Cost prediction method and system applied to smelting industry

Abstract

The invention discloses a cost prediction method and system applied to the smelting industry, and relates to the technical field of data analysis, and the method comprises the steps: setting a standard line of each aspect of content according to the characteristics of the various aspects of content; the standard line of each parameter of each aspect of content is comprehensively set by combining historical data, content characteristics of each aspect and smelting process simulation conditions, and a reliable basis is provided for subsequent stable conditions and unstable conditions and prediction. The difference between the standard line and the actual line of each aspect content is compared to distinguish a stable part and an unstable part on the actual line, and a prediction strategy is selected according to the unstable part, so that targeted prediction is performed, and the change condition of each aspect of the smelting process is better predicted. For abnormal conditions, additional cost is analyzed, comprehensive cost is calculated, the accuracy and adaptability of smelting cost prediction are improved, the management condition of the smelting cost is optimized, and efficient operation of smelting process tasks is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a cost prediction method and system applied to the metallurgical industry. Background Technology

[0002] Given the complexity of modern smelting industry and the need for data-driven decision-making, the smelting process involves multiple stages, including ore preparation, smelting, and refining. Each stage generates a large amount of data, including raw material consumption, process parameters, equipment status, quality inspection, and energy consumption. With the development of big data and artificial intelligence technologies, this data is effectively collected, cleaned, integrated, and analyzed to reveal the cost drivers in the smelting process.

[0003] In existing technologies, smelting costs are often predicted based on the target parameters of the smelting plan. However, the complex and variable operating conditions of the smelting process lead to significant discrepancies between the actual and planned smelting conditions, resulting in poor accuracy and low adaptability in smelting cost prediction. Therefore, accurate prediction of the smelting process is necessary to ensure the precision of cost forecasting. Therefore, how to improve the accuracy and adaptability of smelting cost prediction is a technical problem that needs to be solved. Summary of the Invention

[0004] The purpose of this invention is to address the problems of poor accuracy and low adaptability in smelting cost prediction in existing technologies due to the inability to accurately predict and analyze various operating conditions during the smelting process. Therefore, this invention proposes a cost prediction method applicable to the smelting industry, which includes: Obtain the smelting process, define the various aspects involved in the smelting process, and set the standard line for each aspect based on the characteristics of these aspects. Data on various aspects of the smelting process over a given period of time is obtained, and actual lines for each aspect are established. The differences between the standard line and the actual line for each aspect are compared to distinguish the stable and unstable parts on the actual line. Based on the unstable part, a forecasting strategy is selected, and the actual line is predicted for a future period of time according to the forecasting strategy to obtain the forecast line; Determine the matching relationship between each aspect of the smelting process and its cost, and combine this with forecast lines to predict smelting costs, thereby optimizing smelting cost management.

[0005] In some embodiments of this application, the various aspects involved in the smelting process include smelting process proportions, smelting equipment status, and smelting emissions.

[0006] In some embodiments of this application, standard lines for each aspect are set based on the characteristics of multiple aspects, including, The types of parameters involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions are determined respectively. Historical data on parameters related to three aspects—smelting process proportions, smelting equipment status, and smelting emissions—were collected, and the historical data of the parameters were matched according to the category of smelting task. The historical data of parameters for smelting tasks under the same category are statistically analyzed for characteristic values, including the mean, median, standard deviation, extreme value group, and outlier group. The extreme value group is the set of points that are close to the threshold boundary and occur frequently, and the outlier group is the set of points that exceed the threshold boundary. Calculate the correlation between each parameter type and the quality of smelting products, calculate the distance from the outlier group to the threshold boundary and the distance from the outlier group to the median value, combine the two types of distances and correlation to determine the baseline coefficient of the parameter, and set the initial baseline of the parameter based on the baseline coefficient, mean and standard deviation. Find the point on the extreme value group that is closest to the initial baseline, and adjust the initial baseline accordingly. The correlation between the parameter types involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions was analyzed. A smelting process simulation model was established to optimize the initial baseline. The optimized initial baseline was used as the standard line for each parameter type involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions.

