An online monitoring method and system for metal continuous casting process

Abstract

The present application relates to the field of metal continuous casting monitoring technology, and specifically to an online monitoring method and system for a metal continuous casting process, the method comprising: determining the consistency index of various friction sequences in the metal continuous casting process; determining the temperature fluctuation index of the metal continuous casting process; based on the distribution of the consistency index of the friction sequence containing the friction data with the same corresponding position order as all temperature subsequences in the cluster with the largest overall change index, and the degree of dispersion of all the consistency indexes, combined with the temperature fluctuation index, determining the abnormal index of the metal continuous casting process, so as to monitor the state of the metal continuous casting process. The present application improves the accuracy of monitoring the health status of the metal continuous casting process by jointly analyzing the friction data and temperature data in the continuous casting process.

Description

Technical Field

[0001] The present application relates to the technical field of metal continuous casting monitoring, and in particular to an online monitoring method and system for a metal continuous casting process. Background Art

[0002] Continuous casting is a process in which liquid metal is converted into ingots by forced cooling. Compared with traditional casting methods, continuous casting has a shorter process flow, a higher degree of automation, higher productivity, and better energy and cost savings. In the continuous casting process of metals, longitudinal cracks are a common and most harmful defect in continuous casting slabs. When longitudinal cracks occur in slabs, the amount of finishing and cost will be increased at the least; in severe cases, bonding and leakage accidents will occur, which will not only interrupt the entire continuous casting production process, but also damage continuous casting equipment such as the crystallizer and the secondary cooling zone. Since the longitudinal crack defect is formed inside the crystallizer due to the uneven cooling of the billet shell, it has become a common defect detection method to detect the continuous casting process by analyzing the operating parameters of the crystallizer.

[0003] The thermocouple temperature measurement method buries a certain number of thermocouples on the copper plate of the crystallizer to obtain the temperature change of the copper plate, and predicts the longitudinal cracks by analyzing the temperature change. The method is simple to implement, highly real-time and easy to maintain. However, there are limitations in relying solely on temperature changes as a monitoring indicator for the continuous casting process. The existing technology is relatively simple in feature extraction when analyzing the health status of the metal continuous casting process. It is difficult to fully reflect the complex working conditions of the continuous casting process by relying solely on temperature data to monitor the health status of the metal continuous casting process, resulting in large errors, thereby reducing the accuracy of monitoring the health status of the metal continuous casting process. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide an online monitoring method and system for a metal continuous casting process. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides an online monitoring method for a metal continuous casting process, the method comprising the following steps:

[0006] Obtain various friction force data at all acquisition times within a preset period during the metal continuous casting process;

[0007] Obtain the temperature average of the temperature data of all thermocouples on the crystallizer at each acquisition moment during the metal continuous casting process, and the temperature averages at all acquisition moments constitute a temperature sequence;

[0008] Various friction force data in various preset time periods constitute various friction force sequences. According to the correlation between various friction force sequences and all other similar friction force sequences, the difference between the first-order difference sequence of various friction force sequences and the first-order difference sequence of other similar friction force sequences, and the difference between various friction force sequences and all other similar friction force sequences about the maximum amplitude, the consistency index of various friction force sequences is determined;

[0009] Clustering multiple temperature subsequences obtained by dividing the temperature sequence to obtain two clusters, and determining the overall change index of each cluster based on the average distribution of differences between all adjacent elements in each temperature subsequence in each cluster;

[0010] Determining the temperature fluctuation index of the metal continuous casting process based on the difference of the overall variation index of the two clusters and the difference of all temperature subsequences in each cluster with all temperature subsequences in the other cluster;

[0011] Based on the distribution of the consistency index of the friction sequence containing the friction data having the same corresponding position sequence as all temperature subsequences in the cluster with the largest overall variation index, and the degree of dispersion of all the consistency indexes, combined with the temperature fluctuation index, the abnormal index of the metal continuous casting process is determined to monitor the state of the metal continuous casting process.

