A control method for smelting furnace production

Abstract

The present invention relates to the field of data processing technology, and more specifically, the present invention relates to a control method for smelting furnace production, comprising the following steps: obtaining the temperature value of the smelting furnace at each moment, determining the temperature stability at each moment; identifying the first moment, the observation value and the first time period; determining the number of observation values ​​used when obtaining the initial prediction value according to the proportion of the observation values ​​in the first time period that are the same as the first moment; determining the importance of each observation value; obtaining the initial prediction value; predicting the subsequent moments according to the initial prediction value, and adjusting the temperature of the smelting furnace according to the prediction value. The present invention improves the accuracy of temperature prediction by weighted summation of the observation values; based on more accurate initial prediction values, it is possible to achieve more precise control of the temperature of the smelting furnace, thereby optimizing the overall production process; precise temperature control helps to improve the quality of the smelting product, while reducing energy consumption and improving production efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically, to a control method for smelting furnace production. Background Art

[0002] Metal 3D printing is a cumulative manufacturing technology that uses laser sintering to fuse metal powders together to form a specific shape. In this process, certain requirements are placed on the quality of the metal powder. Metal powder is produced using a melting furnace, which melts the metal raw materials and sprays the liquid metal into fine metal particles through atomization to form metal powder. In order for the melting furnace to maintain the temperature required by the metal powder production process and ensure the normal production of metal powder, it is usually necessary to predict the next temperature of the melting furnace and adjust the energy supply system based on the gap between the predicted temperature and the temperature set value, thereby achieving control of the melting furnace temperature and regulating the melting furnace production process.

[0003] The patent application document with publication number CN117195470A discloses a production control method for a smelting furnace and a depleted electric furnace based on a mathematical model, including the steps of: obtaining process parameters for controlling the height of the molten pool surface of the depleted electric furnace; establishing a mathematical model of the height of the molten pool surface of the matte copper in the depleted electric furnace according to the process parameters; obtaining the real-time height and real-time process parameters of the molten pool surface of the matte copper in the depleted electric furnace, inputting the real-time height and real-time process parameters into the mathematical model, and obtaining predicted molten pool surface height data; comparing the predicted molten pool surface height data with a preset value, and regulating the real-time process parameters according to the comparison result.

[0004] However, the above application documents are mainly aimed at the prediction and control of the molten pool surface height of the depleted electric furnace, and do not predict the subsequent temperature of the smelting furnace, so it cannot ensure that the smelting furnace can always maintain the required temperature range. When making predictions, it is usually necessary to manually determine the initial prediction value, and predict the subsequent temperature based on the initial prediction value and the observed value. Therefore, improper selection of the initial prediction value will affect the accuracy of all subsequent temperature prediction values, thereby affecting the control of the smelting furnace temperature and the process of producing metal powder in the smelting furnace. Summary of the invention

[0005] In order to solve the problem that improper selection of initial prediction values ​​will affect the accuracy of all subsequent temperature prediction values, thereby affecting the control of the smelting furnace temperature, the present invention proposes a control method for smelting furnace production, the method comprising the following steps:

[0006] Obtain the temperature value of the smelting furnace at each moment; determine the temperature stability at each moment, the temperature stability is negatively correlated with the normalized result of the temperature standard deviation at the moment and several previous moments, the temperature mean and the absolute value of the difference between the preset temperature value; in response to the temperature stability of several moments before any first moment being greater than the observation threshold, determine the moment as the first moment, the temperature values ​​of the first and all subsequent moments are observation values, and the first and several subsequent moments are regarded as the first period; according to the proportion of the observation values ​​in the first period that are the same as the first moment, determine the number of observation values ​​used when obtaining the initial prediction value ; Determine the importance of each observation value. In response to the temperature stability at the time corresponding to the observation value being less than the importance threshold, the importance is equal to the observation value and the following The mean value of the temperature stability at the corresponding moment of the observation is positively correlated with the observation and the subsequent The absolute value of the difference between the mean of the observations and the preset temperature value is negatively correlated. Otherwise, the importance is positively correlated with the temperature stability at the corresponding moment of the observation, and negatively correlated with the absolute value of the difference between the observation and the preset temperature value. The initial prediction value is obtained by weighted summing up the observations in the first period. The weight of each observation in the first period is positively correlated with the importance of the observation, and negatively correlated with the absolute value of the difference between the observation and the preset temperature. The importance of each observation is negatively correlated; the subsequent moments are predicted based on the initial prediction value, and the smelting furnace temperature is adjusted based on the prediction value.