[0007] In some embodiments of this application, the correlation between parameter types involved in three aspects—smelting process proportions, smelting equipment status, and smelting emissions—is analyzed to establish a smelting process simulation model to optimize the initial baseline, including... The relationships between parameters in three aspects—smelting process proportions, smelting equipment status, and smelting emissions—were analyzed. Based on these relationships, a smelting process simulation model was constructed. The initial baseline was then optimized by adjusting the output of the smelting process simulation model.

[0008] In some embodiments of this application, the difference between the standard line and the actual line for each aspect is compared to distinguish stable and unstable portions on the actual line, including: The actual line is a curve showing the changes of all parameters involved in the three aspects of smelting process proportion, smelting equipment status, and smelting emissions over time. The position of the standard line is marked on the curve. The portion of the curve that lies within the standard line is called the stable portion, and the portion of the curve that lies outside the standard line is called the unstable portion. Determine the volatility characteristics of each unstable segment and the ratio of the duration of the unstable segment to the duration of the stable segment. Integrate the volatility characteristics to obtain the volatility value of each unstable segment and determine the volatility index of the unstable segment. ; in, For the first A volatility indicator for the unstable portion of each parameter. For the first The number of unstable parts of each parameter For the first The combination weights corresponding to each unstable part of time. For the first The parameter of the first The fluctuation value of each unstable part , These are two conversion factors, for The maximum value in, This represents the ratio of the duration of the unstable portion to the duration of the stable portion. For the first The first constant of the parameters.

[0009] In some embodiments of this application, a prediction strategy is selected based on the unstable portion, including: Forecasting strategies include two types: individual forecasting and comprehensive forecasting. The choice between individual and comprehensive forecasting is determined by the volatility indicators of the unstable part. Individual forecasting forecasts separately for both the stable and unstable parts, while comprehensive forecasting forecasts the stable and unstable parts as a whole.

[0010] In some embodiments of this application, a prediction strategy is used to predict the actual line over a future period to obtain a predicted line, including... All parameter types are grouped into individual prediction groups and comprehensive prediction groups based on prediction strategies. For the parameter types of individual prediction groups, the first model and the second model are used to predict the stable and unstable parts respectively, and the prediction lines are obtained. For the parameter types of the comprehensive prediction group, a third model is used to comprehensively predict the stable and unstable parts to obtain the prediction line.

[0011] In some embodiments of this application, a prediction line is used to predict smelting costs, including: Identify the normal and abnormal parts of the prediction line. For the normal part, predict the cost by matching the relationship between various aspects and costs. For the abnormal part, identify the abnormal causes, multiple additional cost types and the degree of abnormal impact in the abnormal part, predict the direct cost by matching the cost with multiple aspects, and determine the total cost based on the direct cost, abnormal causes, multiple additional cost types and the degree of abnormal impact. ; in, The total cost of the abnormal portion. For direct costs, This is an abnormal time. The quantity of additional costs due to abnormal causes. For the cost value of the j-th additional cost type, To determine the degree of abnormal impact, This represents the impact coefficient of abnormal costs. This represents the abnormal cost impact coefficient, which is obtained by mapping the degree of abnormal impact.

[0012] Correspondingly, this application also provides a cost prediction system for the smelting industry, including, The first module is used to acquire the smelting process, define the various aspects involved in the smelting process, and set the standard line for each aspect based on the characteristics of the various aspects. The second module is used to acquire data on various aspects of the smelting process over a period of time, establish actual lines for each aspect, and compare the differences between the standard lines and actual lines for each aspect to distinguish the stable and unstable parts on the actual lines. The third module is used to select a prediction strategy based on the unstable part, and to make a prediction on the actual line for a future period of time according to the prediction strategy, so as to obtain the prediction line. The fourth module is used to determine the matching relationship between each aspect of the smelting process and its cost, and to combine this with the forecast line to predict smelting costs, thereby optimizing smelting cost management.

