Electric furnace production management system based on big data

Through big data technology, real-time analysis and prediction of electric furnace temperature fluctuations, combined with the nuclear density estimation method to identify outliers and dynamically adjust production parameters, the shortcomings of existing systems in temperature control and abnormal detection are solved, and the stability of electric furnace production is improved. Efficiency reduces costs and risks.

CN118938842AActive Publication Date: 2024-11-12DAZHOU HANGDA STEEL & IRON CO LTD

Patent Information

Application Number
CN202411226437.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-11-12
Estimated Expiration
2044-09-03

AI Technical Summary

Technical Problem

When dealing with high variability and complex production environments, the existing electric furnace production management system is difficult to accurately predict and adjust temperature fluctuations, resulting in reaction lag, unable to adjust production parameters in time, affecting production efficiency and product quality, and failing to effectively identify data. Implicit patterns and trends in it often lead to misjudgment or misjudgment, increasing the risk of production interruptions and equipment damage.

Method used

Using a big data-based electric furnace production management system, the temperature data is extracted from the electric furnace sensor through real-time temperature analysis module, a temperature control model is constructed, and the Gaussian process regression is used to predict future temperature fluctuations, and the statistical distribution of production parameters is analyzed in combination with the nuclear density estimation method. Adjust and optimize operating parameters and optimize electric furnace production in real time.

Benefits of technology

Achieve more precise temperature control, improve product quality and energy efficiency, effectively identify outliers, reduce production interruptions and equipment failure risks, enhance the stability and reliability of the production line, and ensure the continuity and efficiency of the steelmaking process. , reducing energy consumption and production costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938842A_ABST
    Figure CN118938842A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of data management, in particular to an electric furnace production management system based on big data, and the system comprises a real-time temperature analysis module which extracts the current and historical temperature data of an electric furnace from an electric furnace sensor, calculates the change trend data and time sequence attributes of the temperature of the electric furnace, and constructs a temperature control model; and generating an electric furnace temperature trend analysis result. In the invention, the Gaussian process regression is utilized to predict the temperature of the electric furnace, so that the temperature control is more accurate, the optimal temperature condition in the material processing process is favorably maintained, and the product quality and the energy efficiency are improved. Besides, production parameters such as statistical distribution of temperature, pressure and current are analyzed by adopting a kernel density estimation method, abnormal values are effectively identified from historical and real-time data, and production interruption and equipment failure risks caused by parameter deviation are reduced. And the stability and reliability of the production line are obviously enhanced. Through combination of historical adjustment records and real-time data analysis, an operation parameter adjustment strategy of the electric furnace is optimized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an electric furnace production management system based on big data. Background Art

[0002] The field of data management technology involves collecting, storing, protecting, verifying and processing data to ensure data accessibility, reliability and timeliness. The field uses a variety of technologies and systems, including database management systems, data warehouses, data lakes and various data processing frameworks. The process of transforming data from its original form to useful information is widely used in business intelligence, data analysis, information recovery and data-driven decision support systems. It helps organizations optimize their business processes and improve decision-making quality while ensuring data security and compliance.

[0003] Among them, the electric furnace production management system refers to a system used for short-process electric furnace steelmaking, which is used to monitor and control each stage of the steelmaking process to improve production efficiency and product quality. The system usually includes real-time data acquisition, processing and analysis functions to help operators adjust production parameters and respond to changes in production in real time, thereby ensuring the stability of the steelmaking process and optimizing output. Through the electric furnace production management system, steel mills can manage the production process more effectively, reduce energy consumption, reduce production costs, and improve the consistency and quality of the final product.

[0004] The existing electric furnace production management system has limitations when dealing with production environments with high variability and complexity. Especially in terms of response to temperature fluctuations and anomaly detection, it is difficult to accurately predict and adjust complex production variables, which often leads to delayed response and inability to adjust production parameters in a timely manner, affecting production efficiency and product quality. In addition, the shortcomings of traditional systems in anomaly detection, such as the failure to effectively identify implicit patterns and trends in data, often lead to misjudgment or omission of abnormal conditions, which not only increases the risk of production interruption, but may also cause serious equipment damage and safety accidents, which need to be improved. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose an electric furnace production management system based on big data.

[0006] In order to achieve the above purpose, the present invention adopts the following technical scheme: The electric furnace production management system based on big data includes:

[0007] The real-time temperature analysis module extracts the current and historical temperature data of the electric furnace from the electric furnace sensor, calculates the temperature change trend data and time series attributes of the electric furnace, builds a temperature control model, and generates the temperature trend analysis results of the electric furnace;

[0008] The Gaussian temperature prediction module uses the electric furnace temperature trend analysis result to predict the future temperature fluctuation of the electric furnace through Gaussian process regression, adjusts the heating power and the insulation time according to the prediction result, and obtains the adjusted electric furnace temperature data;

[0009] The production parameter monitoring module collects the pressure and current data of the electric furnace, combines the adjusted electric furnace temperature data, uses the kernel density estimation method to analyze the statistical distribution of the electric furnace temperature parameters, identifies potential deviations and evaluates the possibility of outliers, and generates anomaly detection results;

[0010] The dynamic adjustment optimization module combines the abnormal detection results and historical adjustment records to evaluate the current production line status, optimize the operating parameters of the electric furnace in real time, and obtain the electric furnace production optimization results.

[0011] As a further solution of the present invention, the steps for acquiring the change trend data and time series attributes of the electric furnace temperature are specifically as follows:

[0012] Extract current and historical temperature data from the electric furnace sensor, assuming that the electric furnace temperature data set is D =

[0013] {d1,d2,…,d n}, using the formula:

[0014]

[0015] Calculate the temperature change rate Δd at the i-th time point i , get the temperature change data;

[0016] Where i = 2 to n, d i represents the temperature at the i-th time point, t i represents the i-th time point, Δd i represents the temperature change rate at the i-th time point, λ is the attenuation factor, and e represents the natural exponent;

[0017] Based on the temperature change data, the formula is used:

[0018]

[0019] The changing trend data S at the i-th time point is calculated by combining the moving average method with the weighting factor i , get the changing trend data of the electric furnace temperature;

[0020] Among them, S i represents the change trend data at the i-th time point, i represents the time point, j represents the index of each time point in the time interval from i-k+1 to i, k represents the length of the average interval, and w j is the weighting factor;

[0021] Based on the change trend data of the electric furnace temperature and the electric furnace temperature data set D, the formula is adopted:

[0022] T i =α·d i +β·S i +γ·|d i -S i |

[0023] Calculate the time series attribute T of the i-th time point combining the original data and trend data i ;

[0024] Among them, T i represents the time series attribute of the ith time point combining the original data and trend data, α and β are weight coefficients, and γ is used to adjust the sensitivity to the difference between the two. i -S i |Represents the absolute difference between the original data and the trend data.

