation model, and output a converged comprehensive evaluation model.

[0023] The indicator extraction unit uses enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters as independent variables to calculate the unit flue gas investment cost, operating cost, emission reduction benefits, and carbon emissions throughout the entire life cycle of the treatment technology.

[0024] The comprehensive evaluation unit is used to acquire data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, unit flue gas investment cost, operating cost and emission reduction benefits, and carbon emissions throughout the entire life cycle of treatment technology from the comprehensive evaluation model. This data is then integrated into an analysis dataset. Based on the comprehensive evaluation model, the analysis dataset is comprehensively analyzed to output the comprehensive evaluation results of the enterprise's corresponding treatment technology.

[0025] Preferably, when pre-constructing the comprehensive evaluation model based on technology, economy, and environment, a recurrent neural network model is used as the basic architecture, and it also includes an input layer and an output layer. An autoencoder is set between the input layer and the recurrent neural network model. The autoencoder includes an encoding layer and a decoding layer. The autoencoder is used to identify the correlation indicators of enterprise pollutant emission data, treatment technology parameters, equipment operating parameters, unit flue gas investment cost, operating cost and emission reduction benefits, and carbon emission of treatment technology throughout its entire life cycle, and to label the correlation indicators. The recurrent neural network model is used to extract multi-scale features from the labeled correlation indicators and to concatenate and fuse the multi-scale features to obtain a fused feature vector. A regression scorer is set between the recurrent neural network model and the output layer. A BiLSTM model is introduced into the regression scorer. The adaptive weights of the indicator features in the fused feature vector are calculated based on the BiLSTM model, and the comprehensive evaluation value of the treatment technology is determined by combining the attribute attention mechanism with the sigmoid function. A level discriminator is set between the regression scorer and the output layer. The level discriminator determines the enterprise control level based on the gradient inversion mechanism combined with the comprehensive evaluation value of the treatment technology.

[0026] Preferably, the method of iteratively training the comprehensive evaluation model using a modeling sample set includes:

[0027] Traverse the basic database, technical database, and case library, retrieve the modeling sample set based on the basic database, technical database, and case library, and divide the modeling sample set into training set and test set;

[0028] Load a pre-built comprehensive evaluation model, and preset the training rounds, joint loss function, tuning strategy and hyperparameters of the comprehensive evaluation model;

[0029] Obtain the training set, use the training set to iteratively train the comprehensive evaluation model, and output a converged comprehensive evaluation model;

[0030] Obtain the test set, use the test set as input, execute the comprehensive evaluation model, output the test results, and use the mean squared error to calculate the expected value of the difference between the test results and the true results;

[0031] The formula for calculating the mean square error is as follows:

[0032]

[0033] in, These are the test results and the actual results, respectively. This represents the number of samples in the test set.

[0034] Determine whether the expected value exceeds the preset expected threshold;

[0035] If the expected value exceeds the preset expected threshold, the model hyperparameters are adjusted using the RMSprop method, and the comprehensive evaluation model is iteratively trained.

[0036] If the expected value does not exceed the preset expected threshold, the convergent comprehensive evaluation model is output.

[0037] Preferably, the method for comprehensive analysis of the dataset based on the comprehensive evaluation model includes:

[0038] Identify and analyze the correlation indicators of enterprise pollutant emission data, treatment technology parameters, equipment operating parameters, unit flue gas investment cost, operating cost and emission reduction benefits, and carbon emission of treatment technology throughout its entire life cycle in the centralized enterprise data set, and label the correlation indicators.

[0039] Multi-scale features are extracted from the association indicators of the labels, and the multi-scale features are concatenated and fused to obtain a fused feature vector;

[0040] The adaptive weights of the indicator features in the fused feature vector are calculated based on the BiLSTM model, and the comprehensive evaluation value of the governance technology is determined by combining the attribute attention mechanism with the sigmoid function.

[0041] The adaptive weights of the indicator features are calculated using the following formula:

[0042]

[0043]

[0044]

[0045] in, Indicates adaptive weights, For attention scores based on attribute attention mechanisms, This represents a learnable weight vector, while It is the hyperbolic tangent function. These are the weight matrix and the fused feature vector, respectively. For bias terms, These represent the forward hidden state and the backward hidden state at time step t, respectively.

[0046] The comprehensive evaluation value of the treatment technology is calculated using the following formula:

[0047]

[0048]

[0049] in, This is a comprehensive evaluation value for governance technologies. For the rating weight vector, This is a weighted fusion feature vector;

[0050] The enterprise's management and control level is determined based on a comprehensive evaluation value that combines the gradient reversal mechanism with governance technology.