[0007] The beneficial effects are: by accurately predicting the temperature of the melting furnace, the heating or cooling of the melting furnace can be adjusted more finely to ensure that the temperature is stable within the ideal range; by predicting temperature changes in advance and responding in time, unnecessary energy consumption can be effectively reduced; stable temperature conditions are conducive to improving the consistency and quality of metal powder, thereby improving the performance of the final product; more stable temperature control reduces the risk of damage to the internal components of the melting furnace due to temperature fluctuations, and reduces the frequency and cost of maintenance.

[0008] Furthermore, the temperature stability at each moment satisfies the following relationship:

[0009] ; In the formula, For the Temperature stability at each moment, is the preset temperature value, For the The average temperature of the moment and several previous moments, For the The standard deviation of temperature at a certain moment and several moments before that, For the The maximum value among the moments before and before, is an exponential function.

[0010] The beneficial effect is that by accurately calculating the temperature stability and adjusting the temperature control strategy of the melting furnace accordingly, the accuracy of temperature control and the quality of the product can be significantly improved, while reducing energy consumption and maintenance costs.

[0011] Furthermore, the number of observations used to obtain the initial prediction value is determined based on the proportion of observations in the first period that are the same as those at the first moment. , including: in response to the proportion of the same observation values ​​in the first time period as in the first moment being less than the judgment threshold, then It is positively correlated with the total number of observations in the first period and the mean of temperature stability, and negatively correlated with the proportion of observations in the first period that are different from the first moment and the mean of the difference between all observations and the observations at the first moment. Otherwise, Positively correlated with the total number of observations in the first period.

[0012] The beneficial effect is that by dynamically adjusting the number of observation values ​​used to obtain the initial prediction value, the accuracy of temperature prediction can be improved, the temperature of the smelting furnace can be effectively controlled, product quality can be ensured and energy consumption can be reduced.

[0013] Furthermore, the Satisfies the following relationship:

[0014] ; In the formula, The number of observations to use when obtaining initial forecasts. is the number of observations in the first period that are identical to the first moment, is the total number of observations in the first period, To determine the threshold, is the observed value at the first moment, For the first period observations, is the mean value of the temperature stability in the first period, is an exponential function, is the maximum value function.

[0015] Furthermore, the importance of each observation value is expressed as follows:

[0016] ; In the formula, No. The importance of an observation, For the The temperature stability at the time corresponding to the observation value is is an important threshold, For the observations and after The mean value of the temperature stability at the time corresponding to the observation value, For the observations and the following The mean of the observations, For the observations, is the preset temperature value of the melting furnace, The number of observations to use when obtaining initial forecasts.

[0017] The beneficial effect is that by accurately calculating the importance of each observation value, the accuracy and reliability of temperature prediction can be improved, thereby more effectively controlling the temperature of the smelting furnace.

[0018] Furthermore, the initial prediction value satisfies the following relationship:

[0019] ; In the formula, is the initial prediction value, No. The weight of the observations, For the observations, The number of observations to use when obtaining initial forecasts.

[0020] The beneficial effect is that the initial prediction value is calculated by weighted summation, and the prediction value is optimized according to the weight of each observation value, so the importance of different observation values ​​is quantified, making the prediction result more accurate and representative.

[0021] Furthermore, the weights of the observations in the first time period satisfy the following relationship:

[0022] ; In the formula, No. The weight of the observations, No. The importance of an observation, The number of observations to use when obtaining initial forecasts.