[0013] Compared with the prior art, the beneficial effects of this invention are as follows: 1. Standard lines are set for each aspect based on its characteristics. This involves combining historical data, the specific characteristics of each aspect, and simulations of the smelting process to comprehensively set the standard lines for each parameter within each aspect. This provides a reliable foundation for subsequent predictions of stable and unstable conditions. The differences between the standard lines and actual lines for each aspect are compared to distinguish the stable and unstable portions of the actual line. Prediction strategies are selected based on the unstable portions to identify the current stable and unstable conditions, enabling targeted predictions and better forecasting of changes in each aspect of the smelting process.

[0014] 2. Determine the matching relationship between each aspect of the smelting process and its cost, combine this with the forecast line to predict smelting costs, identify normal and abnormal situations on the forecast line, analyze the additional costs incurred in abnormal situations, calculate the comprehensive cost, improve the accuracy and adaptability of smelting cost prediction, optimize smelting cost management, and ensure the efficient operation of smelting process tasks. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a cost prediction method for the metallurgical industry proposed in this invention. Figure 2 This is a schematic diagram of the smelting cost prediction system based on data analysis proposed in this invention. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0017] Reference Figure 1 A cost prediction method applied to the metallurgical industry includes the following steps: Step S101: Obtain the smelting process, define the various aspects involved in the smelting process, and set the standard line for each aspect based on its characteristics.

[0018] In some embodiments of this application, the various aspects involved in the smelting process include smelting process proportions, smelting equipment status, and smelting emissions.

[0019] In this embodiment, the smelting process encompasses key stages such as ore dressing, crushing, leaching, flotation, heat treatment, and electrolysis. These stages are crucial for ensuring the quality and performance of the final product. The smelting equipment includes various smelting equipment such as blast furnaces, electric furnaces, converters, and refining furnaces, as well as related auxiliary equipment such as crushers, grinding mills, and classifiers. The operating status and performance of the equipment directly affect the efficiency of the smelting process and the quality of the product. Regarding emissions, pollutants such as waste gas, wastewater, and waste residue generated during the smelting process have a negative impact on the environment. Therefore, emissions are also a key aspect that needs to be monitored during the smelting process. All three major aspects mentioned above are closely related to cost, and there is a strong correlation between them. Setting standard lines for each aspect refers to setting standard parameter ranges (standard lines, similar to thresholds) for all parameters under each aspect.

[0020] In some embodiments of this application, standard lines for each aspect are set based on the characteristics of multiple aspects, including, The types of parameters involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions are determined respectively. Historical data on parameters related to three aspects—smelting process proportions, smelting equipment status, and smelting emissions—were collected, and the historical data of the parameters were matched according to the category of smelting task. The historical data of parameters for smelting tasks under the same category are statistically analyzed for characteristic values, including the mean, median, standard deviation, extreme value group, and outlier group. The extreme value group is the set of points that are close to the threshold boundary and occur frequently, and the outlier group is the set of points that exceed the threshold boundary. Calculate the correlation between each parameter type and the quality of smelting products, calculate the distance from the outlier group to the threshold boundary and the distance from the outlier group to the median value, combine the two types of distances and correlation to determine the baseline coefficient of the parameter, and set the initial baseline of the parameter based on the baseline coefficient, mean and standard deviation. Find the point on the extreme value group that is closest to the initial baseline, and adjust the initial baseline accordingly. The correlation between the parameter types involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions was analyzed. A smelting process simulation model was established to optimize the initial baseline. The optimized initial baseline was used as the standard line for each parameter type involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions.

[0021] In this embodiment, the parameters related to the smelting process proportions include raw material input, reaction temperature, and the proportion of different raw materials. The parameters related to the smelting equipment status include equipment temperature, pressure, rotation speed, material flow rate, air permeability, and tuyeres. The parameters related to smelting emissions include exhaust gas emission concentrations (such as sulfur dioxide, nitrogen oxides, and particulate matter) and wastewater emission concentrations (such as chemical oxygen demand, ammonia nitrogen, and heavy metals).