[0051] On the other hand, the present invention also provides an intelligent evaluation method for pollutant treatment technologies in the steel industry, the method comprising:

[0052] It collects real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise demand information, and presents a visually appealing intelligent recommendation queue of treatment technologies.

[0053] Collect technical parameters for pollutant treatment in the steel industry, upload these parameters to the corresponding database, and store and update the technical parameters in the database.

[0054] A comprehensive evaluation model based on technology, economy, and environment is pre-constructed, a modeling sample set is captured, and the comprehensive evaluation model is iteratively trained using the modeling sample set. The converged comprehensive evaluation model is output. The comprehensive evaluation model is executed with enterprise pollutant emission data, treatment technology parameters, and equipment operation parameters as inputs, and the comprehensive evaluation results of the corresponding treatment technologies of the enterprise are output.

[0055] Obtain the comprehensive evaluation results of the enterprise's governance technologies, match the comprehensive evaluation results with application cases in the case library, and generate an intelligent recommendation queue of governance technologies that meet the enterprise's needs.

[0056] Compared with the prior art, the embodiments of this application have the following main advantages:

[0057] In this embodiment of the invention, a comprehensive evaluation model based on technology, economy, and environment is constructed. This model comprehensively assesses the treatment technology from multiple dimensions, including technical feasibility, economic cost-effectiveness, and environmental impact. By collecting real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs from the enterprise side, the model ensures the real-time nature, accuracy, and relevance of the data used in the assessment. Based on this rich data, more scientific and credible evaluation results are obtained. Compared with traditional evaluation methods based on experience and simple calculations, this system can more accurately grasp the actual effects and potential problems of treatment technologies. Furthermore, it can provide personalized evaluation and recommendation services based on factors such as the size of different enterprises, production processes, pollutant emission characteristics, and economic conditions, providing strong support for enterprises' technology selection and management decisions.

[0058] In this embodiment of the invention, an intelligent evaluation module is provided, comprising a model building unit, an indicator extraction unit, and a comprehensive evaluation unit. The comprehensive evaluation model comprehensively considers multiple key dimensions of pollutant treatment technologies in the steel industry, including technical feasibility, economic cost-effectiveness, and environmental impact. Compared to traditional single-dimensional evaluation methods, this multi-dimensional comprehensive evaluation approach more accurately reflects the true performance and potential value of treatment technologies. The indicator extraction unit uses enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters as independent variables to calculate key indicators such as unit flue gas investment cost, operating cost, emission reduction benefits, and carbon emissions throughout the entire life cycle of the treatment technology. This comprehensive indicator extraction method provides enterprises with detailed performance analysis of treatment technologies, helping them to deeply understand the performance of various technologies across different dimensions. The comprehensive evaluation unit integrates enterprise pollutant emission data, treatment technology parameters, equipment operating parameters, and various cost-effectiveness indicators into an analysis dataset, and performs comprehensive analysis on this data based on the comprehensive evaluation model, outputting the enterprise's corresponding comprehensive evaluation results for treatment technologies. This comprehensive evaluation method provides enterprises with comprehensive decision support, helping them evaluate the applicability of treatment technologies from multiple perspectives and select the most suitable technical solution for their own needs.

[0059] In this embodiment of the invention, by extracting multi-scale features of labeled and associated indicators, the dynamic change patterns of indicators under different time dimensions can be captured. After multi-scale feature extraction, feature vectors of different time scales and types are integrated into a unified fused feature vector through a splicing and fusion operation. This fusion operation breaks down information silos of single feature dimensions, enabling the model to utilize both static parameters and dynamic performance simultaneously, thereby more accurately characterizing the comprehensive performance of governance technologies. Furthermore, the adaptive weights of indicator features in the fused feature vector, calculated based on the BiLSTM model, can automatically adjust the weight allocation according to the dynamic changes in the data. This adaptability allows the model to better cope with noise and outliers in the data, improving its robustness. Simultaneously, by combining attribute attention mechanisms with the sigmoid function, the model's focus on important features can be further strengthened, thereby improving the accuracy and reliability of the evaluation results. Comprehensive evaluation value determination: The comprehensive evaluation value of governance technologies determined by the above method can comprehensively reflect the comprehensive performance of governance technologies in multiple dimensions such as technology, economy, and environment. This comprehensive evaluation method not only considers the optimal solution of a single dimension but also takes into account the balance and coordination between multiple dimensions, providing steel enterprises with a more comprehensive and scientific decision-making basis. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the intelligent evaluation system for pollutant treatment technology in the steel industry provided by the present invention.

[0061] Figure 2 The diagram shows the intelligent recommendation queue interface for governance technologies generated by the association recommendation module to meet enterprise needs.

