Smelting method based on large-tonnage double-frequency induction melting electric furnace

Abstract

The invention relates to a smelting method based on a large-tonnage double-frequency induction smelting electric furnace, and belongs to the technical field of metal smelting. 2, a detection module detects temperature data of temperature and time changes and uploads the temperature data to a control module, and the control module takes the temperature data of the same production batch as historical data; 3, the control module predicts the predicted value of the temperature changing along with the time during the smelting production by using a machine learning algorithm according to the historical data; 4, the control module judges that the difference value between the detection value of the detection module and the predicted value at the same proceeding time reaches a deviation threshold value, the step 5 is executed, and otherwise, the step 6 is executed; 5, the control module switches the induction furnace into a low-frequency mode according to the time when the predicted value reaches the switching threshold value; sixthly, when the control module judges that the detection value reaches the switching threshold value, the induction furnace is switched to be in a low-frequency mode; the device has the characteristics of being accurate in switching and considering both the smelting effect and the smelting efficiency.

Description

Technical Field

[0001] The invention belongs to the technical field of metal smelting, and in particular relates to a smelting method based on a large-tonnage dual-frequency induction smelting electric furnace. Background Art

[0002] Induction furnace is a common equipment used for smelting metal. During the smelting process, a certain smelting temperature and stirring strength are required to ensure that the metal is completely melted into liquid and that the composition of each part of the molten metal is uniform.

[0003] A general method of smelting metal using an induction melting furnace, such as a method for extracting vanadium and dephosphorizing molten iron disclosed in Chinese patent application CN118957308A, comprises heating molten iron with an induction furnace, adding a vanadium extracting and dephosphorizing agent, adjusting the power supply parameters of the induction furnace to a heat preservation gear, using the heat preservation gear current to provide an electromagnetic stirring force for slag-metal reaction, and after the reaction is completed, removing the vanadium slag to obtain low-phosphorus semi-steel and transporting it to a steelmaking converter for steelmaking; it is provided that the vanadium extracting and dephosphorizing agent is added after the molten iron is heated, the power supply parameters of the induction furnace are adjusted to a heat preservation gear, and stirring is performed after the smelting is completed, so as to improve the quality of the semi-steel, make it more in line with the use requirements, and increase the adaptability of the product.

[0004] In the above structure, although the insulation gear current is used to provide electromagnetic stirring force for slag-metal reaction, the switching of the two gears is completed, and the characteristics of the two frequencies of the induction furnace are utilized. However, when the switching time is too early, it will cause the steel to be switched from the medium-frequency mode with stronger heating ability to the low-frequency mode with stronger stirring ability before it is completely melted into molten steel, which will cause incomplete melting. When the switching timing is too late, it will cause the stirring to not be switched after the steel is completely melted into molten steel. The switching timing depends to a large extent on whether the temperature of the metal to be smelted in the furnace reaches the threshold value as shown by the temperature detection result of the metal to be smelted. Switching when the threshold value is reached can ensure to a certain extent that it is switched to the low-frequency mode when the melting is complete. However, for high-temperature molten metal, the temperature measurement is likely to be inaccurate, which leads to the probability of inaccurate switching timing of the temperature measurement result. For this reason, a smelting method based on a large-tonnage dual-frequency induction smelting electric furnace is required, which has accurate switching and takes into account both smelting effect and efficiency. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a smelting method based on a large-tonnage dual-frequency induction smelting furnace, which has the characteristics of accurate switching timing when switching between medium frequency and low frequency, and taking into account both smelting effect and efficiency.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A smelting method based on a large-tonnage dual-frequency induction smelting electric furnace comprises the following steps:

[0008] Step 1: Carry out a new smelting production;

[0009] Step 2: The detection module detects the temperature data of the temperature change over time and uploads it to the control module. The control module archives and organizes the temperature data of the same production batch as historical data;

[0010] Step 3: The control module uses a machine learning algorithm to predict the predicted value of the temperature change over time during this smelting production based on historical data, and calculates the time when the predicted value reaches the switching threshold;

[0011] Step 4: The control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the deviation threshold. If the judgment result is yes, execute step 5, otherwise execute step 6;

[0012] Step 5: The control module switches the induction furnace from the medium frequency mode to the low frequency mode according to the time when the switching threshold is reached in the predicted value, and executes step 7;

[0013] Step 6: The control module determines whether the detection value reaches the switching threshold. If the determination result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode, and step 7 is executed;

[0014] Step 7: Repeat steps 1 to 6.

