A high-precision chromium alloy continuous casting forming monitoring method and system

Abstract

This invention discloses a high-precision monitoring method and system for continuous casting of chromium alloys, relating to the field of chromium alloy forming monitoring technology. The method includes the following steps: acquiring factors affecting the quality of continuously cast chromium alloy parts; obtaining multiple sets of continuous casting sample data; analyzing the multiple sets of continuous casting sample data and establishing a Logistic regression model; monitoring process parameters during the continuous casting process in real time and determining whether they are within the standard forming parameter range; and analyzing abnormal forming parameters using the Logistic regression model to determine whether there is a risk of substandard quality in the continuously cast chromium alloy parts. The advantages of this invention are: establishing a Logistic regression model based on the factors affecting the quality of continuously cast chromium alloy parts allows for rapid identification of the risk of substandard quality, achieving intelligent and high-precision monitoring of continuous casting of chromium alloys.

Description

Technical Field

[0001] The present invention relates to the technical field of chromium alloy forming monitoring, and particularly to a high-precision chromium alloy continuous casting forming monitoring method and system. Background Art

[0002] Chromium alloy is an alloy composed of chromium as the base and other elements added, belonging to refractory alloys. Compared with metallic nickel, metallic chromium has a high melting point, a large specific strength, and has good oxidation resistance and corrosion resistance to high sulfur, diesel fuel, and seawater.

[0003] In the existing technical solutions for monitoring the continuous casting forming of chromium alloys, the detection is usually carried out by presetting threshold values for the forming process parameters. As long as the actual processing process parameters do not exceed the preset threshold range, it is determined as qualified. On the one hand, this method lacks an analysis of the forming influence index of each forming process parameter on the casting, and on the other hand, it lacks a targeted analysis of the forming process parameters. It has a low sensitivity to abnormal values of the forming process parameters during the forming process, and it is difficult to quickly discover forming abnormalities in a timely manner, which is likely to cause a decline in the forming quality of the casting. Summary of the Invention

[0004] To solve the above technical problems, a high-precision chromium alloy continuous casting forming monitoring method and system are provided. The technical solution of the present invention solves the problems of the above existing technical solutions that on the one hand, it lacks an analysis of the forming influence index of each forming process parameter on the casting, and on the other hand, it lacks a targeted analysis of the forming process parameters. It has a low sensitivity to abnormal values of the forming process parameters during the forming process, and it is difficult to quickly discover forming abnormalities in a timely manner, which is likely to cause a decline in the forming quality of the casting.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A high-precision chromium alloy continuous casting forming monitoring method includes the following steps:

[0007] Obtain the influencing factors that affect the quality of chromium alloy continuous casting formed parts, and the influencing factors include chromium alloy melting temperature, chromium alloy pouring speed, and casting forming die temperature;

[0008] Set different chromium alloy melting temperatures, chromium alloy pouring speeds, and casting forming die temperatures according to a preset gradient for chromium alloy continuous casting forming tests, and at the same time obtain the forming quality of the chromium alloy continuous casting formed parts during the tests, and obtain multiple groups of chromium alloy continuous casting forming sample data;

[0009] Analyze multiple groups of chromium alloy continuous casting forming sample data and establish a Logistic regression model;

[0010] The chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature are monitored in real time during the continuous casting process of chromium alloy, and compared with the preset standard values ​​to determine whether they are within the standard forming parameter range.

[0011] If yes, output a normal molding parameter signal; otherwise, output an abnormal molding parameter signal.

[0012] The abnormal forming parameters are analyzed using a Logistic regression model to determine whether there is a risk of substandard quality in the chromium alloy continuous casting parts. If so, a first-level alarm signal is output; otherwise, a second-level alarm signal is output.

[0013] Preferably, the analysis of multiple sets of chromium alloy continuous casting sample data and the establishment of a Logistic regression model specifically includes the following steps:

[0014] The forming quality of chromium alloy continuous casting parts was classified into qualified and unqualified categories, and several sets of qualified chromium alloy continuous casting sample data and several sets of unqualified chromium alloy continuous casting sample data were statistically analyzed.

[0015] A ternary logistic regression model was established to evaluate the forming quality of chromium alloy continuously cast parts based on chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature.

[0016] The coefficients in the ternary Logistic regression model were estimated using the maximum likelihood method based on multiple sets of chromium alloy continuous casting sample data to obtain the model regression coefficients.