[0023] Furthermore, predicting the subsequent moments according to the initial predicted values ​​and adjusting the temperature of the smelting furnace according to the predicted values ​​is to predict the subsequent moments according to the initial predicted values ​​by a single exponential smoothing method, and adjust the temperature of the smelting furnace according to the difference between the predicted result and the preset temperature value of the smelting furnace to complete the regulation of the smelting furnace production.

[0024] The present invention has the following beneficial effects:

[0025] The present invention determines the temperature stability at each moment through the temperature value of the smelting furnace at each moment, and determines the first moment according to the size of the temperature stability, effectively avoiding the influence of temperature data that has little to do with the temperature maintenance stage of the smelting furnace on the temperature control of the smelting furnace; determines the number of observation values ​​used for the initial prediction value according to the observation value and its temperature stability of the first time period, so that the calculated initial prediction value is more representative, and the prediction value obtained based on the initial prediction value has higher accuracy; calculates the initial prediction value according to the temperature stability at the corresponding moment of the determined number of observation values ​​and the difference between the observation value and the temperature setting value, better smoothes the random fluctuation of the temperature data, reduces the influence of abnormal values, makes the accuracy of the prediction of the smelting furnace temperature more stable, and further makes the control of the smelting furnace temperature more accurate. The production process of the smelting furnace is regulated by the present invention, which is beneficial to the accurate control of the temperature of the smelting furnace and improves the quality of the metal powder produced by the smelting furnace. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0027] Figure 1 It is a flow chart of the steps of a control method for smelting furnace production according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0029] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0030] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0031] The embodiment of the present invention discloses a control method for smelting furnace production, referring to Figure 1, comprising steps S1 to S6:

[0032] S1: Obtain the temperature value of the smelting furnace at each moment and determine the temperature stability at each moment.

[0033] It should be noted that during the production process of the smelting furnace, the temperature gradually rises from room temperature to the set value. The temperature in this process will rise from 20 to 30 degrees to thousands of degrees, which has a large span. When predicting, the temperature at the next moment is usually predicted based on all data and recent data. The initial temperature of the smelting furnace has no reference significance when the temperature of the smelting furnace is controlled to be close to the preset temperature value. Therefore, the temperature stability is calculated based on the temperature values ​​of the smelting furnace at each moment.

[0034] Specifically, the temperature stability at each moment satisfies the following relationship:

[0035] ;

[0036] In the formula, For the Temperature stability at each moment, is the preset temperature value, For the The average temperature of the moment and several previous moments, For the The standard deviation of temperature at a certain moment and several moments before that, For the The maximum value among the moments before and before, is an exponential function.

[0037] in, For the The difference between the average temperature at a certain moment and the previous moments and the preset temperature value is calculated by dividing The range of the difference between the mean temperature and the preset temperature value is adjusted by , The larger the The smaller the The more likely the temperature value at the moment is in the rising stage from room temperature to the melting furnace temperature setting value, the The lower the temperature stability at each moment, The smaller the The larger the The more likely the temperature value at the moment is in the stage of maintaining the melting furnace temperature, the The higher the temperature stability at each moment; For the The standard deviation of temperature at a certain moment and several moments before that is calculated by dividing by The way to Normalize it and get , The larger the The greater the temperature change in the local area around the moment, the The lower the temperature stability at each moment, The smaller the The smaller the temperature change in the local area around the moment, the The higher the temperature stability at each moment.

[0038] Implementers can set the preset temperature value of the smelting furnace according to actual implementation conditions. For example, when the metal is titanium, the preset temperature value is 1660°C.

[0039] S2: Identify the first moment, the observation value, and the first period.

[0040] It should be noted that when the temperature stability at several moments before any moment is greater than the observation threshold, the temperature value at that moment is near the preset temperature value of the smelting furnace. The temperature data value at that moment is used as the first observation value of the initial prediction value, and the temperature values ​​after the first observation value are also observation values.

[0041] In response to the temperature stability of several moments before any first moment being greater than the observation threshold, the moment is identified as the first moment, the temperature values ​​of the first and subsequent moments are observed values, and the first and subsequent moments are regarded as the first time period.