[0022] In this embodiment, matching historical parameter data according to the category of smelting task (smelting requirement target) means grouping tasks with the same or similar smelting requirements target into one category, and classifying historical parameters with the same or similar smelting requirements target, which may have many commonalities and similarities. The extreme value group is a set of points that are close to the threshold boundary and occur frequently. Here, the threshold boundary is a past threshold, which may have errors of being too large or too small, and is not suitable for complex and variable smelting conditions. Values ​​that occur frequently near the boundary (within the boundary) are taken as the maximum or minimum value group. The correlation between each parameter type and the smelting product quality is calculated, which can be described by the Pearson correlation coefficient. In the smelting process, different parameters have different correlations with the production process and product quality. For key parameters, it may be necessary to select a smaller multiple to set a more stringent baseline range to ensure the stability of the production process and product quality. The distances from outlier clusters to the threshold boundary and from outlier clusters to the median are calculated separately. These two types of distances describe the distances from outliers to the boundary and to the center. When setting the baseline range, the influence of outliers and outliers should also be considered, as they will affect the multiplier.

[0023] In this embodiment, the baseline coefficient (multiple, which may not be positive) of the parameter is determined by combining two types of distance and correlation. The initial baseline of the parameter is set based on the baseline coefficient, mean, and standard deviation. The baseline range can be set as the mean plus or minus a certain multiple of the standard deviation. The point closest to the initial baseline is found in the extrema group, and the initial baseline is adjusted accordingly. The point closest to the initial baseline found in the extrema group replaces the point of the initial baseline.

[0024] In some embodiments of this application, the correlation between parameter types involved in three aspects—smelting process proportions, smelting equipment status, and smelting emissions—is analyzed to establish a smelting process simulation model to optimize the initial baseline, including... The relationships between parameters in three aspects—smelting process proportions, smelting equipment status, and smelting emissions—were analyzed. Based on these relationships, a smelting process simulation model was constructed. The initial baseline was then optimized by adjusting the output of the smelting process simulation model.

[0025] In this embodiment, the proportions in the smelting process, such as the ratio of raw materials and the amount of auxiliary materials added, directly affect the operating efficiency of the smelting equipment. A reasonable proportion can optimize the operating state of the equipment, improve production efficiency, and enhance equipment stability. For example, in steel smelting, increasing the proportion of sinter can optimize the thermal regime and reduction process within the blast furnace, making the blast furnace operation more stable and reducing equipment failures and downtime. The operating state of the equipment also affects the adjustment of the proportions. When equipment ages, wears, or malfunctions, it may be necessary to adjust the proportions to adapt to the actual operating capacity of the equipment. For example, if the permeability of the blast furnace decreases, it may be necessary to reduce the proportion of sinter to lower the pressure load within the blast furnace and ensure the smooth progress of the smelting process. The proportions also affect emission levels: the proportions in the smelting process directly affect the amount and type of emissions generated. Different proportions lead to different chemical reactions and material transfer processes, resulting in different emissions. For example, in nickel smelting, elements such as sulfur and nitrogen in the raw materials react with oxygen at high temperatures to generate harmful gases such as sulfur dioxide and nitrogen oxides. By optimizing the proportions, the generation of these harmful gases can be reduced. Emission parameters are closely related to the raw material ratio and equipment status during the smelting process. Inappropriate ratios or equipment malfunctions may lead to excessive emissions and environmental pollution. Therefore, the types of parameters involved in the three aspects of smelting process ratios, smelting equipment status, and smelting emissions are complexly interrelated.

[0026] In this embodiment, a mathematical model describing the smelting process is established based on the complex interrelationships among the parameter types involved in the three aspects. Numerical calculations and simulations are performed using computer software (such as ANSYS, COMSOL Multiphysics, MATLAB, etc.). Parameters such as smelting process proportions and equipment status are used as input variables for the model. The operation and emissions of the smelting process are simulated. The model outputs curves of key parameter changes, predicted product quality values, and predicted emission concentration values. Preliminary baseline values ​​are substituted into the smelting process simulation model. The model is run to observe whether the operation and emissions of the smelting process meet expectations. Adjustment methods can include stepwise approximation, sensitivity analysis, etc. (adjustment methods). For example, the baseline value of a certain parameter is gradually adjusted, and the changes in the simulation results are observed until the most suitable baseline value is found.

[0027] Step S102: Obtain data on multiple aspects of the smelting process over a current period of time, establish actual lines for each aspect, and compare the differences between the standard lines and actual lines for each aspect to distinguish the stable and unstable parts on the actual lines.

