Intelligent smelting process parameter optimization management method and system

Abstract

The present invention discloses an intelligent smelting process parameter optimization management method and system, which relates to the technical field of smelting process data processing, including determining the initial time lag of process parameters according to process parameter input and smelting raw material output, and considering the time lag characteristics of process parameters when adjusting. The smelting state of smelting equipment is defined, and the working state of smelting equipment is defined to provide a reliable basis for the subsequent optimization of process parameters. The nonlinear relationship between process parameter input and smelting raw material output and the coupling relationship between process parameters are determined, and the nonlinear relationship is analyzed by considering process parameter input, smelting state, and smelting raw material output, and described by a polynomial regression model. The initial time lag is adjusted in combination with the nonlinear relationship and the coupling relationship. The process parameters are adjusted stably, quickly and reasonably, the accuracy and adaptability of the process parameter optimization adjustment are improved, and the efficient and safe implementation of the smelting process is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of smelting process data processing, and in particular to an intelligent smelting process parameter optimization management method and system. Background Art

[0002] With the rapid development of the economy and the intensification of market competition, the metal smelting industry is facing pressures in resources, environment, energy consumption, etc. while pursuing efficient and low-cost production. The application of intelligent technology has become an important direction for the transformation and upgrading of the industry. By introducing advanced sensing technology, data analysis and processing technology, artificial intelligence technology, etc., real-time monitoring, precise control and optimized management of the smelting process are realized, which has become the key to improving production efficiency, reducing energy consumption, reducing environmental pollution, and improving product quality. The intelligent smelting process parameter optimization management solution aims to accurately control and optimize the key parameters in the smelting process through intelligent means to promote the green development and sustainable development of the metal smelting industry.

[0003] In the prior art, the smelting process is implemented only according to preset fixed smelting process parameters, without considering complex relationships such as the relationship between the process parameters and the smelting raw material output, the coupling relationship between the process parameters and the state of the smelting equipment. This results in poor accuracy and low adaptability in the optimization and adjustment of the process parameters, and cannot guarantee the efficiency and safety of the smelting process.

[0004] Therefore, how to improve the accuracy and adaptability of process parameter optimization and adjustment is a technical problem that needs to be solved. Summary of the invention

[0005] The purpose of the present invention is to solve the problems of poor accuracy and low adaptability of process parameter optimization adjustment in the prior art, and to propose an intelligent smelting process parameter optimization management method, which includes:

[0006] Obtain the smelting process flow, determine the process parameter input, smelting equipment and smelting raw material output of each link in the smelting process flow;

[0007] Collect historical data of process parameter input, smelting equipment related data and smelting raw material output, determine the preliminary time lag of process parameters based on process parameter input and smelting raw material output, and define the smelting state of smelting equipment;

[0008] Determine the nonlinear relationship between process parameter input and smelting raw material output and the coupling relationship between process parameters, adjust the initial time lag of process parameters through the nonlinear relationship and coupling relationship, and obtain the time lag of process parameters;

[0009] The process parameters of each link in the smelting process are optimized based on the smelting state, nonlinear relationship, coupling relationship and time lag of process parameters of the smelting equipment.

[0010] In some embodiments of the present application, the preliminary time lag of the process parameters is determined according to the process parameter input and the smelting raw material output, including:

[0011] According to the process parameter input and smelting raw material output, the data scatter diagrams of the two changing over time are constructed respectively;

[0012] Conduct spectrum analysis and wavelet analysis on the process parameter input and smelting raw material output of each link in the smelting process to obtain multiple fluctuation parameters, and determine the fluctuation index by combining multiple fluctuation parameters;

[0013] Each link in the smelting process is divided into multiple fluctuation stages according to the fluctuation index range. Each fluctuation stage corresponds to a fluctuation range. The smoothing weights of process parameter input and smelting raw material output are determined according to the median of the fluctuation range.

[0014] Performing weighted average smoothing processing on the data scatter diagrams of the process parameter input and the smelting raw material output according to the smoothing weights, and obtaining the curve diagrams of the process parameter input and the smelting raw material output;

[0015] Preliminary time lags of process parameters are determined based on respective curve graphs of process parameter input and smelting raw material output.