[0012] Preferably, the method for determining the consistency index of the various friction force sequences is:

[0013] The friction force data includes friction tension data and friction pressure data. All friction tension data in each preset time period constitute each friction tension sequence. All friction pressure data in each preset time period constitute each friction pressure sequence. The mean value of the correlation between each friction tension sequence and all other friction tension sequences is analyzed and recorded as the first mean value of each friction tension sequence.

[0014] Determine the first difference mean of each friction tension sequence based on the difference between the first-order difference sequence of each friction tension sequence and the first-order difference sequence of all other friction tension sequences;

[0015] Determine a second difference mean of each friction tension sequence based on the difference of each friction tension sequence with all other friction tension sequences with respect to the maximum amplitude;

[0016] The first mean of each friction tension sequence is positively correlated with the consistency index of each friction tension sequence, and the first difference mean and the second difference mean of each friction tension sequence are negatively correlated with the consistency index of each friction tension sequence;

[0017] For each friction pressure sequence, the consistency index of each friction pressure sequence is obtained according to the method for obtaining the consistency index of each friction tension sequence.

[0018] Preferably, the first difference mean of each friction and tension sequence is the mean of the difference between the first-order difference sequence of each friction and tension sequence and the first-order difference sequence of all other friction and tension sequences.

[0019] Preferably, the second difference mean value of each friction and tension sequence is the mean value of the difference between the maximum friction and tension data in each friction and tension sequence and the maximum friction and tension data in all other friction and tension sequences.

[0020] Preferably, the overall change index of each cluster is the mean of the absolute values ​​of all elements in the first-order difference sequence of each temperature subsequence.

[0021] Preferably, the expression of the temperature fluctuation index of the metal continuous casting process is: Where, E represents the temperature fluctuation index of the metal continuous casting process, F m,n It represents the difference between the overall change index of the mth cluster and the overall change index of the nth cluster; represents the difference between the i-th temperature subsequence in the m-th cluster and the j-th temperature subsequence in the n-th cluster; M represents the number of all temperature subsequences in the m-th cluster; N represents the number of all temperature subsequences in the n-th cluster.

[0022] Preferably, the method for determining the abnormal index of the metal continuous casting process is:

[0023] Based on the distribution of consistency index of friction force sequence of friction force data with the same position order as all temperature subsequences in the cluster with the largest overall variation index, the average consistency of the metal continuous casting process is determined;

[0024] The abnormal index expression of the metal continuous casting process is: Wherein, K represents the abnormal index of the metal continuous casting process; L represents the discrete degree of the consistency index of all friction force sequences in the metal continuous casting process; E represents the temperature fluctuation index of the metal continuous casting process; R represents the average consistency of the metal continuous casting process.

[0025] Preferably, the method for determining the average consistency of the metal continuous casting process is:

[0026] The cluster with the largest overall change index is recorded as the detection cluster, and the average consistency of the metal continuous casting process is the mean of the consistency index of the friction data with the same corresponding position sequence as all temperature subsequences in the detection cluster.

[0027] Preferably, the monitoring of the state of the metal continuous casting process includes:

[0028] If the normalized value of the abnormality index of the metal continuous casting process is greater than or equal to a preset value, the metal continuous casting process is abnormal and the monitoring system issues a forecast. Otherwise, the metal continuous casting process is normal. In a second aspect, an embodiment of the present application also provides an online monitoring system for a metal continuous casting process, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of the online monitoring method for a metal continuous casting process described in any one of the above items are implemented.