[0025] As a further solution of the present invention, the steps for obtaining the results of the electric furnace temperature trend analysis are specifically as follows:

[0026] Use the time series attribute T of the i-th time point i , construct the temperature control model, using the formula:

[0027] M i =η·T i +ξ

[0028] Analyze the relationship between time series attributes and temperature, and output the temperature trend value M i ;

[0029] Among them, M i represents the temperature trend value, i.e., the output of the model at time point i, η is the model coefficient, which is used to convert the time series attributes into a temperature control trend, and ξ is the intercept, which represents the baseline temperature level;

[0030] Based on the temperature trend value M i , using the formula:

[0031] A i =θ·M i

[0032] Output the dynamic changes of temperature trends by integrating the model, and output the furnace temperature trend analysis results A i ;

[0033] Here, θ is a scaling factor used to adjust the model output.

[0034] As a further solution of the present invention, the step of obtaining the predicted future temperature fluctuation of the electric furnace is specifically as follows:

[0035] Based on the electric furnace temperature trend analysis result A i , using the formula:

[0036]

[0037] Carry out furnace temperature prediction and obtain the future temperature fluctuation data P of the furnace i ;

[0038] Among them, P i is the future temperature fluctuation data of the electric furnace, t represents the predicted time point, τ represents the current time point, l represents the length parameter of time, σ represents the scale parameter of the Gaussian process, ω regulates the periodic fluctuation of temperature prediction, T represents the periodic duration, which is used to calculate the parameters of the periodic function, and the sin function is used to simulate the periodic change of temperature over time;

[0039] Based on the future temperature fluctuation data P of the electric furnace i , combined with the current heating demand, using the formula:

[0040] F i =P i +δ·(T target -T current ) 2

[0041] Adjust the prediction results to get the temperature adjustment value F i ;

[0042] Among them, F i is the temperature adjustment value, T target and T current represent the target and current temperatures respectively, and δ is the adjustment factor used to adjust according to the square of the temperature difference.

[0043] As a further solution of the present invention, the step of obtaining the adjusted electric furnace temperature data is specifically as follows:

[0044] Based on the future temperature fluctuation data P of the electric furnace i The obtained temperature adjustment value F i , using the formula:

[0045]

[0046] Evaluate the current heating power demand of the electric furnace and obtain the adjusted heating power P new ;

[0047] Among them, P new Represents the adjusted heating power, P currentrepresents the current heating power, κ is the sensitivity coefficient of adjusting the power according to the temperature difference, θ controls the curvature of the influence of the temperature difference on the power adjustment, T current is the current temperature;

[0048] Based on the adjusted heating power P new and temperature adjustment value F i , using the formula:

[0049]

[0050] Adjust the holding time and integrate the adjusted electric furnace temperature data P new and t new ;

[0051] Among them, t new represents the adjusted holding time, t current is the current insulation time, ρ is the coefficient for adjusting the time according to the temperature difference and power, T target is the target temperature.

[0052] As a further solution of the present invention, the step of obtaining the statistical distribution of the temperature parameters of the electric furnace is specifically as follows:

[0053] Obtain the adjusted electric furnace temperature data P new and t new , and collect the furnace pressure P and current I data to construct the initial data set D'={P,I,P new ,t new}, using the formula:

[0054]

[0055] Integrate the data of each electric furnace operation cycle i to generate the electric furnace operation data set D i ;

[0056] Among them, P i and I i represent the i-th observation value of pressure and current data respectively, and is the i-th observed value of the adjusted electric furnace temperature and holding time, i is the index, referring to the observation period;

[0057] Based on the electric furnace operation data set D i , using the formula:

[0058]

[0059] Calculate the data set D i Medium electric furnace temperature P new The kernel density of , and the kernel density analysis result f(x) is obtained;

[0060] Where f(x) is the result of kernel density analysis, K is the kernel function, h is the bandwidth parameter, α, β, γ, δ are weight coefficients, which respectively represent the contribution of furnace temperature, holding time, pressure and current data in kernel density estimation, n is the total number of data points; x is the variable, which is the point used to evaluate density;

[0061] Based on the kernel density analysis result f(x), the formula is adopted:

[0062] s(x)=∫(f(x)-μ) 2 dx

[0063] Calculate the discrete degree s(x) of the density function to reveal the statistical distribution of the furnace temperature parameters;

[0064] Among them, s(x) represents the degree of discreteness of the density function, μ is the mean of the estimated density function f(x), which represents the average probability density of all observations, and ∫ represents the integration operation, which is used to calculate the entire function (f(x)-μ) 2 The sum over all x, dx is the differential element, representing the infinitesimal change in x during the integration process.

[0065] As a further solution of the present invention, the step of obtaining the abnormality detection result is specifically:

[0066] The discrete degree S(x) of the density function is used as the threshold standard, and the density function f(x) of each data point is i ) for comparison, according to the formula:

[0067]

[0068] Calculate the deviation D of the i-th data point i ';

[0069] Among them, f(x i ) is the density value of the i-th data point in the kernel density estimation, μ f is the average value of the density of all data points, β adjusts the sensitivity of the deviation calculation, D i 'Indicates the deviation of the i-th data point;

[0070] Based on the deviation D of the i-th data point i ', using the formula:

[0071]

[0072] Calculate the ratio of the deviation of each data point to the variance of the density function to obtain the abnormality E of the i-th data point i ;

[0073] Where π is the ratio of the circumference of a circle to the tangent -1 The results are normalized to the interval [0,1], E i Represents the abnormality of the i-th data point;

[0074] Based on the abnormality E of the i-th data point i , using the formula:

[0075] M={x i |E i >α'·γ}

[0076] By setting the anomaly threshold α', data points with an anomaly higher than α' are screened and marked as potential deviations or outliers, generating anomaly detection results M;

[0077] Among them, α' is the preset abnormality threshold, γ is the coefficient of adjusting the threshold, and x i is the original data point, and M is the anomaly detection result, that is, the set of outliers.

