Optimal energy consumption prediction and energy efficiency intelligent diagnosis method in steel production process

Abstract

The invention discloses an optimal energy consumption prediction and energy efficiency intelligent diagnosis method in a steel production process, and belongs to the technical field of steel production processes, and the method comprises the following steps: building a steel production database based on the steel production process by taking a heat or a batch as a unit, and carrying out the preprocessing of data in the steel production database; establishing a steel generation process mechanism model; building a black box model based on the correlation analysis and the neural network model, and training the black box model; based on coupling of the steel production process mechanism model and the trained black box model, establishing an energy consumption prediction model for performing energy consumption prediction on the steel production process energy consumption factors; energy consumption of energy consumption factors in the steel generation process is determined based on an energy consumption prediction model through a machine learning method, then optimal energy consumption in the steel generation process is determined through an energy consumption calculation formula in the steel production process, and historical optimal energy consumption is determined according to historical data.

Description

Technical Field

[0001] The invention belongs to the technical field of steel production process, and relates to an optimal energy consumption prediction and energy efficiency intelligent diagnosis method for steel production process. Background Art

[0002] In the current industrial production environment, especially in high-energy-consuming industries such as steel production, effective energy management and significant reduction of energy consumption have become core issues for improving corporate competitiveness and promoting sustainable development. Traditionally, energy consumption management relies on manual experience and intuitive judgment, which is not only inefficient but also difficult to accurately evaluate and optimize energy consumption levels. With the deepening of scientific research, the current mechanism analysis of typical industrial production processes, that is, the study of its processes based on physical and chemical methods such as thermodynamics and material balance, has been deeply studied. Based on the mechanism, a corresponding energy consumption model can be established. In addition, with the rapid development of information technology, especially the widespread application of industrial big data and artificial intelligence technology, new solutions are provided for energy consumption management. Industrial big data contains massive data generated in the production process. These data contain rich information and can be used to reveal the energy consumption laws in the production process, optimize production parameters, and thus achieve accurate management of energy consumption. Data-driven methods use algorithms such as machine learning and deep learning to efficiently process and analyze large amounts of data, discover the complex correlation between energy consumption and various production parameters, and achieve prediction and optimization of energy consumption and energy efficiency.

[0003] Mechanism analysis (e.g., based on physicochemical methods such as thermodynamics and material balance) and data-driven methods (e.g., using machine learning and deep learning algorithms to analyze production data) have been used in the analysis of industrial production energy consumption, but each has its own limitations. Although the mechanism analysis model can provide in-depth theoretical insights, its uncertainty increases significantly when faced with many "black box" links in the actual production process. On the contrary, although the data-driven method can efficiently process and analyze a large amount of data and predict energy consumption and energy efficiency trends, the instability and reliability of the data often affect the precision and accuracy of the model, making the prediction results deviate from reality. Therefore, how to couple mechanism analysis with data-driven methods and build a comprehensive model that can not only deeply understand the energy consumption mechanism but also efficiently use data is a key issue that needs to be urgently addressed in the current field of industrial production energy efficiency management. Summary of the invention

[0004] In order to solve the above problems, the technical solution adopted by the present invention is: a method for predicting the optimal energy consumption of a steel production process, comprising the following steps:

[0005] S1: Based on the steel production process in units of heats or batches, a steel production database is established, and data in the steel production database is preprocessed;

[0006] S2: Establish a mechanism model for the steel production process;

[0007] S3: Build a black-box model based on correlation analysis and neural network model, and train the black-box model;

[0008] S4: Based on the mutual complementarity and coupling of the steel production process mechanism model and the trained black-box model, establish an energy consumption prediction model for predicting the energy consumption factors in the steel production process;

[0009] S5: Determine the energy consumption of the energy consumption factors in the steel production process based on the energy consumption prediction model through machine learning methods, and then determine the optimal energy consumption in the steel production process through the energy consumption calculation formula of the steel production process and the historical best energy consumption based on historical data.

[0010] Furthermore: The establishment of the steel production database includes material consumption data, product data, composition parameters, physical property parameters, operation parameters, and technical parameter information. The construction requirements of the database correspond one by one to the product furnace charges or batches.