[0015] As a preferred technical solution of the present invention, the step three also includes: the control module uses the ARIMA algorithm to determine the algorithm parameters (p, d, q) according to the power consumption data archived by time, and uses the ARIMA algorithm to predict the temperature change over time during this smelting production.

[0016] As a preferred technical solution of the present invention, step five also includes: the control module determines whether the average of the predicted value and the detected value at the same time in the predicted value reaches the switching threshold. When the judgment result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode, and step seven is executed.

[0017] As a preferred technical solution of the present invention, the step 2 also includes: a plurality of temperature sensors in the detection module detect temperature data of temperature changes over time and upload the temperature data group to the control module, and the control module archives and organizes the temperature data of the same production batch as historical data; the step 3 also includes: the control module determines whether the temperature variance in the temperature data group uploaded five times in a row exceeds the variance threshold, and when the judgment result is yes, the control module upwardly corrects the deviation threshold, and the step 4 also includes: the control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the upwardly corrected deviation threshold, when the judgment result is yes, execute step 5, otherwise execute step 6.

[0018] As a preferred technical solution of the present invention, the step three also includes: the control module is pre-input with a variance threshold F0, the control module determines whether the temperature variance in the temperature data group uploaded five times in a row exceeds the variance threshold, and when the judgment result is yes, the control module increases the deviation threshold by A1 times, wherein A1=F / F0×c; the step four also includes: the control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the deviation threshold after increasing by A1 times, when the judgment result is yes, execute step five, otherwise execute step six.

[0019] As a preferred technical solution of the present invention, the step five also includes: the control module switches the induction furnace from the medium frequency mode to the low frequency mode after a delay according to the time when the switching threshold is reached in the predicted value, and executes step seven; the step six also includes: the control module determines whether the detection value reaches the switching threshold, and when the judgment result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode after a delay, and executes step seven.

[0020] As a preferred technical solution of the present invention, step one also includes: performing a new smelting production, inputting the melting point of this smelting, and the control module increases the delay when the melting point exceeds the melting point threshold, and reduces the delay when the melting point is lower than the melting point threshold.

[0021] As a preferred technical solution of the present invention, the step three also includes: after predicting the predicted value of the temperature change over time during this smelting production, the control module determines the heating rate, and corrects the delay downward when the heating rate exceeds the rate threshold, and corrects the delay upward when the heating rate is lower than the rate threshold.

[0022] The beneficial effects of the present invention are:

[0023] (1) By making the control module calculate the predicted value of the temperature change over time based on historical data, and when the difference between the predicted values ​​at the time corresponding to the detected value reaches the deviation threshold, when there is a high probability of inaccurate detection, the detected value is stopped from being accepted, and the predicted value is used to guide the switching timing, thereby reducing the impact of inaccurate temperature measurement on the switching timing; at the same time, when the detection result is relatively accurate, the detected value is used to guide the switching timing. Compared with the predicted value that represents a typical situation but has a gap with the actual temperature, the detected value is closer to the actual temperature, thereby improving the accuracy of the switching timing;

[0024] (2) By making the control module determine whether the average of the predicted value and the detected value at the same time in the predicted value reaches the switching threshold, the average of the predicted value and the detected value is used. Compared with a single detected value or a predicted value, the average of the detected value and the predicted value can make the obtained value closer to the actual temperature value to a greater extent or with a greater probability, thereby making the switching time point based on the average closer to the time point when the metal to be smelted is completely smelted, thereby improving the accuracy of the switching timing;

[0025] (3) By making the control module judge whether the temperature variance in the temperature data group uploaded five times continuously exceeds the variance threshold, and when the judgment result is yes, the control module upwardly corrects the deviation threshold, so that when the detection value is easily misjudged as exceeding the deviation threshold under normal circumstances, the control module relaxes the judgment standard and adopts the detection value closer to the actual situation to determine the switching timing; when the probability of the detection value being misjudged is low, the judgment standard is tightened to avoid the influence of the inaccurate detection value on the switching timing;

[0026] (4) By making the control module correct the delay downward when the heating rate exceeds the threshold, when the heating rate is lower than the threshold, it means that the heating efficiency of the metal is slow after the alternating electric field is passed through, and the probability of the metal being fully melted is low, and the delay needs to be extended. The delay is corrected upward to ensure the melting effect. When the metal is subjected to the same alternating electric field and the temperature is higher, it means that the heating rate is fast and the metal is more likely to be fully melted. At this time, the threshold is corrected downward to ensure the operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0028] Figure 1 This is a control loop block diagram of the present invention; DETAILED DESCRIPTION

[0029] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0030] See also Figure 1 , a smelting method based on a large-tonnage dual-frequency induction smelting furnace, comprising the following steps:

[0031] Step 1: Carry out a new smelting production;

[0032] Specifically, the production is carried out in batches. In each batch of production, the composition, melting point and target product of the metal to be smelted are the same. When any one or more of the composition, melting point or target product changes, it is recorded as a new batch of production.