[0078] As a further solution of the present invention, the steps for obtaining the electric furnace production optimization result are specifically as follows:

[0079] Collect historical adjustment records H and integrate them with the anomaly detection results M according to the formula:

[0080]

[0081] Evaluate the operating status of the current production line and calculate the production line status index SI after evaluation;

[0082] Among them, n represents the number of data points involved in the calculation, M i Represents the value of the abnormal data point, H i represents the value of the historical data point, and SI represents the production line status index after evaluation;

[0083] According to the production line status index SI after the evaluation, according to the formula:

[0084] R(t)=SI×k+(1-k)×Rprev

[0085] Adjust the operating parameters of the electric furnace in real time to obtain the adjustment function R(t);

[0086] Among them, k is used to adjust the influence of real-time feedback, Rprev is the operating parameter of the previous cycle, and R(t) is the adjustment function, that is, the adjusted operating parameter of the current cycle;

[0087] Based on the adjustment function R(t), the formula is adopted:

[0088]

[0089] Perform optimization operations to obtain the optimization result function O(t) and generate the optimization result of electric furnace production;

[0090] Among them, μ R is the average value of R(t), which is used to standardize the comparison between the current operating parameters and the historical performance. O(t represents the optimization result function, e R (t) is used to reflect the impact of the increase in operating parameter R(t) on production results, 1+ is an adjustment term used to adjust the current operating parameter R(t) and its average μ R for comparison.

[0091] Compared with the prior art, the advantages and positive effects of the present invention are:

[0092] In the present invention, Gaussian process regression is used to predict the temperature of the electric furnace, making the temperature control more accurate, which helps to maintain the optimal temperature conditions during material processing, thereby improving product quality and energy efficiency. In addition, the kernel density estimation method is used to analyze the statistical distribution of production parameters such as temperature, pressure and current, effectively identifying outliers from historical and real-time data, reducing the risk of production interruptions and equipment failures caused by parameter deviations. The stability and reliability of the production line are significantly enhanced. By combining historical adjustment records and real-time data analysis, the operation parameter adjustment strategy of the electric furnace is optimized, ensuring the continuity and efficiency of the steelmaking process, while reducing energy consumption and production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 is a system flow chart of the present invention;

[0094] Figure 2 A flow chart of the steps for obtaining the change trend data and time series attributes of the electric furnace temperature of the present invention;

[0095] Figure 3 A flow chart of the steps for obtaining the temperature trend analysis results of an electric furnace according to the present invention;

[0096] Figure 4 A flow chart of the steps for obtaining the prediction of the future temperature fluctuation of the electric furnace according to the present invention;

[0097] Figure 5 A flow chart of the steps for obtaining the adjusted electric furnace temperature data of the present invention;

[0098] Figure 6 A flow chart of the acquisition steps for analyzing the statistical distribution of the temperature parameters of an electric furnace according to the present invention;

[0099] Figure 7 A flowchart of the steps for obtaining abnormality detection results of the present invention;

[0100] Figure 8The present invention is a flow chart of the steps for obtaining the electric furnace production optimization results. DETAILED DESCRIPTION

[0101] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0102] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0103] See also Figure 1 , the electric furnace production management system based on big data includes:

[0104] The real-time temperature analysis module extracts the current and historical temperature data of the electric furnace from the electric furnace sensor, calculates the temperature change trend data and time series attributes of the electric furnace, builds a temperature control model, and generates the temperature trend analysis results of the electric furnace;

[0105] The Gaussian temperature prediction module uses the furnace temperature trend analysis results to predict the future furnace temperature fluctuations through Gaussian process regression, adjusts the heating power and insulation time according to the prediction results, and obtains the adjusted furnace temperature data;

[0106] The production parameter monitoring module collects the pressure and current data of the electric furnace, combines it with the adjusted temperature data of the electric furnace, uses the kernel density estimation method to analyze the statistical distribution of the temperature parameters of the electric furnace, identifies potential deviations and evaluates the possibility of outliers, and generates anomaly detection results;

[0107] The dynamic adjustment and optimization module combines the abnormal detection results and historical adjustment records to evaluate the current production line status, optimize the operating parameters of the electric furnace in real time, and obtain the electric furnace production optimization results.

[0108] The results of the electric furnace temperature trend analysis include the comparison of the high and low temperature points in the history of the electric furnace, the temperature values ​​at key time points and the periodic temperature change characteristics. The adjusted electric furnace temperature data include the maximum temperature, minimum temperature and temperature fluctuation range at future time points. The anomaly detection results include the statistical number of temperature anomalies, the duration of anomalies and their potential impact on the production process. The electric furnace production optimization results include the optimal heating rate, insulation time and expected energy-saving effect adjusted based on real-time data.

[0109] See also Figure 2 , the specific steps for obtaining the trend data and time series attributes of the electric furnace temperature are as follows:

[0110] Extract current and historical temperature data from the electric furnace sensor, assuming that the electric furnace temperature data set is D =

[0111] {d1,d2,…,d n}, using the formula:

[0112]

[0113] Calculate the temperature change rate Δd at the i-th time point i , get the temperature change data;

[0114] Where i = 2 to n, d i represents the temperature at the i-th time point, t i represents the i-th time point, Δd i represents the temperature change rate at the i-th time point, λ is the attenuation factor used to adjust the sensitivity of temperature change, and e represents the natural exponent used to modulate the nonlinear effect of the difference;

[0115] Based on the temperature change data, the formula is used:

[0116]

[0117] The changing trend data S at the i-th time point is calculated by combining the moving average method with the weighting factor i , get the changing trend data of the electric furnace temperature;

[0118] Among them, S i represents the change trend data at the i-th time point, i represents the time point, j represents the index of each time point in the time interval from i-k+1 to i, k represents the length of the average interval, and w j is a weighting factor that reflects the relative importance of each time point;

[0119] Based on the changing trend data of the electric furnace temperature and the electric furnace temperature data set D, the formula is used:

[0120] T i =α·di +β·S i +γ·|d i -S i |

[0121] Calculate the time series attribute T of the i-th time point combining the original data and trend data i ;

[0122] Among them, T i represents the time series attribute of the i-th time point combining the original data and trend data, α and β are weight coefficients used to balance the original temperature data d i and trend data S i The influence of γ strengthens the sensitivity to the difference between the two, |d i -S i |Represents the absolute difference between the original data and the trend data, used to highlight abnormalities or drastic changes.

