Negative pressure isolation type diversion system and method for eutectoid tank blanking pipe

Abstract

The invention relates to the technical field of data processing, and discloses a eutectoid tank blanking pipe negative pressure isolation type diversion system and method, and the system comprises a gas film module which is used for obtaining the current blanking data, determining the control parameters of a porous throttler according to the current blanking data, and controlling the gas film thickness of the porous throttler during blanking according to the control parameters; the calculation module is used for acquiring the change condition of the content of silicon entering the molten salt system in the blanking process, and calculating a gas film thickness correction value according to the change condition of the silicon content; and the correction module is used for correcting the gas film thickness of the porous throttler according to the gas film thickness correction value to obtain the final gas film thickness. According to the invention, negative pressure interference in the blanking process of the aluminum-silicon oxide can be effectively counteracted, so that the aluminum-silicon oxide can sufficiently enter a molten salt system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, more particularly, to a eutectoid tank underflow pipe negative pressure isolation type flow guide system and method. BACKGROUND

[0002] High-alumina fly ash and high-alumina coal gangue are one of the main solid wastes generated in the process of coal resource development and utilization. High-alumina fly ash and high-alumina coal gangue are used to extract aluminum-silicon oxides mainly composed of alumina and silicon dioxide, and then the aluminum-silicon oxides are used as raw materials to produce aluminum-silicon alloy liquid, which is an effective way to utilize high-alumina fly ash and high-alumina coal gangue.

[0003] In the prior art, when producing aluminum-silicon alloy liquid, due to the large drop of the underflow pipe of the eutectoid tank material box, under the action of the purification negative pressure in the experimental environment, part of the aluminum-silicon oxides is prone to external circulation and cannot enter the molten salt system in sufficient amount, resulting in uncontrollable silicon content of the eutectoid aluminum-silicon alloy. SUMMARY

[0004] The present application provides a eutectoid tank underflow pipe negative pressure isolation type flow guide system and method to solve the problem that the aluminum-silicon oxides are easily disturbed by the purification negative pressure during underflow in the prior art, comprising: A gas film module is configured to obtain current underflow data, determine control parameters of a multi-hole restrictor according to the current underflow data, and control the gas film thickness of the multi-hole restrictor during underflow according to the control parameters; a calculation module is configured to obtain the change of silicon content entering the molten salt system during underflow, and calculate a gas film thickness correction value according to the change of silicon content; and a correction module is configured to correct the gas film thickness of the multi-hole restrictor according to the gas film thickness correction value to obtain a final gas film thickness.

[0005] Further, the gas film module determines the control parameters of the multi-hole restrictor according to the current underflow data, and controls the gas film thickness of the multi-hole restrictor during underflow according to the control parameters, including: obtaining material characteristic data and underflow operation parameters, determining pressure parameters of each point of the flow guide pipe according to the material characteristic data and the underflow operation parameters; clustering each point of the flow guide pipe according to the pressure parameters, determining the cluster centers of the corresponding clusters of each point according to the clustering results; determining interference parameters according to the cluster centers of the points, determining the gas film thickness control values of the corresponding point regions according to the interference parameters, and controlling the multi-hole restrictor according to the gas film thickness control values.

[0006] Further, the pressure parameters of each point of the flow guide pipe are determined according to the material characteristic data and the blanking operation parameters, including: obtaining historical blanking data, determining historical material characteristic data, historical blanking operation parameters and corresponding pressure distribution parameters according to the historical blanking data, establishing a training sample set according to the historical material characteristic data, the historical blanking operation parameters and the corresponding pressure distribution parameters, establishing an initial pressure evaluation model according to the training sample set and training the initial pressure evaluation model to obtain a trained pressure evaluation model, inputting the current material characteristic data and the blanking operation parameters into the trained pressure evaluation model to obtain the pressure distribution parameters in the current flow guide pipe, and determining the pressure parameters of each point of the flow guide pipe according to the pressure distribution parameters in the current flow guide pipe.