[0028] In this embodiment, sensors and monitoring equipment are used to acquire real-time data on three aspects: smelting process, smelting equipment, and emissions. The data is cleaned, organized, and analyzed to ensure its accuracy and reliability. Based on the acquired data, actual lines representing the smelting process, smelting equipment, and emissions are plotted. These actual lines reflect the actual operating status of the smelting process over a given period.

[0029] In some embodiments of this application, the difference between the standard line and the actual line for each aspect is compared to distinguish stable and unstable portions on the actual line, including: The actual line is a curve showing the changes of all parameters involved in the three aspects of smelting process proportion, smelting equipment status, and smelting emissions over time. The position of the standard line is marked on the curve. The portion of the curve that lies within the standard line is called the stable portion, and the portion of the curve that lies outside the standard line is called the unstable portion. Determine the volatility characteristics of each unstable segment and the ratio of the duration of the unstable segment to the duration of the stable segment. Integrate the volatility characteristics to obtain the volatility value of each unstable segment and determine the volatility index of the unstable segment. ; in, For the first A volatility indicator for the unstable portion of each parameter. For the first The number of unstable parts of each parameter For the first The combination weights corresponding to each unstable part of time. For the first The parameter of the first The fluctuation value of each unstable part , These are two conversion factors, for The maximum value in, This represents the ratio of the duration of the unstable portion to the duration of the stable portion. For the first The first constant of the parameters.

[0030] In this embodiment, the unstable portion describes anomalies or abnormal fluctuations outside the standard line, while the stable portion describes normal operation within the standard line. Fluctuation characteristics include fluctuation range, fluctuation amplitude, and variance. The maximum and minimum deviations between the unstable portion curve and the standard line are calculated; the difference between these two values ​​represents the fluctuation range. A larger fluctuation range indicates a higher degree of fluctuation or abnormality in the unstable portion. The difference between the maximum and minimum values ​​of the unstable portion curve within each fluctuation period is calculated; this represents the fluctuation amplitude. A larger fluctuation amplitude indicates more severe fluctuations in the unstable portion. These fluctuation characteristics are then weighted and summed to obtain the fluctuation value for each unstable portion.

[0031] It should be noted that each parameter may have multiple small unstable parts. This represents the correction to the sum of the fluctuation values ​​of the unstable portion, which has the greatest impact, based on the ratio of the duration of the unstable portion to the duration of the stable portion. It exists to balance the correction function.

[0032] Step S103: Select a prediction strategy based on the unstable part, and make a prediction on the actual line for a future period of time according to the prediction strategy to obtain the prediction line.

[0033] In this embodiment, the volatility index of the unstable component affects the prediction. When the volatility index is small, both the stable and unstable components are predicted as a whole. When the volatility index is large, the stable and unstable components are predicted separately.

[0034] In some embodiments of this application, a prediction strategy is selected based on the unstable portion, including: Forecasting strategies include two types: individual forecasting and comprehensive forecasting. The choice between individual and comprehensive forecasting is determined by the volatility indicators of the unstable part. Individual forecasting forecasts separately for both the stable and unstable parts, while comprehensive forecasting forecasts the stable and unstable parts as a whole.

[0035] In some embodiments of this application, a prediction strategy is used to predict the actual line over a future period to obtain a predicted line, including... All parameter types are grouped into individual prediction groups and comprehensive prediction groups based on prediction strategies. For the parameter types of individual prediction groups, the first model and the second model are used to predict the stable and unstable parts respectively, and the prediction lines are obtained. For the parameter types of the comprehensive prediction group, a third model is used to comprehensively predict the stable and unstable parts to obtain the prediction line.

[0036] In this embodiment, when the fluctuation index is large in the smelting process, we can adopt a strategy of predicting the unstable and stable parts separately. For the unstable part, since its changes are complex and difficult to describe with linear relationships, we can use more complex models, such as deep learning models (first model), to capture its nonlinear changes. Deep learning models have strong fitting and adaptive learning capabilities, and can adapt well to the changing characteristics of the unstable part. For the stable part, since its changes are relatively regular, we can use simple linear models or time series models (second model) for prediction. These models have high efficiency and accuracy when processing stable data. When the fluctuation index is small, we can consider combining the unstable and stable parts for prediction. At this time, the data of the unstable and stable parts can be merged, and a unified prediction model, such as support vector machine or random forest, can be used for prediction (third model). These models perform well when processing mixed data, and can comprehensively consider the influence of the unstable and stable parts to obtain more accurate prediction results.

