Steel desulfurization technology multidimensional comprehensive evaluation method and system

Abstract

The present application relates to a kind of steel desulfurization technology multidimensional comprehensive evaluation method and system, method includes: obtaining multidimensional original monitoring information, the original monitoring information is preprocessed, and standardization preprocessed monitoring data set is obtained;Standardization preprocessed monitoring data set is used to build multidimensional comprehensive representation system, multidimensional comprehensive representation system is input into multidimensional evaluation model, and evaluation score is obtained;Multidimensional evaluation model is obtained using training set training;The index subjective weight of multidimensional comprehensive representation system and the objective weight of each secondary index are calculated by multidimensional evaluation model, and the closeness of each desulfurization technology scheme and positive ideal solution, negative ideal solution is calculated based on approximation ideal solution distance method, and short board analysis scheme is obtained.The present application can comprehensively, objectively, quantitatively evaluate the comprehensive performance of different steel desulfurization technologies, and provide accurate decision support for steel enterprises desulfurization process selection, existing desulfurization system upgrading, environmental protection operation optimization.

Description

Technical Field

[0001] This invention relates to the fields of air pollution control, industrial environmental protection technology evaluation, and steel energy conservation and carbon reduction, and in particular to a multi-dimensional comprehensive evaluation method and system for steel desulfurization technology. Background Technology

[0002] The steel industry is a key sector for industrial air pollutant emissions in my country. Sulfur dioxide (SO2) in flue gas is a major air pollutant, and direct emissions can cause environmental problems such as acid rain and smog, seriously harming the ecological environment and human health. Currently, various desulfurization technologies, including wet desulfurization, dry desulfurization, semi-dry desulfurization, ammonia desulfurization, and circulating fluidized bed desulfurization, are widely used in various processes of steel production. Different desulfurization technologies vary significantly in terms of desulfurization efficiency, operating costs, environmental friendliness, site requirements, and process compatibility.

[0003] The current evaluation methods for steel desulfurization technology in the industry have obvious shortcomings: the evaluation dimensions are too limited; qualitative indicators are difficult to quantify; the models have poor adaptability; and there is a lack of consideration for the entire life cycle.

[0004] Therefore, there is an urgent need for a comprehensive evaluation system and method for steel desulfurization technology that covers four dimensions: technology, economy, environmental protection, and operation, integrates multiple algorithm models, combines qualitative and quantitative methods, and considers the entire life cycle. This system and method can comprehensively quantify the overall performance of different desulfurization technologies and provide a scientific basis for steel enterprises to select desulfurization processes, optimize existing systems, and manage environmental compliance. Summary of the Invention

[0005] To address the problems of existing technologies, such as single evaluation dimensions, subjective qualitative indicators, low model accuracy, and lack of full life-cycle consideration, the purpose of this invention is to provide a multi-dimensional comprehensive evaluation method and system for steel desulfurization technology. This aims to solve the problem that existing steel desulfurization technology evaluations often focus on a single desulfurization efficiency indicator, neglecting key dimensions such as unit operation stability, process adaptability, economic cost, secondary pollution, and site adaptability. This results in biased evaluation results and a lack of scientific basis for technology selection. The invention constructs a multi-level, multi-dimensional evaluation indicator system, integrates various classic evaluation algorithms, and achieves objective, comprehensive, and quantitative evaluation of steel desulfurization technology, thus overcoming the shortcomings of existing evaluation technologies.

[0006] To achieve the above objectives, the present invention provides the following solution: A multi-dimensional comprehensive evaluation method for steel desulfurization technology includes: Obtain multi-dimensional raw monitoring information, preprocess the raw monitoring information to obtain a standardized preprocessed monitoring dataset; A multi-dimensional comprehensive characterization system is constructed using the standardized preprocessed monitoring dataset. The multi-dimensional comprehensive characterization system is then input into a multi-dimensional evaluation model to obtain an evaluation score. The multi-dimensional evaluation model is trained using a training set. The training set includes historical monitoring data and evaluation sample labels under different desulfurization processes and different operating conditions. The subjective weights of the indicators and the objective weights of each secondary indicator in the multi-dimensional comprehensive characterization system are calculated using the multi-dimensional evaluation model. Based on the approximation ideal solution distance method, the closeness of each desulfurization technology scheme to the positive and negative ideal solutions is calculated to obtain the shortcoming analysis scheme.