[0123] The calculation process is as follows:

[0124] Assume that the temperature data D obtained from the electric furnace sensor is D = {1500, 1520, 1535, 1550, 1570}, and the time point t = {1, 2, 3, 4, 5}.

[0125] For the formula:

[0126]

[0127] Where λ is the attenuation factor, set to 0.05. This factor helps adjust for the effects of rapid temperature changes.

[0128] For d2=1520, d1=1500, t2=2, t1=1:

[0129]

[0130] Δd3, Δd4, and Δd5 are calculated similarly.

[0131] As a result, the temperature change rate array ΔD={7.36, 7.21, 6.86, 7.61} is obtained.

[0132] The calculation of short-term trends sets the averaging interval k = 2 and the time weighting factor w = {0.5, 1}.

[0133] For the formula:

[0134]

[0135] Where i=3, j is 2 and 3, and the weighting factor w={0.5,1}:

[0136] S4 and S5 are calculated similarly.

[0137] As a result, we get the short-term trend array S = {7.26, 7.11, 7.52}.

[0138] Calculation of Time Series Properties Using previous data and assumed model parameters, time series properties can be calculated.

[0139] For the formula:

[0140] T i =α·d i +β·S i +γ·|d i -S i |

[0141] Assume α=0.6, β=0.3, γ=0.1.

[0142] Among them, d3=1535, S3=7.26:

[0143] T3=

[0144] 0.6·1535+0.3·7.26+0.1·|1535-7.26|=921+2.18+152.77≈1075.95

[0145] As a result, the time series attribute array T = {1075.95,...} is obtained.

[0146] See also Figure 3 , the specific steps for obtaining the results of the electric furnace temperature trend analysis are:

[0147] The time series attribute T at the i-th time point i , construct the temperature control model, using the formula:

[0148] M i =η·T i +ξ

[0149] Analyze the relationship between time series attributes and temperature, and output the temperature trend value M i ;

[0150] Among them, M i represents the temperature trend value, i.e., the output of the model at time point i, η is the model coefficient, which is used to convert the time series attributes into a temperature control trend, and ξ is the intercept, which represents the baseline temperature level;

[0151] Based on the temperature trend value M i , using the formula:

[0152] A i=θ·M i

[0153] Output the dynamic changes of temperature trends by integrating the model, and output the furnace temperature trend analysis results A i ;

[0154] Where θ is a scaling factor used to adjust the model output to match the current temperature control requirements.

[0155] The calculation process is as follows:

[0156] For the formula:

[0157] M i =η·T i +ξ

[0158] The time series attribute T i =1075.95 is input into this model. η is the model coefficient, which is assumed to be obtained through historical data analysis and is set to 0.95 (this value is assumed to be the standard conversion factor obtained in past electric furnace production). ξ is the intercept, which represents the baseline temperature level and is set to 20 (this value is assumed to be based on the basic temperature state of the electric furnace without material or operation).

[0159] Calculate the model output M3:

[0160] M3=0.95·1075.95+20M3=1022.1525+20M3=1042.1525

[0161] The result M3=1042.1525 indicates that under the current input conditions, the control model of the electric furnace expects the output temperature trend.

[0162] For the formula:

[0163] A i =θ·M i

[0164] θ—scaling factor, adjusted according to the actual temperature control requirements of the electric furnace, set to 1 (this value is assumed to be the standardized output and directly reflects the output of the model as the final trend analysis result).

[0165] Generate the electric furnace temperature trend analysis result A3:

[0166] A3=1·1042.1525A3=1042.1525

[0167] The result A3=1042.1525 represents the final temperature trend analysis result of the electric furnace temperature control model under the given time series attributes, which can be directly used to judge and adjust the temperature setting of the electric furnace.

[0168] See also Figure 4 , the specific steps for obtaining the predicted future temperature fluctuations of the electric furnace are:

[0169] Based on the results of electric furnace temperature trend analysis A i , using the formula:

[0170]

[0171] Carry out furnace temperature prediction and obtain the future temperature fluctuation data P of the furnace i ;

[0172] Among them, t represents the prediction time point, τ represents the current time point, which is used to calculate the square of the time interval and is reflected in the prediction in the form of an exponential; l is the length parameter of time, which controls the impact of the time interval on the prediction; σ is the scale parameter of the Gaussian process, which determines the amplitude of the change of the entire prediction; ω adjusts the periodic fluctuation of the temperature prediction so that the model can adapt to the periodic temperature changes; T represents the duration of this periodicity, which is used to calculate the parameters of the periodic function; the sin function is used to simulate the periodic changes of temperature over time;

[0173] Based on the future temperature fluctuation data of the electric furnace P i , combined with the current heating demand, using the formula:

[0174] F i =P i +δ·(T target -T current ) 2

[0175] Adjust the prediction results to get the temperature adjustment value F i ;

[0176] Among them, T target and T current represent the target and current temperatures respectively, and δ is the adjustment factor used to make fine adjustments based on the square of the temperature difference, thereby more accurately controlling the sensitivity of the temperature adjustment.

[0177] The calculation process is as follows:

[0178] For the formula:

[0179]

[0180] Where t is the target time point for prediction, assuming t = 5 hours. τ is the current time point, assuming τ = 4 hours. l is the time length parameter, which controls the effect of the time interval on the prediction, assuming l = 1 hour. σ is the scale parameter of the Gaussian process, which determines the amplitude of the entire prediction, assuming σ = 1.0. ω is the period adjustment coefficient, which enables the model to adapt to periodic temperature changes, assuming ω = 0.5. T is the period length, which indicates the duration of periodic changes, assuming T = 24 hours. A i is the current temperature trend analysis result. Given A from the previous analysis i =1042.1525.