[0037] Understandably, a large volatility index indicates significant nonlinear changes and uncertainties in the data. In this case, dividing the data into stable and unstable components for separate prediction can better capture the different characteristics of each component. Conversely, a small volatility index indicates relatively stable overall data changes with minimal differences between components. In this case, merging the stable and unstable data and using a unified prediction model can improve the accuracy and efficiency of the prediction. Flexibility: Flexible selection of prediction strategies based on the magnitude of the volatility index allows for better adaptation to different data characteristics and prediction needs. Accuracy: Predicting the unstable and stable components separately, or combining both, can more accurately capture the patterns of data change and improve prediction accuracy. Efficiency: Using a simple model for the stable component improves computational efficiency; using a complex model for the unstable component ensures the capture of its nonlinear changes.

[0038] Step S104: Determine the matching relationship between each aspect of the smelting process and its cost, and combine this with the forecast line to predict the smelting cost, thereby optimizing smelting cost management.

[0039] In this embodiment, raw material costs are as follows: Raw materials are one of the main costs in the smelting process, and their price, quality, and supply stability directly affect smelting costs. For example, in steel smelting, price fluctuations of raw materials such as iron ore and coke significantly impact smelting costs. Auxiliary material costs are also included: Various auxiliary materials, such as limestone and dolomite, are used in the smelting process, and their costs vary depending on the smelting process ratios. Energy consumption is another important component of costs in the smelting process, including electricity and fuel. Different smelting process ratios lead to different energy consumption levels, thus affecting costs.

[0040] In some embodiments of this application, a prediction line is used to predict smelting costs, including: Identify the normal and abnormal parts of the prediction line. For the normal part, predict the cost by matching the relationship between various aspects and costs. For the abnormal part, identify the abnormal causes, multiple additional cost types and the degree of abnormal impact in the abnormal part, predict the direct cost by matching the cost with multiple aspects, and determine the total cost based on the direct cost, abnormal causes, multiple additional cost types and the degree of abnormal impact. ; in, The total cost of the abnormal portion. For direct costs, This is an abnormal time. The quantity of additional costs due to abnormal causes. For the cost value of the j-th additional cost type, To determine the degree of abnormal impact, This represents the impact coefficient of abnormal costs. This represents the abnormal cost impact coefficient, which is obtained by mapping the degree of abnormal impact.

[0041] In this embodiment, the main influencing factors of normal costs are analyzed, such as raw material prices, equipment efficiency, and energy consumption. Relevant historical data are collected to establish a matching relationship model between each influencing factor and cost. This can be achieved through methods such as regression analysis and time series analysis.

[0042] In this embodiment, anomalies often generate additional costs that cannot be fully determined by matching relationships. Matching relationships determine direct costs, such as poor equipment operation or malfunctions, which can lead to additional costs due to other factors. By deeply analyzing the data and smelting process of the anomalies, the causes of the anomalies can be identified, such as equipment failure, raw material quality issues, and operational errors. The direct costs generated by the anomalies (such as raw material consumption and energy consumption) can still be predicted through matching relationships between various factors and costs. The total cost of the anomalies is obtained by adding the direct costs to the additional costs. Simultaneously, the impact of the degree of anomaly on the total cost is considered, and appropriate adjustments are made.

[0043] In this embodiment, the degree of impact of an anomaly can be described by analyzing the curve characteristics (amplitude, slope, etc.) of the abnormal part of the curve, and the correlation between this parameter and other parameters can be considered to comprehensively determine the degree of impact of the anomaly, thereby adjusting the cost.