[0011] Furthermore: The process of building a black-box model based on correlation analysis and neural network model and training the black-box model is as follows:

[0012] S301: Determine the parameters of the neural network model and optimize the parameters of the neural network model;

[0013] S302: Build a black-box model based on correlation analysis and neural network model for analyzing the mechanism of the steel production process;

[0014] S303: Update and optimize the black-box model according to the on-site data of the steel production process, so as to better adapt to the changes in the process conditions of the steel production process and the uncertainties in industrial production, and obtain a trained black-box model.

[0015] Furthermore: The process of establishing an energy consumption prediction model for predicting the energy consumption factors in the steel production process based on the mutual complementarity and coupling of the steel production process mechanism model and the trained black-box model is as follows:

[0016] Identify the association between the steel production process mechanism model and the trained black-box model, determine the input-output relationship, and determine the coupling points between the steel production process mechanism model and the trained black-box model;

[0017] Based on the coupling points, establish a coupling mechanism between the steel production process mechanism model and the trained black-box model, and determine the coupling constraint conditions;

[0018] Combine the data between the mechanism model of the steel production process and the trained black-box model, and through a dynamic feedback mechanism, adjust the parameters in real time for energy consumption prediction to obtain an energy consumption prediction model.

[0019] Furthermore, the energy consumption calculation formula for the steel production process is expressed as follows:

[0020]

[0021] Where: Q all : Predicted value of energy consumption; Q use_i : Represents the amount of energy consumed, Q rec_i : Represents the amount of energy recovered, k: Represents the number of types of energy consumption, m: Represents the number of types of energy recovery; Q use_elec : Power consumption; Q use_cog Coke oven gas consumption; Q use_lng : Natural gas consumption; Q use_water : Water consumption; Q use_lqi : Low-pressure steam consumption; Q use_mqi : Medium-pressure steam consumption; Q use_n2 : Nitrogen consumption; Q use_jn2 High-pressure nitrogen consumption; Q use_ar : Argon consumption; Q use_o2 : Oxygen consumption; Q rec_lqi : Low-pressure steam recovery amount; Q rec_zha : Slag heat recovery.

[0022] Furthermore, the machine learning method adopts the particle swarm algorithm or the diploid genetic algorithm in the genetic algorithm.

[0023] Furthermore, the energy consumption of the energy consumption factors in the steel production process is determined based on the energy consumption prediction model through the machine learning method, and then the optimal energy consumption in the steel production process is determined through the energy consumption calculation formula of the steel production process, and the historical optimal energy consumption is determined according to the historical data as follows:

[0024] Optimize the energy consumption prediction model through the particle swarm algorithm or the diploid genetic algorithm in the genetic algorithm of machine learning. Determine the range of variation of different factors in the production process, and the optimal energy consumption in the production process can be determined through iterative calculation;

[0025] Optimize the historical production data to find the optimal energy consumption value that can meet the production conditions, and obtain the historical optimal energy consumption.

[0026] According to the energy efficiency intelligent diagnosis method of any one of the above-mentioned optimal energy consumption prediction methods for the steel production process, it includes the following steps:

[0027] Based on the characteristics of each process in the steel production process, the best energy consumption, and the energy consumption prediction model, an intelligent energy consumption diagnosis model is established to analyze the first-level, second-level, and third-level influencing factors in the steel production process;

[0028] Based on the intelligent energy consumption diagnosis model, analyze the energy consumption and energy efficiency in the actual steel production process, and determine the reasons for the energy consumption fluctuations in the actual production process.

[0029] Furthermore: The process of analyzing the energy consumption and energy efficiency in the actual steel production process based on the intelligent energy consumption diagnosis model to determine the reasons for the energy consumption fluctuations in the actual steel production process is as follows:

[0030] Determine the best working conditions in the steel production process. According to the determined best energy consumption, determine the values of the energy-consuming factor variables in the energy consumption prediction model as the reference values to obtain the reference working conditions under the best energy consumption state;

[0031] Conduct actual energy efficiency calculations. According to the collected data from actual production, substitute them into the intelligent energy consumption diagnosis model to calculate the actual energy efficiency value;

[0032] Compare the data of the best working conditions and the actual working conditions in the steel production process. Substitute the variables with differences into the intelligent energy consumption diagnosis model for calculation, and through sensitivity analysis and regression analysis of their impacts on the consumption and recovery of various energy media and the total energy consumption, realize the single-factor analysis of the energy consumption impact in the steel production process;

[0033] Through the results of the single-factor analysis, analyze the reasons for the energy consumption fluctuations in the actual steel production process.