[0016] In some embodiments of the present application, the preliminary time lag of the process parameters is determined based on the respective curves of the process parameter input and the smelting raw material output, including:

[0017] Calculate the mutual information between the process parameter input and the smelting raw material output, and draw a mutual information change curve of the mutual information between the process parameter input and the smelting raw material output over time according to their respective curve graphs;

[0018] Identify the obvious change points on the curve graph of the process parameter input, determine the time point when the change is obvious, and mark the time point when the change is obvious on the mutual information change curve;

[0019] Determine the maximum value of the mutual information within a preset time range after the time point at which the mutual information change curve changes significantly, and record the time point at which the mutual information has the maximum value as the end time point;

[0020] Determine a mutual information change rate obvious point within a preset time range after a time point at which the change on the mutual information change curve is obvious, and record the time point at which the mutual information change rate is obvious as the starting time point. The mutual information change rate obvious point is the earliest point at which the mutual information change rate exceeds the preset change rate;

[0021] The preliminary time lag of each process parameter was determined based on the start time point and the end time point.

[0022] In some embodiments of the present application, the smelting state of the smelting equipment is defined, including:

[0023] The data related to the smelting equipment include operating status parameters and production process parameters. Each of the operating status parameters and the production process parameters is evaluated, and the evaluation index of each parameter is determined. The smelting status level of the smelting equipment is evaluated according to the evaluation index of the operating status parameters and the production process parameters, and the smelting status is described by the smelting status level.

[0024] ;

[0025] in, For the The smelting state level of each smelting equipment, For the The number of operating status parameters of smelting equipment, For the The combined weight of the operating state parameters, For the The first smelting equipment The evaluation index of the operating status parameters is For the The number of production process parameters of each smelting equipment, For the The combined weights of the production process parameters, For the The first smelting equipment The evaluation index of production process parameters, , Respectively The first constant and the second constant of a smelting equipment, [] is the rounding symbol.

[0026] In some embodiments of the present application, determining the nonlinear relationship between process parameter input and smelting raw material output and the coupling relationship between process parameters includes:

[0027] The nonlinear relationship between process parameter input and smelting raw material output is described by a polynomial regression model, and the model parameters of the polynomial regression model include polynomial order, intercept term, error term and polynomial coefficient;

[0028] Preset the intercept term, determine the initial error term according to the smelting state level, determine the polynomial order according to the discrete degree of the process parameters, determine the polynomial coefficients and adjust the initial error term through the least square method or cross-validation based on the intercept term, the initial error term and the polynomial order, so as to determine the polynomial regression model;

[0029] ;

[0030] in, For the Polynomial regression of process parameters and smelting raw material output, is the intercept term, is the polynomial order, For the The coefficient of the second term, For the The process parameters of The size of the secondary order, is the adjusted error term;

[0031] The polynomial regression model integrating all process parameters describes the nonlinear relationship between process parameter input and smelting raw material output.

[0032] In some embodiments of the present application, determining the nonlinear relationship between the process parameter input and the smelting raw material output and the coupling relationship between the process parameters also includes:

[0033] The correlation coefficients between the process parameter inputs are calculated, and training sets and test sets are constructed. The coupled neural network model is trained based on the correlation coefficients, training sets and test sets. The coupled relationship between the process parameters is described by the coupled neural network model.

[0034] In some embodiments of the present application, the initial time hysteresis of the process parameters is adjusted by the nonlinear relationship and the coupling relationship to obtain the time hysteresis of the process parameters, including:

[0035] The process parameters are screened through coupling relationships. On the basis of preliminary time lags, the smelting process is simulated according to the nonlinear relationship and coupling relationship of the screened process parameters, and the time lag of the process parameters is adjusted.

[0036] In some embodiments of the present application, the process parameters of each link in the smelting process are optimized based on the smelting state, nonlinear relationship, coupling relationship and time lag of the process parameters of the smelting equipment, including:

[0037] Determine the target range and actual value of the process parameters in each link of the smelting process flow, record the process parameters whose actual values ​​are outside the target range as the process parameters to be adjusted, and quantify the strength of the nonlinear relationship and coupling relationship of the process parameters to be adjusted;

[0038] The adjustment step is determined based on the smelting state of the smelting equipment, the nonlinear relationship of the process parameters to be adjusted, and the strength of the coupling relationship. The actual value of the process parameters to be adjusted is brought closer to the target range by combining the adjustment step and the time lag of the process parameters to be adjusted.