[0007] Optionally, obtaining the standardized preprocessed monitoring dataset includes: Obtain multi-dimensional raw monitoring information; the raw monitoring information includes: flue gas parameters, equipment operating parameters, environmental monitoring parameters, and operation and maintenance data; The original monitoring information is then subjected to standardization, filtering and noise reduction, and feature selection processes in sequence to obtain the standardized preprocessed monitoring dataset.

[0008] Optionally, constructing the multi-dimensional comprehensive representation system includes: A primary evaluation dimension is constructed to evaluate technical performance, economic cost, environmental benefits, and operational adaptability. Secondary evaluation indicators corresponding to the primary evaluation dimensions are extracted from the standardized preprocessed monitoring dataset. A mapping relationship between the primary evaluation dimensions and the secondary evaluation indicators is established to form the multi-dimensional comprehensive characterization system.

[0009] Optionally, the secondary evaluation indicators corresponding to the primary evaluation dimension include: Technical performance dimensions: desulfurization efficiency, desulfurizing agent utilization rate, system response speed, load adaptability range, and equipment processing capacity; Economic cost dimensions: unit investment cost for flue gas desulfurization, unit operating cost for flue gas desulfurization, desulfurizing agent consumption, equipment maintenance cost, and labor operation and maintenance cost; Environmental benefits dimensions: secondary pollutant emissions, desulfurization by-product resource utilization rate, wastewater reuse rate, solid waste harmless treatment rate, and carbon emissions per unit of desulfurization. Operational compatibility dimensions: unit safety and stability rate, process compatibility, equipment footprint, adaptability to old plant renovation, and level of automation control.

[0010] Optionally, calculating the subjective weights of the indicators and the objective weights of each secondary indicator in the multi-dimensional comprehensive representation system includes: The hierarchical structure of indicators is constructed based on the analytic hierarchy process, and pairwise comparison evaluation matrices are built to calculate eigenvectors and complete consistency checks to obtain the subjective weights of the indicators. The objective weights of each secondary indicator are then calculated based on the entropy weight method.

[0011] Optionally, obtaining the shortcoming analysis scheme includes: The subjective weights of the aforementioned indicators and the objective weights of each secondary indicator are combined using a linear weighting method to obtain the combined weights. The initial index matrix that has been quantified in the multi-dimensional comprehensive representation system is standardized to obtain a standardized judgment matrix. The standardized judgment matrix is ​​then multiplied by the combined weights to obtain a weighted standardized matrix. Traverse the weighted standardized matrix, extract the set of maximum values ​​of each indicator as the positive ideal solution, extract the set of minimum values ​​of each indicator as the negative ideal solution, and calculate the distances of each desulfurization technology scheme to the positive ideal solution and the distances to the negative ideal solution to obtain the comprehensive closeness. The desulfurization technology solutions are ranked from highest to lowest based on their overall similarity, and the shortcomings are identified by combining the scores of each individual indicator to obtain the shortcomings analysis solution.

[0012] Optionally, extracting the set of maximum values ​​for each indicator as the positive ideal solution and extracting the set of minimum values ​​for each indicator as the negative ideal solution includes: ; ; in, For the positive ideal solution, For a negative ideal solution, For the m-th desulfurization technology scheme, the weighted standardized matrix element corresponding to the j-th evaluation index.

[0013] Optionally, calculating the distances to the positive ideal solution and the negative ideal solution for each desulfurization technology scheme includes: ; ; in, The distance from each desulfurization technology scheme to the ideal solution. To calculate the distance to the negative ideal solution for each desulfurization technology scheme, For the matrix elements within the weighted normalization matrix, For the i-th desulfurization technology, the weighted calculated value of the j-th evaluation index is given.

[0014] Optionally, obtaining the overall closeness includes: ; in, For overall relevance.

[0015] To achieve the above objectives, the present invention also provides a multi-dimensional comprehensive evaluation system for steel desulfurization technology, comprising: The preprocessing module is used to acquire multi-dimensional raw monitoring information, preprocess the raw monitoring information, and obtain a standardized preprocessed monitoring dataset. The characterization system construction module is used to construct a multi-dimensional comprehensive characterization system using the standardized preprocessed monitoring dataset. The multi-dimensional evaluation module is used to input the multi-dimensional comprehensive characterization system into the multi-dimensional evaluation model to obtain the evaluation score; the multi-dimensional evaluation model is obtained by training a training set; the training set includes: historical monitoring data and evaluation sample labels under different desulfurization processes and different operating conditions. The subjective weights of the indicators and the objective weights of each secondary indicator in the multi-dimensional comprehensive characterization system are calculated using the multi-dimensional evaluation model. Based on the approximation ideal solution distance method, the closeness of each desulfurization technology scheme to the positive and negative ideal solutions is calculated to obtain the shortcoming analysis scheme.

