A converter energy consumption and carbon emission diagnosis and optimization method based on production rhythm
By combining long short-term memory networks and clustering algorithms with entropy weighting, the problem of accurately assessing energy consumption and carbon emissions in converter steelmaking was solved, achieving optimization of energy efficiency and carbon emissions in the converter process and providing intelligent diagnosis and optimization suggestions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHEASTERN UNIV CHINA
- Filing Date
- 2026-05-15
- Publication Date
- 2026-07-24
AI Technical Summary
Existing technologies struggle to accurately assess energy consumption and carbon emissions during converter steelmaking, making it difficult to identify energy efficiency bottlenecks and key carbon emission points. Furthermore, existing systems fail to deeply integrate production rhythm with the energy efficiency and carbon emissions of the smelting process for optimization.
The oxygen consumption, converter gas recovery, and steam recovery of the converter process are predicted by using a long short-term memory network. The baseline operating conditions are determined by combining a clustering algorithm and the contribution of influencing factors is constructed by using the entropy weight method, so as to realize intelligent diagnosis and optimization of converter operating conditions.
It enables precise diagnosis and optimization of energy consumption and carbon emissions in the converter process, improves energy efficiency and reduces carbon emissions, and provides scientific guidance for production optimization.
Smart Images
Figure CN122198269B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent diagnostic technology for iron and steel metallurgical processes, and in particular to a method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm. Background Technology
[0002] As a major carbon emitter, the steel industry urgently needs to transform into a green and low-carbon sector. Converter steelmaking, a core process in steel production, consumes enormous amounts of energy and generates significant carbon emissions, making it a key bottleneck restricting the green development of the steel industry.
[0003] Accurate assessment of energy efficiency and carbon emissions requires knowledge of converter operating conditions during the blowing process. However, the harsh environment of the converter—high temperature, dust, and strong interference—makes continuous and accurate direct measurement of these parameters extremely difficult and expensive. In practice, constant theoretical values, fixed empirical coefficients, or simple average distribution methods are often used for estimation, leading to significant deviations between the calculated results and actual operating conditions, making it difficult to accurately identify energy efficiency bottlenecks and key carbon emission points. Existing research mainly focuses on constructing energy efficiency and carbon emission evaluation methods and indicator systems, failing to identify energy efficiency bottlenecks or establish sophisticated causal models. Existing converter smelting rhythm analysis systems primarily predict process times by collecting signals from overhead cranes and tilting mechanisms to achieve just-in-time production. However, these methods mainly optimize the connection between logistics and time, failing to deeply couple and dynamically optimize the production rhythm with the inherent energy efficiency and carbon emission intensity of the smelting process.
[0004] Therefore, a method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm is needed. Summary of the Invention
[0005] In view of this, the present invention provides a method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm. By determining the minimum energy consumption and carbon emissions of the converter process, the baseline operating conditions are determined, and the current operating conditions are optimized by adjusting key influencing factors, thereby achieving simultaneous optimization of converter process energy consumption and converter process carbon emissions.
[0006] Therefore, the present invention provides the following technical solution:
[0007] A method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm, comprising: Using the converter process production rhythm parameters, operating parameters, and material structure parameters as inputs, the predicted values of oxygen consumption, converter gas recovery, and steam recovery are obtained through a long short-term memory network. The energy consumption and carbon emissions of the current converter are calculated based on the predicted values of oxygen consumption, converter gas recovery, and steam recovery. Baseline operating conditions are obtained based on production data using clustering algorithms; A deviation index is constructed between the current converter operating condition and the benchmark operating condition, and the converter operating condition is diagnosed based on the deviation index to determine whether the current operating condition is an abnormal condition. The key influencing factors of converter operating conditions are ranked according to their contribution; the contribution includes: The degree of influence of each key influencing factor on each evaluation indicator is multiplied by the entropy weight of the corresponding evaluation indicator, and the product results are summed to obtain the contribution of each key influencing factor. The converter operating conditions are optimized by combining the ranking of the key influencing factors and the diagnostic results.
[0008] The production rhythm parameters of the converter process include: blowing time, interval time, and smelting cycle. The operating parameters include: molten iron temperature and molten steel temperature; The material structure parameters include: iron-to-steel ratio and the amount of scrap steel added.
[0009] The production data includes: molten iron charge and temperature, oxygen blowing amount and blowing time, scrap steel addition amount, converter gas recovery amount, steam recovery amount, molten steel weight and temperature, slag weight, and unit molten steel energy consumption intensity and carbon emission intensity. The unit steel energy consumption intensity and carbon emission intensity refer to the energy consumption and carbon emission of a unit of steel, which are the ratio of total energy consumption or total carbon emission to the amount of steel.