[0007] Further, the points of the flow guide pipe are clustered according to the pressure parameters, including: establishing a sample data set according to the pressure parameters of each point, randomly selecting k cluster centers in the sample data set, calculating the Euclidean distance of the pressure parameters in the sample data set to the initial cluster centers, dividing each point to the corresponding cluster according to the Euclidean distance of the pressure parameters in the sample data set to the initial cluster centers, calculating the mean value of the pressure parameters in each cluster, and recalculating the cluster centers according to the mean value of the pressure parameters in each cluster; iteratively calculating the cluster centers of each cluster until the cluster centers no longer change or the number of iterations reaches a preset maximum number of iterations, and obtaining the clustering results of each point.

[0008] Further, the gas film thickness control value of the corresponding point region is determined according to the interference parameter, and the multi-hole restrictor is controlled according to the gas film thickness control value, including: obtaining a preset allowable interference parameter, calculating the difference between the interference parameter of any point and the preset allowable interference parameter, and judging whether the difference between the interference parameter and the preset allowable interference parameter is greater than a first preset threshold; if the difference between the interference parameter and the preset allowable interference parameter is greater than the first preset threshold, a first gas film thickness is set as the gas film thickness control value of the corresponding point region; if the difference between the interference parameter and the preset allowable interference parameter is less than or equal to the first preset threshold, it is judged whether the difference between the interference parameter and the preset allowable interference parameter is greater than a second preset threshold; if the difference between the interference parameter and the preset allowable interference parameter is greater than the second preset threshold, a second gas film thickness is set as the gas film thickness control value of the corresponding point region; if the difference between the interference parameter and the preset allowable interference parameter is less than or equal to the second preset threshold, a third gas film thickness is set as the gas film thickness control value of the corresponding point region; the gas film thickness control values of all point regions are obtained, and the multi-hole restrictor is controlled to generate a gas film according to the gas film thickness control values of all point regions.

[0009] Further, the computing module is configured to calculate the gas film thickness correction value according to the silicon content variation, including: drawing a silicon content variation curve according to the silicon content variation, performing curve fitting on the silicon content variation curve, and determining a silicon content fitting curve according to the curve fitting result; obtaining a silicon content standard growth value in a preset time period, determining a predicted time when the silicon content reaches the silicon content standard growth value according to the silicon content fitting curve; calculating a difference value between the preset time period and the predicted time, and determining the gas film thickness correction value according to the difference value; obtaining a fluctuation of the silicon content variation curve, adjusting the gas film thickness correction value according to the fluctuation of the silicon content variation curve, and obtaining a final gas film thickness correction value.

[0010] Further, the obtaining of the fluctuation of the silicon content variation curve includes: dividing the silicon content variation curve into fluctuation windows, segmenting the silicon content variation curve according to the fluctuation windows to obtain a plurality of silicon content curve segments, calculating silicon content average values of the silicon content curve segments, and drawing an average value variation curve according to the silicon content average values of all the silicon content curve segments.

[0011] Further, the dividing of the silicon content variation curve into fluctuation windows includes: extracting inflection points of the silicon content variation curve, setting a window between adjacent two inflection points as an initial fluctuation window, calculating a variance of the silicon content in the initial fluctuation window, performing negative correlation mapping and normalization processing on the variance of the silicon content in the initial fluctuation window to obtain a stability index of the initial fluctuation window, obtaining a preset standard stability threshold, and determining whether the stability index of the initial fluctuation window is greater than or equal to the preset standard stability index; if the stability index of the initial fluctuation window is greater than or equal to the preset standard stability index, the corresponding initial fluctuation window is determined as the fluctuation window of the silicon content variation curve; if the stability index of the initial fluctuation window is less than the preset standard stability index, the initial fluctuation window is equally divided into two sub-fluctuation windows, and the stability index of each sub-fluctuation window is calculated; the sub-fluctuation windows are iteratively divided according to the stability index, and the stability index is calculated until the stability index of all the sub-fluctuation windows is greater than or equal to the preset standard stability index, and the dividing of the silicon content variation curve into fluctuation windows is completed.

[0012] Further, the correction module is configured to correct the gas film thickness of the multi-hole throttler according to the gas film thickness correction value to obtain a final gas film thickness, including: correcting the gas film thickness of the multi-hole throttler according to a correction formula, the correction formula being, , wherein, is a corrected gas film thickness control value, is a correction value of the gas film thickness before correction, is the gas film thickness correction value. This is a preset range adjustment coefficient. It is a natural exponential function.