[0039] ;

[0040] in, For the The adjustment step length of the process parameters to be adjusted, is the smelting state of the smelting equipment, Adjust the step length based on It represents the basic adjustment step obtained by mapping the smelting state of the smelting equipment. , are the contribution weights of nonlinear relationship and coupling relationship respectively, , Respectively The strength of the nonlinear relationship and coupling relationship of the process parameters to be adjusted, For the The preset constants of the process parameters to be adjusted.

[0041] Correspondingly, the present application also provides an intelligent smelting process parameter optimization management system, including:

[0042] The process module is used to obtain the smelting process flow and determine the process parameter input, smelting equipment and smelting raw material output of each link in the smelting process flow;

[0043] A determination module is used to collect the historical data of process parameter input, smelting equipment related data and smelting raw material output, determine the preliminary time lag of process parameters according to the process parameter input and smelting raw material output, and define the smelting state of smelting equipment;

[0044] The analysis module is used to determine the nonlinear relationship between the process parameter input and the smelting raw material output and the coupling relationship between the process parameters, and adjust the initial time lag of the process parameters through the nonlinear relationship and the coupling relationship to obtain the time lag of the process parameters;

[0045] The optimization module is used to optimize the process parameters of each link in the smelting process based on the smelting status, nonlinear relationship, coupling relationship and time lag of the process parameters of the smelting equipment.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. Determine the initial time lag of process parameters based on process parameter input and smelting raw material output, taking into account the time lag characteristics of process parameters when adjusting. Define the smelting state of smelting equipment and define the working state of smelting equipment, providing a reliable basis for the subsequent optimization of process parameters.

[0048] 2. Determine the nonlinear relationship between process parameter input and smelting raw material output, as well as the coupling relationship between process parameters. Consider process parameter input, smelting state, and smelting raw material output to analyze the nonlinear relationship, and describe it through a polynomial regression model. Adjust the initial time lag in combination with the nonlinear relationship and coupling relationship. Determine the adjustment step size based on the smelting state, nonlinear relationship, coupling relationship, and time lag of process parameters of the smelting equipment, adjust the process parameters stably, quickly, and reasonably, improve the accuracy and adaptability of process parameter optimization adjustment, and ensure the efficient and safe progress of the smelting process. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the process flow of the intelligent smelting process parameter optimization management method proposed in the present invention;

[0050] Figure 2 This is a schematic diagram of the structure of the intelligent smelting process parameter optimization management system proposed in the present invention. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0052] Reference Figure 1 ,The intelligent smelting process parameter optimization management method includes,the following steps,

[0053] Step S101, obtaining the smelting process flow, and determining the process parameter input, smelting equipment and smelting raw material output of each link in the smelting process flow.

[0054] In this embodiment, smelting is the process of extracting metal elements from ore, which is a key link in the production of metal materials. The smelting process mainly includes ore dressing, crushing, grinding, flotation, smelting and refining.

[0055] Ore dressing: remove impurities from the raw ore through physical or chemical methods to increase the metal content. Physical methods include gravity separation, magnetic separation, electrostatic separation, etc., and chemical methods include leaching, cyanidation, etc.

[0056] Crushing and Grinding: Crushing the raw ore into smaller particles and further improving the fineness of the ore through grinding for subsequent processing.

[0057] Flotation: Use the bubble method to separate useful mineral particles from impurities in the ore pulp after grinding.

[0058] Smelting: The ore after flotation is heated and the metal elements are extracted through chemical reaction or physical separation.

[0059] Refining: Fine processing of the metal obtained from smelting to remove impurities and improve purity.

[0060] In this embodiment, the process parameter input includes pressure, furnace temperature, crusher speed, etc., the smelting equipment includes crusher, ball mill, magnetic separator, flotation machine, blast furnace, converter, electric furnace, etc., and the smelting raw material output includes raw material output quantity and quality, etc.

[0061] Step S102, collect the historical data of process parameter input, smelting equipment related data and smelting raw material output, determine the preliminary time lag of the process parameters according to the process parameter input and smelting raw material output, and define the smelting state of the smelting equipment.

[0062] In this embodiment, time lag refers to the fact that the control of process parameters will be reflected in the output of smelting raw materials after a period of time. Because the smelting process is a complex reaction process that combines chemistry and physics, the influence of process parameters at time t1 may not be reflected until time t2. The initial time lag is a relatively vague and inaccurate time lag, which is obtained by analyzing the historical data of process parameter input, smelting equipment related data and smelting raw material output. It will be adjusted later based on the coupling relationship between process parameters and the nonlinear relationship between process parameters and smelting raw material output.

