A seamless steel tube blank blanking abnormal condition monitoring and control system

Abstract

The present application relates to the technical field of working condition monitoring control, in particular to a seamless steel tube blank blanking abnormal working condition monitoring and control system, the system comprises a processor and a memory, the processor executes the computer program stored in the memory to realize the following steps: obtaining the fitness index value corresponding to the initial weight factor combination of the historical sample seamless steel tube set, using the particle swarm optimization algorithm to iterate the initial weight factor combination, calculating the fitness index value corresponding to the new weight factor combination obtained by each iteration, stopping iteration until the elbow inflection point of the fitness index value appears, selecting the maximum fitness index value as the optimal index value of the historical sample seamless steel tube set, and the weight factor combination corresponding to the maximum fitness index value as the optimal weight factor combination, and monitoring and controlling the blanking production process of the tube based on the optimal index value and the optimal weight factor combination. And the present application can improve the effect of abnormal monitoring and control of the blanking production process of the tube.

Description

Technical Field

[0001] This invention relates to the field of working condition monitoring and control technology, specifically to a monitoring and control system for abnormal working conditions during seamless steel pipe billet cutting. Background Technology

[0002] Seamless steel pipes typically require billet cutting during production to meet the length requirements of subsequent rolling or processing steps. Currently, during the billet cutting stage of seamless steel pipe production, in order to promptly detect and address equipment failures, operational errors, or environmental anomalies, and to prevent production interruptions, quality defects, or safety accidents, it is usually necessary to monitor and control abnormal operating conditions during the billet cutting process.

[0003] In the seamless steel pipe billet blanking production stage, abnormal operating conditions are usually monitored and controlled using traditional methods, such as comparing the monitored equipment current with the threshold. However, seamless steel pipe billet blanking is a complex production process. Not only are there multiple production steps involved in the blanking process, but there are also differences in billet specifications. This leads to poor effectiveness of monitoring and controlling abnormal operating conditions in seamless steel pipe billet blanking using traditional methods, such as false identification or missed identification. Therefore, how to improve the effectiveness of monitoring and controlling abnormal operating conditions in seamless steel pipe billet blanking has become an urgent problem to be solved. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a monitoring and control system for abnormal conditions during seamless steel pipe billet cutting. The specific technical solution adopted is as follows:

[0005] One embodiment of the present invention provides a monitoring and control system for abnormal conditions during seamless steel pipe billet cutting, including a processor and a memory. The processor executes a computer program stored in the memory to perform the following steps:

[0006] Obtain the historical sample seamless steel pipe set, the billet specification parameters of the historical sample seamless steel pipe in the historical sample seamless steel pipe set, and the current data sequence corresponding to the target billet blanking production process of the historical sample seamless steel pipe;

[0007] Based on the initial weight factor combination, the differences in billet specification parameters between different historical seamless steel pipes in the historical sample seamless steel pipe set, and the differences in current data sequences corresponding to the target billet blanking production process of different historical seamless steel pipes, the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The initial weight factor combination is iterated using the particle swarm optimization algorithm, and the fitness index value corresponding to the new weight factor combination obtained in each iteration is calculated until the fitness index value reaches the elbow inflection point and the iteration stops. The maximum fitness index value is selected as the optimal index value under the historical sample seamless steel pipe set, and the weight factor combination corresponding to the maximum fitness index value is selected as the optimal weight factor combination.

[0008] Based on the optimal index values ​​and optimal weight factor combinations of the historical sample seamless steel pipe set, the target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled.

[0009] Beneficial Effects: This invention first obtains a historical sample seamless steel pipe set, along with the billet specification parameters of the historical sample seamless steel pipes within the set and the current data sequence corresponding to the target billet blanking production process of the historical sample seamless steel pipes. Then, based on the initial weight factor combination, the differences in billet specification parameters between different historical sample seamless steel pipes within the historical sample seamless steel pipe set, and the differences in the current data sequences corresponding to the target billet blanking production process of different historical sample seamless steel pipes, the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The initial weight factor combination is iterated using a particle swarm optimization algorithm, and the fitness index value corresponding to the new weight factor combination obtained in each iteration is calculated until the fitness index value reaches an elbow inflection point, at which point the iteration stops. The maximum fitness index value is selected as the optimal index value under the historical sample seamless steel pipe set, and the weight factor combination corresponding to the maximum fitness index value is selected as the optimal weight factor combination. Finally, based on the optimal index value and the optimal weight factor combination under the historical sample seamless steel pipe set, the target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled. Furthermore, this invention uses the optimal index value and optimal weight factor combination obtained by jointly analyzing the billet specification parameters of historical seamless steel pipes and the current data sequence corresponding to the target billet blanking production process of historical seamless steel pipes as the basis for monitoring and controlling the target billet blanking production process of the seamless steel pipe to be monitored. This can improve the effect of abnormal monitoring and control of the seamless steel pipe billet blanking production process, such as improving the ability to identify abnormal working conditions of billet blanking. Attached Figure Description