[0013] To achieve the above objectives, the present invention also provides a negative pressure isolation flow guiding method for the feed pipe of a eutectoid tank, comprising: Obtain the current feeding data, determine the control parameters of the porous throttle based on the current feeding data, and control the gas film thickness of the porous throttle during feeding according to the control parameters; obtain the change in silicon content entering the molten salt system during the feeding process, calculate the gas film thickness correction value based on the change in silicon content; correct the gas film thickness of the porous throttle according to the gas film thickness correction value to obtain the final gas film thickness.

[0014] The beneficial effects of this invention are as follows: By applying the above technical solutions, this invention improves the material feeding environment into the molten salt system by setting up a porous throttling device and distributing the gas film thickness control, offsetting the purification negative pressure, and enabling aluminum and silicon oxides to enter the molten salt system in sufficient quantity. At the same time, the gas film thickness is corrected by silicon content feedback, ensuring the long-term, high-precision stability of the final product composition. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This invention provides a schematic diagram of a negative pressure isolation flow guiding system for a eutectoid tank feed pipe according to an embodiment of the present invention. Figure 2 The diagram shows the overall flow chart of a negative pressure isolation flow guiding method for the feed pipe of a eutectoid tank according to an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] This application provides a negative pressure isolation flow guiding system for the feed pipe of a eutectoid tank, such as... Figure 1 As shown, it includes: The gas film module is configured to obtain current discharging data, determine control parameters of the porous restrictor according to the current discharging data, and control the gas film thickness of the porous restrictor during discharging according to the control parameters; the calculation module is configured to obtain the change of the silicon content in the molten salt system during discharging, and calculate a gas film thickness correction value according to the change of the silicon content; and the correction module is configured to correct the gas film thickness of the porous restrictor according to the gas film thickness correction value, and obtain a final gas film thickness.

[0019] In the embodiment, a high-temperature-resistant flexible guide pipe is additionally arranged at the discharging pipe of the eutectic tank, so that the material is discharged along the guide pipe; a porous restrictor is arranged on the inner wall of the guide pipe, a gas film is generated by the porous restrictor to isolate the material by negative pressure, and the gas film thickness of the porous restrictor is corrected by real-time monitoring of the change of the silicon content of the aluminum-silicon alloy, so as to ensure that the material is sufficiently discharged into the molten salt system.

[0020] In some embodiments of the present application, the gas film module determines control parameters of the porous restrictor according to current discharging data, and controls the gas film thickness of the porous restrictor during discharging according to the control parameters, which includes: obtaining material characteristic data and discharging operation parameters, determining pressure parameters of each point of the guide pipe according to the material characteristic data and the discharging operation parameters; clustering each point of the guide pipe according to the pressure parameters, determining the cluster center of each point corresponding to the cluster according to the clustering result; determining the interference parameter according to the cluster center of the point, determining the gas film thickness control value of the corresponding point area according to the interference parameter, and controlling the porous restrictor according to the gas film thickness control value.

[0021] In the embodiment, the material characteristic data is the density, particle size distribution of aluminum-silicon oxide, and the discharging operation parameters are the discharging drop and the discharging frequency; a plurality of point areas are uniformly distributed on the guide pipe; the pressure parameters of each point of the guide pipe are determined according to the material characteristic data and the discharging operation parameters; each point is divided into a corresponding cluster according to the pressure parameters; the cluster center value corresponding to the cluster is determined as the interference parameter; and the gas film thickness of the porous restrictor corresponding to each point area is set through the interference parameter.

[0022] In some embodiments of the present application, the determination of the pressure parameters of each point of the guide pipe according to the material characteristic data and the discharging operation parameters includes: obtaining historical discharging data, determining historical material characteristic data, historical discharging operation parameters and corresponding pressure distribution parameters according to the historical discharging data, establishing a training sample set according to the historical material characteristic data, the historical discharging operation parameters and the corresponding pressure distribution parameters, establishing an initial pressure evaluation model according to the training sample set and training the initial pressure evaluation model to obtain a trained pressure evaluation model, inputting the current material characteristic data and the discharging operation parameters into the trained pressure evaluation model to obtain the pressure distribution parameters in the current guide pipe, and determining the pressure parameters of each point of the guide pipe according to the pressure distribution parameters in the current guide pipe.