[0010] The process of obtaining baseline operating conditions based on production data through clustering algorithms includes: Based on production data, a clustering algorithm was used to obtain three clusters with production rhythm as the horizontal axis and unit steel energy consumption as the vertical axis. These clusters are: a stable rhythm and low energy consumption cluster, a high energy consumption and slow rhythm cluster, and a high energy consumption and fast rhythm cluster. The baseline operating condition is the central operating condition of a cluster with stable rhythm and low energy consumption.
[0011] The diagnostic method includes using the deviation between the energy consumption and carbon emissions of the current converter operating condition and the corresponding indicators of the benchmark operating condition as diagnostic indicators for the converter operating condition. When the diagnostic indicators exceed the preset threshold, the current converter operating condition is determined to be an abnormal operating condition; otherwise, it is determined to be a normal operating condition.
[0012] The key influencing factors include: oxygen consumption, waste heat recovery, molten iron temperature, molten steel temperature, steel grade factors, and slag weight.
[0013] The converter operating conditions are optimized by combining the ranking of the key influencing factors and the diagnostic results, including: taking the values of each key influencing factor under the baseline operating conditions as target values; If the diagnostic result indicates an abnormal operating condition, a prompt will be triggered, suggesting that the key influencing factor with the greatest contribution be adjusted to the target value.
[0014] The energy consumption of the converter process includes:
[0015] In the formula, Energy consumption of the converter process during the statistical period; Oxygen consumption converted to standard coal equivalent Nitrogen consumption equivalent to standard coal consumption, Argon gas consumption equivalent to standard coal consumption, The amount of standard coal equivalent of coke oven gas consumption. Electricity consumption converted to standard coal equivalent, Water consumption converted to standard coal equivalent, Steam consumption converted to standard coal equivalent Converter gas recovery equivalent to standard coal content, This is the equivalent of standard coal equivalent for steam recovery. This refers to the output of molten steel.
[0016] Carbon emissions from the converter process include:
[0017] in, For carbon emissions from molten iron, For carbon emissions from scrap steel, For limestone carbon emissions, For iron ore carbon emissions, For carbon emissions from coke oven gas, For electricity carbon emissions, For nitrogen and carbon emissions, To consume steam carbon emissions, For oxygen and carbon emissions, For carbon output from molten steel, For slag carbon output, For converter gas carbon recovery, For steam carbon recovery; This refers to the output of molten steel.
[0018] Advantages and positive effects of the present invention: This method uses a long short-term memory network to obtain predicted values for oxygen consumption, converter gas recovery, and steam recovery based on the converter process's production rhythm parameters, operating parameters, and material structure parameters. Based on these predicted values, it calculates the current energy consumption and carbon emissions of the converter process. This enables intelligent diagnosis of converter operating conditions and, combined with a clustering algorithm, determines a baseline for converter operating conditions. Finally, it adjusts influencing factors based on the current and baseline operating conditions of the converter to optimize its operation.
[0019] This method constructs a predictive model for energy consumption and recovery in the converter process. First, it analyzes the correlation between converter production rhythm and oxygen consumption and waste heat recovery. Then, it uses a long short-term memory neural network algorithm to predict oxygen consumption, converter gas recovery, and steam recovery, ensuring that the relative error between the predicted and actual values is low and within a reasonable range.
[0020] This method uses a clustering algorithm to identify a benchmark operating condition as a reference, constructs the index deviation between the current operating condition and the benchmark operating condition, and realizes intelligent diagnosis of converter operating conditions. Furthermore, it constructs contribution degree through entropy weight method and ranks the influencing factors to determine the optimization priority, adjusts the influencing factors, and realizes the optimization and adjustment of converter operating conditions. Attached Figure Description
[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0022] Figure 1 This is a flowchart of the converter energy consumption and carbon emission diagnosis and optimization method based on production rhythm, as described in an embodiment of the present invention. Figure 2 This is a schematic diagram of the material and energy flow structure in the converter process of an iron and steel enterprise, as an example. Detailed Implementation
[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.
[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0025] This invention provides a method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm. Using converter process production rhythm parameters, operating parameters, and material structure parameters as inputs, a Long Short-Term Memory (LSTM) network is used to obtain predicted values for oxygen consumption, converter gas recovery, and steam recovery. Based on these predicted values, energy consumption and carbon emissions for the current converter operating condition are calculated. A baseline operating condition is obtained using a clustering algorithm based on production data. Key influencing factors of the converter operating condition are ranked according to their contribution. The converter operating condition is optimized based on the difference between the current and baseline operating conditions, combined with the ranked key influencing factors. This method achieves coupled optimization of converter process energy consumption and carbon emissions.