[0062] Figure 3 The diagram shows the interface for uploading governance technical parameters on the enterprise side.

[0063] Figure 4 A schematic diagram of the case library page is shown.

[0064] Figure 5 The diagram shows the interface for technical evaluators to adjust the weights and quantification methods of governance technical indicators.

[0065] Figure 6 The diagram illustrates the process of iteratively training the comprehensive evaluation model using a modeling sample set.

[0066] Figure 7 A schematic diagram illustrating the implementation process of a comprehensive analysis method for datasets based on a comprehensive evaluation model is shown.

[0067] Figure 8 A schematic diagram illustrating the implementation process of an intelligent assessment method for pollutant treatment technologies in the steel industry is shown.

[0068] In the diagram: 100 - Enterprise side, 200 - Data management module, 210 - Basic database, 220 - Technical database, 230 - Case library, 300 - Intelligent evaluation module, 310 - Model building unit, 320 - Indicator extraction unit, 330 - Comprehensive evaluation unit, 400 - Related recommendation module, 500 - System management module, 510 - System login unit, 520 - System administrator unit, 530 - Technician management unit, 540 - Enterprise management unit. Detailed Implementation

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0070] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0071] To address the problem that existing technology assessments typically rely on simple cost-benefit analyses, considering only economic factors and neglecting feasibility and environmental impact, thus failing to comprehensively reflect the overall performance of pollution control technologies, we propose an intelligent assessment system and method for pollution control technologies in the steel industry. In short, the system consists of an enterprise terminal (100), a data management module (200), an intelligent assessment module (300), an association and recommendation module (400), and a system management module (500). During operation, the enterprise terminal (100) first collects real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs. Then, the data management module (200) collects the pollution control technology parameters for the steel industry and uploads them to the corresponding database. The intelligent assessment module (300), using the enterprise's pollutant emissions data, treatment technology parameters, and equipment operating parameters as input, executes a comprehensive evaluation model and outputs the comprehensive assessment result for the corresponding treatment technology for the enterprise. Finally, the association and recommendation module (400) obtains the comprehensive assessment result for the corresponding treatment technology for the enterprise, matches it with application cases in the case library (230), and generates an intelligent recommendation queue of treatment technologies that meets the enterprise's needs. In this embodiment of the invention, a comprehensive evaluation model based on technology, economy, and environment is constructed. This model comprehensively assesses the treatment technology from multiple dimensions, including technical feasibility, economic cost-effectiveness, and environmental impact. By collecting real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs from the enterprise's perspective, the model ensures the real-time nature, accuracy, and relevance of the data used in the assessment. Based on this rich data, more scientific and credible evaluation results are derived. Compared with traditional evaluation methods based on experience and simple calculations, this system can more accurately grasp the actual effects and potential problems of treatment technologies. Furthermore, it can provide personalized evaluation and recommendation services based on factors such as the size of different enterprises, production processes, pollutant emission characteristics, and economic conditions, providing strong support for enterprises' technology selection and management decisions.

[0072] This invention provides an intelligent evaluation system for pollutant treatment technologies in the steel industry. Figure 1 This diagram illustrates the structure of an intelligent evaluation system for pollutant control technologies in the steel industry. The system specifically includes:

[0073] The enterprise-side 100 is used to collect real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs, and to visualize and present a smart recommendation queue of treatment technologies. Figure 3The diagram shows the interface for enterprises to upload governance technology parameters. Enterprises can fill in the relevant parameters of a governance technology as required through this module, and click "Generate Technology Evaluation" to obtain relevant information about using the governance technology, such as the corresponding operating costs, investment costs, emission reduction benefits and carbon emissions.

[0074] The data management module 200 is used to collect technical parameters for pollutant treatment in the steel industry, upload the technical parameters for pollutant treatment in the steel industry to the corresponding database, store and update the technical parameters in the database;

[0075] The data management module 200 includes:

[0076] Basic database 210 is used to store basic enterprise information and enterprise management levels;

[0077] Technical database 220 is used to store enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters;

[0078] Case Library 230 stores application cases related to pollutant treatment technologies in the steel industry. Figure 4 The illustration shows page 230 of the case library. Case library 230 compiles information on denitrification devices from different industries and companies, forming a single case library. Companies can use fuzzy search to find relevant cases and suitable denitrification technologies. Furthermore, this case library 230 is linked to the association recommendation module 400, which uses an association algorithm to achieve intelligent recommendation of cases.

[0079] Among them, the basic database 210, the technical database 220, and the case study database 230 can use MySQL or Redis databases. MySQL and Redis databases have advantages such as high performance, high reliability, and ease of use. MySQL is an open-source relational database management system that supports multiple users, multiple threads, and multiple storage engines. It supports CRUD operations on data using SQL and is widely used in web development.