[0042] Implementers can set the observation threshold according to actual implementation, for example, 0.75.

[0043] S3: Determine the number of observations used to obtain the initial forecast value based on the proportion of observations in the first period that are the same as the first moment. .

[0044] It should be noted that when obtaining the initial prediction value, in order to make the subsequent temperature prediction value obtained based on the initial prediction value more accurately reflect the temperature of the smelting furnace at the next moment, it is usually necessary to calculate the initial prediction value through multiple observation values. Which observation values ​​are used to calculate the initial prediction value needs to be determined manually, which has certain subjective factors. Sometimes the artificially selected observation values ​​may not be conducive to the prediction of the subsequent temperature, thereby affecting the control of the smelting furnace temperature. Therefore, it is necessary to determine the number of observation values ​​used for the initial prediction value based on the temperature stability of the first time period.

[0045] Specifically, the number of observations used for the initial prediction value is obtained according to the proportion of the observations in the first period that are the same as those at the first moment. , Satisfies the following relationship:

[0046] ;

[0047] In the formula, The number of observations to use when obtaining initial forecasts. is the number of observations in the first period that are identical to the first moment, is the total number of observations in the first period, To determine the threshold, is the observed value at the first moment, For the first period observations, is the mean value of the temperature stability in the first period, is an exponential function, is the maximum value function.

[0048] in, When , it means that most of the observations in the first period are different from the observations at the first moment. There may be obvious mutation points in the observations. Fewer observations should be used to calculate the initial prediction value to quickly respond to temperature changes. In summary, The bigger the The smaller it is, the higher the temperature stability in the first period is, and the smaller the possibility of obvious mutation points in the observed values ​​is. In this case, more observed values ​​should be used to obtain the initial prediction value. The more hours, The larger the value, the lower the temperature stability in the first period, and there may be obvious mutation points in the observed values, so fewer observed values ​​should be used to obtain the initial prediction value; The ratio of observations in the first period that are different from those at the first moment. The larger the value, the more likely it is that there is an obvious mutation point in the observations, and fewer observations should be used to calculate the initial prediction value. The smaller the value, the less likely it is that there is an obvious mutation point in the observations, and more observations should be used to calculate the initial prediction value. is the mean of the difference between all observations in the first period and the observations at the first moment. The larger the value, the more likely it is that there is an obvious mutation point in the observations, and fewer observations should be used to obtain the initial prediction value. The smaller the value, the less likely it is that there is an obvious mutation point in the observations, and more observations should be used to obtain the initial prediction value. At the same time, for the convenience of calculation, by dividing The way to Normalize. When , it means that most of the observations in the first period are the same as the observations at the first moment, then the overall fluctuation of the observations near the first observation is small, and enough observations should be used to calculate the initial prediction value to reduce the impact of random temperature fluctuations. Therefore, the total number of observations in the first period is used as the number of observations used to obtain the initial prediction value.

[0049] Implementers can set the judgment threshold according to actual implementation conditions, for example, 0.5.

[0050] S4: Determine the importance of each observation.

[0051] It should be noted that since the temperature data is constantly changing, it is necessary to determine the importance of each observation so that different weights can be given when calculating the initial prediction value.

[0052] Specifically, the importance of each observation value is expressed as follows:

[0053] ;

[0054] In the formula, No. The importance of an observation, For the The temperature stability at the time corresponding to the observation value is is an important threshold, For the observations and after The mean value of the temperature stability at the time corresponding to the observation value, For the observations and the following The mean of the observations, For the observations, is the preset temperature value of the melting furnace, The number of observations to use when obtaining initial forecasts.