[0029] This application has at least the following beneficial effects:

[0030] The present application determines the consistency index of various friction force sequences in the metal continuous casting process by analyzing the correlation between each friction force sequence and some other friction force sequences, the difference between the changing trend of all friction force data in each friction force sequence and the changing trend of all friction force data in some other friction force sequences, and the difference between each friction force sequence and some other friction force sequences in terms of amplitude, so as to judge the operating status of the continuous casting machine; determines the overall change index of each clustering cluster by analyzing the changing trend of all temperature means in each temperature subsequence in each clustering cluster in the metal continuous casting process; determines the temperature fluctuation index of the metal continuous casting process based on the difference in the overall change index of the two clustering clusters and the difference between all temperature subsequences in each clustering cluster and all temperature subsequences in another clustering cluster, and judges whether there are problems such as longitudinal crack defects in the ingot during the metal continuous casting process; determines the abnormal index of the metal continuous casting process by comprehensively analyzing the friction force data and temperature data, so as to accurately predict potential equipment failures or process out of control, thereby achieving preventive maintenance.

[0031] This application takes into account that the existing thermocouple temperature measurement method only monitors the continuous casting process through temperature changes, which leads to large errors. Therefore, through a joint analysis of multi-dimensional data during the continuous casting process of aluminum plates, the continuous casting process is monitored more comprehensively, thereby improving the accuracy of monitoring the health status of the metal continuous casting process. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 A flowchart of the steps of an online monitoring method for a metal continuous casting process provided by one embodiment of the present application;

[0034] Figure 2A flow chart of consistency-related index extraction provided for one embodiment of the present application;

[0035] Figure 3 A schematic diagram of an abnormality index extraction process provided for one embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the online monitoring method and system for a metal continuous casting process proposed in the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0038] The specific scheme of an online monitoring method and system for a metal continuous casting process provided by the present application is described in detail below with reference to the accompanying drawings.

[0039] See also Figure 1 , which shows a flow chart of the steps of an online monitoring method for a metal continuous casting process provided by an embodiment of the present application, the method comprising the following steps:

[0040] Step S1: Obtaining friction force data at all sampling moments within a preset sampling time in a metal continuous casting process, and temperature data of each thermocouple on the crystallizer at all sampling moments.

[0041] In this embodiment, the friction data generated by the relative movement between the ingot and the crystallizer are obtained by means of a hydraulic vibration device.

[0042] 19 rows of thermocouples are arranged at equal distances on the two wide copper plates of the crystallizer, and 1 row of thermocouples are arranged on each of the two narrow copper plates. Three rows of thermocouples are arranged at equal distances along the pouring direction, for a total of 120 thermocouples. The temperature data of the crystallizer is obtained through the crystallizer copper plates where temperature measuring thermocouples are buried.

[0043] Set the data collection frequency to f and the preset collection time to t. Upload all the collected friction data and all the temperature data to the monitoring system.

[0044] Among them, the friction force data includes two categories: friction tension data and friction pressure data. During the negative slip time, the friction force between the ingot shell and the copper plate is called friction pressure; during the positive slip time, the friction force between the ingot shell and the copper plate is called friction tension.

[0045] It should be noted that the values ​​of the acquisition frequency f and the preset acquisition time are both manually set. In this embodiment, the acquisition frequency f is 10 Hz, and the preset acquisition time t is 180 s. The implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0046] Step S2: Determine the consistency index of each type of friction sequence based on the correlation between each type of friction sequence and all other similar friction sequences, the difference between the first-order difference sequence of each type of friction sequence and the first-order difference sequence of other similar friction sequences, and the difference between each type of friction sequence and all other similar friction sequences about the maximum amplitude.

[0047] During the continuous casting of aluminum plates, due to the vibration of the crystallizer, the friction between the ingot and the copper plate of the crystallizer changes periodically. If there is no problem during the continuous casting process, under normal working conditions, the fluctuation of the friction tension and friction pressure in each cycle is small, and the periodicity of the friction is strong.

[0048] Under normal conditions of metal continuous casting, the speed at which the billet is pulled out of the crystallizer is usually kept constant, that is, the changes in the friction pressure and friction pull of the crystallizer are relatively stable, and the friction force shows a periodic and stable change over time. However, during the continuous casting process, if the cooling of the crystallizer or the secondary cooling zone is uneven, the thickness of the billet shell will be uneven, which will affect the friction between the billet shell and the crystallizer, and the stability of the friction force data will be destroyed.