[0034] A method for predicting the best energy consumption and intelligent diagnosis of energy efficiency in the steel production process provided by the present invention aims to optimize the energy consumption of typical industrial production processes by combining industrial big data, conduct intelligent analysis of energy efficiency to improve process energy efficiency, and promote digital transformation and energy conservation and emission reduction.

[0035] Based on the current industrial big data, the present invention establishes a production database by collecting relevant production data, establishes a mechanism model according to the current theoretical research, improves the deficiencies of the mechanism model through machine learning and deep learning methods, and establishes a mechanism and data-driven coupled energy consumption prediction model. On the basis of the energy consumption prediction model, determine the best energy consumption as the reference energy consumption through intelligent optimization algorithms or historical optimization methods, compare the actual production data with the best energy consumption, substitute them into the energy consumption prediction model for factor analysis, and finally obtain the intelligent diagnosis results of the three-level energy consumption and energy efficiency.

[0036] Compared with the prior art, the present invention proposes a method for predicting the optimal energy consumption and intelligent diagnosis of energy efficiency in the steel production process. Starting from industrial big data, it avoids the disadvantages of simply analyzing the energy consumption mechanism or establishing an energy consumption model by means of data-driven methods, transforms historical data into data wealth, and establishes an energy consumption prediction model that is more in line with the actual production. In addition, this application also proposes a method for determining the optimal energy consumption and an idea for intelligent diagnosis of energy efficiency. Through the method of the present invention, energy efficiency analysis can be well-founded, that is, supported by specific data, reliable theoretical analysis, and operation suggestions can also be given. The energy management unit and on-site operators can clearly determine the current production situation, which is convenient for adjustment to achieve the purpose of energy conservation, emission reduction, cost reduction and efficiency improvement. The present invention has a high level of intelligence and automation, has good operability for production enterprises, and can produce good effects in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0038] Figure 1 It is a flowchart of the present invention;

[0039] Figure 2 It is a flowchart of the data processing of the present invention;

[0040] Figure 3 It is a flowchart of neural network prediction;

[0041] Figure 4 It is a flowchart of the diploid genetic algorithm;

[0042] Figure 5 It is a usage method exemplified by the electric arc furnace steelmaking process. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments. To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way limits the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0044] Figure 1 is a schematic flowchart of the present invention;

[0045] The objective of the present invention is to establish an optimal energy consumption prediction method for the steel production process, including the following steps:

[0046] S1: Based on the steel production process in units of heats or batches, establish a steel production database and preprocess the data in the steel production database;

[0047] S2: Based on the data in the preprocessed steel production database, establish a mechanism model for the steel production process;

[0048] S3: Construct a black-box model based on correlation analysis and neural network models and train the black-box model;

[0049] S4: Couple the mechanism model of the steel production process with the trained black-box model to establish an energy consumption prediction model for predicting the energy consumption factors in the steel production process;

[0050] S5: Determine the energy consumption of the energy consumption factors in the steel production process based on the energy consumption prediction model through machine learning methods, and then determine the optimal energy consumption in the steel production process through the energy consumption calculation formula of the steel production process and determine the historical optimal energy consumption according to historical data.

[0051] The steps S1 / S2 / S3 / S4 / S5 are executed in sequence;

[0052] Furthermore: Arrange sampling points for the steel production process and establish a production database. Select a suitable production line for the target production process, and arrange sensors around the data information such as material consumption data, product data, composition parameters, physical property parameters, operation parameters, and technical parameters in the production process to collect data such as flow rate, temperature, and pressure required by the model, and store the data in the database. The construction requirements of the database correspond one by one to the product heats or batches;

[0053] Taking steel production as an example, it is necessary to establish a production database based on heats or batches, and map various parameters one by one.