[0080] The intelligent assessment module 300 pre-builds a comprehensive evaluation model based on technology, economy, and environment, captures a modeling sample set, uses the modeling sample set to iteratively train the comprehensive evaluation model, and outputs a converged comprehensive evaluation model. Taking enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters as input, it executes the comprehensive evaluation model and outputs the comprehensive evaluation results of the enterprise's corresponding treatment technology.

[0081] In this embodiment of the invention, the intelligent evaluation module 300 includes:

[0082] Model building unit 310 is used to pre-build a comprehensive evaluation model based on technology, economy and environment, capture a modeling sample set, use the modeling sample set to iteratively train the comprehensive evaluation model, and output a converged comprehensive evaluation model.

[0083] The indicator extraction unit 320 uses enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters as independent variables to calculate the unit flue gas investment cost, operating cost, emission reduction benefits, and carbon emissions throughout the entire life cycle of the treatment technology.

[0084] The comprehensive evaluation unit 330 is used to acquire data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, unit flue gas investment cost, operating cost and emission reduction benefits, and carbon emissions throughout the entire life cycle of treatment technology from the comprehensive evaluation model, and integrate them into an analysis dataset. Based on the comprehensive evaluation model, the analysis dataset is comprehensively analyzed to output the comprehensive evaluation results of the enterprise's corresponding treatment technology.

[0085] In this embodiment of the invention, an intelligent evaluation module 300 is provided. The intelligent evaluation module 300 consists of a model building unit 310, an indicator extraction unit 320, and a comprehensive evaluation unit 330. The comprehensive evaluation model can comprehensively consider multiple key dimensions of pollutant treatment technologies in the steel industry, including technical feasibility, economic cost-effectiveness, and environmental impact. This multi-dimensional comprehensive evaluation method, compared to traditional single-dimensional evaluation methods, can more accurately reflect the true performance and potential value of treatment technologies. The indicator extraction unit 320 can use enterprise pollutant emission data, treatment technology parameters, and equipment operating parameters as independent variables to calculate key indicators such as unit flue gas investment cost, operating cost, emission reduction benefits, and carbon emissions throughout the entire life cycle of the treatment technology. This comprehensive indicator extraction method can provide enterprises with detailed performance analysis of treatment technologies, helping them to deeply understand the performance of various technologies in different dimensions. The comprehensive evaluation unit 330 can integrate enterprise pollutant emission data, treatment technology parameters, equipment operating parameters, and various cost-effectiveness indicators into an analysis dataset, and perform comprehensive analysis on this data based on the comprehensive evaluation model, outputting the enterprise's corresponding comprehensive evaluation results for treatment technologies. This comprehensive evaluation approach can provide enterprises with comprehensive decision support, helping them to assess the applicability of governance technologies from multiple perspectives and select the most suitable technology solution for their own needs.

[0086] The associated recommendation module 400 is used to obtain the comprehensive evaluation results of the enterprise's governance technologies, match these results with application cases in the case library 230, and generate an intelligent recommendation queue of governance technologies that meets the enterprise's needs. Figure 2The diagram shows the intelligent recommendation queue interface for governance technologies generated by the association recommendation module 400, which meets the needs of enterprises. When government departments / enterprises have not invested in the corresponding denitrification devices, relevant industry parameters can be filled in the association recommendation module 400. Clicking "View Recommendation Results" will provide recommendations for feasible pollution prevention and control technologies, fugitive control measures for the entire production process, and related case studies, covering the entire process from source emission reduction and process control to end-of-pipe treatment.

[0087] The system management module 500 is communicatively connected to the enterprise terminal 100, the data management module 200, and the intelligent evaluation module 300, and is used to manage and control the enterprise terminal 100, the data management module 200, and the intelligent evaluation module 300.

[0088] In this embodiment of the invention, the system management module 500 includes:

[0089] The system login unit 510 is used to collect user login verification information and verify the user login verification information;

[0090] The system administrator unit 520 is used to maintain the basic database 210 and to update, modify, and delete data in the basic database 210.

[0091] The technician management unit 530 is used to verify technician information and support technical evaluators in adjusting the weights and quantification methods of governance technical indicators. Figure 5 The diagram illustrates the interface for technical evaluators to adjust the weights and quantification methods of governance technology indicators. Technical evaluators or relevant experts assign weights to the indicators, configure them, and store the final indicator weights and quantification results for use by the underlying algorithms of the governance technology evaluation and intelligent recommendation modules. The indicator weight allocation can be adjusted according to relevant policies and standards to achieve the most suitable governance technology recommendations.