[0010] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a flowchart of a method for monitoring and controlling abnormal conditions during seamless steel pipe billet cutting according to the present invention. Detailed Implementation

[0012] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the protection scope of the embodiments of the present invention.

[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0014] This embodiment provides a monitoring and control system for abnormal conditions during seamless steel pipe billet cutting, including a processor and a memory. The processor executes a computer program stored in the memory to implement a method for monitoring and controlling abnormal conditions during seamless steel pipe billet cutting, such as... Figure 1 As shown, the method for monitoring and controlling abnormal conditions during seamless steel pipe billet cutting includes the following steps:

[0015] Step S001: Obtain the historical sample seamless steel pipe set, the billet specification parameters of the historical sample seamless steel pipes in the historical sample seamless steel pipe set, and the current data sequence corresponding to the target billet blanking production process of the historical sample seamless steel pipes.

[0016] Since this embodiment relies on the management specifications of normal historical seamless steel pipes and the current signals of the production equipment used in each blanking process when monitoring abnormal conditions during billet cutting, this embodiment will first obtain the billet specifications of normal historical seamless steel pipes and the current signals of the production equipment used in each blanking process of normal historical seamless steel pipes. The specific acquisition process is as follows:

[0017] First, a predetermined number of normal historical seamless steel pipes are selected and recorded as historical sample seamless steel pipes. The set of all historical sample seamless steel pipes is recorded as the historical sample seamless steel pipe set. It is required that the billet specification parameter sequences of any two historical sample seamless steel pipes in the set are different. The billet specification parameter sequence of the historical sample seamless steel pipe is composed of all the billet specification parameters of the historical sample seamless steel pipe. The billet specification parameter types at the same position in the billet specification parameter sequences of different historical sample seamless steel pipes are the same. Normal historical seamless steel pipes refer to seamless steel pipes that have completed the entire billet cutting process and where no abnormalities were detected during the entire billet cutting process, or seamless steel pipes marked as having normal billet cutting conditions. Historical seamless steel pipes can be selected from the database of seamless steel pipe manufacturers or factories. Since enterprises automatically mark the normal or abnormal status of the seamless steel pipe billet cutting process, seamless steel pipes that are not marked as abnormal in billet cutting or those marked as normal in billet cutting can be selected from the database. In specific applications, implementers can set a preset number of values ​​based on actual conditions such as calculation volume and monitoring accuracy, such as setting the preset number to 50. Then, all billet specification parameters of each historical sample seamless steel pipe are obtained. In this embodiment, the billet specification parameter types include the length, diameter, wall thickness, and weight of the billet, that is, the billet specification parameter types and quantities corresponding to all historical sample seamless steel pipes are consistent.