[0063] In some embodiments of the present application, the preliminary time lag of the process parameters is determined according to the process parameter input and the smelting raw material output, including:

[0064] According to the process parameter input and smelting raw material output, the data scatter diagrams of the two changing over time are constructed respectively;

[0065] Conduct spectrum analysis and wavelet analysis on the process parameter input and smelting raw material output of each link in the smelting process to obtain multiple fluctuation parameters, and determine the fluctuation index by combining multiple fluctuation parameters;

[0066] Each link in the smelting process is divided into multiple fluctuation stages according to the fluctuation index range. Each fluctuation stage corresponds to a fluctuation range. The smoothing weights of process parameter input and smelting raw material output are determined according to the median of the fluctuation range.

[0067] Performing weighted average smoothing processing on the data scatter diagrams of the process parameter input and the smelting raw material output according to the smoothing weights, and obtaining the curve diagrams of the process parameter input and the smelting raw material output;

[0068] Preliminary time lags of process parameters are determined based on respective curve graphs of process parameter input and smelting raw material output.

[0069] In this embodiment, the smelting process itself is a complex physical and chemical process, which is affected by many factors (such as raw material composition, operating conditions, equipment status, etc.). Changes in these factors may cause process fluctuations, resulting in random fluctuations in process parameters and smelting raw material output data. In order to minimize the impact of these fluctuations, the fluctuation process is analyzed and targeted smoothing is performed. Spectral analysis is a technique for converting time domain signals into frequency domain signals. Through spectral analysis, we can identify the main frequency components in the signal to understand the periodic characteristics of the fluctuation. Wavelet analysis is a multi-resolution analysis method that can analyze signals at different scales. Through wavelet analysis, we can simultaneously obtain global and local information of the signal, so as to more accurately understand the changes in the signal. Multiple fluctuation parameters include fluctuation amplitude, fluctuation frequency, fluctuation phase, etc., and these fluctuation parameters are combined to determine a fluctuation index to describe the degree of fluctuation, and exponential weighting is performed according to the degree of fluctuation. The exponential smoothing rule smoothes by assigning different weights to data at different time points (different fluctuation intervals).

[0070] In some embodiments of the present application, the preliminary time lag of the process parameters is determined based on the respective curves of the process parameter input and the smelting raw material output, including:

[0071] Calculate the mutual information between the process parameter input and the smelting raw material output, and draw a mutual information change curve of the mutual information between the process parameter input and the smelting raw material output over time according to their respective curve graphs;

[0072] Identify the obvious change points on the curve graph of the process parameter input, determine the time point when the change is obvious, and mark the time point when the change is obvious on the mutual information change curve;

[0073] Determine the maximum value of the mutual information within a preset time range after the time point at which the mutual information change curve changes significantly, and record the time point at which the mutual information has the maximum value as the end time point;

[0074] Determine a mutual information change rate obvious point within a preset time range after a time point at which the change on the mutual information change curve is obvious, and record the time point at which the mutual information change rate is obvious as the starting time point. The mutual information change rate obvious point is the earliest point at which the mutual information change rate exceeds the preset change rate;

[0075] The preliminary time lag of each process parameter was determined based on the start time point and the end time point.

[0076] In this embodiment, mutual information is a concept based on Shannon's information entropy theory, which is used to measure the degree of association between two variables. On the curve graph of the process parameter input, the obvious change point is identified, and the time point of obvious change is determined. The obvious change point is the point where the process parameter changes greatly, which is convenient for the analysis of time lag. The point at which the earliest mutual information change rate exceeds the preset change rate is taken as the starting time node, the time point of the maximum mutual information value is recorded as the end time point, and the time length between the starting time point and the end time point is taken as the length of the initial time lag of the process parameter. Find the time point when the rate of change begins to increase significantly. This point usually marks the rapid rise of the mutual information value and can be used as the beginning of the lag time period.

[0077] In some embodiments of the present application, the smelting state of the smelting equipment is defined, including:

[0078] The data related to the smelting equipment include operating status parameters and production process parameters. Each of the operating status parameters and the production process parameters is evaluated, and the evaluation index of each parameter is determined. The smelting status level of the smelting equipment is evaluated according to the evaluation index of the operating status parameters and the production process parameters, and the smelting status is described by the smelting status level.