[0017] Test the significance of the model regression coefficients to determine whether the model regression coefficients meet the significance requirements;

[0018] Test the significance of the model to determine whether the model is statistically significant.

[0019] Preferably, the ternary logistic regression model is:

[0020]

[0021] In the formula, y = 1 represents that the chromium alloy continuous casting part is a qualified product, and y = 0 represents that the chromium alloy continuous casting part is an unqualified product.

[0022] P represents the predicted probability of the ternary logistic regression model;

[0023] T is the melting temperature of the chromium alloy;

[0024] v represents the casting speed of the chromium alloy;

[0025] t is the temperature of the casting mold;

[0026] α, β1, β2 and β3 are all coefficients of the ternary logistic regression model.

[0027] Preferably, the standard forming parameter ranges for the chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature are calculated as follows:

[0028] The system outputs control commands to the continuous casting forming system according to the optimal process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature.

[0029] The actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature are obtained by detecting the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the continuous casting process.

[0030] Obtain actual process data on chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature for several sets of qualified chromium alloy continuous casting parts.

[0031] The actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature for several groups of qualified chromium alloy continuous casting parts were analyzed to obtain the standard forming parameter ranges for chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature.

[0032] Preferably, the analysis of the actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature for several groups of qualified chromium alloy continuously cast parts specifically includes the following steps:

[0033] Arrange the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data of several qualified chromium alloy continuous casting parts in ascending order.

[0034] Set the detection level to obtain the critical value bp(n) for kurtosis test;

[0035] Calculate the kurtosis test value bk(n) for each chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data;

[0036] Determine whether the kurtosis test value bk(n) of the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data is greater than the critical value bp(n) of the kurtosis test. If yes, then remove the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data. If no, then do not respond.

[0037] After calculating the average and standard deviation of the remaining chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data, the standard forming parameter range for the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data is determined to be: in, S is the average of the remaining chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data, and S is the standard deviation of the remaining chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data.

[0038] Preferably, the kurtosis test value bk(n) is calculated as follows:

[0039]

[0040] In the formula, n is the number of the currently calculated chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data sorted from smallest to largest.

[0041] The average value of all chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data.

[0042] x i This refers to the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data arranged in ascending order before n.

[0043] Preferably, the step of analyzing abnormal forming parameters using a Logistic regression model to determine whether there is a risk of substandard quality in chromium alloy continuously cast parts specifically includes the following steps:

[0044] Substitute the abnormal forming chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature into the ternary logistic regression model of forming quality - chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature to obtain the non-conforming probability value under the current process parameters.

[0045] The non-conformance probability value under the current process parameters is compared with the preset threshold. If it is greater than the preset threshold, the risk of non-conformance is determined to be high, and a first-level alarm signal is output. If not, the risk of non-conformance is determined to be low, and a second-level alarm signal is output.

[0046] Preferably, the preset threshold value range is set to 0.15-0.3.

[0047] Furthermore, a high-precision monitoring system for continuous casting of chromium alloys is proposed to realize the high-precision monitoring method for continuous casting of chromium alloys as described above, including:

[0048] The control module is used to output control signals to each component, thereby controlling each component.

[0049] The process parameter monitoring module is electrically connected to the control module. The process parameter monitoring module is used to collect the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the continuous casting process of chromium alloy, and feed them back to the control module.

[0050] The data processing module is electrically connected to the control module. The data processing module is used to analyze multiple sets of chromium alloy continuous casting sample data, establish a Logistic regression model, monitor the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the chromium alloy continuous casting process, and compare them with preset standard values ​​to determine whether they are within the standard forming parameter range. The abnormal forming parameters are analyzed through the Logistic regression model to determine whether there is a risk of unqualified chromium alloy continuous casting parts.

[0051] The data storage module is electrically connected to the control module and the data processing module, and is used to store the Logistic regression model.

[0052] The signal output module is electrically connected to the control module and the data processing module, and is used to output alarm signals.

[0053] Optionally, the process parameter monitoring module includes:

[0054] A first temperature sensor is used to detect the melting temperature of chromium alloy.

[0055] A flow sensor is used to detect the casting speed of chromium alloy;

[0056] The second temperature sensor is used to detect the temperature of the casting mold.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] This invention proposes a high-precision monitoring method for continuous casting of chromium alloys. Based on the influencing factors of the quality of continuously cast chromium alloy parts, a Logistic regression model is established to obtain the influence index of each step of the forming process parameters on the quality of the cast parts. Then, by substituting the actual process parameters in the continuous casting process of chromium alloys into the Logistic regression model, the risk of non-conforming quality of continuously cast chromium alloy parts can be obtained, thus realizing intelligent and high-precision monitoring of continuous casting of chromium alloys.