[0033] Meanwhile, the induction melting furnace at least comprises a furnace body and a coil arranged around the furnace body. When an alternating current is passed through the coil, an alternating magnetic field is periodically generated in the furnace body, so that the metal block in the furnace body generates an eddy current electric field due to the alternating magnetic field, thereby utilizing the electrothermal effect to melt the metal. The coil is electrically connected to a control module, and the control module controls the frequency and operating power of the coil.

[0034] When a certain smelting production starts, the metal to be smelted is placed in the furnace body, and then the coil is supplied with medium frequency current to smelt the metal to be smelted, and after the smelting is completed, it is switched to supply low frequency current for stirring;

[0035] After step 1 is completed, proceed to step 2;

[0036] Step 2: The detection module detects the temperature data of the temperature change over time and uploads it to the control module. The control module archives and organizes the temperature data of the same production batch as historical data;

[0037] Specifically, since it is impossible to collect enough data as historical data when a certain batch of production just starts, in this embodiment, when the temperature data of the temperature time change is detected by the detection module in the first four productions of a certain batch of production and uploaded to the control module, the control module archives and organizes the temperature data of the same production batch as historical data. When it is the fifth time, step one is executed. At this time, in step two, the temperature data of the past four productions are archived and organized as historical data. When the nth production is subsequently carried out, in step two, the temperature data of the past n-1 productions are archived and organized as historical data.

[0038] After step 2 is completed, proceed to step 3;

[0039] Step 3: The control module uses a machine learning algorithm to predict the predicted value of the temperature change over time during this smelting production based on historical data, and calculates the time when the predicted value reaches the switching threshold;

[0040] Specifically, the control module determines the temperature value that changes with time in the current production based on past historical data, and draws a chart of the temperature change over time as a prediction value, with the starting point of the chart being the time point when production starts;

[0041] After step 3 is completed, step 4 is executed: the control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the deviation threshold. If the judgment result is yes, step 5 is executed, otherwise step 6 is executed;

[0042] Specifically, the control module starts timing synchronously when performing each production, and then obtains the number of seconds from the start of production once per second, and finds the temperature value corresponding to the current number of seconds in the prediction value chart as the current prediction value. At the same time, the detection module counts the temperature of the metal to be smelted once per second and uploads it to the control module as the detection value. After receiving the detection value each time, the control module determines whether the difference between the detection value and the current prediction value exceeds the deviation threshold. When the deviation threshold is exceeded for five consecutive judgments, the control module determines that the difference between the detection value and the prediction value at the same time reaches the threshold.

[0043] Although the prediction result of the current prediction value cannot fully represent the actual temperature value, it can represent the typical temperature value at the current time point. When the detected value deviates greatly from the predicted value, it means that there is a high probability of inaccurate detection. At this time, it is necessary to stop using the detected value in this production and use the predicted value to guide the switching timing. At this time, execute step 5;

[0044] Step 5: The control module switches the induction furnace from the medium frequency mode to the low frequency mode according to the time when the switching threshold is reached in the predicted value, and executes step 7;

[0045] At this time, the control module stops accepting the detection value in this production, and obtains the number of seconds corresponding to the point where the temperature value reaches the switching threshold from the predicted value chart. Then, when the control module timing display reaches this number of seconds, it means that the metal in the furnace body has a high probability of reaching the target temperature and is completely melted. The control module instructs the induction furnace to switch from the medium frequency mode to the low frequency mode, and stops accepting the detection value, and uses the predicted value to guide the switching timing;

[0046] In the above process, if the deviation threshold is not exceeded for five consecutive judgments, the control module continues to receive the detection value uploaded by the detection module. When the detection value reaches the switching threshold, it means that the detection module detects that the metal in the furnace body has reached the target temperature and is completely melted, and the detection result is credible. At this time, the control module instructs the induction furnace to switch from the medium frequency mode to the low frequency mode, and completes the switching timing according to the detection value.