[0023] In this embodiment, the discharging data in the historical discharging process is collected, the material characteristic data, the discharging operation parameters and the corresponding pressure distribution parameters in the historical discharging process are taken as a group of sample data, the total material characteristic data, the discharging operation parameters and the respective corresponding pressure parameters in the historical discharging process are counted and preprocessed, the preprocessing at least includes one of the following: data cleaning, normalization, feature construction and feature extraction, the training sample set is established through the preprocessed data, the material characteristic data and the discharging operation parameters in the training sample set are taken as input variables, the pressure parameters are taken as output variables, the pressure evaluation model is obtained through the deep learning neural network model established by the training sample set, the network structure is ResNet50, 80% of the training sample set is taken as a training set and 20% is taken as a test set, the mean square error loss function is used, the Adam is used as the optimizer, and finally the trained pressure evaluation model is obtained through the training.

[0024] In some embodiments of the present application, the clustering of each point of the draft tube according to the pressure parameter comprises: establishing a sample data set according to the pressure parameters of each point, randomly selecting k cluster centers in the sample data set; calculating the Euclidean distance of the pressure parameters in the sample data set to the initial cluster center, and dividing each point to the corresponding cluster according to the Euclidean distance of the pressure parameters in the sample data set to the initial cluster center; calculating the mean value of the pressure parameters in each cluster, and recalculating the cluster center according to the mean value of the pressure parameters in each cluster; iteratively calculating the cluster center of each cluster until the cluster center no longer changes or the number of iterations reaches a preset maximum number of iterations, and obtaining the clustering result of each point.

[0025] In this embodiment, the k-means clustering algorithm is used to cluster each point according to the pressure parameter, and each point is divided into a corresponding cluster. The k value of this embodiment is set by the number of points, and the more the number of points, the higher the corresponding k value.

[0026] In some embodiments of the present application, the determination of the gas film thickness control value of the corresponding point region according to the interference parameter, and the control of the multi-hole restrictor according to the gas film thickness control value, comprises: obtaining a preset allowable interference parameter, calculating the difference between the interference parameter of any point and the preset allowable interference parameter, and judging whether the difference between the interference parameter and the preset allowable interference parameter is greater than a first preset threshold; if the difference between the interference parameter and the preset allowable interference parameter is greater than the first preset threshold, the first gas film thickness is set as the gas film thickness control value of the corresponding point region; if the difference between the interference parameter and the preset allowable interference parameter is less than or equal to the first preset threshold, it is judged whether the difference between the interference parameter and the preset allowable interference parameter is greater than a second preset threshold; if the difference between the interference parameter and the preset allowable interference parameter is greater than the second preset threshold, the second gas film thickness is set as the gas film thickness control value of the corresponding point region; if the difference between the interference parameter and the preset allowable interference parameter is less than or equal to the second preset threshold, the third gas film thickness is set as the gas film thickness control value of the corresponding point region; the gas film thickness control values of all point regions are obtained, and the multi-hole restrictor is controlled to generate a gas film according to the gas film thickness control values of all point regions.

[0027] In the present embodiment, the first gas film thickness > the second gas film thickness > the third gas film thickness, the gas film thickness control value of the corresponding point is set by the difference between the interference parameter and the preset allowable interference parameter, the greater the difference, the higher the corresponding gas film thickness control value, and the multi-hole restrictor is controlled in a distributed gas film.

[0028] In some embodiments of the present application, the calculation module calculates the gas film thickness correction value according to the silicon content variation, comprising: drawing a silicon content variation curve according to the silicon content variation, curve fitting the silicon content variation curve, and determining a silicon content fitting curve according to the curve fitting result; obtaining a silicon content standard growth value in a preset period, determining a prediction time when the silicon content reaches the silicon content standard growth value according to the silicon content fitting curve; calculating the difference between the preset period and the prediction time, and determining the gas film thickness correction value according to the difference between the preset period and the prediction time; obtaining the fluctuation of the silicon content variation curve, adjusting the gas film thickness correction value according to the fluctuation of the silicon content variation curve, and obtaining the final gas film thickness correction value.