[0059] This solution calculates the standard forming parameter ranges for chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature based on the actual process values ​​of chromium alloy continuous casting. It can identify fluctuations in normal processing parameters, avoid false alarms, greatly ensure the accuracy of alarms during the chromium alloy continuous casting process, and facilitate real-time monitoring of the chromium alloy continuous casting process by staff. Attached Figure Description

[0060] Figure 1 This is a flowchart of steps S100-S600 of the high-precision chromium alloy continuous casting forming monitoring method proposed in this invention;

[0061] Figure 2 This is a flowchart of steps S301-S305 of the high-precision chromium alloy continuous casting forming monitoring method proposed in this invention;

[0062] Figure 3 This is a flowchart of steps S401-S404 of the high-precision chromium alloy continuous casting forming monitoring method proposed in this invention.

[0063] Figure 4 This is a flowchart of steps S405-S409 of the high-precision chromium alloy continuous casting forming monitoring method proposed in this invention;

[0064] Figure 5 This is a flowchart of steps S601-S602 of the high-precision chromium alloy continuous casting forming monitoring method proposed in this invention;

[0065] Figure 6 The diagram shows the structural block of the high-precision chromium alloy continuous casting and forming monitoring system proposed in this invention. Detailed Implementation

[0066] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0067] Reference Figure 1 As shown, a high-precision monitoring method for continuous casting of chromium alloys includes the following steps:

[0068] S100. Obtain the influencing factors that affect the quality of chromium alloy continuous casting parts. The influencing factors include chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature.

[0069] S200. According to the preset gradient, different chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature are set to carry out chromium alloy continuous casting molding test, and at the same time, the molding quality of chromium alloy continuous casting molding parts in the test is obtained, and multiple sets of chromium alloy continuous casting molding sample data are obtained.

[0070] S300. Analyze multiple sets of chromium alloy continuous casting sample data and establish a Logistic regression model;

[0071] S400: Real-time monitoring of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during continuous casting of chromium alloy, and comparison with preset standard values ​​to determine whether they are within the range of standard forming parameters.

[0072] S500: If yes, output a normal molding parameter signal; otherwise, output an abnormal molding parameter signal.

[0073] S600: Analyze the abnormal forming parameters using a Logistic regression model to determine whether there is a risk of substandard quality in the chromium alloy continuous casting parts. If yes, output a first-level alarm signal; otherwise, output a second-level alarm signal.

[0074] This solution establishes a Logistic regression model based on the influencing factors of the quality of chromium alloy continuous casting parts, obtains the influence index of each step of the forming process parameters on the quality of the casting parts, and then, by substituting the actual process parameters in the chromium alloy continuous casting process into the Logistic regression model, the risk of non-conforming chromium alloy continuous casting parts can be obtained, thus realizing intelligent and high-precision monitoring of chromium alloy continuous casting.

[0075] This solution implements a dual alarm mechanism for the continuous casting process of chromium alloys based on the dual monitoring of abnormal fluctuations in forming parameters and unqualified forming, which greatly ensures the accuracy of alarms in the continuous casting process of chromium alloys and makes it easier for staff to monitor the continuous casting process of chromium alloys in real time.

[0076] Please see Figure 2 As shown, the analysis of multiple sets of chromium alloy continuous casting sample data and the establishment of a Logistic regression model specifically include the following steps:

[0077] S301. Classify the forming quality of chromium alloy continuous casting parts into qualified and unqualified categories, and compile several sets of qualified chromium alloy continuous casting sample data and several sets of unqualified chromium alloy continuous casting sample data respectively.

[0078] S302. Establish a ternary logistic regression model for the forming quality of chromium alloy continuous casting parts, based on chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature.

[0079] S303. Estimate the coefficients in the ternary Logistic regression model using the maximum likelihood method based on multiple sets of chromium alloy continuous casting and forming sample data to obtain the model regression coefficients.

[0080] S304. Test the significance of the regression coefficients of the model and determine whether the regression coefficients of the model meet the significance requirements;

[0081] S305. Test the significance of the model to determine whether the model is statistically significant.

[0082] The ternary logistic regression model is as follows:

[0083]

[0084] In the formula, y = 1 represents that the chromium alloy continuous casting part is a qualified product, and y = 0 represents that the chromium alloy continuous casting part is an unqualified product.