[0181] Compute the square of the time interval:

[0182] time_diff = (t-τ) 2 =(5-4) 2 =1

[0183] Calculate the exponential part:

[0184]

[0185] Compute the sine part:

[0186]

[0187] Calculate P i :

[0188] P i =σ·exp_part·(A i +sin_part)=1.0·0.606·(1042.1525+0.25)≈631.5

[0189] For the formula:

[0190] F i =P i +δ·(T target -T current ) 2

[0191] Among them, P i is the result from step 1, P i ≈631.5. T target is the target temperature, assuming it is 1500 degrees. current is the current actual temperature, assuming it is 1450 degrees. δ is the adjustment factor, which is fine-tuned according to the square of the temperature difference, assuming δ = 0.1.

[0192] Calculate the square of the temperature difference:

[0193] temp_diff_sq=(T target -T current ) 2 =(1500-1450) 2 =2500

[0194] Calculate F i :

[0195] F i =P i +δ·temp_diff_sq=631.5+0.1·2500=631.5+250=881.5

[0196] P i Indicates the temperature fluctuation during the forecast, which is calculated to be about 631.5. F i is the final temperature adjustment value, which is about 881.5, indicating that after prediction, the heating power needs to be adjusted according to actual demand to reach the target temperature of 1500 degrees.

[0197] See also Figure 5 , the specific steps for obtaining the adjusted electric furnace temperature data are:

[0198] Based on the future temperature fluctuation data of the electric furnace P i The obtained temperature adjustment value F i , using the formula:

[0199]

[0200] Evaluate the current heating power demand of the electric furnace and obtain the adjusted heating power P new ;

[0201] Among them, P new Represents the adjusted heating power, P current represents the current heating power, κ is the sensitivity coefficient for adjusting the power according to the temperature difference, θ controls the curvature of the influence of the temperature difference on the power adjustment, ensuring faster adjustment when the temperature difference is large, T current is the current actual temperature.

[0202] Based on the adjusted heating power P new and temperature adjustment value F i , using the formula:

[0203]

[0204] Adjust the holding time and integrate the adjusted electric furnace temperature data P new and t new ;

[0205] Among them, t new represents the adjusted holding time, tcurrent is the current insulation time, ρ is the coefficient for adjusting the time according to the temperature difference and power, T target is the target temperature, ensuring that the adjustment time reacts to the appropriate sensitivity and responds quickly to the ideal state.

[0206] The calculation process is as follows:

[0207] For the formula:

[0208]

[0209] F i =881.5, assuming P current = 100 kW is the current heating power, based on the common settings of a typical medium-sized electric furnace. current =1450 degrees is the current actual temperature, which is consistent with the common operating temperature of steel smelting. κ=0.05 is the power adjustment coefficient, ensuring that the adjustment is not too drastic. θ=0.1 is the parameter for adjusting the temperature difference sensitivity, providing smooth adjustment dynamics.

[0210] Calculation of temperature difference: ΔT=F i -T current =881.5-1450=-568.5

[0211] Calculation of logarithmic logic function:

[0212] logistic_part=1+e -θ·ΔT =1+e -0.1·(-568.5) ≈1+e 56.85

[0213] e 56.85 The value of is very large. In practical applications, it should be considered that this value may cause numerical overflow. Here, a large number is used to approximate it, such as 10 20 .

[0214] Calculation of the new heating power:

[0215]

[0216] Since the logarithmic logic part is very large, the influence is minimized, resulting in practically no change in the heating power.

[0217] For the formula:

[0218]

[0219] Assume t current =30 minutes, is the current holding time, based on the standard operating cycle. ρ=1.5 is the time adjustment factor, ensuring that the time adjustment is appropriate. T target =1500 degrees is the target temperature.new = 100 kW is the result from step 1.

[0220] Calculation of the absolute value of the target temperature difference:

[0221] temp_diff=|T target -F i |=|1500-881.5|=618.5

[0222] Adjustment calculation of holding time:

[0223]

[0224] Therefore, the adjusted holding time is approximately 42 minutes.

[0225] Conclusion: P new Maintaining approximately 100 kW, the logarithmic logic function effectively prevents over-adjustment when the temperature difference is large. new It is about 42 minutes, which means that according to the new temperature prediction and the current power, the insulation time needs to be increased to reach the target temperature.

[0226] See also Figure 6 , the steps for obtaining the statistical distribution of the temperature parameters of the electric furnace are as follows:

[0227] Get the adjusted electric furnace temperature data P new and t new , and collect the furnace pressure P and current I data to construct the initial data set D'={P,I,P new ,t new}, using the formula:

[0228]

[0229] Integrate the data of each electric furnace operation cycle i to generate the electric furnace operation data set D i ;

[0230] Among them, P i and I i represent the i-th observation value of pressure and current data respectively, and is the i-th observed value of the adjusted electric furnace temperature and holding time, where i is an index indicating a specific observation period;

[0231] Based on the electric furnace operation data set D i , using the formula:

[0232]

[0233] Calculate the data set D i Medium electric furnace temperature P newThe kernel density of , and the kernel density analysis result f(x) is obtained;

[0234] Among them, K is the kernel function, which is used to smooth the data points; h is the bandwidth parameter, which controls the width of the kernel function and affects the smoothness of the estimate; α, β, γ, δ are weight coefficients, which respectively represent the contribution of the furnace temperature, holding time, pressure and current data in the kernel density estimation; n is the total number of data points, which indicates the number of data points involved in the kernel density estimation; x is the variable, which is used to evaluate the density point;

[0235] Based on the kernel density analysis result f(x), the formula is used:

[0236] s(x)=∫(f(x)-μ) 2 dx

[0237] Calculate the discrete degree s(x) of the density function to reveal the statistical distribution of the furnace temperature parameters;

[0238] Among them, s(x) represents the degree of dispersion (variance) of the density function, which is used to measure the central tendency and dispersion tendency in the statistical distribution. μ is the mean of the estimated density function f(x), which represents the average probability density of all observations. This analysis helps to identify abnormal fluctuations and deviations in the temperature distribution. ∫ represents the integration operation, which is used to calculate the entire function (f(x)-μ) 2 The sum over all x's gives the total variance. dx is the differential element, which represents the infinitesimal change in x during the integration.