[0016] The beneficial effects of this invention are as follows: This invention breaks through the limitations of traditional single desulfurization efficiency indicators and constructs a multi-dimensional indicator system covering technical performance, economic cost, environmental benefits, and operational adaptability. It is adaptable to the actual working conditions of different processes such as steel sintering, ironmaking, and steelmaking, and can simultaneously meet the needs of various business scenarios such as process selection for new projects, evaluation of existing desulfurization system upgrades, and evaluation of equipment operation and maintenance effects.

[0017] This invention standardizes and quantifies qualitative indicators such as process compatibility, and combines LCA lifecycle assessment with a full lifecycle cost model for calculation, effectively reducing subjective bias caused by expert evaluation and improving the standardization of the evaluation process and the credibility of the results. It employs an AHP-entropy weighting method combined with the TOPSIS approximation of the ideal solution distance method, taking into account both industry process experience and field measurement data, to output a comparative ranking of overall technical similarity and merits, thereby improving the model's versatility and the accuracy of evaluation calculations.

[0018] This invention relies on LCA to complete the collaborative accounting of pollution reduction and carbon reduction in the entire desulfurization process. It sets up a dynamic iteration model, an analysis mechanism for the sensitivity of indicators and the uncertainty range of results, and can update model parameters according to changes in production conditions and environmental standards. It can identify highly sensitive control indicators, locate the shortcomings of each participating technology, and provide enterprises with clear directions for optimization and improvement. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a multi-dimensional comprehensive evaluation method for steel desulfurization technology according to an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] like Figure 1 As shown in the figure, this embodiment discloses a multi-dimensional comprehensive evaluation method for steel desulfurization technology, including: acquiring multi-dimensional raw monitoring information; preprocessing the raw monitoring information to obtain a standardized preprocessed monitoring dataset; constructing a multi-dimensional comprehensive characterization system using the standardized preprocessed monitoring dataset; inputting the multi-dimensional comprehensive characterization system into a multi-dimensional evaluation model to obtain an evaluation score; training the multi-dimensional evaluation model using a training set; the training set includes: historical monitoring data and evaluation sample labels under different desulfurization processes and different operating conditions; calculating the subjective weights of the indicators and the objective weights of each secondary indicator of the multi-dimensional comprehensive characterization system through the multi-dimensional evaluation model; and calculating the closeness of each desulfurization technology scheme to the positive and negative ideal solutions based on the approximation ideal solution distance method.

[0024] Specifically, this embodiment discloses a multi-dimensional comprehensive evaluation method for steel desulfurization technology, including: Step 1: Multi-dimensional monitoring information collection and preprocessing; acquire four types of multi-dimensional monitoring information from the steel desulfurization system: flue gas parameters, equipment operating parameters, environmental monitoring parameters, and operation and maintenance data. Standardize, filter and reduce noise, and perform feature selection on the raw monitoring data to obtain a standardized preprocessed monitoring dataset. Data collection scope: Through IoT sensors, online environmental monitoring platforms, and manual record-keeping, multi-dimensional monitoring information is collected throughout the entire steel desulfurization system, specifically categorized into four main types: Flue gas parameters: inlet / outlet SO2 concentration, flue gas flow rate, flue gas temperature, dust content, and oxygen content; Equipment operating parameters: desulfurization tower load, induced draft fan / circulating pump power, number of equipment start-ups and shutdowns, fault duration, and instantaneous consumption of desulfurizing agent; Environmental monitoring parameters: COD of discharged wastewater, suspended solids, amount of desulfurization waste residue generated, concentration of fugitive dust, and equipment operating noise; Operation and maintenance data: daily / monthly total consumption of desulfurizing agent, frequency of replacement of spare parts, duration of manual inspection, and equipment maintenance costs.