[0047] After step 5 or step 6 is completed, execute step 7, that is, return to step 1 and restart a new production process;

[0048] By enabling the control module to calculate the predicted value of the temperature change over time based on historical data, and when the difference between the detected value and the predicted value at the same time reaches the deviation threshold, when there is a high probability of inaccurate detection, the detected value is stopped from being accepted, and the predicted value is used to guide the switching timing, thereby reducing the impact of inaccurate temperature measurement on the switching timing; at the same time, when the detection result is relatively accurate, the detected value is used to guide the switching timing. Compared with the predicted value that represents the typical situation but has a gap with the actual temperature, the detected value is closer to the actual temperature, thereby improving the accuracy of the switching timing.

[0049] In the above step 3, the control module uses a machine learning algorithm to predict the predicted value of the temperature change over time during this smelting production based on historical data, and calculates the time when the switching threshold is reached in the predicted value. Specifically, the control module uses the ARIMA algorithm to determine the algorithm parameters (p, d, q) based on the power consumption data archived by time, and uses the ARIMA algorithm to predict the temperature change over time during this smelting production;

[0050] Specifically, the following steps are included:

[0051] The control module receives the historical data for preprocessing. First, the data of temperature changes over time in the past several productions are connected in sequence. Then, the unit root test is used to check the stability of the data. In this embodiment, the test method is ADF test.

[0052] Then the control module determines whether the historical data is stable. If the judgment result is no, the historical data is differentially operated to achieve stability. First, the original data is differentially operated once to determine whether it is stable. If the judgment result is no, the differential operation is repeated until the data reaches a stable state, and the required number of differentials d is recorded;

[0053] The control module then calculates the autocorrelation function (ACF) and partial autocorrelation function (PACF) according to different p and q values, analyzes the ACF and PACF graphs under different p and q values, and selects the combination that obtains the minimum information criterion (such as AIC or BIC), the autoregressive order p, and the sliding average order q;

[0054] At this time, the control module determines the algorithm parameters (p, d, q) and establishes the ARIMA (p, d, q) model. After the establishment, the established ARIMA model is used to obtain the historical data and the trend graph of temperature change over time. The trend graph of temperature change over time from the start of production to the time point when the temperature reaches the switching threshold is intercepted as the predicted value.

[0055] In the above process, when it is judged that the detected value and the predicted value have a large deviation, although the predicted value representing the typical situation can be closer to the actual value with a higher probability, the detected value also represents the actual temperature. In this case, it is necessary to use both the detected value and the predicted value and calculate the average of the two data.

[0056] By making the control module determine whether the average of the predicted value and the detected value at the corresponding time in the predicted value reaches the switching threshold, the average of the predicted value and the detected value is used. Compared with a single detected value or predicted value, the average of the detected value and the predicted value can make the obtained value closer to the actual temperature value to a greater extent or with a greater probability, thereby making the switching time point based on the average value closer to the time point when the metal to be smelted is completely smelted, thereby improving the accuracy of the switching timing.

[0057] The temperature detection by a single temperature sensor may lead to inaccurate temperature detection. Therefore, step 2 also includes: a plurality of temperature sensors in the detection module detect temperature data of temperature changes over time and upload the temperature data group to the control module, and the control module archives and organizes the temperature data of the same production batch as historical data;

[0058] Specifically, a plurality of temperature sensors synchronously upload temperature data once per second. After receiving a plurality of temperature data uploaded by the plurality of temperature sensors, the control module takes an average of the plurality of temperature data as the detection value.

[0059] In the above process, when the variance of several temperature data received by the detection module is large, it means that the temperature of several parts of this batch of production has a large difference due to random factors. At this time, even if there is no inaccurate temperature detection, the deviation between the detection value and the predicted value will exceed the deviation threshold. At this time, the control module will stop accepting the detection value that better represents the actual situation when there is actually no inaccurate temperature detection, and use the predicted value instead. Therefore, when the variance of several temperature data received by the detection module is large, it is necessary to relax the judgment standard of whether the temperature detection is inaccurate;

[0060] To this end, step three also includes: the control module determines whether the temperature variance in the temperature data group uploaded five times continuously exceeds the variance threshold, and when the judgment result is yes, the control module corrects the deviation threshold upward, and step four also includes: the control module determines whether the difference between the detection value of the detection module and the predicted value under the same time reaches the upwardly corrected deviation threshold, when the judgment result is yes, execute step five, otherwise execute step six;