[0029] In the present embodiment, the silicon content fitting curve is obtained by curve fitting the silicon content variation curve based on the least square method, the silicon content growth is predicted through the silicon content fitting curve, the difference between the preset period and the prediction time when the silicon content reaches the silicon content standard growth value is set as the gas film thickness correction value, the fluctuation parameter is calculated according to the fluctuation of the silicon content variation curve, the standardized processing of the fluctuation parameter is carried out, the standardized processing of the fluctuation parameter is carried out, the standardized processing of the fluctuation parameter is carried out, and the standardized processing of the fluctuation parameter is carried out.

[0030] In some embodiments of the present application, the fluctuation of the silicon content change curve is obtained by: dividing the silicon content change curve into fluctuation windows, segmenting the silicon content change curve according to the fluctuation windows to obtain a plurality of silicon content curve segments, calculating the average silicon content of each silicon content curve segment, and drawing an average value change curve according to the average silicon content of all silicon content curve segments; and calculating the absolute value of the slope of adjacent average silicon contents in the average value change curve, and calculating the average value of all absolute values of the slope to obtain the fluctuation parameter of the silicon content change curve.

[0031] In the present embodiment, the average absolute value of the slope of the average silicon content in each fluctuation window is calculated from the silicon content change curve, and the average absolute value of the slope is taken as the fluctuation parameter of the silicon content change curve.

[0032] In some embodiments of the present application, the silicon content change curve is divided into fluctuation windows by: extracting inflection points of the silicon content change curve, setting a window between adjacent two inflection points as an initial fluctuation window, calculating the variance of the silicon content in the initial fluctuation window, performing a negative correlation mapping on the variance of the silicon content in the initial fluctuation window and normalizing the variance to obtain a stability index of the initial fluctuation window, obtaining a preset standard stability threshold, and determining whether the stability index of the initial fluctuation window is greater than or equal to the preset standard stability index; if the stability index of the initial fluctuation window is greater than or equal to the preset standard stability index, the corresponding initial fluctuation window is determined as the fluctuation window of the silicon content change curve; and if the stability index of the initial fluctuation window is less than the preset standard stability index, the initial fluctuation window is divided into two sub-fluctuation windows, and the stability index of each sub-fluctuation window is calculated; the sub-fluctuation windows are iteratively divided according to the stability index and the stability index is calculated until the stability index of all sub-fluctuation windows is greater than or equal to the preset standard stability index, and the fluctuation window of the silicon content change curve is divided.

[0033] In the present embodiment, in order to improve the accuracy of the fluctuation parameter calculation, a window with stable silicon content change should be selected as the fluctuation window, therefore, an initial fluctuation window is divided between adjacent two inflection points of the silicon content change curve, the variance of the silicon content in the initial fluctuation window is calculated, the variance is negatively correlated by an inverse proportional calculation method, the normalized variance after the negative correlation is processed to obtain the stability index of the corresponding initial fluctuation window, the initial fluctuation window with a stability index greater than or equal to the preset standard stability index is determined as the fluctuation window, the initial fluctuation window with a stability index less than the preset standard stability index is divided into two sub-fluctuation windows and the stability index of the sub-fluctuation window is recalculated, and if the stability index is still less than the preset standard stability index, the division is continued until the stability index of all sub-fluctuation windows is greater than or equal to the preset standard stability index, and the fluctuation window division result of the silicon content change curve is obtained.

[0034] In some embodiments of the present application, the correction module is configured to correct the gas film thickness of the multi-hole throttler according to the gas film thickness correction value to obtain a final gas film thickness, including: correcting the gas film thickness of the multi-hole throttler according to a correction formula, wherein the correction formula is, , wherein, is a corrected gas film thickness control value, is a gas film thickness correction value before correction, is a gas film thickness correction value, is a preset range adjustment coefficient, is a natural exponential function.

[0035] In the present embodiment, the gas film thickness of the multi-hole throttler is corrected by the correction formula, and the preset range adjustment coefficient is set to 2, so as to correct the gas film thickness correction value.