[0025] Step 2: Construction of a multi-dimensional comprehensive characterization system; Based on four primary evaluation dimensions—technical performance, economic cost, environmental benefits, and operational adaptability—a primary evaluation characterization system is constructed. Corresponding secondary evaluation indicators are extracted from the pre-processed monitoring dataset, and a mapping relationship between primary and secondary indicators is established to form a complete multi-dimensional comprehensive characterization system. This invention abandons the single-index evaluation model and establishes four primary evaluation dimensions: technical performance, economic cost, environmental benefits, and operational adaptability, along with 14 secondary indicators, forming a hierarchical, multi-dimensional comprehensive characterization system. The indicators are quantified and mapped, as shown in Table 1.

[0026] Table 1 Indicator mapping and quantification rules: Quantitative indicators: The preprocessed monitoring data is used directly, and the index dispersion is calculated by combining the entropy weight method to complete the preliminary quantification. Environmental indicators: Life cycle assessment (LCA) is adopted to trace the entire process of desulfurizing agent production, transportation, equipment energy consumption, and by-product disposal, and to calculate pollutant emissions and carbon footprint; Qualitative indicators (process compatibility, modification adaptability, etc.): A 5-level assignment method (excellent=5, good=4, average=3, poor=2, poor=1) is used to convert them into 01 interval values ​​to achieve unified quantification; Economic cost indicators: Using a full life cycle cost model, the total investment and disposal expenses of the project throughout its entire life cycle are statistically analyzed to obtain quantitative values ​​for each cost indicator.

[0027] A nonlinear mapping relationship between primary and secondary indicators is established, and an arctangent decay model is used to fit the correlation between the indicators, forming a complete multi-dimensional comprehensive characterization system.

[0028] Step 3: Multi-dimensional evaluation model training and construction; pre-construct a comprehensive evaluation model integrating the analytic hierarchy process (AHP), entropy weight method, and distance method for approximating the ideal solution; collect historical monitoring data and evaluation samples under different desulfurization processes and operating conditions as the training set; iteratively train until the model converges, and solidify the model parameters. More specifically: 1. Overall Model Architecture: The multi-dimensional evaluation model of this invention is divided into five layers: input layer → hierarchical analysis layer → weight fusion layer → comprehensive evaluation layer → output layer. It integrates three major algorithms: Analytic Hierarchy Process (AHP), entropy weight method, and TOPSIS method, taking into account both subjective process experience and objective data patterns.

[0029] Input layer: Loads a multi-dimensional comprehensive representation system; Hierarchical analysis layer: Based on the analytic hierarchy process, a hierarchical structure of indicators is built, a pairwise comparison and evaluation matrix is ​​constructed, eigenvectors are calculated and consistency checks are performed (CR < 0.1 is considered acceptable), and the subjective weights of the indicators are obtained; Weight fusion layer: The objective weights of each secondary indicator are calculated based on the entropy weight method, and the subjective weights and objective weights are fused by the linear weighting method to obtain the combined weights, thus avoiding the defects of single weighting. Comprehensive evaluation layer: Integrates the TOPSIS algorithm to complete index positiveization, matrix standardization, ideal solution calculation, and closeness calculation; Output layer: Outputs model evaluation scores, indicator weights, and technology ranking results.

[0030] 2. Model training process: Historical application data of desulfurization technology under different working conditions in the steel industry and measured evaluation samples are selected as training sets and input into the model in batches for iterative training: After each round of training, the model output results are compared with the actual human comprehensive evaluation results, and the model parameters are corrected; when the evaluation error of 5 consecutive rounds of training is less than 1%, the model is judged to have converged and the training is completed.

[0031] Step 4: Comprehensive evaluation and scheme ranking analysis; Load the trained evaluation model and multi-dimensional comprehensive representation system, use the combined weighting method to determine the weight of each secondary indicator, calculate the closeness of each participating desulfurization technology to the positive ideal solution and negative ideal solution based on the distance method of approximating the ideal solution, and output the comprehensive evaluation results, the ranking of the advantages and disadvantages of the technologies and the conclusion of the shortcoming analysis in combination with the dynamic weight adjustment mechanism.