[0061] Specifically, step three also includes: the control module pre-inputs a variance threshold F0, the control module determines whether the temperature variance in the temperature data group uploaded five times continuously exceeds the variance threshold, and when the judgment result is yes, the control module increases the deviation threshold by A1 times, where A1=F / F0×c; step four also includes: the control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the deviation threshold after increasing by A1 times, when the judgment result is yes, execute step five, otherwise execute step six;

[0062] When the variance F is large, it means that there is a high probability that the deviation between the detection value and the predicted value still exceeds the deviation threshold when there is no inaccurate temperature detection. It is necessary to relax the judgment standard of whether the detection is inaccurate. At this time, the value of A1=F / F0×c is large, and the control module increases the deviation threshold by A1 times. Under the same circumstances, the difference between the detection value and the predicted value is more difficult to be judged as exceeding the deviation threshold. When the difference between the detection value and the predicted value is easily misjudged as exceeding the deviation threshold under normal circumstances, the judgment standard is relaxed;

[0063] When the variance F is small, it means that the deviation between the detected value and the predicted value still exceeds the deviation threshold. There is a high probability that the temperature detection is inaccurate, and it is necessary to narrow the judgment standard of whether the temperature detection is inaccurate. At this time, the value of A1=F / F0×c is small, and the control module increases the deviation threshold by A1 times. Under the same circumstances, the difference between the detected value and the predicted value is more likely to be judged as exceeding the deviation threshold. When the difference between the detected value and the predicted value is easily misjudged as exceeding the deviation threshold, when the detected value is misjudged as exceeding the deviation threshold under normal circumstances, the judgment standard is tightened;

[0064] By making the control module judge whether the temperature variance in the temperature data group uploaded five times continuously exceeds the variance threshold, and when the judgment result is yes, the control module corrects the deviation threshold upward, so that when the detection value is easily misjudged as exceeding the deviation threshold under normal circumstances, the control module relaxes the judgment standard and adopts the detection value that is closer to the actual situation to decide the switching timing; when the probability of the detection value being misjudged is low, the judgment standard is tightened to avoid the influence of the inaccurate detection value on the switching timing.

[0065] In the above process, when the detected value or predicted value of the temperature reaches the threshold, there is still a probability that there are unmelted steel blocks in the furnace body. At this time, it is necessary to delay for a certain time and switch to the low-frequency mode when the steel is completely melted. For this reason, step five also includes: the control module switches the induction furnace from the medium frequency mode to the low frequency mode after a delay according to the time when the switching threshold is reached in the predicted value, and executes step seven; the step six also includes: the control module determines whether the detected value reaches the switching threshold, and when the judgment result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode after a delay, and executes step seven;

[0066] After setting the delay, the equipment will be in the medium frequency mode for a longer time to melt the steel block, reducing the probability of incomplete melting and improving the melting effect and the accuracy of the switching timing;

[0067] In different situations, different delays are required. For example, when smelting metal with a higher melting point, the metal melting point is higher. When the temperature detection value or predicted value reaches the threshold, there is a greater probability that there is more unmelted metal. At this time, the delay needs to be extended so that the equipment is in the medium frequency mode for a longer time to melt the steel block and ensure the smelting effect. When the metal melting point is lower, when the temperature detection value or predicted value reaches the threshold, there is a lower probability or degree of unmelted metal. At this time, there is no need to extend the delay and switch to the low frequency mode as soon as possible to improve efficiency.

[0068] To this end, step one also includes: performing a new smelting production, inputting the melting point of this smelting, and the control module increases the delay when the melting point exceeds the threshold, and reduces the delay when the melting point is lower than the threshold;

[0069] By setting the control module to increase the delay when the melting point exceeds the threshold and reduce the delay when the melting point is below the threshold, sufficient melting is ensured when melting metals with higher melting points and the operating efficiency is improved when melting metals with lower melting points.

[0070] In use, different metals have different effects when smelted in an induction furnace. Some metals generate higher temperatures when the same alternating electric field is passed through. The faster the heating rate, the more likely the metal is to be fully smelted. At this time, the delay needs to be corrected downward and the low-frequency mode needs to be switched as soon as possible. When the metal heating rate is slow, it means that the probability of the metal being fully smelted is low. At this time, the delay needs to be extended to ensure the smelting effect. For this reason, step three also includes: after predicting the predicted value of the temperature change over time during this smelting production, the control module determines the heating rate, and when the heating rate exceeds the threshold, the delay is corrected downward, and when the heating rate is lower than the threshold, the delay is corrected upward. Then, in steps five and six, the delay that is corrected upward or downward is used;

[0071] By making the control module correct the delay downward when the heating rate exceeds the threshold, when the heating rate is lower than the threshold, it means that the heating efficiency of the metal is slow after the alternating electric field is passed, and the probability of the metal being fully melted is low, and the delay needs to be extended. The delay is corrected upward to ensure the melting effect. When the metal is subjected to the same alternating electric field and the temperature is higher, it means the heating rate is faster and the metal is more likely to be fully melted. At this time, the threshold is corrected downward to ensure operating efficiency.