[0044] Correspondingly, this application also provides a cost prediction system for the smelting industry, such as... Figure 2 As shown, including, The first module is used to acquire the smelting process, define the various aspects involved in the smelting process, and set the standard line for each aspect based on the characteristics of the various aspects. The second module is used to acquire data on various aspects of the smelting process over a period of time, establish actual lines for each aspect, and compare the differences between the standard lines and actual lines for each aspect to distinguish the stable and unstable parts on the actual lines. The third module is used to select a prediction strategy based on the unstable part, and to make a prediction on the actual line for a future period of time according to the prediction strategy, so as to obtain the prediction line. The fourth module is used to determine the matching relationship between each aspect of the smelting process and its cost, and to combine this with the forecast line to predict smelting costs, thereby optimizing smelting cost management.

[0045] Compared with the prior art, the beneficial effects of this invention are as follows: 1. Standard lines are set for each aspect based on its characteristics. This involves combining historical data, the specific characteristics of each aspect, and simulations of the smelting process to comprehensively set the standard lines for each parameter within each aspect. This provides a reliable foundation for subsequent predictions of stable and unstable conditions. The differences between the standard lines and actual lines for each aspect are compared to distinguish the stable and unstable portions of the actual line. Prediction strategies are selected based on the unstable portions to identify the current stable and unstable conditions, enabling targeted predictions and better forecasting of changes in each aspect of the smelting process.

[0046] 2. Determine the matching relationship between each aspect of the smelting process and its cost, combine this with the forecast line to predict smelting costs, identify normal and abnormal situations on the forecast line, analyze the additional costs incurred in abnormal situations, calculate the comprehensive cost, improve the accuracy and adaptability of smelting cost prediction, optimize smelting cost management, and ensure the efficient operation of smelting process tasks.

[0047] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this 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 (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0048] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0049] Those skilled in the art will understand that the modules in the system of the implementation scenario can be distributed throughout the system of the implementation scenario as described, or they can be modified to reside in one or more systems different from this implementation scenario. The modules of the above-mentioned implementation scenario can be merged into one module, or they can be further divided into multiple sub-modules.

[0050] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A cost prediction method applied to the metallurgical industry, characterized in that, include, Obtain the smelting process, define the various aspects involved in the smelting process, and set the standard line for each aspect based on the characteristics of these aspects. Data on various aspects of the smelting process over a given period of time is obtained, and actual lines for each aspect are established. The differences between the standard line and the actual line for each aspect are compared to distinguish the stable and unstable parts on the actual line. Based on the unstable part, a forecasting strategy is selected, and the actual line is predicted for a future period of time according to the forecasting strategy to obtain the forecast line; Determine the matching relationship between each aspect of the smelting process and its cost, and combine this with forecast lines to predict smelting costs, thereby optimizing smelting cost management.

2. The cost prediction method applied to the metallurgical industry according to claim 1, characterized in that, The smelting process involves many aspects, including smelting process proportions, smelting equipment status, and smelting emissions.

3. The cost prediction method applied to the metallurgical industry according to claim 2, characterized in that, Set standard lines for each aspect of the content based on its characteristics, including: The types of parameters involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions are determined respectively. Historical data on parameters related to three aspects—smelting process proportions, smelting equipment status, and smelting emissions—were collected, and the historical data of the parameters were matched according to the category of smelting task. The historical data of parameters for smelting tasks under the same category are statistically analyzed for characteristic values, including the mean, median, standard deviation, extreme value group, and outlier group. The extreme value group is the set of points that are close to the threshold boundary and occur frequently, and the outlier group is the set of points that exceed the threshold boundary. Calculate the correlation between each parameter type and the quality of smelting products, calculate the distance from the outlier group to the threshold boundary and the distance from the outlier group to the median value, combine the two types of distances and correlation to determine the baseline coefficient of the parameter, and set the initial baseline of the parameter based on the baseline coefficient, mean and standard deviation. Find the point on the extreme value group that is closest to the initial baseline, and adjust the initial baseline accordingly. The correlation between the parameter types involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions was analyzed. A smelting process simulation model was established to optimize the initial baseline. The optimized initial baseline was used as the standard line for each parameter type involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions.