[0018] Next, the equipment operating current data at each monitoring moment during the complete stage of the billet blanking production process of each historical sample seamless steel pipe production is obtained. The time sequence composed of the equipment operating current data at each monitoring moment during the complete stage of the billet blanking production process of each historical sample seamless steel pipe production is recorded as the original data sequence corresponding to the complete stage of the billet blanking production process of each historical sample seamless steel pipe. In this embodiment, the billet blanking production process includes feeding and conveying, length positioning, sawing, conveyor roller, weighing device, and warehousing. For any billet, the billet has completed the entire billet blanking production process after continuously completing feeding and conveying, length positioning, sawing, conveyor roller, weighing device, and warehousing. The specific explanation of obtaining the original data sequence corresponding to the complete stage of the billet blanking production process of any seamless steel pipe is as follows: when the seamless steel pipe production enters the billet blanking production process, the current sensor is used to monitor the billet blanking process. The operating current of key equipment in the production process is collected, and the operating current data collected at each monitoring moment is recorded as the equipment operating current data at each monitoring moment in the complete stage of the seamless steel pipe billet blanking production process. Then, the time sequence composed of all the equipment current data collected in the complete stage of the seamless steel pipe billet blanking production process is recorded as the original data sequence corresponding to the complete stage of the seamless steel pipe billet blanking production process. The complete stage of any seamless steel pipe billet blanking production process refers to the time period from the start time of entering the billet blanking production process to the last moment of the billet blanking production process during the production of the seamless steel pipe. That is, the time range of any complete stage of the billet blanking production process refers to the continuous time period composed of the start time of the process and the last moment of the completion of all operations of the process. For example, the complete stage of any seamless steel pipe sawing process refers to the time period from the start time of the billet entering the sawing process to the end time of the sawing process during the production of the seamless steel pipe.The key equipment in the material feeding and conveying process is the feeding rack mechanism and the conveyor roller conveyor. The feeding rack mechanism is an electromagnetic chuck or a robotic arm drive motor. The execution sequence of the material feeding and conveying process is to first pass through the feeding rack mechanism and then through the conveyor roller conveyor. The operating current of the electromagnetic chuck or robotic arm drive motor and the operating current of the conveyor roller conveyor drive motor are collected. The key equipment in the length positioning process is a servo length positioning mechanism, and the operating current of the servo length positioning mechanism motor is collected. The key equipment in the sawing process is a saw or sawing machine, and the operating current of the saw or sawing machine is collected. The key equipment in the conveyor roller conveyor is a drive device, and the operating current of the drive device motor is collected. The key equipment in the weighing device is a weighing sensor, and the operating current of the weighing sensor is collected. The key equipment in the warehousing and storage process is a stacker crane, and the operating current of the stacker crane is collected. The current collection time interval or collection frequency needs to be set by the implementer according to the actual situation. For example, in this embodiment, the current collection time interval or the time interval between adjacent monitoring moments can be set to 0.1 seconds.

[0019] Automated billet blanking systems are typically installed on production lines. These systems generally include multiple processes such as a loading rack mechanism, a servo-controlled length-setting mechanism, a sawing device, a conveyor roller conveyor, a weighing device, and a storage mechanism. The automatic control system enables continuous operations such as billet loading, length-setting, and sawing. For example, the billet is first conveyed from the rack to the length-measuring position for length measurement. Then, it enters the length-setting station via the conveyor roller conveyor, where the servo-controlled length-setting mechanism sets the length and the billet is fixed by a clamping mechanism. The sawing device then completes the length-setting sawing process. After sawing, the billet enters the weighing station via the conveyor roller conveyor for weight detection. Finally, based on the length and storage location information, the material is conveyed to the corresponding storage location, thus completing the entire billet blanking process. Therefore, seamless steel pipe billet blanking production is completed by an automated billet blanking system. Seamless steel pipe billet blanking refers to the process of precisely cutting steel ingots or solid round steel (called billets) into fixed-length blanks that meet the requirements of subsequent piercing processes at the initial stage of the seamless steel pipe production process.

[0020] After obtaining the original data sequences corresponding to the complete stages of the billet blanking production process for each historical seamless steel pipe sample, the original data sequences are preprocessed, and the preprocessed sequences are recorded as the current data sequences corresponding to each billet blanking production process for each historical seamless steel pipe sample. Data preprocessing includes noise reduction filtering, outlier removal, and other processes, which are known techniques.

[0021] Since the impact of various specifications and parameters on equipment load varies in different billet blanking production processes of seamless steel pipes, this embodiment will model and analyze different billet blanking production process stages to ensure the accuracy of abnormal operating condition monitoring and the monitoring and control effect. Since the abnormal operating condition monitoring method for each billet blanking production process in this embodiment is the same, this embodiment will describe the abnormal operating condition monitoring process of any billet blanking production process as an example for ease of understanding. For example, the abnormal operating condition monitoring process of the sawing process can be described as an example and recorded as the target billet blanking production process. That is, the current data used later is the data under the same billet blanking production process stage.