[0085] P represents the predicted probability of the ternary logistic regression model;

[0086] T is the melting temperature of the chromium alloy;

[0087] v represents the casting speed of the chromium alloy;

[0088] t is the temperature of the casting mold;

[0089] α, β1, β2 and β3 are all coefficients of the ternary logistic regression model.

[0090] This scheme establishes a ternary logistic regression model based on the chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature, which affect the quality of chromium alloy continuously cast parts. The parameters of the ternary logistic regression model are calculated through multiple experimental data. The influence indicators of chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature on the quality of chromium alloy continuous casting are established. This facilitates the subsequent monitoring of forming parameters, and allows for the judgment of non-compliance risks for abnormal parameters, thereby achieving high-precision monitoring of chromium alloy continuous casting.

[0091] It is understandable that although this scheme uses the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature as the main factors affecting the quality of chromium alloy continuous casting parts in a ternary logistic regression model, this scheme can also perform logistic regression model calculations for other influencing factors.

[0092] Please see Figure 3 As shown, the calculation method for the standard forming parameter ranges of chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature is as follows:

[0093] S401. Output control commands to the continuous casting system according to the optimal process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature.

[0094] S402. Implement testing of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during continuous casting forming process to obtain the actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature.

[0095] S403. Obtain actual process data of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature for several sets of qualified chromium alloy continuous casting parts.

[0096] S404. Analyze the actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature for several groups of qualified chromium alloy continuous casting parts, and obtain the standard forming parameter ranges for chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature.

[0097] Based on actual process data during the continuous casting process of chromium alloys, the standard forming parameter ranges for chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature are calculated. This enables fluctuation analysis of standard forming parameters in actual production processes. In subsequent forming monitoring, it allows for the identification of fluctuations in normal processing parameters, avoiding false alarms.

[0098] Please see Figure 4 As shown, the analysis of actual process data on chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature for several groups of qualified chromium alloy continuously cast parts includes the following steps:

[0099] S405. Arrange the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data of several qualified chromium alloy continuous casting parts in ascending order.

[0100] S406. Set the detection level to obtain the critical value bp(n) for kurtosis testing. Under normal circumstances, the detection level ranges from 0.01 to 0.1. In this scheme, the detection level is set to 0.05.

[0101] S407. Calculate the kurtosis test value bk(n) for each chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data.

[0102] S408. Determine whether the kurtosis test value bk(n) of the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data is greater than the critical value bp(n) of the kurtosis test. If yes, then remove the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data. If no, then do not respond.

[0103] S409. The average value and standard deviation of the remaining chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data are calculated. The standard forming parameter range for the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data is then determined. in, S is the average of the remaining chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data, and S is the standard deviation of the remaining chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data.

[0104] The method for calculating the kurtosis test value bk(n) is as follows:

[0105]

[0106] In the formula, n is the number of the currently calculated chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data sorted from smallest to largest.

[0107] The average value of all chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data.

[0108] x i This refers to the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data arranged in ascending order before n.

[0109] This scheme calculates the standard forming parameter range by using the kurtosis test to remove outliers from the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data. For the non-outliers, the mean and standard deviation are calculated. According to the normal distribution theory, the fluctuations in process parameters under normal forming conditions fall within a certain range. Within this range, using this range as the standard forming parameter range can greatly improve the accuracy of detection, increase the sensitivity to abnormal values ​​in the forming process, and enable timely and rapid detection of abnormal states in chromium alloy casting forming, thereby avoiding a decline in the forming quality of castings and preventing property losses.

[0110] Please see Figure 5 As shown, the analysis of abnormal forming parameters using a Logistic regression model to determine whether there is a risk of substandard quality in chromium alloy continuously cast parts includes the following steps:

[0111] S601. Substitute the abnormal forming chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature into the ternary logistic regression model of forming quality - chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature to obtain the non-conforming probability value under the current process parameters.

[0112] S601. Compare the non-conformance probability value under the current process parameters with the preset threshold to determine whether it is greater than the preset threshold. If it is, the non-conformance risk is determined to be high, and a first-level alarm signal is output. If not, the non-conformance risk is determined to be low, and a second-level alarm signal is output.

[0113] The preset threshold value range is set to 0.15-0.3.