[0026] Example like Figure 1 As shown, a method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm includes the following steps: S1. Using the converter process production rhythm parameters, operating parameters, and material structure parameters as inputs, the predicted values of oxygen consumption, converter gas recovery, and steam recovery are obtained through a long short-term memory network. 1. Conduct on-site data collection based on the converter production structure to obtain converter data, including: Production rhythm parameters, operating parameters, and material structure parameters for the converter process; The production rhythm parameters for the converter process include: blowing time, interval time, and smelting cycle. Operating parameters include: molten iron temperature and molten steel temperature; Material structure parameters include: iron-to-steel ratio and the amount of scrap steel added.
[0027] 2. Preprocess the collected data; In the actual production process of converters, special situations may arise where sensors repeatedly transmit the same data to the database. Because this data is completely identical and repetitive, deduplication is performed based on the primary key. For each primary key value, only the first occurrence of the record is retained, and other duplicate records are deleted. Data loss is a common problem in data analysis and modeling. For some non-critical missing data, a deletion method based on the model's unified time granularity is used to save data processing time. For the missing critical data, average interpolation is used to fill in the missing data, thus fully restoring the true situation of the data. In actual production, the time granularity requirements for various data differ, resulting in collected data including multiple time granularity formats such as seconds, minutes, hours, and shifts. To unify standards and facilitate research, the converter production batch is used as the benchmark for time granularity.
[0028] 3. Based on the preprocessed data, the predicted values of oxygen consumption, converter gas recovery, and steam recovery are obtained through a long short-term memory network. The converter production rhythm parameters and raw material conditions, through energy and material transfer in the smelting process, ultimately affect oxygen consumption and waste heat recovery. Combining relevant calculation parameters from the converter energy consumption mechanism model, variables such as smelting time, blowing time, interval time, oxygen consumption, converter gas recovery, and steam recovery were selected. Grey relational analysis was used to quantitatively evaluate the correlation between each variable and the target predicted value. The results show that the selected variables have a strong correlation with the target predicted value and can be used as effective key characteristic parameters.
[0029] Construct a time series sample set and divide the sample set into a training set and a test set according to a preset ratio, with the training set accounting for 70% of the total sample set and the test set accounting for 30% of the total sample set.
[0030] A prediction model for converter waste heat recovery and oxygen consumption is constructed based on a long short-term memory (LSTM) neural network model. The LSTM neural network model includes at least an input layer, a hidden layer, and an output layer. Training data is input into the prediction model for training. By adjusting parameters such as the number of hidden layer nodes and the number of model iterations, the prediction error is made to meet industrial accuracy requirements, thus obtaining the trained prediction model.
[0031] The test set data is input into the trained prediction model to obtain the corresponding prediction results. By comparing the prediction results with the actual measurement data, the prediction error index is calculated to verify the accuracy and stability of the prediction model in converter energy prediction.
[0032] S2. Calculate the current converter operating condition based on the predicted oxygen consumption, the predicted converter gas recovery, and the predicted steam recovery. Energy consumption in the converter process includes:
[0033] In the formula, Energy consumption of the converter process during the statistical period; Oxygen consumption converted to standard coal equivalent Nitrogen consumption equivalent to standard coal consumption, Argon gas consumption equivalent to standard coal consumption, The amount of standard coal equivalent of coke oven gas consumption. Electricity consumption converted to standard coal equivalent, Water consumption converted to standard coal equivalent, Steam consumption converted to standard coal equivalent Converter gas recovery equivalent to standard coal content, This is the equivalent of standard coal equivalent for steam recovery. This refers to the output of molten steel.
[0034] Carbon emissions from the converter process include:
[0035] in, For carbon emissions from molten iron, For carbon emissions from scrap steel, For limestone carbon emissions, For iron ore carbon emissions, For carbon emissions from coke oven gas, For electricity carbon emissions, For nitrogen and carbon emissions, To consume steam carbon emissions, For oxygen and carbon emissions, For carbon output from molten steel, For slag carbon output, For converter gas carbon recovery, For steam carbon recovery; This refers to the output of molten steel.
[0036] S3. Obtain baseline operating conditions based on production data using clustering algorithms; Production data, including: The amount and temperature of molten iron charged, the amount and time of oxygen blowing, the amount of scrap steel added, the amount of converter gas recovered, the amount of steam recovered, the weight and temperature of molten steel, the weight of slag, and the energy consumption and carbon emission intensity per unit of molten steel.