[0049] Therefore, in order to judge whether there are abnormal conditions in the metal continuous casting process, so as to timely optimize the control strategy, improve the quality and efficiency of the casting and reduce the loss in the continuous casting process, based on the correlation between various friction sequences and all other similar friction sequences, the difference between the first-order difference sequence of various friction sequences and the first-order difference sequence of other similar friction sequences, and the difference between various friction sequences and all other similar friction sequences about the maximum amplitude, the consistency index of various friction sequences is determined to judge the operating status of the continuous casting machine, specifically:

[0050] (1) First, the acquisition time corresponding to when the friction data is 0 is recorded as the segmentation time. Since the friction data is divided into two categories, the friction sequences composed of the corresponding friction data are also divided into two categories. The friction tension data of all acquisition times within each adjacent segmentation time constitute each friction tension sequence, and all friction pressure data within each adjacent segmentation time constitute each friction pressure sequence.

[0051] Since the number and change trend of friction data from the beginning of friction data collection to the first segmentation time are not stable, and the number and change trend of friction data from the last segmentation time to the last collection time are not stable enough, the friction data of these two parts are not analyzed.

[0052] (2) Secondly, the mean of the correlation between each friction tension sequence and all other friction tension sequences is analyzed and recorded as the first mean of each friction tension sequence.

[0053] It should be noted that there are many methods for measuring the correlation between sequences. In this embodiment, the correlation between the friction tension sequences is measured by calculating the Pearson correlation coefficient between the friction tension sequences. In actual application, as other implementation methods, implementers may also use covariance, Spearman rank correlation coefficient and other methods to measure the correlation between sequences. This embodiment does not impose any special restrictions.

[0054] The calculation process of the Pearson correlation coefficient is a well-known technique, and the specific calculation process will not be described in detail.

[0055] (3) Further, the difference between the first-order difference sequence of each friction tension sequence and the first-order difference sequence of all other friction tension sequences is analyzed and the average is taken, which is recorded as the first difference mean of each friction tension sequence.

[0056] It should be noted that there are many methods for measuring the differences between sequences. In this embodiment, the differences between first-order difference sequences are measured by calculating the DTW distance between the first-order difference sequences. As other implementation methods, the implementer may also use other methods for measuring the differences between sequences, such as Euclidean distance, Manhattan distance, and the inverse of cosine similarity. This embodiment does not impose any special restrictions.

[0057] The process of obtaining the first-order difference sequence and the process of calculating the DTW distance are both well-known technologies, and the specific processes will not be described in detail.

[0058] (4) Further, the difference between the maximum friction tension data in each friction tension sequence and the maximum friction tension data in all other friction tension sequences is analyzed and the average is obtained, which is recorded as the second difference mean of each friction tension sequence;

[0059] It should be noted that there are many methods for measuring the differences between numerical values. In this embodiment, the absolute value of the difference between the maximum friction tension data in each friction force sequence and the maximum friction tension data in all other friction force sequences is taken as a method for measuring the difference between the maximum friction tension data. As other implementation methods, implementers may also use other methods for measuring the difference between data such as ratios, and this embodiment does not impose any special restrictions.

[0060] The first mean of each friction tension sequence is positively correlated with the consistency index of each friction tension sequence, and the first difference mean and the second difference mean of each friction tension sequence are negatively correlated with the consistency index of each friction tension sequence.

[0061] It should be understood that a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship may be an additive relationship or a multiplicative relationship, etc., which is determined by the actual application and this application does not impose any special restrictions. A negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It may be a subtractive relationship or a division relationship, etc., which is determined by the actual application.