[0054] Table 1 shows the material parameters of Furnace No. 1 in the steel production process

[0055]

[0056] Table 2 shows the production parameters of the 1st heat in the steel production process

[0057]

[0058]

[0059] Table 3 shows the operation parameters of the 1st heat in the steel production process

[0060]

[0061] Furthermore: Preprocess the data in the steel production database to obtain effective analysis data for the steel production process. Since there are noises and errors in the collection of the steel production process, missing values and outliers will be generated. It is necessary to process the missing values and outliers in the collected original data. For missing values, the median filling method is used to complete the missing values, and for outliers, the box plot method is used for processing. The process is as Figure 2 shown, and the specific process is as follows:

[0062] Read the specific historical data of the steel production process;

[0063] Judge whether the historical data of the steel production process is complete. When it is judged that the historical data of the steel production process is complete, collect the remaining complete data;

[0064] When it is judged that the historical data of the steel production process is incomplete, judge whether the missing value can be supplemented. If the missing value cannot be supplemented, delete the data. If the missing value can be supplemented, supplement it based on the median method;

[0065] Based on the complete data, judge whether there are outliers by setting a threshold. When there are abnormal data, delete the abnormal data;

[0066] When there are no abnormal data, integrate all the data to obtain the preprocessed steel production data.

[0067] Step 2: Establishment of the mechanism model for the steel production process. The establishment of the mechanism model for the steel production process is a key step in understanding and optimizing the energy consumption and energy efficiency of typical industrial production processes and their sub-processes. Relevant experts at home and abroad have conducted in-depth and extensive research on its mechanism. These studies have not only deepened the understanding of the essence of industrial production processes but also provided a solid theoretical foundation for constructing reliable mechanism models. Taking steel production as an example, its sub-processes, such as the coking process, sintering process, ironmaking process, steelmaking process, etc., usually involve complex physical, chemical, and thermodynamic processes. According to the latest relevant research, a mechanism model can be established.

[0068] The expression of the mechanism model for the steel production process is as follows:

[0069] aA + bB → cC + dD ΔH

[0070] ΔU = Q - W

[0071] ∑Q in = ∑Q out

[0072] Q = mcΔT

[0073] M material,element = M product,element

[0074] Where: A, B, C, D: Reactants and products of chemical reactions; a, b, c, d: Coefficients; ΔH: Enthalpy change; U: Internal energy; Q: Thermal energy; W: Work; Q in : Input energy; Q out : Output energy; m: Mass; c: Specific heat; T: Temperature; M material,element : Mass of raw material elements; M product,element : Mass of product elements.

[0075] Regarding the uncertain parameters involved in the mechanism model and the adaptability of the black-box model, etc., it is necessary to improve the mechanism model through a data-driven model. The black-box model refers to the process where the mechanism of the steel production process is complex and unclear and has not been fully understood yet. The mechanism model is a model that can explain various processes;

[0076] Taking the electric arc furnace steelmaking in the steel production process as an example, most of the reaction processes involved in this process can be found in the latest literature to establish material balance and heat balance analysis models. However, when combined with the production process, some parameters in the black-box model are uncertain due to environmental influences and need to be further adjusted in combination with production reality.

[0077] Figure 3 It is a schematic diagram of the neural network prediction process;

[0078] Mechanistic models often face challenges such as simplifying assumptions, parameter uncertainties, "black box" problems, and adaptability issues when reflecting the actual situation of industrial production. Neural network algorithms can effectively solve these problems through parameter estimation and optimization, constructing high-precision black box prediction models, anomaly detection and fault diagnosis, as well as adaptive learning and optimization. This enables the integration of mechanistic and data-driven models, significantly improving the accuracy and practicality of mechanistic models and providing strong support for optimizing industrial production processes and enhancing energy efficiency.

[0079] Taking the example of electric arc furnace steelmaking in the steel production process, key parameters and factors such as the prediction of element burn-off rate, carbon addition, tapping temperature, and prediction of molten steel carbon content (input parameters of the energy consumption prediction model) can be better determined through neural network algorithms.

[0080] The process of constructing a black box model based on correlation analysis and neural network model and training the black box model is as follows:

[0081] S301: Determine the parameters of the neural network model and optimize the parameters of the neural network model;

[0082] S302: Construct a black box model based on correlation analysis and neural network model for analyzing the mechanism of the steel production process;

[0083] S303: Update and optimize the black box model according to the on-site data of the steel production process, so as to better adapt to the changes in process conditions in the steel production process and the uncertainties in industrial production, and obtain a trained black box model.

[0084] Furthermore, the process of determining the parameters of the neural network model and optimizing the parameters of the neural network model is as follows:

[0085] Neural network algorithms can learn the relationship between parameters and output variables from a large amount of data in the steel production process, thus better determining the parameters in the mechanistic model and optimizing them according to real-time data. The parameter determination and optimization process involves regression algorithms and optimization algorithms, and the process is as follows:

[0086] (1) Data preparation: Determine the independent and dependent variables of the neural network algorithm, and collect and preprocess the data in the steel production process;

[0087] (2) Model selection: Select a suitable regression model such as linear regression, polynomial regression, logarithmic regression model, etc. according to the data situation in the steel production process.