[0092] Enterprise Management Unit 540 interacts with Enterprise Terminal 100 to obtain the governance technology parameters uploaded by Enterprise Terminal 100. Based on enterprise pollutant emission data, governance technology parameters, and equipment operating parameters, it assesses the enterprise's control level. It's worth noting that Enterprise Management Unit 540 can visualize core data content via a map, primarily including the distribution of enterprise control levels, enterprise governance technology analysis, and feasible pollution prevention technologies. The first two parts can be viewed on a map. The distribution of enterprise control levels includes the number and industry of the surveyed A, B, C, and D level enterprises. The distribution of enterprise governance technologies includes commonly used NOs of the surveyed enterprises. x The categories of governance technologies and their corresponding proportions.

[0093] In this embodiment of the invention, a comprehensive evaluation model based on technology, economy, and environment is constructed. This model comprehensively assesses the treatment technology from multiple dimensions, including technical feasibility, economic cost-effectiveness, and environmental impact. By collecting real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise needs from the enterprise's perspective, the model ensures the real-time nature, accuracy, and relevance of the data used in the assessment. Based on this rich data, more scientific and credible evaluation results are derived. Compared to traditional assessment methods based on experience and simple calculations, this system can more accurately grasp the actual effects and potential problems of governance technologies. It can also provide personalized assessment and recommendation services based on factors such as the size of different enterprises, production processes, pollutant emission characteristics, and economic conditions, providing strong support for enterprises' technology selection and management decisions. Furthermore, the system's infrastructure adopts the Ruoyi-Vue system architecture. Ruoyi-Vue is an open-source enterprise-level rapid development platform based on Vue.js and Spring Boot, providing a complete front-end and back-end separation solution. Utilizing Ruoyi's code generation function, it quickly generates front-end and back-end CRUD functions based on database tables. This intelligent assessment platform is built on the Ruoyi-Vue framework, and the system adopts the classic MVC three-tier architecture design, following the B / S architecture. The system architecture is divided into a view layer (View), a controller layer (Controller), a business logic layer (Service), and a data persistence layer (Dao). The view layer uses Vue.js as the front-end framework and Element-UI as the UI component library. The front-end page uses the HTTP protocol to send requests to the controller layer. The controller and business logic layers use Spring Boot as the backend framework and MyBatis-Plus as the persistence layer framework. The controller receives requests from the frontend, parses and validates the data, and then sends the data to the service layer for business logic processing. Finally, the data persistence layer accesses the database server to perform database operations, and after processing, returns the results to the controller, which then sends the results to the frontend for page processing.

[0094] In this embodiment of the invention, the pre-constructed comprehensive evaluation model based on technology, economy, and environment is based on a recurrent neural network model, and also includes an input layer and an output layer. An autoencoder is set between the input layer and the recurrent neural network model. The autoencoder includes an encoding layer and a decoding layer. The autoencoder is used to identify the correlation indicators of enterprise pollutant emission data, treatment technology parameters, equipment operating parameters, unit flue gas investment cost, operating cost and emission reduction benefits, and carbon emission of treatment technology throughout its entire life cycle, and to label the correlation indicators. The recurrent neural network model is used to extract multi-scale features from the labeled correlation indicators and to concatenate and fuse the multi-scale features to obtain a fused feature vector. A regression scorer is set between the recurrent neural network model and the output layer. A BiLSTM model is introduced into the regression scorer. The adaptive weights of the indicator features in the fused feature vector are calculated based on the BiLSTM model, and the comprehensive evaluation value of the treatment technology is determined by combining the attribute attention mechanism with the sigmoid function. A level discriminator is set between the regression scorer and the output layer. The level discriminator determines the enterprise control level based on the gradient inversion mechanism combined with the comprehensive evaluation value of the treatment technology.

[0095] This invention provides a method for iteratively training a comprehensive evaluation model using a modeling sample set. Figure 6 This diagram illustrates the implementation process of an iterative training method for a comprehensive evaluation model using a modeling sample set. The iterative training method specifically includes:

[0096] S101, Traverse the basic database 210, technical database 220, and case library 230, capture the modeling sample set based on the basic database 210, technical database 220, and case library 230, and divide the modeling sample set into a training set and a test set, wherein the sample ratio of the training set and the test set is 6:1;

[0097] S102, load the pre-built comprehensive evaluation model, and preset the training rounds, joint loss function, optimization strategy and hyperparameters of the comprehensive evaluation model;

[0098] S103, Obtain the training set, use the training set to iteratively train the comprehensive evaluation model, and output a converged comprehensive evaluation model;

[0099] S104: Obtain the test set, use the test set as input, execute the comprehensive evaluation model, output the test results, and use the mean square error to calculate the expected value of the difference between the test results and the true results.