[0062] Preferably, the expression of the consistency index of each friction tension sequence in this embodiment is: In the formula, A e represents the consistency index of the e-th friction tension sequence; B e represents the first mean value of the e-th friction tension sequence; C e represents the first difference mean of the e-th friction tension sequence; D e represents the second difference mean of the e-th friction tension sequence; τ represents a preset constant greater than 0, which is used to avoid the denominator being 0. In this embodiment, the value of τ is 0.01. The implementer can also set it by himself according to the specific situation, on the premise of ensuring that the denominator is not 0 and does not excessively affect the calculation result of the consistency index. This embodiment does not impose any special restrictions.

[0063] In actual application, as another implementation method, the consistency index of each friction tension sequence is determined by: the ratio of the first mean of each friction tension sequence to the exponential function value with a natural constant as the base and the sum of the first difference mean and the second difference mean of the corresponding friction tension sequence as the independent variable.

[0064] If the operating conditions during the continuous casting of the aluminum plate are normal, the friction tension sequences have strong consistency, the first mean of the friction tension sequence is larger, and the difference between the first-order difference sequences of all the friction tension sequences is smaller, that is, the first difference mean of the friction tension sequence is smaller, the difference in the maximum value is smaller, that is, the second difference mean of the friction tension sequence is smaller, and thus the consistency index of the friction tension sequence is larger; conversely, if the operating conditions during the continuous casting of the aluminum plate are abnormal, the consistency between the friction tension sequences is weaker, the first mean of the friction tension sequence is smaller, and the difference between the first-order difference sequences of all the friction tension sequences is larger, that is, the first difference mean of the friction tension sequence is larger, the difference in the maximum value is larger, that is, the second difference mean of the friction tension sequence is larger, and thus the consistency index of the friction tension sequence is smaller.

[0065] For each friction pressure sequence, the consistency index of each friction tension sequence is obtained according to the method for obtaining the consistency index of each friction tension sequence. Because all friction sequences are divided into two categories: friction tension sequence and friction pressure sequence, the consistency index of all friction sequences can be obtained.

[0066] Preferably, the consistency correlation index extraction flow chart provided in this embodiment is as follows: Figure 2 shown.

[0067] Step S3: Based on the average distribution of differences between all adjacent elements in each temperature subsequence in each cluster, determine the overall change index of each cluster; based on the difference in the overall change index of the two clusters, and the difference between all temperature subsequences in each cluster and all temperature subsequences in the other cluster, determine the temperature fluctuation index of the metal continuous casting process.

[0068] During the continuous casting of aluminum plates, if there is a problem with the continuous casting process, resulting in longitudinal cracks on the surface of the ingot, the indentation and thermal contraction of the ingot shell will increase the air gap between the copper plate of the crystallizer and the ingot, which will hinder heat transfer and cause changes in the temperature inside the crystallizer. As the air gap increases, the temperature measured by the thermocouple will drop; then as the crack position moves downward and heals (Note: when the friction force is less than 0, the ingot is subjected to compressive stress, and the cracks on the surface of the shell may heal), the shell will be close to the copper plate of the crystallizer again under static pressure, and the temperature measured by the thermocouple will rise (Note: the temperature changes very quickly, in seconds, the temperature fluctuation can reach about 20°C).

[0069] The average value of the temperature data of all thermocouples on the crystallizer at each collection time is recorded as the temperature mean at each collection time, and the temperature mean of all collection times constitutes a temperature sequence. If the continuous casting process of the aluminum plate is normal, the temperature mean in the temperature sequence will be relatively stable, and the difference between the two clusters will not be large; if an abnormality occurs during the continuous casting process, resulting in longitudinal crack defects in the ingot, the collected crystallizer temperature data will show a data trend of stability-significant decline-gradual recovery-stable.

[0070] Therefore, in order to determine whether there are longitudinal crack defects in the ingot during the metal continuous casting process, the temperature fluctuation index of the metal continuous casting process is determined based on the average distribution of the differences between all adjacent elements in each temperature subsequence in each cluster, and the differences between all temperature subsequences in each cluster and all temperature subsequences in another cluster. Specifically, it is:

[0071] (1) First, the average value of the temperature data of all thermocouples on the crystallizer at each sampling time during the metal continuous casting process is recorded as the temperature mean value at each sampling time during the metal continuous casting process. The temperature mean values ​​at all sampling times constitute the temperature sequence of the metal continuous casting process.