[0037] Clustering algorithms are used to classify the converter's operating conditions. Similarity aggregation of operating conditions is achieved by minimizing the differences between samples within each cluster. The K-means clustering algorithm is employed, and the objective function is expressed as:
[0038] In the formula, For the number of operating condition categories, Indicates the first Sample set of similar working conditions These are the corresponding cluster centers. No. The feature vectors corresponding to the converter production data are obtained. By iteratively updating the cluster centers, several typical converter operating conditions are finally obtained.
[0039] Based on the clustering results, the actual operating conditions of current production are mapped to the corresponding operating condition categories, and compared with the benchmark operating conditions in terms of energy efficiency and carbon emissions, so as to quantitatively assess the degree of deviation between the actual operating conditions and the advanced operating level.
[0040] The contribution of operating condition evaluation is obtained by integrating the unit steel molten energy consumption evaluation index and the carbon emission intensity evaluation index using the entropy weight method.
[0041] 1) Using the baseline operating condition as a reference, compare it with the current operating condition and calculate the degree of influence of each influencing factor on the evaluation index:
[0042] In the formula, Number the influencing factors. Number the evaluation indicators (energy consumption, carbon emissions). For the first under the reference working condition Individual indicator values.
[0043] 2) Construct a factor-indicator change rate matrix:
[0044] Where m is the number of factors and n is the number of indicators.
[0045] To avoid direction cancellation, take the absolute value:
[0046] 3) Dimensionless processing of the indicators: Range standardization is performed on each column to obtain the standardized matrix Z:
[0047] 4) Calculate the weighting, transforming the values of each evaluation object under each indicator into a probability distribution form to prepare for the subsequent calculation of entropy. The formula is as follows:
[0048] 5) Calculate the entropy value of the evaluation index, expressed by the formula:
[0049] 6) Calculate the variation index of the evaluation index, expressed by the formula:
[0050] 7) Calculate the entropy weight of the evaluation index, expressed by the formula:
[0051] 8) Calculate the contribution of each indicator by weighted summation, expressed by the formula:
[0052] S4. Optimize the converter operating conditions based on the difference between the current converter operating conditions and the benchmark operating conditions, combined with the ranking of key influencing factors.
[0053] Key influencing factors include: oxygen consumption, waste heat recovery, molten iron temperature, molten steel temperature, steel grade factors, and slag weight.
[0054] The key influencing factors of converter operation are ranked according to their contribution.
[0055] The deviation between the energy consumption and carbon emissions of the current converter operating condition and the corresponding indicators of the benchmark operating condition is used as the converter operating condition diagnostic indicators. When the diagnostic indicators exceed the preset threshold, the current converter operating condition is determined to be an abnormal operating condition; otherwise, it is determined to be a normal operating condition.
[0056] The values of each key influencing factor under the baseline operating condition are used as the target values; if the diagnosis result is an abnormal operating condition, a prompt is triggered, suggesting that the key influencing factor with the greatest contribution be adjusted to the target value.
[0057] Specifically: If the difference between the energy consumption of the converter process in the current converter operating condition and the energy consumption of the converter process in the benchmark operating condition is greater than the preset threshold, a prompt will be triggered, and the key influencing factor with the greatest contribution to adjustment will be indicated. If the difference between the carbon emissions of the converter process in the current converter operating condition and the carbon emissions of the converter process in the benchmark operating condition is greater than a preset threshold, a prompt will be triggered, and the key influencing factor with the greatest contribution to adjustment will be indicated.
[0058] S5. Output diagnostic conclusions in the form of intelligent diagnostic reports.
[0059] The report includes: an assessment of the energy efficiency and carbon emission levels of the converter process; an analysis of the contribution of key influencing factors to energy efficiency and carbon emissions; and actionable operational recommendations based on metallurgical mechanisms and expert experience, such as production rhythm optimization, parameter adjustment strategies, and equipment maintenance improvement directions. This report provides data support and decision-making basis for energy conservation and carbon reduction in the converter process.
[0060] This method not only deepens the understanding of energy efficiency and carbon emissions in converter production, but also constructs a more scientific and comprehensive evaluation system for converter production processes by combining a comprehensive analytical framework model that integrates production rhythm and energy efficiency / carbon emissions. This evaluation system, through a hierarchical approach, meticulously examines the status of each indicator, making the energy efficiency evaluation and carbon emission accounting of equipment, production processes, and technologies in converter production more accurate and detailed. This multi-level, multi-dimensional evaluation method not only helps identify potential areas for energy efficiency improvement and carbon emission reduction, but also provides clear guidance for enterprises to optimize their energy efficiency and carbon emissions.