[0088] (3) Model training: Use the training data set to train the regression model to find a suitable regression model.

[0089] (4) Model evaluation: Use the test data set to evaluate the reliability of the regression model and calculate its error.

[0090] (5) Model optimization: Adjust the model parameters or select a more appropriate regression model according to the results.

[0091] Black-box model construction. For the "black box" in the mechanism model, machine learning algorithms can build a high-precision prediction model based on data, thereby making up for the shortcoming that the mechanism model cannot perform modeling analysis on it and improving the integrity of the model. Black-box model construction requires the help of correlation analysis and neural network models.

[0092] Step 4021: Correlation analysis to explore the influencing factors of the model.

[0093] (1) Determine the correlation analysis factors and determine the relevant data. According to the literature and production conditions, determine the range of influencing factors and comparison values. Define the comparison value sequence as the mother sequence, define the influencing factor sequence as the child sequence, and establish the mother sequence matrix as Y = [y 1 , y 2 , y 3 , …, y m T , and establish the child sequence matrix as:

[0094]

[0095] where: x: influencing factor; y: comparison value; n, m: number of elements,

[0096] (2) Preprocess the data in the iron and steel production database. Different elements have different dimensions and data ranges, so they need to be normalized to remove the dimension and unified into an approximate range, focusing on their changes and trends.

[0097]

[0098] where: mean value; normalization result.

[0099] (3) Calculate the grey correlation coefficient. Calculate the correlation coefficient between each index in the child sequence and the mother sequence. Denote:

[0100] a = min i min k |x 0 (k) - x i (k)|

[0101] b = max i max k |x 0 (k) - x​i (k)|

[0102] Structure:

[0103]

[0104] where a, b: calculation reference coefficients; ρ: resolution coefficient; z k.j , ξ j (k): correlation coefficient.

[0105] (4) Calculate the correlation degree.

[0106]

[0107] where r j is the correlation degree.

[0108] Step 4022: Neural network prediction. Neural network prediction is a relatively popular prediction method in current industrial applications. Based on the grey correlation analysis in Step 4021, after determining the relevant influencing factors, combined with the algorithm of neural network prediction, a mathematical model (product generation amount, temperature prediction, composition prediction, etc. models) that conforms to the actual production situation can be obtained better. Neural network prediction includes six steps, namely data normalization, data partitioning, network setting, data training, data verification, and testing generalization ability. Its process is as Figure 3 shown.

[0109] Adaptive optimization. The establishment of the mechanism model needs to conform to the actual production. There are differences between the mechanism analysis in the literature and the actual production, and the model in the literature needs to be optimized to improve its applicability. Machine learning algorithms have good adaptive capabilities and can update and optimize the mechanism model according to the on-site steel production data, so as to better adapt to the changes in process conditions and the uncertainties in industrial production.

[0110] Abnormal diagnosis and fault analysis. For abnormal situations and fault problems in the industrial process, machine learning can perform data analysis well and discover problems in time. The clustering algorithm in machine learning algorithms has good usability.

[0111] Step 5: Based on the mutual complementation and coupling of the mechanism model of the steel production process and the trained black-box model, establish an energy consumption prediction model for predicting the energy consumption factors in the steel production process.

[0112] For the steel production process, through literature reading, summarize the current research results and construct a mechanism model.

[0113] Based on the energy consumption calculation formula, according to the needs of the production site, clarify the data source of the consumption amount of the medium, that is, data collection or the predicted value of the mechanism model or the predicted value of the black-box model.

[0114] Build a black-box model. For the uncertain content in the mechanism model and the content required for energy consumption prediction but lacking a mechanism model, adopt the neural network prediction method to establish a black-box model that better conforms to the actual production.