[0072] The temperature series is used as the input of the variable-length subsequence clustering algorithm in time series. The temperature series is divided into multiple temperature subsequences, and all temperature subsequences are clustered. The number of clusters is set to 2, and two clusters are obtained.

[0073] Among them, since the sequence length in the variable length subsequence algorithm in the time series changes dynamically, in order to prevent the sequence length from being too short or too long and affecting the analysis effect, the preset value range of the length of the temperature subsequence in this embodiment is 15-70, and the implementer can also set it by himself according to the specific situation, and this embodiment does not make special restrictions. In addition, the number of clusters is set to 2 in order to distinguish between temperature data with steady changes and temperature data with non-steady changes, so the number of clusters is set to 2.

[0074] It should be understood that the variable length subsequence algorithm in the time series is a well-known technology, and the specific process will not be repeated here.

[0075] (2) Furthermore, the overall change index of each cluster is the mean of the absolute values ​​of all elements in the first-order difference sequence of each temperature subsequence.

[0076] (3) The expression of the temperature fluctuation index in the metal continuous casting process is: Where, E represents the temperature fluctuation index of the metal continuous casting process; F m,n It represents the difference between the overall change index of the mth cluster and the overall change index of the nth cluster; represents the difference between the i-th temperature subsequence in the m-th cluster and the j-th temperature subsequence in the n-th cluster; M represents the number of all temperature subsequences in the m-th cluster; N represents the number of all temperature subsequences in the n-th cluster; where m=1, n=1.

[0077] If the operating conditions during the continuous casting of aluminum plates are normal, the change trend of the temperature mean in the temperature sequence is relatively stable, the smaller the difference between the temperature subsequences in the two clusters, the smaller the difference in the overall change index of the two clusters during the metal continuous casting process, and the smaller the temperature fluctuation index of the metal continuous casting process; conversely, if the operating conditions during the continuous casting of aluminum plates are abnormal, the greater the oscillation amplitude of the change trend of the temperature mean in the temperature sequence, the greater the difference between the temperature subsequences in the two clusters, the greater the difference in the overall change index of the two clusters during the metal continuous casting process, and the greater the temperature fluctuation index of the metal continuous casting process.

[0078] Step S4: Based on the distribution of the consistency index of the friction sequence containing the friction data having the same corresponding position sequence as all temperature subsequences in the cluster with the largest overall variation index, and the degree of dispersion of all the consistency indexes, combined with the temperature fluctuation index, determine the abnormality index of the metal continuous casting process.

[0079] In order to more comprehensively evaluate the health of the continuous casting process, by combining multiple indicators of temperature and friction, more accurately and timely discover problems in the continuous casting process, therefore, based on the distribution of the consistency index of the friction sequence of the friction data with the same position order as all temperature subsequences in the cluster with the largest overall change index, and the discrete degree of all the consistency indexes, combined with the temperature fluctuation index, the abnormal index of the metal continuous casting process is determined to predict potential equipment failures or process out of control, thereby achieving preventive maintenance, specifically:

[0080] The cluster with the largest overall change index is recorded as the detection cluster, and the mean of the consistency index of the friction data with the same corresponding position sequence as all temperature subsequences in the detection cluster is analyzed and recorded as the average consistency of the metal continuous casting process.

[0081] The expression of the abnormal index of the metal continuous casting process is: Wherein, K represents the abnormal index of the metal continuous casting process; L represents the discrete degree of the consistency index of all friction force sequences in the metal continuous casting process; E represents the temperature fluctuation index of the metal continuous casting process; R represents the average consistency of the metal continuous casting process.