[0100] The formula for calculating the mean square error is as follows:

[0101]

[0102] in, These are the test results and the actual results, respectively. This represents the number of samples in the test set.

[0103] S105, determine whether the expected value exceeds the preset expected threshold;

[0104] S106. If the expected value exceeds the preset expected threshold, the model hyperparameters are adjusted using the RMSprop method, and the comprehensive evaluation model is iteratively trained.

[0105] S107. If the expected value does not exceed the preset expected threshold, output the converged comprehensive evaluation model.

[0106] This invention provides a method for comprehensive analysis of datasets based on a comprehensive evaluation model. Figure 7 This diagram illustrates the implementation flow of a comprehensive analysis method for datasets based on a comprehensive evaluation model. Specifically, this method includes:

[0107] S201 identifies and analyzes the correlation indicators of enterprise pollutant emission data, treatment technology parameters, equipment operating parameters, unit flue gas investment cost, operating cost and emission reduction benefits, and carbon emission of treatment technology throughout its entire life cycle, and marks the correlation indicators.

[0108] S202, extract multi-scale features from the associated indicators of the labels, and concatenate and fuse the multi-scale features to obtain a fused feature vector.

[0109] S203, based on the BiLSTM model, calculates the adaptive weights of the indicator features in the fused feature vector, and determines the comprehensive evaluation value of governance technology by combining the attribute attention mechanism with the sigmoid function;

[0110] The adaptive weights of the indicator features are calculated using the following formula:

[0111]

[0112]

[0113]

[0114] in, Indicates adaptive weights, For attention scores based on attribute attention mechanisms, This represents a learnable weight vector, while It is the hyperbolic tangent function. These are the weight matrix and the fused feature vector, respectively. For bias terms, These represent the forward hidden state and the backward hidden state at time step t, respectively.

[0115] The comprehensive evaluation value of the treatment technology is calculated using the following formula:

[0116]

[0117]

[0118] in, This is a comprehensive evaluation value for governance technologies. For the rating weight vector, This is a weighted fusion feature vector;

[0119] S204 determines the enterprise's management and control level based on a comprehensive evaluation value that combines the gradient reversal mechanism with governance technology.

[0120] In this embodiment of the invention, by extracting multi-scale features of labeled and associated indicators, the dynamic change patterns of indicators under different time dimensions can be captured. After multi-scale feature extraction, feature vectors of different time scales and types are integrated into a unified fused feature vector through a splicing and fusion operation. This fusion operation breaks down information silos of single feature dimensions, enabling the model to utilize both static parameters and dynamic performance simultaneously, thereby more accurately characterizing the comprehensive performance of governance technologies. Furthermore, the adaptive weights of indicator features in the fused feature vector, calculated based on the BiLSTM model, can automatically adjust the weight allocation according to the dynamic changes in the data. This adaptability allows the model to better cope with noise and outliers in the data, improving its robustness. Simultaneously, by combining attribute attention mechanisms with the sigmoid function, the model's focus on important features can be further strengthened, thereby improving the accuracy and reliability of the evaluation results. Comprehensive evaluation value determination: The comprehensive evaluation value of governance technologies determined by the above method can comprehensively reflect the comprehensive performance of governance technologies in multiple dimensions such as technology, economy, and environment. This comprehensive evaluation method not only considers the optimal solution of a single dimension but also takes into account the balance and coordination between multiple dimensions, providing steel enterprises with a more comprehensive and scientific decision-making basis.

[0121] On the other hand, embodiments of the present invention also provide an intelligent evaluation method for pollutant treatment technologies in the steel industry. Figure 8 This diagram illustrates the implementation process of an intelligent assessment method for pollutant treatment technologies in the steel industry. The intelligent assessment method specifically includes:

[0122] S10 collects real-time data on enterprise pollutant emissions, treatment technology parameters, equipment operating parameters, and enterprise demand information, and presents a visually intelligent recommendation queue of treatment technologies.

[0123] S20: Collect technical parameters for pollutant treatment in the steel industry, upload the technical parameters for pollutant treatment in the steel industry to the corresponding database, store and update the technical parameters in the database;

[0124] S30: Pre-build a comprehensive evaluation model based on technology, economy and environment, capture a modeling sample set, use the modeling sample set to iteratively train the comprehensive evaluation model, and output a converged comprehensive evaluation model. Take the enterprise's pollutant emission data, treatment technology parameters and equipment operation parameters as input, execute the comprehensive evaluation model, and output the comprehensive evaluation results of the enterprise's corresponding treatment technology.