[0032] Furthermore, the preprocessing of multi-dimensional monitoring information in step 1 specifically includes: Content 1: Max-Min Normalization Processing; The Max-Min Normalization algorithm is used to perform a linear transformation on the original multi-dimensional monitoring information to eliminate the differences in the dimensions of different indicators and obtain standardized monitoring information. Content 2: Joint filtering and noise reduction processing; a spatial smoothing operator combined with an adaptive filtering algorithm is used to reduce noise in standardized monitoring information, remove random fluctuations and abnormal interference data, and generate filtered information; Content 3: Feature Filtering and Dataset Generation; Fast Fourier Transform is performed on the filtered information to extract data features. The filtering threshold is set by combining feature sparsity and data compression ratio to remove redundant data and finally obtain the preprocessed monitoring dataset. The multi-dimensional monitoring information specifically includes: flue gas parameters such as SO2 concentration, desulfurizing agent consumption, flue gas flow rate, and flue gas temperature; equipment operating parameters such as desulfurization tower operating load, fan power, equipment failure rate, and system start-up and shutdown duration; environmental monitoring parameters such as dust emission concentration, wastewater discharge, solid waste generation, and noise levels; and maintenance data such as manual maintenance time, spare parts wear and tear, and reagent replenishment frequency. More specifically: Data preprocessing workflow: Content 1: Max-Min Standardization: Using the formula: ; Where x is the original measured sample value of the indicator. This represents the maximum value of all samples for this indicator. x' is the minimum value of all samples of the indicator, and x' is the standardized value of the indicator, which ranges from [0,1].

[0033] The original data is normalized, and all index values ​​are mapped to the [0,1] interval to eliminate the influence of different dimensions such as pressure, concentration, and cost.

[0034] Content 2: Joint Filtering Noise Reduction: Spatial smoothing operators are used to handle local abrupt changes in time-series data, combined with adaptive filtering algorithms to remove outliers caused by sensor faults and instantaneous operating condition fluctuations. The noise reduction objective function is: ; Where f(D) represents the original monitoring data, and D(i) represents the data after noise reduction. , These are the weighting coefficients for spatial smoothing and adaptive filtering, respectively.

[0035] Content 3: Feature Filtering: Perform Fast Fourier Transform (FFT) on the denoised data to convert the time domain data into frequency domain data and extract core features; set thresholds based on feature sparsity and data compression ratio to remove invalid and redundant data, and finally obtain a preprocessed monitoring dataset with uniform format and reliable data.

[0036] Furthermore, the multi-dimensional comprehensive representation system in step 2 is divided into primary indicators and secondary indicators: The primary indicators include the dimensions of technical performance, economic cost, environmental benefits, and operational adaptability. The corresponding secondary evaluation indicators are as follows: (1) Technical performance dimensions: desulfurization efficiency, desulfurizing agent utilization rate, system response speed, load adaptability range, and equipment processing capacity; (2) Economic cost dimension: unit investment cost of flue gas desulfurization, unit operation cost of flue gas desulfurization, desulfurizing agent consumption, equipment maintenance cost, and labor operation and maintenance cost; (3) Environmental benefits dimension: secondary pollutant emissions, desulfurization by-product resource utilization rate, wastewater reuse rate, solid waste harmless treatment rate, and carbon emissions per unit of desulfurization; (4) Operational compatibility dimensions: unit safety and stability rate, process compatibility, equipment footprint, adaptability to old plant renovation, and level of automation control; A mapping relationship between primary dimensions and secondary evaluation indicators is established. Among them, the quantitative indicators of the technical performance dimension are quantified using measured data combined with the entropy weight method; the indicators of the economic cost dimension are quantified through ledger accounting combined with the full life cycle cost model; the indicators of the environmental benefits dimension are calculated using life cycle assessment (LCA); and the quantitative indicators of the operational adaptability dimension are quantified using measured statistical methods, while the qualitative indicators are quantified using a 5-level rating method.

[0037] Furthermore, the multi-dimensional evaluation model in step 3 includes an input layer, a hierarchical analysis layer, a weight fusion layer, a comprehensive evaluation layer, and an output layer; The hierarchical analysis layer builds a hierarchical structure of indicators based on the analytic hierarchy process, constructs a pairwise comparison and evaluation matrix, and completes hierarchical single sorting and consistency verification. The weight fusion layer combines the subjective weights obtained by the analytic hierarchy process (AHP) with the objective weights obtained by the entropy weighting method to calculate the combined weights, thus eliminating the subjectivity or data limitations of single weighting. The comprehensive evaluation layer introduces the distance method to approximate the ideal solution, completes the positiveization, standardization and weighting matrix construction of the indicators, and calculates the Euclidean distance and closeness between each scheme and the positive and negative ideal solutions; During model training, historical desulfurization monitoring data and known technology evaluation results are input, and the model parameters are iteratively optimized. When the evaluation error is less than the set threshold for three consecutive times, the model is considered to have converged.