[0082] It should be noted that there are many methods for measuring the degree of discreteness of a set of data. In this embodiment, the degree of discreteness of the consistency index is measured by calculating the variance of the consistency index of all friction force sequences. The implementer may also use other methods for measuring the degree of discreteness such as standard deviation, coefficient of dispersion or interquartile range, and this embodiment does not impose any special restrictions.

[0083] If the degree of dispersion of the consistency index is greater, the average consistency of the metal continuous casting process is smaller, which means that the possibility of consistency between the friction tension sequence or the friction pressure sequence is smaller, that is, the possibility of periodicity of the friction data is smaller. At the same time, if the temperature fluctuation index is larger, the abnormality index of the metal continuous casting process is larger; conversely, if the degree of dispersion of the consistency index is smaller, the average consistency of the metal continuous casting process is larger, which means that the possibility of consistency between the friction tension sequence or the friction pressure sequence is larger, that is, the possibility of periodicity of the friction data is larger. At the same time, if the temperature fluctuation index is smaller, the abnormality index of the metal continuous casting process is smaller.

[0084] Preferably, the schematic diagram of the abnormal index extraction process provided in this embodiment is as follows: Figure 3 shown.

[0085] Step S5: Based on the abnormal index of the metal continuous casting process, the state of the metal continuous casting process is monitored.

[0086] The abnormal index of the metal continuous casting process is normalized. If the normalized result of the abnormal index of the metal continuous casting process is greater than or equal to the preset value, then in the continuous casting process of the aluminum plate, serious longitudinal grain defects appear in the shell of the ingot. At this time, the monitoring system will make a timely forecast to avoid danger; conversely, if the normalized result of the abnormal index of the metal continuous casting process is less than the preset value, it means that the working conditions of the continuous casting process of the aluminum plate are normal and casting can continue.

[0087] At this point, based on the difference in the overall change index of the two clusters, and the difference between all temperature subsequences in each cluster and all temperature subsequences in the other cluster, the temperature fluctuation index of the metal continuous casting process is determined, and whether the ingot has problems such as longitudinal cracks in the metal continuous casting process is judged by the size of the temperature fluctuation index; based on the distribution of the consistency index of the friction sequence of the friction data with the same corresponding position sequence as all temperature subsequences in the cluster with the largest overall change index, and the degree of discreteness of all the consistency indices, combined with the temperature fluctuation index, the abnormal index of the metal continuous casting process is determined, and the prediction of equipment failure or process out of control is completed, thereby achieving preventive maintenance.

[0088] Based on the same inventive concept as the above-mentioned method, an embodiment of the present application also provides an online monitoring system for a metal continuous casting process, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned methods for online monitoring of a metal continuous casting process are implemented.

[0089] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0090] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0091] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. An online monitoring method for a metal continuous casting process, characterized in that: The method comprises the following steps: Obtain various friction force data at all acquisition times within a preset period during the metal continuous casting process; Obtain the temperature average of the temperature data of all thermocouples on the crystallizer at each acquisition moment during the metal continuous casting process, and the temperature averages at all acquisition moments constitute a temperature sequence; Various friction force data in various preset time periods constitute various friction force sequences. According to the correlation between various friction force sequences and all other similar friction force sequences, the difference between the first-order difference sequence of various friction force sequences and the first-order difference sequence of other similar friction force sequences, and the difference between various friction force sequences and all other similar friction force sequences about the maximum amplitude, the consistency index of various friction force sequences is determined; Clustering multiple temperature subsequences obtained by dividing the temperature sequence to obtain two clusters, and determining the overall change index of each cluster based on the average distribution of differences between all adjacent elements in each temperature subsequence in each cluster; Determining the temperature fluctuation index of the metal continuous casting process based on the difference of the overall variation index of the two clusters and the difference of all temperature subsequences in each cluster with all temperature subsequences in the other cluster; Based on the distribution of the consistency index of the friction sequence containing the friction data having the same corresponding position sequence as all temperature subsequences in the cluster with the largest overall variation index, and the degree of dispersion of all the consistency indexes, combined with the temperature fluctuation index, the abnormal index of the metal continuous casting process is determined to monitor the state of the metal continuous casting process.