[0022] Therefore, this embodiment can obtain the historical sample seamless steel pipe set, the billet specification parameters of the historical sample seamless steel pipe in the historical sample seamless steel pipe set, and the current data sequence corresponding to the target billet blanking production process of the historical sample seamless steel pipe through the above process.

[0023] Step S002: Based on the initial weight factor combination, the differences in billet specification parameters between different historical seamless steel pipes in the historical sample seamless steel pipe set, and the differences in current data sequences corresponding to the target billet blanking production process of different historical seamless steel pipes, the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The initial weight factor combination is iterated using the particle swarm optimization algorithm, and the fitness index value corresponding to the new weight factor combination obtained in each iteration is calculated until the fitness index value reaches the elbow inflection point, at which point the iteration stops. The maximum fitness index value is selected as the optimal index value under the historical sample seamless steel pipe set, and the weight factor combination corresponding to the maximum fitness index value is selected as the optimal weight factor combination.

[0024] During the seamless steel pipe billet blanking process, the diversity of billet specifications and the dynamic fluctuations in loads during sawing, feeding, and other processes lead to significant non-steady-state characteristics in the current signals at each stage. Furthermore, in actual production, abnormal conditions such as sudden changes in sawing resistance, instability of the feeding mechanism, or timing discrepancies between processes further exacerbate the signal complexity. These multiple interferences make traditional methods for monitoring abnormal billet blanking conditions ineffective in accurately identifying and effectively monitoring and controlling such conditions. To improve the monitoring and control of abnormal conditions during seamless steel pipe billet blanking, this embodiment will next perform joint modeling and analysis based on the billet specification parameters of historical sample seamless steel pipes and the current data sequence corresponding to the target billet blanking production process stages of historical sample seamless steel pipes. This will enable the monitoring and control of abnormalities in the target billet blanking production process of seamless steel pipes. The specific process is as follows:

[0025] First, an initial weight factor combination is obtained. Based on the initial weight factor combination, the differences in billet specification parameters between different historical seamless steel pipe samples in the historical sample seamless steel pipe set, and the differences in current data sequences corresponding to the target billet blanking production process of different historical seamless steel pipe samples, the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. Then, the initial weight factor combination is iterated using the particle swarm optimization algorithm, and the fitness index value corresponding to the new weight factor combination obtained in each iteration is calculated until the fitness index value reaches an elbow inflection point, at which point the iteration stops. The maximum fitness index value is selected as the optimal index value under the historical sample seamless steel pipe set, and the weight factor combination corresponding to the maximum fitness index value is selected as the historical optimal weight factor combination. The weight factors at the same position in different weight factor combinations are weight factors of the same billet specification parameter type. Furthermore, based on the new weight factor combination, the differences in billet specification parameters between different historical seamless steel pipe samples in the historical sample seamless steel pipe set, and the differences in current data sequences corresponding to the target billet blanking production process of different historical seamless steel pipe samples, the following is obtained: The fitness index value corresponding to the new weight factor combination under the historical sample seamless steel pipe set is obtained by the same method as the initial weight factor combination under the historical sample seamless steel pipe set. The elbow inflection point is determined based on the elbow rule, which is a known technique. The optimal index value under the historical sample seamless steel pipe set can characterize the features presented by the normal target billet blanking production process or the benchmark value under the healthy state of the target billet blanking production process. It is also a quantitative benchmark of the optimal correlation law between specifications and current under the normal working conditions of the historical target billet blanking production process. It represents the optimal data distribution state that the system should have when the target billet blanking production process is normal. The purpose of the above optimization is to find the most stable and reasonable key law between specifications and current based on historical samples, or to make the point to be analyzed show a stable and reasonable trend, so as to establish a reliable normal benchmark, provide a sensitive scale for subsequent abnormal working condition identification, and improve the accuracy and sensitivity of abnormal working condition identification. The process of the particle swarm optimization algorithm iterating on the initial weight factor combination is a known technique. The above-mentioned optimal combination of historical weighting factors can most effectively normalize the fluctuations in normal current caused by specification differences. The more accurate the optimal index value under the historical sample seamless steel pipe set is, the better it can identify abnormal working conditions or distinguish between normal current fluctuations and abnormal working conditions caused by specification differences, or ensure the ability to distinguish between normal current fluctuations and abnormal working conditions caused by specification differences.