[0125] S40: Obtain the comprehensive evaluation results of the enterprise's governance technologies, match the comprehensive evaluation results with the application cases in the case library 230, and generate a smart recommendation queue of governance technologies that meets the enterprise's needs.

[0126] In summary, this invention provides an intelligent evaluation system and method for pollutant treatment technologies in the steel industry. In this embodiment, a comprehensive evaluation model based on technology, economy, and environment is constructed to comprehensively evaluate treatment technologies from multiple dimensions, including technical feasibility, economic cost-effectiveness, and environmental impact. By collecting real-time pollutant emission data, treatment technology parameters, equipment operating parameters, and enterprise needs information from the enterprise, the system ensures the real-time nature, accuracy, and relevance of the data used in the evaluation. Based on this rich data, more scientific and credible evaluation results are obtained. Compared with traditional evaluation methods based on experience and simple calculations, this system can more accurately grasp the actual effects and potential problems of treatment technologies. Furthermore, it can provide personalized evaluation and recommendation services based on factors such as the size of different enterprises, production processes, pollutant emission characteristics, and economic conditions, providing strong support for enterprises' technology selection and management decisions.

[0127] It should be noted that, for the sake of simplicity, the foregoing embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to the present invention. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0128] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can still combine, add, delete, or otherwise adjust the features of the various embodiments of the present invention according to the circumstances without conflict or creative effort, thereby obtaining different technical solutions that do not fundamentally depart from the concept of the present invention. These technical solutions also fall within the scope of protection of the present invention.

Claims

1. An intelligent evaluation system for steel industry pollutant treatment technology, characterized in that, The system comprises: An enterprise end for collecting enterprise pollutant emission data, treatment technology parameters, equipment operation parameters and enterprise demand information in real time, and visualizing a treatment technology intelligent recommendation queue; A data management module for collecting steel industry pollutant treatment technology parameters, uploading the steel industry pollutant treatment technology parameters to a corresponding database, and storing and updating the technology parameters in the database; An intelligent evaluation module for pre-building a comprehensive evaluation model based on technology-economy-environment, grabbing a modeling sample set, iteratively training the comprehensive evaluation model using the modeling sample set, outputting a converged comprehensive evaluation model, inputting enterprise pollutant emission data, treatment technology parameters and equipment operation parameters into the comprehensive evaluation model, executing the comprehensive evaluation model, and outputting a corresponding treatment technology comprehensive evaluation result of the enterprise; An association recommendation module for obtaining the corresponding treatment technology comprehensive evaluation result of the enterprise, matching the treatment technology comprehensive evaluation result with application cases in a case library, and generating a treatment technology intelligent recommendation queue meeting the enterprise demand information.

2. The steel industry pollutant control technology intelligent evaluation system according to claim 1, characterized in that: The system further comprises: A system management module in communication connection with the enterprise end, the data management module and the intelligent evaluation module, for managing and controlling the enterprise end, the data management module and the intelligent evaluation module.

3. The steel industry pollutant control technology intelligent evaluation system according to claim 2, characterized in that: The system management module comprises: A system login unit for collecting user login verification information and checking the user login verification information; A system administrator unit for maintaining a basic database and updating, modifying and deleting the basic database; A technician management unit for verifying technician information and supporting a technician to adjust weights of treatment technology indexes and quantification methods; 4. The steel industry pollutant remediation technology intelligent evaluation system of claim 1, wherein: An enterprise management unit for interacting with the enterprise end, obtaining treatment technology parameters uploaded by the enterprise end, and evaluating an enterprise management level based on enterprise pollutant emission data, treatment technology parameters and equipment operation parameters. The data management module comprises: A basic database for storing enterprise basic information and enterprise management levels; A technology database for storing enterprise pollutant emission data, treatment technology parameters and equipment operation parameters; 5. The steel industry pollutant control technology intelligent evaluation system according to claim 4, characterized in that: A case library for storing application cases associated with steel industry pollutant treatment technologies. The intelligent evaluation module comprises: A model building unit for pre-building a comprehensive evaluation model based on technology-economy-environment, grabbing a modeling sample set, iteratively training the comprehensive evaluation model using the modeling sample set, and outputting a converged comprehensive evaluation model; An index extraction unit for taking enterprise pollutant emission data, treatment technology parameters and equipment operation parameters as independent variables, and respectively calculating unit flue gas investment cost, operation cost and emission reduction benefit, and treatment technology full life cycle carbon emission; A comprehensive evaluation unit for obtaining enterprise pollutant emission data, treatment technology parameters, equipment operation parameters, unit flue gas investment cost, operation cost and emission reduction benefit, and treatment technology full life cycle carbon emission from the comprehensive evaluation model, integrating the data into an analysis data set, comprehensively analyzing the analysis data set based on the comprehensive evaluation model, and outputting a corresponding treatment technology comprehensive evaluation result of the enterprise.