[0114] Based on the Logistic regression model, the probability of non-conforming quality of chromium alloy continuous casting parts is calculated by analyzing the process parameters that exhibit abnormal fluctuations. The presence of non-conforming risk is determined by judging whether the probability exceeds the preset value.

[0115] Regarding the range of preset threshold values, those skilled in the art will understand that a lower preset threshold indicates a higher degree of control over the risk of nonconformity, while a higher preset threshold indicates a lower degree of control over the risk of nonconformity. In actual production testing, the preset threshold can be determined according to actual needs.

[0116] Please see Figure 6 As shown, further, a high-precision monitoring system for continuous casting of chromium alloys is proposed to realize the high-precision monitoring method for continuous casting of chromium alloys as described above, including:

[0117] The control module is used to output control signals to each component, thereby controlling each component.

[0118] The process parameter monitoring module is electrically connected to the control module. The process parameter monitoring module is used to collect the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the continuous casting process of chromium alloy, and feed them back to the control module.

[0119] The data processing module is electrically connected to the control module. The data processing module is used to analyze multiple sets of chromium alloy continuous casting sample data, establish a Logistic regression model, monitor the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the chromium alloy continuous casting process, and compare them with preset standard values ​​to determine whether they are within the standard forming parameter range. The abnormal forming parameters are analyzed through the Logistic regression model to determine whether there is a risk of unqualified chromium alloy continuous casting parts.

[0120] The data storage module is electrically connected to the control module and the data processing module. The data storage module is used to store the Logistic regression model.

[0121] The signal output module is electrically connected to the control module and the data processing module, and is used to output alarm signals.

[0122] The process parameter monitoring module includes:

[0123] The first temperature sensor is used to detect the melting temperature of chromium alloy.

[0124] Flow sensor, used to detect the casting speed of chromium alloy;

[0125] The second temperature sensor is used to detect the temperature of the casting mold.

[0126] During use, the above-mentioned forming monitoring system monitors the continuous casting parameters of chromium alloy in real time through the process parameter monitoring module and feeds them back to the control module. The control module sends the detection data to the data processing module for processing and receives feedback signals from the data processing module. Based on the feedback signals, the control module outputs control signals to the signal output module to output different alarm signals.

[0127] In summary, the advantages of this invention are: by establishing a Logistic regression model based on the influencing factors of the quality of chromium alloy continuous casting parts, the risk of non-conforming quality of chromium alloy continuous casting parts can be quickly identified, thus realizing intelligent and high-precision monitoring of chromium alloy continuous casting.

[0128] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A high-precision chromium alloy continuous casting forming monitoring method, characterized by, Includes the following steps: To identify the factors affecting the quality of chromium alloy continuously cast parts, the influencing factors include chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature. Chromium alloy continuous casting molding test was conducted by setting different chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature according to the preset gradient. At the same time, the molding quality of chromium alloy continuous casting molded parts in the test was obtained, and multiple sets of chromium alloy continuous casting molding sample data were obtained. A Logistic regression model was established by analyzing multiple sets of chromium alloy continuous casting sample data. The chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature are monitored in real time during the continuous casting process of chromium alloy, and compared with the preset standard values ​​to determine whether they are within the standard forming parameter range. If yes, output a normal molding parameter signal; otherwise, output an abnormal molding parameter signal. The abnormal forming parameters are analyzed using a Logistic regression model to determine whether there is a risk of unqualified quality of chromium alloy continuous casting parts. If yes, a first-level alarm signal is output; otherwise, a second-level alarm signal is output. The analysis of multiple sets of chromium alloy continuous casting sample data and the establishment of a Logistic regression model specifically include the following steps: The forming quality of chromium alloy continuous casting parts was classified into qualified and unqualified categories, and several sets of qualified chromium alloy continuous casting sample data and several sets of unqualified chromium alloy continuous casting sample data were statistically analyzed. A ternary logistic regression model was established to evaluate the forming quality of chromium alloy continuously cast parts based on chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature. The coefficients in the ternary Logistic regression model were estimated using the maximum likelihood method based on multiple sets of chromium alloy continuous casting sample data to obtain the model regression coefficients. Test the significance of the model regression coefficients to determine whether the model regression coefficients meet the significance requirements; Test the significance of the model to determine whether the model is statistically significant; The calculation method for the standard forming parameter ranges of the chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature is as follows: The system outputs control commands to the continuous casting forming system according to the optimal process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature. The actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature are obtained by detecting the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the continuous casting process. Obtain actual process data on chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature for several sets of qualified chromium alloy continuous casting parts. The actual process values ​​of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature of several groups of qualified chromium alloy continuous casting parts were analyzed to obtain the standard forming parameter range of chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature. The analysis of the actual process data of chromium alloy melting temperature, chromium alloy casting speed, and casting mold temperature for several groups of qualified chromium alloy continuous casting parts specifically includes the following steps: Arrange the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data of several qualified chromium alloy continuous casting parts in ascending order. Set the detection level to obtain the critical value bp(n) for kurtosis test. Calculate the kurtosis test value bk(n) for each chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data. Determine whether the kurtosis test value bk(n) of the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data is greater than the critical value bp(n) of the kurtosis test. If yes, then remove the chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data. If no, then do not respond. The average value and the standard deviation of the remaining chromium alloy melting temperature data or chromium alloy pouring speed data or casting mold temperature data are calculated, and the standard forming parameter range of the chromium alloy melting temperature data or chromium alloy pouring speed data or casting mold temperature data is wherein, is the average value of the remaining chromium alloy melting temperature data or chromium alloy pouring speed data or casting mold temperature data, is the standard deviation of the remaining chromium alloy melting temperature data or chromium alloy pouring speed data or casting mold temperature data.