[0036] Based on the same technical concept, as shown in Figure 2 , the present application also provides a negative pressure isolation type flow guiding method for a eutectic trough discharge pipe, including: S101, obtaining current discharge data, determining the control parameter of the multi-hole throttler according to the current discharge data, and controlling the gas film thickness of the multi-hole throttler during discharge according to the control parameter; S102, obtaining the change of silicon content into the molten salt system during the discharge process, and calculating the gas film thickness correction value according to the change of silicon content; S103, correcting the gas film thickness of the multi-hole throttler according to the gas film thickness correction value to obtain a final gas film thickness.

[0037] By applying the above technical solutions, the present application obtains current discharge data by the gas film module, determines the control parameter of the multi-hole throttler according to the current discharge data, and controls the gas film thickness of the multi-hole throttler during discharge according to the control parameter; the calculation module is configured to obtain the change of silicon content into the molten salt system during the discharge process, and calculate the gas film thickness correction value according to the change of silicon content; the correction module is configured to correct the gas film thickness of the multi-hole throttler according to the gas film thickness correction value to obtain a final gas film thickness. The present application can effectively offset the negative pressure interference in the aluminum-silicon oxide discharge process, so that it can enter the molten salt system in sufficient amount.

[0038] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the same; although the present application has been described in detail with reference to the foregoing examples, those of ordinary skill in the art will understand that they can still modify the technical solutions described in the foregoing examples, or make equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A negative pressure isolation flow guiding system for the feed pipe of a eutectoid tank, characterized in that, include: The air film module is used to acquire the current feeding data, determine the control parameters of the porous throttle based on the current feeding data, and control the air film thickness of the porous throttle during feeding based on the control parameters. The calculation module is used to obtain the change in silicon content entering the molten salt system during the feeding process, and to calculate the correction value for the gas film thickness based on the change in silicon content. The correction module is used to correct the gas film thickness of the porous throttle according to the gas film thickness correction value to obtain the final gas film thickness.

2. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 1, characterized in that, The air film module determines the control parameters of the porous throttle based on the current feeding data, and controls the air film thickness of the porous throttle during feeding according to the control parameters, including: Obtain material characteristic data and feeding operation parameters, and determine the pressure parameters at each point of the guide pipe based on the material characteristic data and feeding operation parameters; Clustering is performed on each point of the guide tube based on the pressure parameters, and the cluster center of the corresponding cluster is determined based on the clustering results. The interference parameters are determined based on the cluster centers of the points, the air film thickness control value for the corresponding point area is determined based on the interference parameters, and the porous throttling device is controlled based on the air film thickness control value.

3. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 2, characterized in that, The process of determining the pressure parameters at each point on the guide pipe based on material characteristic data and feeding operation parameters includes: Acquire historical material feeding data, determine historical material characteristic data, historical material feeding operation parameters and corresponding pressure distribution parameters based on historical material feeding data, and establish a training sample set based on historical material feeding data, historical material feeding operation parameters and corresponding pressure distribution parameters; An initial stress assessment model is established based on the training sample set, and the initial stress assessment model is trained to obtain a trained stress assessment model. Input the current material characteristic data and feeding operation parameters into the trained pressure assessment model to obtain the pressure distribution parameters in the current guide tube. Based on the pressure distribution parameters in the current guide tube, determine the pressure parameters at each point in the guide tube.

4. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 2, characterized in that, The clustering of points in the guide tube based on pressure parameters includes: A sample dataset is established based on the pressure parameters of each location, and k cluster centers are randomly selected from the sample dataset. Calculate the Euclidean distance from the stress parameters in the sample dataset to the initial cluster centers, and assign each point to its corresponding cluster based on the Euclidean distance from the stress parameters in the sample dataset to the initial cluster centers; Calculate the mean stress parameter within each cluster, and recalculate the cluster centers based on the mean stress parameter within each cluster; The cluster centers of each cluster are calculated iteratively until the cluster centers no longer change or the number of iterations reaches the preset maximum number of iterations, thus obtaining the clustering results for each point.

5. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 2, characterized in that, The step of determining the air film thickness control value for the corresponding location area based on the interference parameters, and controlling the porous throttle according to the air film thickness control value, includes: Obtain the preset allowable interference parameter, calculate the difference between the interference parameter at any point and the preset allowable interference parameter, and determine whether the difference between the interference parameter and the preset allowable interference parameter is greater than the first preset threshold. If the difference between the interference parameter and the preset allowable interference parameter is greater than the first preset threshold, the first air film thickness is set to the air film thickness control value of the corresponding point area. If the difference between the interference parameter and the preset allowable interference parameter is less than or equal to the first preset threshold, then determine whether the difference between the interference parameter and the preset allowable interference parameter is greater than the second preset threshold. If the difference between the interference parameter and the preset allowable interference parameter is greater than the second preset threshold, the second air film thickness is set to the air film thickness control value of the corresponding point area. If the difference between the interference parameter and the preset allowable interference parameter is less than or equal to the second preset threshold, the third air film thickness is set to the air film thickness control value of the corresponding point area. Obtain the air film thickness control value for all locations, and control the porous throttling device to generate air film based on the air film thickness control value for all locations.

6. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 1, characterized in that, The calculation module calculates the gas film thickness correction value based on the change in silicon content, including: Based on the changes in silicon content, a curve of silicon content change is plotted, and curve fitting is performed on the curve of silicon content change. Based on the curve fitting results, the fitted curve of silicon content is determined. Obtain the standard growth value of silicon content within a preset time period, and determine the predicted time when the silicon content reaches the standard growth value of silicon content based on the silicon content fitting curve. Calculate the difference between the preset time period and the predicted time, and determine the air film thickness correction value based on the difference between the preset time period and the predicted time. The fluctuation of the silicon content change curve is obtained, and the gas film thickness correction value is adjusted according to the fluctuation of the silicon content change curve to obtain the final gas film thickness correction value.

7. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 6, characterized in that, The fluctuation of the silicon content change curve is obtained, including: The silicon content change curve is divided into fluctuation windows, and the silicon content change curve is further divided according to the fluctuation windows to obtain several silicon content curve segments; Calculate the average silicon content for each silicon content curve segment, and plot the average value variation curve based on the average silicon content for all silicon content curve segments. Calculate the absolute value of the slope of adjacent silicon content average values ​​in the average value change curve, and calculate the average value of all absolute slope values ​​to obtain the fluctuation parameter of the silicon content change curve.

8. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 7, characterized in that, The process of dividing the silicon content variation curve into fluctuation windows includes: Extract the inflection point of the silicon content change curve and set the window between two adjacent inflection points as the initial fluctuation window; Calculate the variance of silicon content in the initial fluctuation window, perform negative correlation mapping on the variance of silicon content in the initial fluctuation window and normalize it to obtain the stability index of the initial fluctuation window. Obtain the preset standard stability threshold and determine whether the stability index of the initial fluctuation window is greater than or equal to the preset standard stability index; If the stability index of the initial fluctuation window is greater than or equal to the preset standard stability index, then the corresponding initial fluctuation window will be determined as the fluctuation window of the silicon content change curve. If the stability index of the initial fluctuation window is less than the preset standard stability index, the initial fluctuation window is divided into two sub-fluctuation windows, and the stability index of each sub-fluctuation window is calculated. The stability index is iterated through the sub-variable fluctuation window and the stability index is calculated until the stability index of all sub-variable windows is greater than or equal to the preset standard stability index, thus completing the division of the fluctuation window for the silicon content change curve.

9. The eutectoid tank feed pipe negative pressure isolation diversion system according to claim 7, characterized in that, The correction module is used to correct the gas film thickness of the porous throttle according to the gas film thickness correction value to obtain the final gas film thickness, including: The gas film thickness of the porous throttling device is corrected according to a correction formula, which is as follows: ， in, To correct the air film thickness control value, To correct the previous film thickness correction value, This is the correction value for the air film thickness. This is a preset range adjustment coefficient. It is a natural exponential function.

10. A negative pressure isolation flow guiding method for the feed pipe of a eutectoid tank, characterized in that, include: Obtain the current feeding data, determine the control parameters of the porous throttle based on the current feeding data, and control the air film thickness of the porous throttle during feeding based on the control parameters; The changes in silicon content entering the molten salt system during the feeding process are obtained, and the correction value for gas film thickness is calculated based on the changes in silicon content. The gas film thickness of the porous throttling device is corrected based on the gas film thickness correction value to obtain the final gas film thickness.

Images