4. The cost prediction method applied to the metallurgical industry according to claim 3, characterized in that, This study analyzes the interrelationships among parameter types involved in three aspects: smelting process proportions, smelting equipment status, and smelting emissions. A smelting process simulation model is then established to optimize the initial baseline, including... The parameter relationships among the parameter types involved in the three aspects of smelting process proportions, smelting equipment status, and smelting emissions are analyzed. Based on the parameter relationships, a smelting process simulation model is constructed, and the initial baseline is optimized through the output and adjustment methods of the smelting process simulation model.

5. The cost prediction method applied to the metallurgical industry according to claim 3, characterized in that, By comparing the differences between the standard line and the actual line for each aspect, the stable and unstable parts on the actual line are distinguished, including... The actual line is a curve showing the changes of all parameters involved in the three aspects of smelting process proportion, smelting equipment status, and smelting emissions over time. The position of the standard line is marked on the curve. The portion of the curve that lies within the standard line is called the stable portion, and the portion of the curve that lies outside the standard line is called the unstable portion. Determine the volatility characteristics of each unstable segment and the ratio of the duration of the unstable segment to the duration of the stable segment. Integrate the volatility characteristics to obtain the volatility value of each unstable segment and determine the volatility index of the unstable segment. ； in, For the first A volatility indicator for the unstable portion of each parameter. For the first The number of unstable parts of each parameter For the first The combination weights corresponding to each unstable part of time. For the first The parameter of the first The fluctuation value of each unstable part , These are two conversion factors, for The maximum value in, This represents the ratio of the duration of the unstable portion to the duration of the stable portion. For the first The first constant of the parameters.

6. The cost prediction method applied to the metallurgical industry according to claim 5, characterized in that, The prediction strategy is selected based on the unstable component, including: Forecasting strategies include two types: individual forecasting and comprehensive forecasting. The choice between individual and comprehensive forecasting is determined by the volatility indicators of the unstable part. Individual forecasting forecasts separately for both the stable and unstable parts, while comprehensive forecasting forecasts the stable and unstable parts as a whole.

7. The cost prediction method applied to the smelting industry according to claim 6, characterized in that, Based on the forecasting strategy, the actual line is predicted for a future period of time, resulting in the predicted line, including... All parameter types are grouped into individual prediction groups and comprehensive prediction groups based on prediction strategies. For the parameter types of individual prediction groups, the first model and the second model are used to predict the stable and unstable parts respectively, and the prediction lines are obtained. For the parameter types of the comprehensive prediction group, a third model is used to comprehensively predict the stable and unstable parts to obtain the prediction line.

8. The cost prediction method applied to the metallurgical industry according to claim 1, characterized in that, The forecast line is used to predict smelting costs, including: Identify the normal and abnormal parts of the prediction line. For the normal part, predict the cost by matching the relationship between various aspects and costs. For the abnormal part, identify the abnormal causes, multiple additional cost types and the degree of abnormal impact in the abnormal part, predict the direct cost by matching the cost with multiple aspects, and determine the total cost based on the direct cost, abnormal causes, multiple additional cost types and the degree of abnormal impact. ； in, The total cost of the abnormal portion. For direct costs, This is an abnormal time. The quantity of additional costs due to abnormal causes. For the cost value of the j-th additional cost type, To determine the degree of abnormal impact, This represents the impact coefficient of abnormal costs. This represents the abnormal cost impact coefficient, which is obtained by mapping the degree of abnormal impact.

9. A cost prediction system applied to the metallurgical industry, characterized in that, The cost prediction method for performing any one of claims 1-8 includes, The first module is used to acquire the smelting process, define the various aspects involved in the smelting process, and set the standard line for each aspect based on the characteristics of the various aspects. The second module is used to acquire data on various aspects of the smelting process over a period of time, establish actual lines for each aspect, and compare the differences between the standard lines and actual lines for each aspect to distinguish the stable and unstable parts on the actual lines. The third module is used to select a prediction strategy based on the unstable part, and to make a prediction on the actual line for a future period of time according to the prediction strategy, so as to obtain the prediction line. The fourth module is used to determine the matching relationship between each aspect of the smelting process and its cost, and to combine this with the forecast line to predict smelting costs, thereby optimizing smelting cost management.

Images