[0115] Identify the association between the mechanism model in the iron and steel production process and the trained black-box model, determine the input-output relationship, and determine the coupling points between the mechanism model in the iron and steel production process and the trained black-box model;

[0116] The input of the energy consumption prediction model comes from on-site collected data, namely the production database;

[0117] The output of the energy consumption prediction model is the energy consumption prediction value;

[0118] Coupling points: (1) The input data required by the mechanism model, which is difficult to collect on-site and lacks actual data, but can be predicted by the black-box model; (2) The environmental parameters and reaction parameters required by the mechanism model, which are difficult to collect in practice but can be predicted by the black-box model; (3) The mechanism model is complex and requires many parameters, but these parameters are difficult to collect or implement, but can be simplified by the black-box model, etc.;

[0119] Based on the coupling points, establish a coupling mechanism between the mechanism model in the iron and steel production process and the trained black-box model, and determine the coupling constraint conditions;

[0120] Coupling constraint conditions: including on-site production data, upper and lower limits of the physical and chemical process, operating conditions of equipment, etc.;

[0121] Establish data combination between the mechanism model in the iron production process and the trained black-box model, and through the dynamic feedback mechanism, adjust the parameters in the coupling model in a timely manner; conduct energy consumption prediction to obtain the energy consumption prediction model.

[0122] Through machine learning modeling, production enterprises can better establish relevant production models that are more consistent with actual production. During actual production, they can couple with existing models to establish an energy consumption prediction model, and can predict the energy consumption under known production conditions. Taking the electric arc furnace steelmaking in the iron and steel production process as an example, the calculation method of the energy consumption prediction model is as follows. Most of the consumption and recovery of different energy media come from the prediction of the model. The energy consumption model is a prediction model rather than simple data collection. Having a complex and accurate energy consumption prediction model provides an analysis basis for the subsequent determination of the best energy consumption and the diagnosis of energy efficiency.

[0123] The energy consumption calculation formula for the iron and steel production process is expressed as follows:

[0124]

[0125] Among them: Among them: Q all : Energy consumption prediction value; Q use_i : Represents the amount of energy consumed, Q rec_i : Represents the amount of energy recovered, k: Represents the number of types of energy consumption, m: Represents the number of types of energy recovery; Q use_elec : Power consumption; Q use_cog Coke oven gas consumption; Q use_lng : Natural gas consumption; Q use_water : Water consumption; Q use_lqi : Low-pressure steam consumption; Q use_mqi : Medium-pressure steam consumption; Q use_n2 : Nitrogen consumption; Q use_hn2 High-pressure nitrogen consumption; Q use_ar : Argon consumption; Q use_o2 : Oxygen consumption; Q rec_lqi : Low-pressure steam recovery amount; Q rec_zha : Slag heat recovery;

[0126] Represents the energy consumption amount from the j-th type of consumed energy to the k-th type of consumed energy; Represents the recovered energy amount from the n-th type of recovered energy to the m-th type of recovered energy.

[0127] Further, the optimal energy consumption is determined. Through the energy consumption prediction model established by coupling mechanism and data-driven, various factors affecting energy consumption in the production process are reflected in the relationship between variables of the energy consumption prediction model. The influence amount and influence rate of different factors on energy consumption can be quantitatively analyzed. Through machine learning particle swarm optimization algorithm, genetic algorithm such as diploid genetic algorithm, etc., the complex energy consumption model can be optimized. By determining the change range of different factors in the production process, the optimal energy consumption in the production process can be determined through computer iterative calculation.

[0128] The optimal energy consumption determined by this method is the model optimal energy consumption. The corresponding production situation may not be achievable in reality. For example, the production operation requirements corresponding to the optimal energy consumption are difficult to meet. In order to combine with the actual production, the historical optimal energy consumption can also be obtained from the actual production. By optimizing the historical production data, the optimal energy consumption value that can meet the production conditions can be found and used as the benchmark for analysis. Both the model optimal energy consumption and the historical optimal energy consumption are the benchmarks for analysis, and both meet the relatively objective requirements and can be selected by production personnel.

[0129] The diploid genetic algorithm is a model optimization algorithm that can be used for reference. The flow chart of the diploid genetic algorithm is as Figure 4 shown. Its main process includes analyzing the problem, determining the chromosome, determining the population, calculating the fitness value, crossover, mutation, dominant and recessive rearrangement, etc. Some key calculation formulas are as follows:

[0130] If the objective function is a maximization problem, the fitness function can be taken as:

[0131] Fit(f(x)) = f(x)

[0132] If the objective function is a minimization problem, the fitness function can be taken as:

[0133]

[0134] For size transformation of the fitness function, the transformation method of the new fitness function f' is:

[0135] f' = af + b

[0136] Where:

[0137] The Monte Carlo method is the most commonly used selection method in genetic algorithms, and the probability of an individual being selected is:

[0138]

[0139] The most commonly used sorting model is the linear sorting model, which was proposed by Baker:

[0140]

[0141] According to any one of the energy efficiency intelligent diagnosis methods for the best energy consumption prediction method in the steel production process described above, it includes the following steps:

[0142] Based on the characteristics of each process in the steel production process, the best energy consumption, and the energy consumption prediction model, establish an energy consumption intelligent diagnosis model for analyzing the first-level, second-level, and third-level influencing factors of the steel production process;

[0143] Based on the energy consumption intelligent diagnosis model, realize the analysis of the energy consumption and energy efficiency in the actual steel production process, and determine the reasons for the energy consumption fluctuations in the actual production process.

[0144] Furthermore, the energy efficiency diagnosis analysis process is a process of analyzing the energy consumption and energy efficiency in the current production process by comparing with the best energy consumption. The energy efficiency intelligent diagnosis should clarify the reasons for the changes in the current energy consumption and energy efficiency compared with the best energy consumption, give corresponding data results support, and give clear reason analysis and operation suggestions to the energy management and control unit and on-site operators. The energy efficiency intelligent diagnosis is based on the energy consumption prediction model, and its steps are as follows:

[0145] Step 701: Determine the best working conditions of the steel production process. According to the determined best energy consumption, determine the values of each variable in the energy consumption prediction model as the reference values to obtain the reference working conditions under the best energy consumption state.

[0146] Step 702: Actual energy efficiency calculation. According to the collected data in actual production, substitute them into the model to calculate the actual value of energy efficiency.

[0147] Step 703: Single-factor analysis.

[0148] Compare the data of the optimal working condition and the actual working condition, substitute the variables with differences into the energy consumption intelligent diagnosis model for calculation, analyze their impacts on the consumption and recovery of various energy media and the total energy consumption. Through single-factor analysis, the impact of single factors on energy efficiency can be refined.

[0149] Taking the electric arc furnace steelmaking in the steel production process as an example, single-factor analysis includes the analysis of factors such as the molten iron tapping temperature, scrap preheating temperature, molten iron composition, gas consumption, scrap consumption, molten iron consumption, oxygen blowing amount, and steam recovery analysis.

[0150] Three-level influencing factor analysis. Based on the results of single-factor analysis, a first-level, second-level, and third-level influencing factor analysis model can be established. Among them, the first-level, second-level, and third-level factors are determined according to the mutual influence among factors. The factors that directly affect energy consumption are first-level factors, and the second-level and third-level factors are indirect factors for the previous-level factors. Taking the electric arc furnace steelmaking in the steel production process as an example, the first-level factors include raw material factors, gas factors, tapping factors, other power medium factors, etc.; the second-level factors include molten iron composition, scrap consumption, molten iron ratio, gas consumption, power medium consumption, etc., and the third-level factors include molten iron carbon content, silicon content, etc.

[0151] Figure 5 This is the usage process example of the electric arc furnace steelmaking process;

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting optimal energy consumption in a steel production process, characterized in that: The following steps are involved: Establish a steel production database based on the steel production process in units of furnaces or batches; Establish a mechanistic model of the steel-making process; Construct a black box model based on correlation analysis and neural network model, and train the black box model; Based on the mutual complementation and coupling of the steel production process mechanism model and the trained black box model, an energy consumption prediction model for predicting energy consumption factors in the steel production process is established; The energy consumption of energy-consuming factors in the steel production process is determined based on the energy consumption prediction model through machine learning methods, and then the optimal energy consumption of the steel production process is determined through the energy consumption calculation formula of the steel production process, and the historical optimal energy consumption is determined based on historical data.

2. The method for predicting optimal energy consumption in a steel production process according to claim 1, characterized in that: The steel production database established includes material consumption data, product data, composition parameters, physical property parameters, operation parameters and technical parameter information. The construction of the database requires a one-to-one correspondence with product heats or batches.

3. The method for predicting optimal energy consumption in a steel production process according to claim 1, characterized in that: The process of building a black box model based on correlation analysis and a neural network model and training the black box model is as follows: S301: Determine and optimize the neural network model parameters; S302: constructing a black box model based on correlation analysis and a neural network model to analyze the mechanism of the steel production process; S303: Update and optimize the black box model according to the field data of the steel production process, so as to adapt to the changes in process conditions of the steel production process and the uncertainties in industrial production, and obtain a trained black box model.