[0055] in, , explain The temperature stability at the moment of observation is small. The less the observed value represents the center of the temperature fluctuation near it, the observations and after The mean value of the temperature stability at the time of observation Calculate the The importance of an observation, The larger the value, the more important the corresponding observation value. The greater the importance of an observation, The smaller the value, the less important the corresponding observation value. The smaller the importance of the observation, the smaller the The smaller the temperature stability at the moment of observation, the The observation value is less representative of the center of temperature fluctuations near it, so through this observation value and the subsequent The difference between the mean of the observed values ​​and the preset temperature value is calculated by dividing Adjust the range of the temperature difference by To calculate the The importance of an observation, The larger the The smaller the importance of an observation, The more hours, the The greater the importance of an observation. , explain The temperature stability of the observed value is relatively large. The more likely the observation is to be at a location where the temperature fluctuation is stable, the Temperature stability of observations The larger the value, the more important the corresponding observation value. The greater the importance of the observation, the The greater the temperature stability around an observation value, The more it represents the difference between the temperature value near it and the preset temperature value, for the convenience of calculation, by dividing By adjusting the range of the temperature difference, we can get , The larger the value, the The greater the distance between the observed value and the temperature setting value, the The more likely an observation is to be in a location with a large temperature fluctuation, the The less important the observation is.

[0056] Implementers can set the importance threshold according to actual implementation conditions, for example, 0.8.

[0057] S5: Get initial prediction value.

[0058] It should be noted that the initial prediction value obtained by the prior art has a certain degree of randomness, and all subsequent prediction values ​​are obtained on the basis of the initial prediction value. Therefore, the size of the initial prediction value affects the calculation of subsequent prediction values, and further affects the control of the smelting furnace temperature. Therefore, the initial prediction value is calculated based on a certain number of observations, their weights and observations.

[0059] The initial prediction value is obtained by weighted summing up the observations in the first period. The weight of each observation in the first period is positively correlated with the importance of the observation and the subsequent The sum of the importance of the observations is negatively correlated.

[0060] Specifically, the initial prediction value satisfies the following relationship:

[0061] ;

[0062] In the formula, is the initial prediction value, No. The weight of the observations, For the observations, The number of observations to use when obtaining initial forecasts.

[0063] Specifically, the weight of each observation value in the first time period satisfies the following relationship:

[0064] ;

[0065] In the formula, No. The weight of the observations, No. The importance of an observation, The number of observations to use when obtaining initial forecasts.

[0066] S6: Predict the subsequent time according to the initial prediction value, and adjust the temperature of the smelting furnace according to the prediction value.

[0067] Specifically, predicting the subsequent moments according to the initial prediction value and adjusting the temperature of the smelting furnace according to the prediction value is to predict the subsequent moments according to the initial prediction value through a single exponential smoothing method, and adjust the temperature of the smelting furnace according to the difference between the predicted result and the preset temperature value of the smelting furnace to complete the regulation of the smelting furnace production.

[0068] Implementers can set the smoothing coefficient of the exponential smoothing method according to actual implementation conditions, for example, 0.9.

[0069] The exponential smoothing method selects the first observation value or the average of the first few observation values ​​in the data sequence to calculate the initial prediction value. It is necessary to manually set the observation values ​​used in calculating the initial prediction value, and use the average of these observation values ​​as the initial prediction value. The present invention determines the number of observation values ​​used in calculating the initial prediction value based on the temperature stability of the observation values ​​and the corresponding moments, and weights the observation values ​​used based on the temperature stability of the observation values ​​used and the size of their values, and performs weighted summation on the observation values ​​used to obtain the initial prediction value. The initial prediction value obtained is used to predict the temperature value of the smelting furnace at subsequent moments using a single exponential smoothing method, and the functional system of the smelting furnace is adjusted based on the gap between the predicted result and the preset temperature value of the smelting furnace, thereby controlling the temperature of the smelting furnace, achieving the regulation of the production of metal powder by the smelting furnace, improving the quality of the produced metal powder, and completing the regulation of the production of the smelting furnace.