[0026] In this embodiment, the initial weight factor combination consists of the initial weight factors of all the above-mentioned billet specification parameter types, and the initial weight factors of all billet specification parameter types are the same and sum to 1; in addition, any weight factor combination consists of the weight factors of all billet specification parameter types, and the sum of all weight factors in any weight factor combination is 1.

[0027] In this embodiment, the specific process of obtaining the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set, based on the initial weight factor combination, the differences in billet specification parameters between different historical sample seamless steel pipes in the historical sample seamless steel pipe set, and the differences in current data sequences corresponding to the target billet blanking production process of different historical sample seamless steel pipes, is as follows:

[0028] First, based on the initial weighting factor combination and the differences in billet specification parameters between different historical seamless steel pipe samples in the historical sample seamless steel pipe set, the comprehensive difference characterization value of specification parameters of different historical seamless steel pipe samples under the initial weighting factor combination is obtained. The specific process for obtaining the comprehensive difference characterization value of specification parameters between the i-th and j-th historical seamless steel pipe samples under the initial weighting factor combination is as follows: Calculate the normalized result of the absolute value of the difference between the billet specification parameters of the i-th and j-th historical seamless steel pipe samples under each billet specification parameter type, and record it as the corresponding billet specification parameter. The difference in specification parameters between the i-th and j-th historical seamless steel pipes under each billet specification parameter type is calculated by multiplying the initial weighting factor of each billet specification parameter type with the difference in specification parameters between the i-th and j-th historical seamless steel pipes under the corresponding specification parameter type, and recording this as the weighted difference in specification parameters between the i-th and j-th historical seamless steel pipes under the corresponding billet specification parameter type. The sum of these weighted differences in specification parameters between the i-th and j-th historical seamless steel pipes under all billet specification parameter types is recorded as the comprehensive difference in specification parameters between the i-th and j-th historical seamless steel pipes under the initial weighting factor combination, expressed as: Where Norm() is the linear normalization function, and M is the total number of billet specification parameter types. The initial weighting factor for the m-th billet specification parameter type, Let be the billet specification parameters of the i-th historical sample seamless steel pipe under the m-th billet specification parameter type. Let be the billet specification parameters of the j-th historical seamless steel pipe under the m-th billet specification parameter type. The billet specification parameters of any historical seamless steel pipe under the m-th billet specification parameter type are the specification parameters belonging to the m-th billet specification parameter type among all the billet specification parameters of that historical seamless steel pipe. The comprehensive difference characterization value of specification parameters between the i-th and j-th historical seamless steel pipes under the initial weight factor combination can measure the difference of all billet specification parameters between the i-th and j-th historical seamless steel pipes under the initial weight factor combination. The larger the comprehensive difference characterization value of specification parameters, the larger the difference of billet specification parameters between the i-th and j-th historical seamless steel pipes.

[0029] In this embodiment, after quantifying the comprehensive difference characterization value of specification parameters, the difference between the current data sequences corresponding to the target billet blanking production process of different historical seamless steel pipe samples in the historical sample seamless steel pipe set is used to obtain the current difference characterization value of different historical seamless steel pipe samples under the initial weight factor combination. Moreover, the current difference characterization value between the i-th and j-th historical seamless steel pipe samples in the historical sample seamless steel pipe set under the initial weight factor combination is the normalized result of the DTW distance between the current data sequences corresponding to the target billet blanking production process of the i-th and j-th historical seamless steel pipe samples. The DTW distance is the dynamic time warping distance, which is also normalized using Norm(). The larger the DTW distance, the greater the difference between the two current data sequences.

[0030] Furthermore, during the production of seamless steel pipe blanks, the current data will fluctuate to a certain extent due to changes in the specifications of the blanks. Therefore, in order to effectively distinguish between current fluctuations caused by specification differences and abnormal operating conditions, this embodiment needs to perform quantitative modeling based on the specification differences of historical samples. Thus, it is necessary to quantify the comprehensive difference characterization value of specification parameters and the current difference characterization value. In other words, the comprehensive difference characterization value of specification parameters and the current difference characterization value are the key parameters for realizing abnormal operating condition monitoring.