2. The method for online monitoring of a metal continuous casting process according to claim 1, characterized in that: The method for determining the consistency index of the various friction force sequences is as follows: The friction force data includes friction tension data and friction pressure data. All friction tension data in each preset time period constitute each friction tension sequence. All friction pressure data in each preset time period constitute each friction pressure sequence. The mean value of the correlation between each friction tension sequence and all other friction tension sequences is analyzed and recorded as the first mean value of each friction tension sequence. Determine the first difference mean of each friction tension sequence based on the difference between the first-order difference sequence of each friction tension sequence and the first-order difference sequence of all other friction tension sequences; Determine a second difference mean of each friction tension sequence based on the difference of each friction tension sequence with all other friction tension sequences with respect to the maximum amplitude; The first mean of each friction tension sequence is positively correlated with the consistency index of each friction tension sequence, and the first difference mean and the second difference mean of each friction tension sequence are negatively correlated with the consistency index of each friction tension sequence; For each friction pressure sequence, the consistency index of each friction pressure sequence is obtained according to the method for obtaining the consistency index of each friction tension sequence.

3. The online monitoring method for a metal continuous casting process according to claim 2, characterized in that: The first difference mean of each friction tension sequence is the mean of the difference between the first-order difference sequence of each friction tension sequence and the first-order difference sequence of all other friction tension sequences.

4. The method for online monitoring of a metal continuous casting process according to claim 2, characterized in that: The second difference mean of each friction and tension sequence is the mean of the difference between the maximum friction and tension data in each friction and tension sequence and the maximum friction and tension data in all other friction and tension sequences.

5. The method for online monitoring of a metal continuous casting process according to claim 1, characterized in that: The overall change index of each cluster is the mean of the absolute values ​​of all elements in the first-order difference sequence of each temperature subsequence.

6. The method for online monitoring of a metal continuous casting process according to claim 1, characterized in that: The expression of the temperature fluctuation index of the metal continuous casting process is: Where, E represents the temperature fluctuation index of the metal continuous casting process, F m,n It represents the difference between the overall change index of the mth cluster and the overall change index of the nth cluster; represents the difference between the i-th temperature subsequence in the m-th cluster and the j-th temperature subsequence in the n-th cluster; M represents the number of all temperature subsequences in the m-th cluster; N represents the number of all temperature subsequences in the n-th cluster.

7. The method for online monitoring of a metal continuous casting process according to claim 1, characterized in that: The method for determining the abnormal index of the metal continuous casting process is: Based on the distribution of consistency index of friction force sequence of friction force data with the same position order as all temperature subsequences in the cluster with the largest overall variation index, the average consistency of the metal continuous casting process is determined; The abnormal index expression of the metal continuous casting process is: Wherein, K represents the abnormal index of the metal continuous casting process; L represents the discrete degree of the consistency index of all friction force sequences in the metal continuous casting process; E represents the temperature fluctuation index of the metal continuous casting process; R represents the average consistency of the metal continuous casting process.

8. The method for online monitoring of a metal continuous casting process according to claim 7, characterized in that: The method for determining the average consistency of the metal continuous casting process is: The cluster with the largest overall change index is recorded as the detection cluster, and the average consistency of the metal continuous casting process is the mean of the consistency index of the friction data with the same corresponding position sequence as all temperature subsequences in the detection cluster.

9. The method for online monitoring of a metal continuous casting process according to claim 1, characterized in that: The monitoring of the state of the metal continuous casting process includes: If the normalized value of the abnormal index of the metal continuous casting process is greater than or equal to a preset value, the metal continuous casting process is abnormal and the monitoring system issues a forecast; otherwise, the metal continuous casting process is normal.

10. An online monitoring system for a metal continuous casting process, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the online monitoring method for a metal continuous casting process as described in any one of claims 1 to 9 are implemented.

Images