[0070] The present invention improves the accuracy of temperature prediction by performing weighted summation of observed values ​​and taking into account temperature stability and importance; based on more accurate initial prediction values, more precise regulation of the smelting furnace temperature can be achieved, thereby optimizing the overall production process; precise temperature control helps to improve the quality of smelting products, while reducing energy consumption and improving production efficiency.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A control method for smelting furnace production, characterized in that: The following steps are involved: Get the temperature value of the smelting furnace at each moment; Determine the temperature stability at each moment, the temperature stability is negatively correlated with the normalized result of the temperature standard deviation at the moment and several previous moments, the temperature mean and the absolute value of the difference between the preset temperature value; in response to the temperature stability at several moments before any first moment being greater than the observation threshold, identify the moment as the first moment, the temperature values ​​of the first and all subsequent moments are observation values, and the first and several subsequent moments are regarded as the first time period; Determine the number of observations used to obtain the initial forecast value based on the proportion of observations in the first period that are the same as the first moment ; Determine the importance of each observation value. In response to the temperature stability at the time corresponding to the observation value being less than the importance threshold, the importance is equal to the observation value and the following The mean value of the temperature stability at the corresponding moment of the observation is positively correlated with the observation and the subsequent The importance is negatively correlated with the absolute value of the difference between the mean of the observed values ​​and the preset temperature value. Otherwise, the importance is positively correlated with the temperature stability at the corresponding moment of the observed value and negatively correlated with the absolute value of the difference between the observed value and the preset temperature value. The initial prediction value is obtained by weighted summing up the observations in the first period. The weight of each observation in the first period is positively correlated with the importance of the observation and the subsequent The importance of each observation is negatively correlated; the subsequent moments are predicted based on the initial prediction value, and the smelting furnace temperature is adjusted based on the prediction value.

2. A control method for smelting furnace production according to claim 1, characterized in that: The temperature stability at each moment satisfies the following relationship: ； In the formula, For the Temperature stability at each moment, is the preset temperature value, For the The average temperature of the moment and several previous moments, For the The standard deviation of temperature at a certain moment and several moments before that, For the The maximum value among the moments before and before, is an exponential function.

3. A control method for smelting furnace production according to claim 1, characterized in that: Determine the number of observations used to obtain the initial forecast value based on the proportion of observations in the first period that are the same as the first moment ,include: In response to the fact that the proportion of observations in the first period that are the same as those at the first moment is less than the judgment threshold, It is positively correlated with the total number of observations in the first period and the mean of temperature stability, and negatively correlated with the proportion of observations in the first period that are different from the first moment and the mean of the difference between all observations and the observations at the first moment. Otherwise, Positively correlated with the total number of observations in the first period.

4. A control method for smelting furnace production according to claim 3, characterized in that: Said Satisfies the following relationship: ； In the formula, The number of observations to use when obtaining initial forecasts. is the number of observations in the first period that are identical to the first moment, is the total number of observations in the first period, To determine the threshold, is the observed value at the first moment, For the first period observations, is the mean value of the temperature stability in the first period, is an exponential function, is the maximum value function.

5. A control method for smelting furnace production according to claim 1, characterized in that: The importance of each observation value is as follows: ； In the formula, For the The importance of an observation, For the The temperature stability at the time corresponding to the observation value is is an important threshold, For the observations and after The mean value of the temperature stability at the time corresponding to the observation value, For the observations and the following The mean of the observations, For the observations, is the preset temperature value of the melting furnace, The number of observations to use when obtaining initial forecasts.

6. A control method for smelting furnace production according to claim 1, characterized in that: The initial prediction value satisfies the following relationship: ； In the formula, is the initial prediction value, For the The weight of the observations, For the observations, The number of observations to use when obtaining initial forecasts.

7. A control method for smelting furnace production according to claim 1, characterized in that: The weights of the observations in the first period satisfy the following relationship: ； In the formula, For the The weight of the observations, For the The importance of an observation, The number of observations to use when obtaining initial forecasts.

8. A control method for smelting furnace production according to claim 1, characterized in that: The method of predicting the subsequent moments according to the initial predicted values ​​and adjusting the temperature of the smelting furnace according to the predicted values ​​is to predict the subsequent moments according to the initial predicted values ​​by a single exponential smoothing method, and adjust the temperature of the smelting furnace according to the difference between the predicted result and the preset temperature value of the smelting furnace to complete the regulation of the smelting furnace production.

Images