[0031] Because the fluctuations in equipment current are mainly caused by the dimensional differences of historical seamless steel pipe samples, this embodiment constructs a two-dimensional mapping space after obtaining the comprehensive dimensional difference characterization value of dimensional parameters and the dimensional difference characterization value of current under the target billet blanking production process. The horizontal axis of the two-dimensional mapping space is the comprehensive dimensional difference characterization value of dimensional parameters, and the vertical axis is the dimensional difference characterization value of current. Then, the comprehensive dimensional difference characterization value of dimensional parameters and the dimensional difference characterization value of current under the initial weight factor combination of different historical seamless steel pipe samples in the historical seamless steel pipe sample set are mapped to the two-dimensional mapping space. That is, the historical seamless steel pipe samples in the historical seamless steel pipe sample set are combined pairwise without repetition to obtain all historical seamless steel pipe sample groups. The two historical seamless steel pipe samples in each historical seamless steel pipe sample group are then combined. The comprehensive difference values ​​of specification parameters and current difference values ​​of historical seamless steel pipe samples under the initial weight factor combination are mapped to a two-dimensional mapping space, and the mapped points are recorded as the points to be analyzed corresponding to the initial weight factor combination under the historical seamless steel pipe set. One historical seamless steel pipe group corresponds to one point to be analyzed. The x-coordinate of the point to be analyzed corresponding to any historical seamless steel pipe group is the comprehensive difference value of specification parameters of two historical seamless steel pipe samples in that group under the initial weight factor combination, and the y-coordinate is the current difference value of two historical seamless steel pipe samples in that group under the initial weight factor combination. That is, the purpose of mapping the comprehensive difference value of specification parameters and the current difference value is to regularize the normal current fluctuations caused by specification differences, so as to effectively distinguish between normal current fluctuations and abnormal operating conditions caused by specification changes, and improve the monitoring and identification capability of abnormal operating conditions.

[0032] When the weights are reasonable or the points to be analyzed under the corresponding weight factor combination can establish a reliable normal benchmark and effectively distinguish between current fluctuations caused by specification differences and abnormal billet cutting conditions, or when the points to be analyzed under the corresponding weight factor combination can more effectively regularize the normal current fluctuations caused by specification differences, the points to be analyzed under the corresponding weight factor combination will exhibit a stable and reasonable trend. That is, the points to be analyzed under the corresponding weight factor combination should be distributed along a certain continuous trend, forming a change curve with a certain stability, monotonicity, and smoothness. In other words, the more stable the distribution of the points to be analyzed under the weight factor combination and the more obvious the monotonicity, the more it indicates that the corresponding weight factor combination can most effectively regularize the normal current fluctuations caused by specification differences. The more regular the normal current fluctuations, the more accurately abnormal operating conditions can be identified; that is, a stable and consistent correlation exists between changes in specifications and current changes. Therefore, the stability and monotonicity of the distribution of the points to be analyzed under different weight factor combinations characterize the suitability of the corresponding weight factor combinations. Thus, this embodiment will next obtain the suitability index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set based on the distribution characteristics of the points to be analyzed in the two-dimensional mapping space under the initial weight factor combination. The specific process for obtaining the suitability index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set based on the distribution characteristics of the points to be analyzed in the two-dimensional mapping space under the initial weight factor combination is as follows:

[0033] First, obtain the upper and lower envelopes of the points to be analyzed in the two-dimensional mapping space. The upper envelope is a broken line formed by connecting the points with the largest ordinates corresponding to each abscissa in the abscissa sequence in ascending order of abscissa. The lower envelope is a broken line formed by connecting the points with the smallest ordinates corresponding to each abscissa in the abscissa sequence in ascending order of abscissa. The abscissa sequence refers to the result of arranging all abscissas appearing in all points to be analyzed in the two-dimensional mapping space from smallest to largest. The point with the largest ordinate corresponding to any abscissa is the point with the largest ordinate value among all points to be analyzed with abscissa value of that abscissa. Then, based on the upper and lower envelopes of the points to be analyzed in the two-dimensional mapping space, obtain the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set. The larger the stability evaluation value, the more stable the distribution of the points to be analyzed under the initial weight factor combination, and the more effectively the weight factor combination can mitigate the influence of specification differences. The more regular the normal current fluctuations, the more accurately abnormal operating conditions can be identified. Then, the points to be analyzed are sorted in ascending order of their horizontal coordinates to obtain a sequence of points to be analyzed. If points have the same horizontal coordinate, the point with the smaller vertical coordinate is placed first. Next, based on the slope between adjacent points in the sequence, the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The larger the monotonicity evaluation value, the more obvious the monotonically increasing characteristic of the distribution of points to be analyzed under the initial weight factor combination, and the more effectively the weight factor combination can regularize the normal current fluctuations caused by specification differences, and the more accurately abnormal operating conditions can be identified. Finally, the product of the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set and the monotonicity evaluation value is calculated and recorded as the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set. The larger the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set, the closer this weight is to the optimal state. It can more effectively regularize the normal current fluctuations caused by specification differences, and can provide a reliable basis for the accurate identification of subsequent abnormal working conditions. It also indicates that the fitness index value analyzed under this weight factor combination can more effectively distinguish between current fluctuations caused by specification differences and abnormal working conditions of billet cutting, or can more effectively identify abnormal working conditions of billet cutting.