[0239] The calculation process is as follows:

[0240] Suppose there are data for three different electric furnace operation cycles:

[0241] Cycle 1: Pressure P1 = 150 psi, Current I1 = 100 A, Adjusted Temperature Insulation time minute.

[0242] Cycle 2: Pressure P2 = 155 psi, Current I2 = 105 A, Adjusted Temperature Insulation time minute.

[0243] Cycle 3: Pressure P3 = 152 psi, Current I3 = 103 A, Adjusted Temperature Insulation time minute.

[0244] Kernel density estimation calculation:

[0245] A standard Gaussian kernel function K is selected as the smoothing kernel function, and the bandwidth h=10 is assumed.

[0246] The specific calculation of formula f(x) is as follows:

[0247]

[0248] Gaussian kernel function K:

[0249]

[0250] Assuming the calculation point x = 1400 (target temperature), P new ,t new ,P, and I are calculated by kernel density estimation:

[0251]

[0252] For each P i , I i Calculate similar values. Substitute these values ​​into f(x) and calculate:

[0253]

[0254] (The specific "a" is calculated based on the actual situation of each parameter)

[0255] The result of f(1400) represents the kernel density estimate of the furnace temperature at x = 1400°C.

[0256] Assume that there is a kernel density estimate of f(x) at x = 1400 ° C, f(1400) ≈ 0.182 (this is an assumed value, which should actually be calculated in the above step 2), and calculate the variance of the density function:

[0257] s(x)=∫(f(x)-μ) 2 dx

[0258] The variance calculation can be handled by a discrete approximation:

[0259] Assume that μ(the mean of f(x)) is the simple average of all values ​​of f(x).

[0260] Further assume that there are estimated values ​​of f(x) at x=1390°C, x=1400°C, and x=1410°C (also hypothetical values): f(1390)=0.150, f(1410)=0.160.

[0261] First calculate the mean μ:

[0262]

[0263] Then calculate the variance s(x):

[0264]

[0265]

[0266] The result of s(x) represents the variance of the kernel density estimate of the electric furnace temperature within the analyzed temperature range. It helps to understand the degree of dispersion of the temperature parameters and thus evaluate the stability and abnormal fluctuations of temperature control.

[0267] See also Figure 7 , the specific steps for obtaining anomaly detection results are:

[0268] The discrete degree S(x) of the density function is used as the threshold standard, and the density function f(x) of each data point is i ) for comparison, according to the formula:

[0269]

[0270] Calculate the deviation D of the i-th data point i ';

[0271] Among them, f(x i ) is the density value of the i-th data point in the kernel density estimation, μ f is the average value of the density of all data points, β adjusts the sensitivity of the deviation calculation, D i 'Indicates the deviation of the i-th data point;

[0272] The deviation D based on the i-th data point i ', using the formula:

[0273]

[0274] Calculate the ratio of the deviation of each data point to the variance of the density function to obtain the abnormality E of the i-th data point i ;

[0275] Where π is the ratio of the circumference of a circle to the tangent -1 The results are normalized to the interval [0,1], E i Represents the abnormality of the i-th data point. The closer the value is to 1, the more likely the point is an outlier.

[0276] Based on the abnormality E of the i-th data point i , using the formula:

[0277] M={x i |E i >α'·γ}

[0278] By setting the anomaly threshold α', data points with an anomaly higher than α' are screened and marked as potential deviations or outliers, generating anomaly detection results M;

[0279] Among them, α' is the preset abnormality threshold, γ is the coefficient of adjusting the threshold, and x i is the original data point, and M is the anomaly detection result, that is, the set of outliers.

[0280] The calculation process is as follows:

[0281] For the formula:

[0282]

[0283] Suppose there are the following data points and their kernel density estimates:

[0284] Data points: x1=100, x2=150, x3=200;

[0285] Density values: f(x1) = 0.2, f(x2) = 0.5, f(x3) = 0.3;

[0286] Assuming the average density μ f =(0.2+0.5+0.3) / 3=0.3333;

[0287] The variance of the density function is assumed to be S(x) = 0.02;

[0288] Balance coefficient β = 1.5;

[0289] Calculate the deviation D1' of x1:

[0290]

[0291] The calculated deviation D1'≈12.17 indicates that the density value of x1 has a large deviation compared to the average density.

[0292] For the formula:

[0293]

[0294] Use the deviation of x1 D1'≈12.17;

[0295]

[0296] Calculation results show that: E1≈0.474 indicates that x1 is high in the anomaly score, and the closer it is to 1, the more likely it is an outlier.

[0297] For the formula:

[0298] A={x i |E i >α'·γ}

[0299] Assume α'=0.5,γ=1, then;

[0300] A={x1|0.474>0.5·1}

[0301] Since 0.474 is not greater than 0.5, x1 is not marked as an anomaly.

[0302] It indicates that under the given threshold setting, no data point exceeds the abnormality standard.

[0303] See also Figure 8 , the specific steps for obtaining the optimization results of electric furnace production are:

[0304] Collect historical adjustment records H and integrate them with the anomaly detection results M according to the formula:

[0305]

[0306] Evaluate the operating status of the current production line and calculate the production line status index SI after evaluation;

[0307] Among them, n represents the number of data points involved in the calculation, M i Represents the value of the abnormal data point, H i Represents the value of the historical data point, SI represents the production line status index after comprehensive evaluation;

[0308] According to the evaluated production line status index SI, according to the formula:

[0309] R(t)=SI×k+(1-k)×R prev

[0310] Adjust the operating parameters of the electric furnace in real time to obtain the adjustment function R(t);

[0311] Among them, k is used to adjust the influence of real-time feedback, R prev is the operating parameter of the previous cycle, and R(t) is the adjustment function, that is, the adjusted operating parameter of the current cycle.