4. The method for predicting optimal energy consumption in a steel production process according to claim 1, characterized in that: The process of establishing an energy consumption prediction model for predicting energy consumption of energy consumption factors in the steel production process based on the mutual complementation and coupling of the steel production process mechanism model and the trained black box model is as follows: Identify the association between the steel production process mechanism model and the trained black box model, determine the input-output relationship, and determine the coupling point between the steel production process mechanism model and the trained black box model; Based on the coupling points, the coupling mechanism between the steel production process mechanism model and the trained black box model is established, and the coupling constraints are determined; The data between the steel production process mechanism model and the trained black box model are combined, and the parameters are adjusted in real time through a dynamic feedback mechanism to predict energy consumption and obtain an energy consumption prediction model.

5. The method for predicting optimal energy consumption in a steel production process according to claim 1, characterized in that: The energy consumption calculation formula of the steel production process is as follows: Where: Q all : predicted energy consumption; Q use_i : Indicates the amount of energy consumed, Q rec_i : represents the amount of energy recovered, k: represents the number of types of energy consumption, m: represents the number of types of energy recovery; Q use_elec : Power consumption; Q use_cog Coke oven gas consumption; Q use_lng : Natural gas consumption; Q use_water : Water consumption; Q use_lqi : Low-pressure steam consumption; Q use_mqi : Medium pressure steam consumption; Q use_n2 : Nitrogen consumption; Q use_hn2 High pressure nitrogen consumption; Q use_ar : Argon consumption; Q use_o2 : oxygen consumption; Q rec_lqi : Low-pressure steam recovery; Q rec_zha : Slag heat recovery.

6. The method for predicting optimal energy consumption in a steel production process according to claim 1, characterized in that: The machine learning method adopts a particle swarm algorithm or a diploid genetic algorithm in a genetic algorithm.

7. The method for predicting optimal energy consumption in a steel production process according to claim 1, characterized in that: The process of determining the energy consumption of the energy consumption factors of the steel production process based on the energy consumption prediction model by the machine learning method, and then determining the optimal energy consumption of the steel production process by the energy consumption calculation formula of the steel production process and determining the historical optimal energy consumption based on historical data is as follows: The energy consumption prediction model is optimized through the particle swarm algorithm of machine learning or the diploid genetic algorithm in the genetic algorithm. The range of variation of different factors in the production process is determined, and the optimal energy consumption in the production process is determined through iterative calculation; By optimizing historical production data, we can find the best energy consumption value that can meet production conditions and obtain the historical best energy consumption.

8. An energy efficiency intelligent diagnosis method for optimal energy consumption prediction in a steel production process according to any one of claims 1 to 7, characterized in that: The following steps are involved: Based on the characteristics of each process in the steel production process, the optimal energy consumption and the energy consumption prediction model, an energy consumption intelligent diagnosis model is established to analyze the primary, secondary and tertiary influencing factors of the steel production process; Based on the intelligent energy consumption diagnosis model, the energy consumption and energy efficiency in the actual steel production process can be analyzed, and the reasons for the energy consumption fluctuations in the actual production process can be determined.

9. The energy efficiency intelligent diagnosis method for optimal energy consumption in a steel production process according to claim 8, characterized in that: The process of analyzing the energy consumption and energy efficiency in the actual steel production process based on the energy consumption intelligent diagnosis model and determining the cause of the energy consumption fluctuation in the actual steel production process is as follows: Determine the optimal working condition of the steel production process, and according to the determined optimal energy consumption, determine the value of the energy consumption factor variable in the energy consumption prediction model as the benchmark value to obtain the benchmark working condition under the optimal energy consumption state; Carry out actual energy efficiency calculation, and bring the actual energy efficiency value into the energy consumption intelligent diagnosis model based on the actual production data collected; Compare the data of the optimal working conditions and actual working conditions of the steel production process, bring the different variables into the energy consumption intelligent diagnosis model for calculation, and use sensitivity analysis and regression analysis to analyze their impact on the recovery of various energy media consumption and the impact on total energy consumption, so as to achieve single factor analysis of the impact of energy consumption in the steel production process; Through the single factor analysis results, the causes of energy consumption fluctuations in the actual steel production process can be analyzed.

Images