[0034] The specific process for obtaining the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set based on the upper and lower envelopes of the points to be analyzed in the two-dimensional mapping space is as follows: Calculate the vertical distance between the upper and lower envelopes at each horizontal coordinate in the horizontal coordinate sequence, and record it as the envelope width under the corresponding horizontal coordinate. That is, the vertical distance between the upper and lower envelopes at any horizontal coordinate refers to the absolute value of the difference between the ordinate values ​​at two positions on the upper and lower envelopes where the horizontal coordinate is that horizontal coordinate; calculate the mean of the envelope widths under all horizontal coordinates in the horizontal coordinate sequence, and record it as the average width between the upper and lower envelopes; perform a negative correlation mapping on the average width and record the mapping result as the first eigenvalue; calculate the mean of the absolute values ​​of the differences between the envelope widths under each horizontal coordinate in the horizontal coordinate sequence and the average width, and perform a negative correlation mapping on the result, and record it as the second eigenvalue. Here, a negative exponential function with a base of constant e is used for the negative correlation mapping; calculate the product of the first eigenvalue and the second eigenvalue, and record it as the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set; the expression for the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is:

[0035]

[0036] Where W is the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set, and exp() is an exponential function with a base of constant e. Z0 is the average width between the upper and lower envelopes, and Z0 is the number of x-coordinates in the x-coordinate sequence. The width of the envelope at the z-th x-coordinate in the x-coordinate sequence; the average width between the upper and lower envelopes. The smaller the value, the narrower the overall distribution bandwidth of the points to be analyzed under the initial weight factor combination, the lower the dispersion, and the more stable the distribution; the absolute value of the difference between the envelope width and the average width of each horizontal axis in the horizontal axis sequence. The smaller the value, the smaller the deviation of the envelope width from the average width at each location, indicating a more stable change in the envelope width and better overall distribution stability of the points being analyzed; because smaller and The smaller the value of W, the larger the value of W. Therefore, the larger the value of W, the more stable the overall distribution of the points to be analyzed under the initial weight factor combination is. This weight factor combination is more effective in regularizing the normal current fluctuations caused by specification differences. It also indicates that the suitability index value corresponding to this weight factor combination under the historical sample seamless steel pipe set is more effective in identifying abnormal operating conditions or in distinguishing between normal current fluctuations and abnormal operating conditions caused by specification differences. Conversely, the smaller the value of W, the more unstable the distribution of the points to be analyzed under the initial weight factor combination is. This indicates that the weight factor combination is less effective in regularizing the normal current fluctuations caused by specification differences and has a poor ability to distinguish between normal current fluctuations and abnormal operating conditions caused by specification differences.

[0037] The specific process for obtaining the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set based on the slope between adjacent points in the analysis point sequence is as follows: The sequence formed by the slopes between adjacent points in the analysis point sequence is denoted as the slope sequence. The frequency of occurrence where the previous slope value is less than the next slope value in the slope sequence is counted and recorded as the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set. The expression for the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is:

[0038]