[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for diagnosing and optimizing converter energy consumption and carbon emissions based on production rhythm, characterized in that, include: Using the converter process production rhythm parameters, operating parameters, and material structure parameters as inputs, the predicted values of oxygen consumption, converter gas recovery, and steam recovery are obtained through a long short-term memory network. The energy consumption and carbon emissions of the current converter are calculated based on the predicted values of oxygen consumption, converter gas recovery, and steam recovery. Baseline operating conditions are obtained based on production data using clustering algorithms; A deviation index is constructed between the current converter operating condition and the benchmark operating condition, and the converter operating condition is diagnosed based on the deviation index to determine whether the current operating condition is an abnormal condition. The key influencing factors of converter operating conditions are ranked according to their contribution. The contribution includes: The degree of influence of each key influencing factor on each evaluation indicator is multiplied by the entropy weight of the corresponding evaluation indicator, and the product results are summed to obtain the contribution of each key influencing factor. The converter operating conditions are optimized by combining the ranking of the key influencing factors and the diagnostic results. The production rhythm parameters of the converter process include: blowing time, interval time, and smelting cycle. The operating parameters include: molten iron temperature and molten steel temperature; The material structure parameters include: iron-to-steel ratio and the amount of scrap steel added; The process of obtaining baseline operating conditions based on production data through clustering algorithms includes: Based on production data, a clustering algorithm was used to obtain three clusters with production rhythm as the x-axis and unit steel energy consumption as the y-axis, including: a stable rhythm and low energy consumption cluster, a high energy consumption and slow rhythm cluster, and a high energy consumption and fast rhythm cluster. The baseline operating condition is the central operating condition of a cluster with stable rhythm and low energy consumption.
2. The method according to claim 1, characterized in that, The production data includes: The amount and temperature of molten iron charged, the amount and time of oxygen blowing, the amount of scrap steel added, the amount of converter gas recovered, the amount of steam recovered, the weight and temperature of molten steel, the weight of slag, and the energy consumption intensity and carbon emission intensity per unit of molten steel. The unit steel energy consumption intensity and carbon emission intensity refer to the energy consumption and carbon emission of a unit of steel, which are the ratio of total energy consumption or total carbon emission to the amount of steel.
3. The method according to claim 1, characterized in that, The diagnostic method includes: The deviation between the energy consumption and carbon emissions of the current converter operating condition and the corresponding indicators of the benchmark operating condition is used as the diagnostic indicators for the converter operating condition. When the diagnostic indicators exceed the preset threshold, the current converter operating condition is determined to be an abnormal operating condition; otherwise, it is determined to be a normal operating condition.
4. The method according to claim 1, characterized in that, The key influencing factors include: Oxygen consumption, waste energy and waste heat recovery, molten iron temperature, molten steel temperature, steel grade factors, and slag weight.
5. The method according to claim 4, characterized in that, The converter operating conditions are optimized by combining the ranking of the key influencing factors and the diagnostic results, including: The values of each key influencing factor under the baseline operating condition are used as the target values; If the diagnostic result indicates an abnormal operating condition, a prompt will be triggered, suggesting that the key influencing factor with the greatest contribution be adjusted to the target value.
6. The method according to claim 1, characterized in that, The energy consumption of the converter operation includes: In the formula, Energy consumption of the converter process during the statistical period; Oxygen consumption converted to standard coal equivalent Nitrogen consumption equivalent to standard coal consumption, Argon gas consumption equivalent to standard coal consumption, The amount of standard coal equivalent of coke oven gas consumption. Electricity consumption converted to standard coal equivalent, Water consumption converted to standard coal equivalent, Steam consumption converted to standard coal equivalent Converter gas recovery equivalent to standard coal content, This is the equivalent of standard coal equivalent for steam recovery. This refers to the output of molten steel.
7. The method according to claim 1, characterized in that, The carbon emissions from the converter operation include: in, For carbon emissions from molten iron, For carbon emissions from scrap steel, For limestone carbon emissions, For iron ore carbon emissions, For carbon emissions from coke oven gas, For electricity carbon emissions, For nitrogen and carbon emissions, To consume steam carbon emissions, For oxygen and carbon emissions, For carbon output from molten steel, For slag carbon output, For converter gas carbon recovery, For steam carbon recovery; This refers to the output of molten steel.