[0038] Furthermore, indicator positiveization, matrix standardization, weighted matrix construction, ideal solution solving, proximity calculation, and scheme ranking are all internal execution sub-steps of the comprehensive evaluation layer of the multi-dimensional evaluation model, and are not processing flows independent of the evaluation model; the specific execution logic within the comprehensive evaluation layer is as follows: Content 1: Positive Indicator Processing. The quantification of all qualitative and quantitative indicators has been completed during the construction phase of the multi-dimensional comprehensive representation system. The input to the comprehensive evaluation layer is the initial indicator matrix that has already been quantified. Content 2: Matrix Standardization; The positiveized index matrix is ​​standardized to eliminate differences in the magnitude of index values, resulting in a standardized judgment matrix; Content 3: Construction of weighted matrix; Multiply the standardized judgment matrix by the combined weights to obtain the weighted standardized matrix; Content 4: Determination of the ideal solution; Extract the optimal value of each index from the weighted standardized matrix as the positive ideal solution, and extract the worst value of each index as the negative ideal solution; Content 5: Proximity Calculation; Calculate the Euclidean distance from each desulfurization technology scheme to the positive ideal solution and the negative ideal solution respectively, and calculate the comprehensive proximity based on the proximity formula; Content 6: Scheme Ranking; Rank the desulfurization technology schemes from highest to lowest based on their overall relevance (overall evaluation score); Combine the scores of each individual indicator to identify the weakest indicators of the participating technologies, and output the ranking of the technologies' strengths and weaknesses, along with the conclusions of the weakness analysis: 1. Indicator Directioning: In this system, economic cost, pollutant emissions, etc., are considered very small indicators (the smaller the value, the better the performance), while desulfurization efficiency, resource utilization rate, etc., are considered very large indicators (the larger the value, the better the performance). The very small indicators are directed using the formula X′=Qmax-Xa to unify the indicator types.

[0039] 2. Matrix Standardization and Weighting: Standardize the positively oriented index matrix to obtain the standardized judgment matrix B = (bij)m×n; combine the standardized matrix with the combined weights. Multiplying j together yields the weighted standardized matrix M.

[0040] Specifically, the method for obtaining the positiveized index matrix is ​​as follows: read the initial index matrix output by the characterization system, where the matrix rows correspond to the desulfurization schemes to be evaluated and the columns correspond to the secondary evaluation indicators; traverse each index column, retain the original value of the extremely large indicators, and use the formula X'=Qmax-Xa to perform element-wise positive transformation of the extremely small indicators; after all columns have been processed, the positiveized index matrix is ​​obtained.

[0041] 3. Determination of positive and negative ideal solutions: Traverse the weighted standardized matrix and extract the set of maximum values ​​for each index as the positive ideal solution. (Optimal state), extract the set of minimum values ​​for each index as the negative ideal solution. (Worst-case scenario): ; ; 4. Euclidean distance and proximity calculation: Calculate the distance from the i-th desulfurization technology to the positive ideal solution. Distance to the negative ideal solution : ; ; Calculate the overall closeness : ; Proximity The value ranges from [0,1]. The larger the value, the better the overall performance of the desulfurization technology.

[0042] Scheme ranking and weakness analysis: All participating steel desulfurization technologies are ranked from highest to lowest according to their relevance; at the same time, based on the scores of each secondary indicator, the weakness indicators of each technology are identified (such as excessively high operating costs, prominent secondary pollution, and poor adaptability to modification), and a special analysis report is generated.

[0043] Furthermore, the method also includes a dynamic iteration step: periodically collecting new operating data of the desulfurization system, re-inputting it into the model for incremental training, updating model parameters and indicator weights, and adapting to the dynamic changes in the production conditions and environmental standards of steel enterprises.

[0044] This embodiment also discloses a multi-dimensional comprehensive evaluation system for steel desulfurization technology, including four core functional modules: a pretreatment module, a characterization system construction module, an evaluation model construction module, and a multi-dimensional evaluation module. The preprocessing module is used to collect multi-dimensional monitoring information of the entire steel desulfurization process, and to complete standardization, filtering and noise reduction, feature selection and preprocessing, and output the preprocessed monitoring dataset. The characterization system construction module is used to build a four-level integrated multi-dimensional comprehensive characterization system, complete the quantification of qualitative and quantitative indicators, and establish the mapping relationship between primary and secondary indicators. The evaluation model building module is used to pre-build a comprehensive evaluation model that integrates the analytic hierarchy process, entropy weight method, and TOPSIS method, and to complete model training and convergence verification using historical data; The multi-dimensional evaluation module is used to load the indicator system and the trained model, complete the indicator weighting, ideal solution calculation, closeness analysis, and output the comprehensive evaluation results and ranking of each desulfurization technology.