6. The steel industry pollutant control technology intelligent evaluation system according to claim 5, characterized in that: The pre-constructed technology-economy-environment comprehensive evaluation model is based on a recurrent neural network model, further comprising an input layer and an output layer, and a self-encoder is arranged between the input layer and the recurrent neural network model, the self-encoder comprising an encoding layer and a decoding layer, the self-encoder being used to identify and mark the associated indexes of the enterprise pollutant emission data, the treatment technology parameters, the equipment operation parameters, the unit flue gas investment cost, the operation cost and emission reduction benefit, and the total life cycle carbon emission of the treatment technology, the recurrent neural network model being used to extract multi-scale features from the marked associated indexes, and the multi-scale features being spliced and fused to obtain a fused feature vector, a regression scorer being arranged between the recurrent neural network model and the output layer, a BiLSTM model being introduced into the regression scorer, the adaptive weight of the index features in the fused feature vector being calculated based on the BiLSTM model, and the comprehensive evaluation value of the treatment technology being determined by combining the sigmoid function with the attribute attention mechanism, and a grade discriminator being arranged between the regression scorer and the output layer, the enterprise management and control grade being determined based on the gradient reversal mechanism and the comprehensive evaluation value of the treatment technology.

7. The steel industry pollutant control technology intelligent evaluation system according to claim 6, characterized in that: The method for iteratively training the comprehensive evaluation model by using the modeling sample set comprises the following steps: Traverse the basic database, the technology database and the case library, capture the modeling sample set based on the basic database, the technology database and the case library, and divide the modeling sample set into a training set and a test set; Load the pre-constructed comprehensive evaluation model, and preset the training rounds of the comprehensive evaluation model, the joint loss function, the tuning strategy and the hyperparameters; Obtain the training set, iteratively train the comprehensive evaluation model by using the training set, and output the converged comprehensive evaluation model; Obtain the test set, input the test set into the comprehensive evaluation model, output the test result, and calculate the expected value of the difference between the test result and the true result by using the mean square error; Determine whether the expected value exceeds the preset expected threshold value; If the expected value exceeds the preset expected threshold value, adjust the model hyperparameters by using the RMSprop method, and continue to iteratively train the comprehensive evaluation model; If the expected value does not exceed the preset expected threshold value, output the converged comprehensive evaluation model.

8. The steel industry pollutant control technology intelligent evaluation system of claim 7, wherein: The method for comprehensively analyzing the analysis data set based on the comprehensive evaluation model comprises the following steps: Identify and mark the associated indexes of the enterprise pollutant emission data, the treatment technology parameters, the equipment operation parameters, the unit flue gas investment cost, the operation cost and emission reduction benefit, and the total life cycle carbon emission of the treatment technology in the analysis data set; Extract multi-scale features from the marked associated indexes, and splice and fuse the multi-scale features to obtain a fused feature vector.

9. The steel industry pollutant remediation technology intelligent evaluation system of claim 8, wherein: The method for comprehensively analyzing the analysis data set based on the comprehensive evaluation model further comprises the following steps: Calculate the adaptive weight of the index features in the fused feature vector based on the BiLSTM model, and determine the comprehensive evaluation value of the treatment technology by combining the sigmoid function with the attribute attention mechanism; Determine the enterprise management and control grade based on the gradient reversal mechanism and the comprehensive evaluation value of the treatment technology.

10. The intelligent evaluation method of the steel industry pollutant treatment technology, which is implemented by using the steel industry pollutant treatment technology intelligent evaluation system according to any one of claims 1-9, characterized in that: The method comprises the following steps: Collect the enterprise pollutant emission data, the treatment technology parameters, the equipment operation parameters and the enterprise demand information in real time, and visually present the intelligent recommendation queue of the treatment technology; Collect the steel industry pollutant treatment technology parameters, upload the steel industry pollutant treatment technology parameters to the corresponding database, store and update the technology parameters in the database; Pre-construct a comprehensive evaluation model based on technology-economy-environment, capture the modeling sample set, use the modeling sample set to iteratively train the comprehensive evaluation model, output the converged comprehensive evaluation model, use the enterprise pollutant emission data, treatment technology parameters and equipment operation parameters as input, execute the comprehensive evaluation model, and output the corresponding treatment technology comprehensive evaluation result of the enterprise; Obtain the corresponding treatment technology comprehensive evaluation result of the enterprise, match the treatment technology comprehensive evaluation result with the application cases in the case library, and generate a treatment technology intelligent recommendation queue that meets the enterprise demand information.

Images