[0079] ;

[0080] in, For the The smelting state level of each smelting equipment, For the The number of operating status parameters of smelting equipment, For the The combined weight of the operating state parameters, For the The first smelting equipment The evaluation index of the operating status parameters is For the The number of production process parameters of each smelting equipment, For the The combined weights of the production process parameters, For the The first smelting equipment The evaluation index of production process parameters, , Respectively The first constant and the second constant of a smelting equipment, [] is the rounding symbol.

[0081] In this embodiment, the operating state parameters include equipment temperature, current voltage, vibration, etc., and the production process parameters include raw material input amount, input rate, energy consumption, etc., which can indirectly reflect the working state of the smelting equipment. Each parameter in the operating state parameters and the production process parameters is evaluated according to a preset interval to obtain an evaluation index.

[0082] In this embodiment, It represents the correction of the relative average evaluation of the production process parameters to the relative average evaluation of the operating status parameters. This is to balance the size of the correction function. This is to balance the size of the smelting status level.

[0083] Step S103, determining the nonlinear relationship between the process parameter input and the smelting raw material output and the coupling relationship between the process parameters, adjusting the initial time lag of the process parameters through the nonlinear relationship and the coupling relationship, and obtaining the time lag of the process parameters.

[0084] In this embodiment, a polynomial regression model is used to describe the nonlinear relationship between process parameter input and smelting raw material output, and a machine learning algorithm (such as a neural network) is used to analyze the coupling relationship between process parameters.

[0085] In some embodiments of the present application, determining the nonlinear relationship between process parameter input and smelting raw material output and the coupling relationship between process parameters includes:

[0086] The nonlinear relationship between process parameter input and smelting raw material output is described by a polynomial regression model, and the model parameters of the polynomial regression model include polynomial order, intercept term, error term and polynomial coefficient;

[0087] Preset the intercept term, determine the initial error term according to the smelting state level, determine the polynomial order according to the discrete degree of the process parameters, determine the polynomial coefficients and adjust the initial error term through the least square method or cross-validation based on the intercept term, the initial error term and the polynomial order, so as to determine the polynomial regression model;

[0088] ;

[0089] in, For the Polynomial regression of process parameters and smelting raw material output, is the intercept term, is the polynomial order, For the The coefficient of the second term, For the The process parameters of The size of the secondary order, is the adjusted error term;

[0090] The polynomial regression model integrating all process parameters describes the nonlinear relationship between process parameter input and smelting raw material output.

[0091] In this embodiment, according to the distribution of smelting process data, the appropriate polynomial order is selected. The intercept term represents the expected value of the dependent variable (smelting raw material output) when all independent variables (process parameters) are 0 (i.e., the raw material situation that has not been processed by the smelting equipment). The error term represents the difference between the model prediction value and the actual value. In the smelting process, due to the influence of various factors such as equipment status, raw material quality, and operating conditions, the actual output may deviate from the model prediction value. By considering the influence of the smelting state, the size and distribution of the error term can be more accurately estimated, thereby improving the prediction accuracy of the model. Therefore, an initial error term is first determined according to the smelting state level. The higher the smelting state level, the smaller the initial error term. The least squares method estimates the coefficients of the model (polynomial coefficients, which determine the degree and direction of the influence of different power terms of the independent variable x on the dependent variable y) by minimizing the residual square sum between the observed value and the model prediction value. The least squares method can provide a closed-form solution for the coefficients with high computational efficiency.

[0092] In some embodiments of the present application, determining the nonlinear relationship between the process parameter input and the smelting raw material output and the coupling relationship between the process parameters also includes:

[0093] The correlation coefficients between the process parameter inputs are calculated, and training sets and test sets are constructed. The coupled neural network model is trained based on the correlation coefficients, training sets and test sets. The coupled relationship between the process parameters is described by the coupled neural network model.