[0312] Based on the adjustment function R(t), the formula is adopted:

[0313]

[0314] Perform optimization operations to obtain the optimization result function O(t) and generate the optimization result of electric furnace production;

[0315] Among them, μ R is the average value of R(t), which is used to standardize the comparison between the current operating parameters and the historical performance. O(t) represents the optimized electric furnace production results. R(t) It reflects the impact of the increase of operating parameter R(t) on production results. is an adjustment term that adjusts the current operating parameter R(t) to its long-term average μ RCompare and adjust the sensitivity of the response in this way. When R(t) is greater than μ R When R(t) is less than or equal to μ, the denominator becomes larger, making the entire function value O(t) tend to 1, indicating that the production state is close to optimization; when R(t) is less than or equal to μ R , the function value O(t) decreases, indicating that further optimization is needed.

[0316] The calculation process is as follows:

[0317] For the formula:

[0318]

[0319] Assume there are 3 data points, the anomaly detection results M and historical adjustment records H are as follows:

[0320] M=[0.9,0.8,0.95]H=[0.85,0.9,0.9]

[0321] The calculation process is as follows:

[0322] |M1-H1|=|0.9-0.85|=0.05

[0323] |M2-H2|=|0.8-0.9|=0.1

[0324] |M3-H3|=|0.95-0.9|=0.05

[0325]

[0326] The calculated status index SI≈0.933 shows that the current production line status is very close to the historical record, indicating that the adjustment strategy maintains good continuity and stability.

[0327] For the formula:

[0328] R(t)=SI×k+(1-k)×R prev

[0329] Assume SI = 0.933. k = 0.5 (adjusted sensitivity coefficient to medium). R prev =50 (parameter value of the previous cycle).

[0330] R(t)=0.933×0.5+(1-0.5)×50=0.4665+25=25.4665

[0331] The obtained R(t)=25.4665 indicates that the electric furnace operation parameters of the current cycle are adjusted on the basis of the previous settings and the current status evaluation to ensure the continuity and stability of production.

[0332] For the formula:

[0333]

[0334] R(t)=25.4665. R =30 (long term average).

[0335]

[0336] Finally, O(t)≈1.393×10 11 The indicated production output after optimization is very high, reflecting a significant improvement in production efficiency.

[0337] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The electric furnace production management system based on big data is characterized by: The system comprises: The real-time temperature analysis module extracts the current and historical temperature data of the electric furnace from the electric furnace sensor, calculates the temperature change trend data and time series attributes of the electric furnace, builds a temperature control model, and generates the temperature trend analysis results of the electric furnace; The Gaussian temperature prediction module uses the electric furnace temperature trend analysis result to predict the future temperature fluctuation of the electric furnace through Gaussian process regression, adjusts the heating power and the insulation time according to the prediction result, and obtains the adjusted electric furnace temperature data; The production parameter monitoring module collects the pressure and current data of the electric furnace, combines the adjusted electric furnace temperature data, uses the kernel density estimation method to analyze the statistical distribution of the electric furnace temperature parameters, identifies potential deviations and evaluates the possibility of outliers, and generates anomaly detection results; The dynamic adjustment optimization module combines the abnormal detection results and historical adjustment records to evaluate the current production line status, optimize the operating parameters of the electric furnace in real time, and obtain the electric furnace production optimization results.

2. The electric furnace production management system based on big data according to claim 1 is characterized in that: The steps for obtaining the change trend data and time series attributes of the electric furnace temperature are specifically as follows: Extract current and historical temperature data from the electric furnace sensor, assuming that the electric furnace temperature data set is D = {d1,d2,…,d n }, using the formula: Calculate the temperature change rate Δd at the i-th time point i , get the temperature change data; Where i = 2 to n, d i represents the temperature at the i-th time point, t i represents the i-th time point, Δd i represents the temperature change rate at the i-th time point, λ is the attenuation factor, and e represents the natural exponent; Based on the temperature change data, the formula is used: The changing trend data S at the i-th time point is calculated by combining the moving average method with the weighting factor i , get the changing trend data of the electric furnace temperature; Among them, S i represents the change trend data at the i-th time point, i represents the time point, j represents the index of each time point in the time interval from i-k+1 to i, k represents the length of the average interval, and w j is the weighting factor; Based on the change trend data of the electric furnace temperature and the electric furnace temperature data set D, the formula is adopted: T i =α·d i +β·S i +γ·|d i -S i | Calculate the time series attribute T of the i-th time point combining the original data and trend data i ; Among them, T i represents the time series attribute of the ith time point combining the original data and trend data, α and β are weight coefficients, and γ is used to adjust the sensitivity to the difference between the two. i -S i |Represents the absolute difference between the original data and the trend data.

3. The electric furnace production management system based on big data according to claim 2 is characterized in that: The steps for obtaining the electric furnace temperature trend analysis result are specifically as follows: Use the time series attribute T of the i-th time point i , construct the temperature control model, using the formula: M i =η·T i +ξ Analyze the relationship between time series attributes and temperature, and output the temperature trend value M i ; Among them, M i represents the temperature trend value, i.e., the output of the model at time point i, η is the model coefficient, which is used to convert the time series attributes into a temperature control trend, and ξ is the intercept, which represents the baseline temperature level; Based on the temperature trend value M i , using the formula: A i =θ·M i Output the dynamic changes of temperature trends by integrating the model, and output the furnace temperature trend analysis results A i ; Here, θ is a scaling factor used to adjust the model output.

4. The electric furnace production management system based on big data according to claim 3 is characterized in that: The steps for obtaining the predicted future temperature fluctuation of the electric furnace are specifically as follows: Based on the electric furnace temperature trend analysis result A i , using the formula: Carry out furnace temperature prediction and obtain the future temperature fluctuation data P of the furnace i ; Among them, P i is the future temperature fluctuation data of the electric furnace, t represents the predicted time point, τ represents the current time point, l represents the length parameter of time, σ represents the scale parameter of the Gaussian process, ω regulates the periodic fluctuation of temperature prediction, T represents the periodic duration, which is used to calculate the parameters of the periodic function, and the sin function is used to simulate the periodic change of temperature over time; Based on the future temperature fluctuation data P of the electric furnace i , combined with the current heating demand, using the formula: F i =P i +δ·(T target -T current ) 2 Adjust the prediction results to get the temperature adjustment value F i ; Among them, F i is the temperature adjustment value, T target and T current represent the target and current temperatures respectively, and δ is the adjustment factor used to adjust according to the square of the temperature difference.