[0045] Furthermore, the preprocessing module includes a standardization unit, a filtering and noise reduction unit, and a feature selection unit; The standardized processing unit executes the maximum-minimum normalization algorithm to unify the dimensions of the original monitoring data. The filtering and noise reduction unit integrates a spatial smoothing operator and an adaptive filtering algorithm to remove outlier data and random noise. The feature selection unit integrates the Fast Fourier Transform algorithm to select effective data features based on thresholds and generate a standard dataset.

[0046] Furthermore, the multi-dimensional evaluation module includes an indicator weighting unit, an ideal solution calculation unit, a closeness calculation unit, and a result output unit; The indicator weighting unit is used to run the analytic hierarchy process and the entropy weighting method to calculate the subjective weight, objective weight and combined weight of the indicator. The ideal solution calculation unit is used to extract the positive and negative ideal solutions from the weighting matrix; The proximity calculation unit is used to calculate the Euclidean distance and overall proximity between each technical solution and the ideal solution; The results output unit is used to visually display the evaluation score, the contribution ratio of indicators, the technology ranking, and the analysis report of the shortcomings indicators.

[0047] Furthermore, the system also includes a data storage module and a visualization module; The data storage module is used to store raw monitoring data, preprocessed datasets, indicator systems, model parameters, and historical evaluation results; The visualization module supports line charts, bar charts, pie charts, and radar charts to display indicator data, weight percentages, and technical scores, and automatically generates editable evaluation and analysis reports.

[0048] In one possible implementation, the first aspect provides a multi-dimensional comprehensive evaluation method for steel desulfurization technology. The core process is: data preprocessing → indicator system construction → model training → comprehensive evaluation and ranking. First, multi-source data from four sources—flue gas, equipment, environment, and operation and maintenance—of the steel desulfurization system are collected and preprocessed for noise reduction and standardization. Second, a multi-dimensional comprehensive representation system of "four primary dimensions + multiple secondary indicators" is built to achieve unified quantification of qualitative and quantitative indicators. Then, a comprehensive evaluation model integrating the analytic hierarchy process (AHP), entropy weight method, and TOPSIS method is constructed, and model training is completed using historical data. Finally, based on combined weighting and the distance method for approximating the ideal solution, the closeness of each desulfurization technology scheme is calculated, achieving ranking of advantages and disadvantages and analysis of shortcomings. The second aspect provides a multi-dimensional comprehensive evaluation system for steel desulfurization technology, corresponding to the above method, with four core modules. Each module is subdivided into independent units, realizing fully automated operation of the entire process, including automatic data collection, intelligent preprocessing, indicator system management, model calculation, and result visualization.

[0049] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A multi-dimensional comprehensive evaluation method for steel desulfurization technology, characterized in that, include: Obtain multi-dimensional raw monitoring information, preprocess the raw monitoring information to obtain a standardized preprocessed monitoring dataset; A multi-dimensional comprehensive characterization system is constructed using the standardized preprocessed monitoring dataset. The multi-dimensional comprehensive characterization system is then input into a multi-dimensional evaluation model to obtain an evaluation score. The multi-dimensional evaluation model is obtained by training a training set, which includes: historical monitoring data and evaluation sample labels under different desulfurization processes and different operating conditions. The subjective weights of the indicators and the objective weights of each secondary indicator in the multi-dimensional comprehensive characterization system are calculated using the multi-dimensional evaluation model. Based on the approximation ideal solution distance method, the closeness of each desulfurization technology scheme to the positive and negative ideal solutions is calculated to obtain the shortcoming analysis scheme.

2. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 1, characterized in that, The standardized preprocessed monitoring dataset obtained includes: Obtain multi-dimensional raw monitoring information; the raw monitoring information includes: flue gas parameters, equipment operating parameters, environmental monitoring parameters, and operation and maintenance data; The original monitoring information is then subjected to standardization, filtering and noise reduction, and feature selection processes in sequence to obtain the standardized preprocessed monitoring dataset.

3. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 1, characterized in that, Constructing the multi-dimensional comprehensive representation system includes: A primary evaluation dimension is constructed to evaluate technical performance, economic cost, environmental benefits, and operational adaptability. Secondary evaluation indicators corresponding to the primary evaluation dimensions are extracted from the standardized preprocessed monitoring dataset. A mapping relationship between the primary evaluation dimensions and the secondary evaluation indicators is established to form the multi-dimensional comprehensive characterization system.

4. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 3, characterized in that, The secondary evaluation indicators corresponding to the primary evaluation dimension include: Technical performance dimensions: desulfurization efficiency, desulfurizing agent utilization rate, system response speed, load adaptability range, and equipment processing capacity; Economic cost dimensions: unit investment cost for flue gas desulfurization, unit operating cost for flue gas desulfurization, desulfurizing agent consumption, equipment maintenance cost, and labor operation and maintenance cost; Environmental benefits dimensions: secondary pollutant emissions, desulfurization by-product resource utilization rate, wastewater reuse rate, solid waste harmless treatment rate, and carbon emissions per unit of desulfurization. Operational compatibility dimensions: unit safety and stability rate, process compatibility, equipment footprint, adaptability to old plant renovation, and level of automation control.

5. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 1, characterized in that, The calculation of the subjective weights of the indicators and the objective weights of each secondary indicator in the multi-dimensional comprehensive representation system includes: The hierarchical structure of indicators is constructed based on the analytic hierarchy process, and pairwise comparison evaluation matrices are built to calculate eigenvectors and complete consistency checks to obtain the subjective weights of the indicators. The objective weights of each secondary indicator are then calculated based on the entropy weight method.

6. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 1, characterized in that, Obtaining the aforementioned shortcoming analysis scheme includes: The subjective weights of the aforementioned indicators and the objective weights of each secondary indicator are combined using a linear weighting method to obtain the combined weights. The initial index matrix that has been quantified in the multi-dimensional comprehensive representation system is standardized to obtain a standardized judgment matrix. The standardized judgment matrix is ​​then multiplied by the combined weights to obtain a weighted standardized matrix. Traverse the weighted standardized matrix, extract the set of maximum values ​​of each indicator as the positive ideal solution, extract the set of minimum values ​​of each indicator as the negative ideal solution, and calculate the distances of each desulfurization technology scheme to the positive ideal solution and the distances to the negative ideal solution to obtain the comprehensive closeness. The desulfurization technology solutions are ranked from highest to lowest based on their overall similarity, and the shortcomings are identified by combining the scores of each individual indicator to obtain the shortcomings analysis solution.

7. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 6, characterized in that, Extracting the set of maximum values ​​for each indicator as the positive ideal solution, and extracting the set of minimum values ​​for each indicator as the negative ideal solution, includes: ； ； in, For the positive ideal solution, For a negative ideal solution, For the m-th desulfurization technology scheme, the weighted standardized matrix element corresponding to the j-th evaluation index.

8. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 6, characterized in that, The calculation of the distances from each desulfurization technology scheme to the positive ideal solution and the negative ideal solution includes: ； ； in, The distance from each desulfurization technology scheme to the ideal solution. To calculate the distance to the negative ideal solution for each desulfurization technology scheme, For the matrix elements within the weighted normalization matrix, For the i-th desulfurization technology, the weighted calculated value of the j-th evaluation index is given.

9. The multi-dimensional comprehensive evaluation method for steel desulfurization technology according to claim 6, characterized in that, Obtaining the overall closeness includes: ； in, For overall relevance.

10. A multi-dimensional comprehensive evaluation system for steel desulfurization technology implemented by the method according to any one of claims 1-9, characterized in that, include: The preprocessing module is used to acquire multi-dimensional raw monitoring information, preprocess the raw monitoring information, and obtain a standardized preprocessed monitoring dataset. The characterization system construction module is used to construct a multi-dimensional comprehensive characterization system using the standardized preprocessed monitoring dataset. The multi-dimensional evaluation module is used to input the multi-dimensional comprehensive characterization system into the multi-dimensional evaluation model to obtain an evaluation score. The multi-dimensional evaluation model is obtained by training a training set, which includes: historical monitoring data and evaluation sample labels under different desulfurization processes and different operating conditions. The subjective weights of the indicators and the objective weights of each secondary indicator in the multi-dimensional comprehensive characterization system are calculated using the multi-dimensional evaluation model. Based on the approximation ideal solution distance method, the closeness of each desulfurization technology scheme to the positive and negative ideal solutions is calculated to obtain the shortcoming analysis scheme.

Images