[0094] In this embodiment, the correlation coefficient is the Spearman correlation coefficient, which is applicable to nonlinear relationships, and the process parameter data set is divided into a training set and a test set. The training set is used to train the model so that the model can learn the laws and patterns in the data; the test set is used to evaluate the performance of the model and test whether the model has good generalization ability. When dividing the training set and the test set, it should be ensured that they are mutually exclusive and the consistency of the data distribution is maintained to avoid introducing additional deviations due to data division. The coupled neural network model is a model that can simulate the complex interactions between multiple variables. During the training process, the process parameters need to be used as input, and the coupling relationship between these parameters is learned through the neural network model. During the training process, optimization methods such as back propagation algorithms can be used to continuously adjust the parameters of the model to minimize the prediction error. The coupled relationship between the process parameters is described by the trained coupled neural network model. The model can output the degree of mutual influence between any two or more process parameters, thereby helping to more deeply understand the interaction of process parameters in the smelting production process.

[0095] In some embodiments of the present application, the initial time hysteresis of the process parameters is adjusted by the nonlinear relationship and the coupling relationship to obtain the time hysteresis of the process parameters, including:

[0096] The process parameters are screened through coupling relationships. On the basis of preliminary time lags, the smelting process is simulated according to the nonlinear relationship and coupling relationship of the screened process parameters, and the time lag of the process parameters is adjusted.

[0097] In this embodiment, when adjusting the time hysteresis of a certain process parameter, the influence of other parameters on the parameter is fully considered. If there is a strong coupling relationship between two parameters, when adjusting the time hysteresis of one parameter, it is necessary to observe the response and change of the other parameter at the same time. The time hysteresis of the selected process parameter is adjusted step by step. After each adjustment, the change of the system output is observed through simulation, and the data before and after the adjustment are recorded.

[0098] Step S104, optimizing the process parameters of each link in the smelting process flow based on the smelting state, nonlinear relationship, coupling relationship and time lag of the process parameters of the smelting equipment.

[0099] In this embodiment, the time lag of the process parameters is used to help control the time situation of the adjustment. When adjusting the process parameters, the time lag is fully considered. Through historical data and real-time monitoring, the response time and effect after parameter adjustment are predicted.

[0100] In some embodiments of the present application, the process parameters of each link in the smelting process are optimized based on the smelting state, nonlinear relationship, coupling relationship and time lag of the process parameters of the smelting equipment, including:

[0101] Determine the target range and actual value of the process parameters in each link of the smelting process flow, record the process parameters whose actual values ​​are outside the target range as the process parameters to be adjusted, and quantify the strength of the nonlinear relationship and coupling relationship of the process parameters to be adjusted;

[0102] The adjustment step is determined based on the smelting state of the smelting equipment, the nonlinear relationship of the process parameters to be adjusted, and the strength of the coupling relationship. The actual value of the process parameters to be adjusted is brought closer to the target range by combining the adjustment step and the time lag of the process parameters to be adjusted.

[0103] ;

[0104] in, For the The adjustment step length of the process parameters to be adjusted, is the smelting state of the smelting equipment, Adjust the step length based on It represents the basic adjustment step obtained by mapping the smelting state of the smelting equipment. , are the contribution weights of nonlinear relationship and coupling relationship respectively, , Respectively The strength of the nonlinear relationship and coupling relationship of the process parameters to be adjusted, For the The preset constants of the process parameters to be adjusted.

[0105] In this embodiment, the target range of the process parameters of each link in the smelting process is determined (comprehensive coupling relationship and nonlinear relationship are used to determine the reasonable target range). The smelting state of the smelting equipment (describing the quality of the smelting state of the equipment) directly affects the selection of the adjustment step. If the equipment is in good condition and can operate stably, then we can choose a larger adjustment step to approach the optimal process parameter combination faster. It represents the adjustment of the basic adjustment step by the sum of the strengths of the nonlinear relationship and coupling relationship of the process parameters to be adjusted. The stronger the strength, the greater and more complex the influence of the process parameters to be adjusted, and the smaller the basic adjustment step needs to be adjusted to slowly approach the target range.

[0106] Correspondingly, the present application also provides an intelligent smelting process parameter optimization management system, such as Figure 2 As shown, including,

[0107] The process module is used to obtain the smelting process flow and determine the process parameter input, smelting equipment and smelting raw material output of each link in the smelting process flow;

[0108] A determination module is used to collect the historical data of process parameter input, smelting equipment related data and smelting raw material output, determine the preliminary time lag of process parameters according to the process parameter input and smelting raw material output, and define the smelting state of smelting equipment;

[0109] The analysis module is used to determine the nonlinear relationship between the process parameter input and the smelting raw material output and the coupling relationship between the process parameters, and adjust the initial time lag of the process parameters through the nonlinear relationship and the coupling relationship to obtain the time lag of the process parameters;

[0110] The optimization module is used to optimize the process parameters of each link in the smelting process based on the smelting status, nonlinear relationship, coupling relationship and time lag of the process parameters of the smelting equipment.