[0072] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A smelting method based on a large-tonnage dual-frequency induction smelting electric furnace, characterized in that: The following steps are included Step 1: Carry out a new smelting production; Step 2: The detection module detects the temperature data of the temperature change over time and uploads it to the control module. The control module archives and organizes the temperature data of the same production batch as historical data; Step 3: The control module uses a machine learning algorithm to predict the predicted value of the temperature change over time during this smelting production based on historical data, and calculates the time when the predicted value reaches the switching threshold; Step 4: The control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the deviation threshold. If the judgment result is yes, execute step 5, otherwise execute step 6; Step 5: The control module switches the induction furnace from the medium frequency mode to the low frequency mode according to the time when the switching threshold is reached in the predicted value, and executes step 7; Step 6: The control module determines whether the detection value reaches the switching threshold. If the determination result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode, and step 7 is executed; Step 7: Repeat steps 1 to 6.

2. A smelting method based on a large-tonnage dual-frequency induction smelting electric furnace according to claim 1, characterized in that: The step three also includes: the control module determines the algorithm parameters (p, d, q) using the ARIMA algorithm according to the power consumption data archived by time, and calculates the predicted value of the temperature change over time during this smelting production using the ARIMA algorithm.

3. The smelting method based on a large-tonnage dual-frequency induction smelting electric furnace according to claim 1 is characterized in that: The step five also includes: the control module determines whether the average of the predicted value and the detected value at the same time in the predicted value reaches the switching threshold. When the judgment result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode, and step seven is executed.

4. The smelting method based on a large-tonnage dual-frequency induction smelting furnace according to claim 1 is characterized in that: The step 2 also includes: a plurality of temperature sensors in the detection module detect temperature data of temperature changes over time and upload the temperature data group to the control module, and the control module archives and organizes the temperature data of the same production batch as historical data; the step 3 also includes: the control module determines whether the temperature variance in the temperature data group uploaded five times in a row exceeds the variance threshold, and when the judgment result is yes, the control module upwardly corrects the deviation threshold, and the step 4 also includes: the control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the upwardly corrected deviation threshold, when the judgment result is yes, execute step 5, otherwise execute step 6.

5. A smelting method based on a large-tonnage dual-frequency induction smelting electric furnace according to claim 4, characterized in that: The step three also includes: the control module is pre-input with a variance threshold F0, the control module determines whether the temperature variance in the temperature data group uploaded five times continuously exceeds the variance threshold, and when the judgment result is yes, the control module increases the deviation threshold by A1 times, wherein A1=F / F0×c; the step four also includes: the control module determines whether the difference between the detection value of the detection module and the predicted value at the same time reaches the deviation threshold after increasing by A1 times, when the judgment result is yes, execute step five, otherwise execute step six.

6. A smelting method based on a large-tonnage dual-frequency induction melting furnace according to claim 1, characterized in that: The step five also includes: the control module switches the induction furnace from the medium frequency mode to the low frequency mode after a delay according to the time when the switching threshold is reached in the predicted value, and executes step seven; the step six also includes: the control module determines whether the detection value reaches the switching threshold, and when the judgment result is yes, the induction furnace is switched from the medium frequency mode to the low frequency mode after a delay, and executes step seven.

7. A smelting method based on a large-tonnage dual-frequency induction smelting electric furnace according to claim 6, characterized in that: The step one also includes: performing a new smelting production, inputting the melting point of this smelting, and the control module increases the delay when the melting point exceeds the melting point threshold, and reduces the delay when the melting point is lower than the melting point threshold.

8. A smelting method based on a large-tonnage dual-frequency induction smelting furnace according to claim 7, characterized in that: The step three also includes: after predicting the predicted value of the temperature change over time during this smelting production, the control module determines the heating rate, and corrects the delay downward when the heating rate exceeds the rate threshold, and corrects the delay upward when the heating rate is lower than the rate threshold.

Images