2. The high-precision chromium alloy continuous casting forming monitoring method according to claim 1, characterized by, The ternary logistic regression model is as follows: ， In the formula, the chromium alloy continuously cast shaped article is a nonconforming product; the chromium alloy continuously cast shaped article is a nonconforming product; P represents the predicted probability of the ternary logistic regression model; T is the melting temperature of the chromium alloy; v represents the casting speed of the chromium alloy; t is the temperature of the casting mold; are coefficients of a ternary logistic regression model.

3. The high-precision chromium alloy continuous casting forming monitoring method according to claim 2, characterized by, The method for calculating the kurtosis test value bk(n) is as follows: In the formula, n is the number of the currently calculated chromium alloy melting temperature data, chromium alloy casting speed data, or casting mold temperature data sorted from smallest to largest. is the average of all chromium alloy melting temperature data or chromium alloy pouring speed data or cast forming mold temperature data; The chromium alloy melting temperature data or the chromium alloy casting speed data or the casting mold temperature data are arranged in order from small to large before n.

4. The high-precision chromium alloy continuous casting forming monitoring method according to claim 3, characterized by, The step of analyzing abnormal forming parameters using a Logistic regression model to determine whether there is a risk of substandard quality in chromium alloy continuously cast parts includes the following steps: Substitute the abnormal forming chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature into the ternary logistic regression model of forming quality - chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature to obtain the non-conforming probability value under the current process parameters. The non-conformance probability value under the current process parameters is compared with the preset threshold. If it is greater than the preset threshold, the risk of non-conformance is determined to be high, and a first-level alarm signal is output. If not, the risk of non-conformance is determined to be low, and a second-level alarm signal is output.

5. The method of claim 4, wherein the high-precision chromium alloy continuous casting forming monitoring method is characterized by, The preset threshold value range is set to 0.15-0.

3.

6. A high-precision chromium alloy continuous casting formation monitoring system for implementing the high-precision chromium alloy continuous casting formation monitoring method according to any one of claims 1 to 5, characterized by, include: The control module is used to output control signals to each component, thereby controlling each component. The process parameter monitoring module is electrically connected to the control module. The process parameter monitoring module is used to collect the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the continuous casting process of chromium alloy, and feed them back to the control module. The data processing module is electrically connected to the control module. The data processing module is used to analyze multiple sets of chromium alloy continuous casting sample data, establish a Logistic regression model, monitor the chromium alloy melting temperature, chromium alloy casting speed and casting mold temperature during the chromium alloy continuous casting process, and compare them with preset standard values ​​to determine whether they are within the standard forming parameter range. The abnormal forming parameters are analyzed through the Logistic regression model to determine whether there is a risk of unqualified chromium alloy continuous casting parts. The data storage module is electrically connected to the control module and the data processing module, and is used to store the Logistic regression model. The signal output module is electrically connected to the control module and the data processing module, and is used to output alarm signals.

7. The high-precision chromium alloy continuous casting forming monitoring system according to claim 6, characterized in that, The process parameter monitoring module includes: A first temperature sensor is used to detect the melting temperature of chromium alloy. A flow sensor is used to detect the casting speed of chromium alloy; The second temperature sensor is used to detect the temperature of the casting mold.

Images