[0111] Compared with the prior art, the present invention has the following beneficial effects:

[0112] 1. Determine the initial time lag of process parameters based on process parameter input and smelting raw material output, taking into account the time lag characteristics of process parameters when adjusting. Define the smelting state of smelting equipment and define the working state of smelting equipment, providing a reliable basis for the subsequent optimization of process parameters.

[0113] 2. Determine the nonlinear relationship between process parameter input and smelting raw material output, as well as the coupling relationship between process parameters. Consider process parameter input, smelting state, and smelting raw material output to analyze the nonlinear relationship, and describe it through a polynomial regression model. Adjust the initial time lag in combination with the nonlinear relationship and coupling relationship. Determine the adjustment step size based on the smelting state, nonlinear relationship, coupling relationship, and time lag of process parameters of the smelting equipment, adjust the process parameters stably, quickly, and reasonably, improve the accuracy and adaptability of process parameter optimization adjustment, and ensure the efficient and safe progress of the smelting process.

[0114] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0115] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present invention.

[0116] Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more systems different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0117] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. Intelligent smelting process parameter optimization management method, characterized in that: include, Obtain the smelting process flow, determine the process parameter input, smelting equipment and smelting raw material output of each link in the smelting process flow; Collect historical data of process parameter input, smelting equipment related data and smelting raw material output, determine the preliminary time lag of process parameters based on process parameter input and smelting raw material output, and define the smelting state of smelting equipment; Determine the nonlinear relationship between process parameter input and smelting raw material output and the coupling relationship between process parameters, adjust the initial time lag of process parameters through the nonlinear relationship and coupling relationship, and obtain the time lag of process parameters; Optimize the process parameters of each link in the smelting process based on the smelting state, nonlinear relationship, coupling relationship and time lag of process parameters of smelting equipment; in, And define the smelting status of smelting equipment, including, The data related to the smelting equipment include operating status parameters and production process parameters. Each of the operating status parameters and the production process parameters is evaluated, and the evaluation index of each parameter is determined. The smelting status level of the smelting equipment is evaluated according to the evaluation index of the operating status parameters and the production process parameters, and the smelting status is described by the smelting status level. ； in, For the The smelting state level of each smelting equipment, For the The number of operating status parameters of smelting equipment, For the The combined weight of the operating state parameters, For the The first smelting equipment The evaluation index of the operating status parameters is For the The number of production process parameters of each smelting equipment, For the The combined weights of the production process parameters, For the The first smelting equipment The evaluation index of production process parameters, , Respectively The first constant and the second constant of a smelting equipment, [] is the rounding symbol; Determine the nonlinear relationship between process parameter input and smelting raw material output as well as the coupling relationship between process parameters, including, The nonlinear relationship between process parameter input and smelting raw material output is described by a polynomial regression model, and the model parameters of the polynomial regression model include polynomial order, intercept term, error term and polynomial coefficient; Preset the intercept term, determine the initial error term according to the smelting state level, determine the polynomial order according to the discrete degree of the process parameters, determine the polynomial coefficients and adjust the initial error term through the least square method or cross-validation based on the intercept term, the initial error term and the polynomial order, so as to determine the polynomial regression model; ； in, For the Polynomial regression of process parameters and smelting raw material output, is the intercept term, is the polynomial order, For the The coefficient of the second term, For the The process parameters of The size of the secondary order, is the adjusted error term; A polynomial regression model integrating all process parameters describes the nonlinear relationship between process parameter input and smelting raw material output; Based on the smelting state, nonlinear relationship, coupling relationship and time lag of process parameters of smelting equipment, the process parameters of each link in the smelting process are optimized, including: Determine the target range and actual value of the process parameters in each link of the smelting process flow, record the process parameters whose actual values ​​are outside the target range as the process parameters to be adjusted, and quantify the strength of the nonlinear relationship and coupling relationship of the process parameters to be adjusted; The adjustment step is determined based on the smelting state of the smelting equipment, the nonlinear relationship of the process parameters to be adjusted, and the strength of the coupling relationship. The actual value of the process parameters to be adjusted is brought closer to the target range by combining the adjustment step and the time lag of the process parameters to be adjusted. ； in, For the The adjustment step length of the process parameters to be adjusted, is the smelting state of the smelting equipment, Adjust the step length based on It represents the basic adjustment step obtained by mapping the smelting state of the smelting equipment. , are the contribution weights of nonlinear relationship and coupling relationship respectively, , Respectively The strength of the nonlinear relationship and coupling relationship of the process parameters to be adjusted, For the The preset constants of the process parameters to be adjusted.