5. The electric furnace production management system based on big data according to claim 4 is characterized in that: The steps for obtaining the adjusted electric furnace temperature data are specifically as follows: Based on the future temperature fluctuation data P of the electric furnace i The obtained temperature adjustment value F i , using the formula: Evaluate the current heating power demand of the electric furnace and obtain the adjusted heating power P new ; Among them, P new Represents the adjusted heating power, P current represents the current heating power, κ is the sensitivity coefficient of adjusting the power according to the temperature difference, θ controls the curvature of the influence of the temperature difference on the power adjustment, T current is the current temperature; Based on the adjusted heating power P new and temperature adjustment value F i , using the formula: Adjust the holding time and integrate the adjusted electric furnace temperature data P new and t new ; Among them, t new represents the adjusted holding time, t current is the current insulation time, ρ is the coefficient for adjusting the time according to the temperature difference and power, T target is the target temperature.

6. The electric furnace production management system based on big data according to claim 5 is characterized in that: The steps for obtaining the statistical distribution of the temperature parameters of the electric furnace are specifically as follows: Obtain the adjusted electric furnace temperature data P new and t new , and collect the furnace pressure P and current I data to construct the initial data set D'={P,I,P new ,t new }, using the formula: The data of each electric furnace operation cycle i is integrated to generate the electric furnace operation data set D i ; Among them, P i and I i represent the i-th observation value of pressure and current data respectively, and is the i-th observed value of the adjusted electric furnace temperature and holding time, i is the index, referring to the observation period; Based on the electric furnace operation data set D i , using the formula: Calculate the data set D i Medium electric furnace temperature P new The kernel density of , and the kernel density analysis result f(x) is obtained; Where f(x) is the result of kernel density analysis, K is the kernel function, h is the bandwidth parameter, α, β, γ, δ are weight coefficients, which respectively represent the contribution of furnace temperature, holding time, pressure and current data in kernel density estimation, n is the total number of data points; x is the variable, which is the point used to evaluate density; Based on the kernel density analysis result f(x), the formula is adopted: s(x)=∫(f(x)-μ) 2 dx Calculate the discrete degree s(x) of the density function to reveal the statistical distribution of the furnace temperature parameters; Among them, s(x) represents the degree of discreteness of the density function, μ is the mean of the estimated density function f(x), which represents the average probability density of all observations, and ∫ represents the integration operation, which is used to calculate the entire function (f(x)-μ) 2 The sum over all x, dx is the differential element, representing the infinitesimal change in x during the integration process.

7. The electric furnace production management system based on big data according to claim 6 is characterized in that: The steps for obtaining the abnormal detection result are specifically as follows: The discrete degree S(x) of the density function is used as the threshold standard, and the density function f(x) of each data point is i ) for comparison, according to the formula: Calculate the deviation D of the i-th data point i '; Among them, f(x i ) is the density value of the i-th data point in the kernel density estimation, μ f is the average value of the density of all data points, β adjusts the sensitivity of the deviation calculation, D i 'Indicates the deviation of the i-th data point; Based on the deviation D of the i-th data point i ', using the formula: Calculate the ratio of the deviation of each data point to the variance of the density function to obtain the abnormality E of the i-th data point i ; Where π is the ratio of the circumference of a circle to the tangent -1 The results are normalized to the interval [0,1], E i Represents the abnormality of the i-th data point; Based on the abnormality E of the i-th data point i , using the formula: M={x i |E i >a'·c} By setting the anomaly threshold α', data points with an anomaly higher than α' are screened and marked as potential deviations or outliers, generating anomaly detection results M; Among them, α' is the preset abnormality threshold, γ is the coefficient of adjusting the threshold, and x i is the original data point, and M is the anomaly detection result, that is, the set of outliers.

8. The electric furnace production management system based on big data according to claim 7 is characterized in that: The steps for obtaining the electric furnace production optimization result are specifically as follows: Collect historical adjustment records H and integrate them with the anomaly detection results M according to the formula: Evaluate the operating status of the current production line and calculate the production line status index SI after evaluation; Among them, n represents the number of data points involved in the calculation, M i Represents the value of the abnormal data point, H i represents the value of the historical data point, and SI represents the production line status index after evaluation; According to the production line status index SI after the evaluation, according to the formula: R(t)=SI×k+(1-k)×R prev Adjust the operating parameters of the electric furnace in real time to obtain the adjustment function R(t); Among them, k is used to adjust the influence of real-time feedback, R prev is the operating parameter of the previous cycle, R(t) is the adjustment function, i.e. the adjusted operating parameter of the current cycle; Based on the adjustment function R(t), the formula is adopted: Perform optimization operations to obtain the optimization result function O(t) and generate the optimization result of electric furnace production; Among them, μ R is the average value of R(t), which is used to standardize the comparison between the current operating parameters and the historical performance. O(t represents the optimization result function, e R (t) is used to reflect the impact of the increase in operating parameter R(t) on production results. is an adjustment term used to adjust the current operating parameter R(t) and its average μ R for comparison.

Citation Information

Patent Citations

  • Hearth thermal state trend prediction method based on time sequence and multiple dimensions of blast furnace

    CN110427715A

  • Anomaly detection method for train traction transformer cooling system

    CN113065219A

  • Wall temperature anomaly identification method and system based on thermal power station furnace tube virtual partition

    CN116398868A

  • Coal mine fully-mechanized mining intelligent management system based on perceptual interaction technology

    CN117449910A

  • Time sequence trend index calculation method and system based on power consumption data

    CN117574175A

Cited By

  • Pre-oxidation furnace temperature real-time monitoring system and method based on cloud platform

    CN119334499A

  • Self-adaptive temperature regulation and control method for heat treatment furnace

    CN119614852A

  • Intelligent battery cover feeding, discharging and vulcanization control system

    CN120080471A

  • Intelligent battery cover feeding and vulcanization control system

    CN120080471B

  • Industrial furnace intelligent temperature control system based on data acquisition

    CN121557752A