2. The intelligent smelting process parameter optimization management method according to claim 1 is characterized in that: Determine the initial time lag of process parameters based on process parameter input and smelting raw material output, including, According to the process parameter input and smelting raw material output, the data scatter diagrams of the two changing over time are constructed respectively; Conduct spectrum analysis and wavelet analysis on the process parameter input and smelting raw material output of each link in the smelting process to obtain multiple fluctuation parameters, and determine the fluctuation index by combining multiple fluctuation parameters; According to the interval to which the fluctuation index belongs, each link in the smelting process is divided into multiple fluctuation stages. Each fluctuation stage corresponds to a fluctuation interval. The smoothing weights of the process parameter input and the smelting raw material output are determined according to the median of the fluctuation interval. Performing weighted average smoothing processing on the data scatter diagrams of the process parameter input and the smelting raw material output according to the smoothing weights, and obtaining the curve diagrams of the process parameter input and the smelting raw material output; Preliminary time lags of process parameters are determined based on respective curve graphs of process parameter input and smelting raw material output.

3. The intelligent smelting process parameter optimization management method according to claim 2 is characterized in that: Determine the initial time lag of process parameters based on the respective curves of process parameter input and smelting raw material output, include, Calculate the mutual information between the process parameter input and the smelting raw material output, and draw a mutual information change curve of the mutual information between the two over time based on the respective curve graphs of the process parameter input and the smelting raw material output; Identify the obvious change points on the curve graph of the process parameter input, determine the time point when the change is obvious, and mark the time point when the change is obvious on the mutual information change curve; Determine the maximum value of the mutual information within a preset time range after the time point at which the mutual information change curve changes significantly, and record the time point at which the mutual information has the maximum value as the end time point; Determine a mutual information change rate obvious point within a preset time range after a time point at which the change on the mutual information change curve is obvious, and record the time point at which the mutual information change rate is obvious as the starting time point. The mutual information change rate obvious point is the earliest point at which the mutual information change rate exceeds the preset change rate; The preliminary time lag of each process parameter was determined based on the start time point and the end time point.

4. The intelligent smelting process parameter optimization management method according to claim 1 is characterized in that: Determine the nonlinear relationship between process parameter input and smelting raw material output and the coupling relationship between process parameters, including, The correlation coefficients between the process parameter inputs are calculated, and training sets and test sets are constructed. The coupled neural network model is trained based on the correlation coefficients, training sets and test sets. The coupled relationship between the process parameters is described by the coupled neural network model.

5. The intelligent smelting process parameter optimization management method according to claim 1 is characterized in that: The initial time lag of the process parameters is adjusted through nonlinear relationships and coupling relationships to obtain the time lag of the process parameters, including: The process parameters are screened through coupling relationships. On the basis of preliminary time lags, the smelting process is simulated according to the nonlinear relationship and coupling relationship of the screened process parameters, and the time lag of the process parameters is adjusted.

6. Intelligent smelting process parameter optimization management system, characterized by: Used to implement the intelligent smelting process parameter optimization management method as described in any one of claims 1 to 5, the system includes: The process module is used to obtain the smelting process flow and determine the process parameter input, smelting equipment and smelting raw material output of each link in the smelting process flow; A determination module is used to collect the historical data of process parameter input, smelting equipment related data and smelting raw material output, determine the preliminary time lag of process parameters according to the process parameter input and smelting raw material output, and define the smelting state of smelting equipment; The analysis module is used to determine the nonlinear relationship between the process parameter input and the smelting raw material output and the coupling relationship between the process parameters, and adjust the initial time lag of the process parameters through the nonlinear relationship and the coupling relationship to obtain the time lag of the process parameters; The optimization module is used to optimize the process parameters of each link in the smelting process based on the smelting status, nonlinear relationship, coupling relationship and time lag of the process parameters of the smelting equipment.

Images