[0039] Where G is the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set, and U is the total number of slopes in the slope sequence. Let u+1 be the slope in the slope sequence. Let be the u-th slope in the slope sequence, and R() be the discriminant function. When the condition is true, that is... The function outputs 1 when the condition is not met. The time function output is 0. The higher the frequency of the previous slope value being less than the next slope value in the slope sequence, that is, the larger G is, the more the point to be analyzed under the initial weight factor combination conforms to the monotonically increasing trend. This indicates that the weight factor combination can more effectively regularize the normal current fluctuations caused by specification differences. It also indicates that the suitability index value corresponding to the weight factor combination under the historical seamless steel pipe set can more effectively identify abnormal working conditions or distinguish between normal current fluctuations and abnormal working conditions caused by specification differences. The smaller G is, the less the point to be analyzed under the initial weight factor combination conforms to the monotonically increasing trend. This further indicates that the weight factor combination is less effective in regularizing the normal current fluctuations caused by specification differences and has a poor ability to distinguish between normal current fluctuations and abnormal working conditions caused by specification differences.

[0040] Furthermore, this embodiment constructs a mapping relationship between the comprehensive differences in specification parameters and current differences, and introduces envelope width stability index and monotonicity index to quantify and evaluate distribution characteristics. This enables adaptive optimization of weight factors in the measurement model, thereby effectively constraining the dispersion of current differences and ensuring their regularity with specification differences. This improves the ability to distinguish between normal fluctuations and abnormal operating conditions, or the ability to distinguish between current fluctuations caused by specification differences and abnormal billet cutting conditions. This provides a more stable and reliable basis for subsequent abnormal monitoring, further enhancing the accuracy and robustness of monitoring abnormal billet cutting conditions.

[0041] Therefore, this embodiment obtains the optimal index value and the optimal combination of historical weight factors under the historical sample seamless steel pipe set through the above process.

[0042] Step S003: Based on the optimal index value and optimal weight factor combination under the historical sample seamless steel pipe set, monitor and control the target billet blanking production process of the seamless steel pipe to be monitored.

[0043] In this embodiment, after obtaining the optimal index value and the optimal weight factor combination under the historical sample seamless steel pipe set, the target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled based on the optimal index value and the optimal weight factor combination under the historical sample seamless steel pipe set. The specific process is as follows:

[0044] First, obtain any seamless steel pipe currently in the target billet blanking production process and denote it as the seamless steel pipe to be monitored under the current target billet blanking production process. Then, denote the new set formed by the historical sample seamless steel pipe set and the seamless steel pipe to be monitored under the current target billet blanking production process as the new sample seamless steel pipe set under the current target billet blanking production process. Obtain the optimal index value under the new sample seamless steel pipe set under the current target billet blanking production process stage. The method for obtaining the optimal index value under the new sample seamless steel pipe set under the current target billet blanking production process stage is the same as the method for obtaining the optimal index value under the historical sample seamless steel pipe set, and therefore will not be described again. However, the optimal index value under the new sample seamless steel pipe set under the current target billet blanking production process stage is obtained through... The parameters related to the seamless steel pipe to be monitored used when calculating the optimal index value include the billet specification parameters of the seamless steel pipe to be monitored and the current data sequence corresponding to the current target billet blanking production process stage of the seamless steel pipe to be monitored. The current data sequence corresponding to the current target billet blanking production process stage of the seamless steel pipe to be monitored refers to the stage from when the seamless steel pipe to be monitored enters the target billet blanking production process to the current target billet blanking production time, that is, the seamless steel pipe to be monitored is currently in the target billet blanking production process stage. The method for obtaining the current data sequence corresponding to the current target billet blanking production process stage of the seamless steel pipe to be monitored is the same as the method for obtaining the current data sequence corresponding to the target billet blanking production process of the aforementioned historical sample seamless steel pipes, and therefore will not be described further.

[0045] Then, the subset sequence corresponding to the historical sample seamless steel pipe set is obtained. The a-th subset in the subset sequence is the set of all historical sample seamless steel pipes remaining after removing the a-th historical sample seamless steel pipe. The fitness index value corresponding to the optimal weight factor combination under each subset in the subset sequence is obtained, and the fitness index value corresponding to the optimal weight factor combination under each subset is recorded as the optimal index value under the corresponding subset. The method for obtaining the fitness index value corresponding to the optimal weight factor combination under any subset is the same as the method for obtaining the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set, so it will not be described in detail. Then, based on the optimal index value under the historical sample seamless steel pipe set, the optimal index value under the new sample seamless steel pipe set under the current target billet blanking production process stage, and the optimal index value under each subset, the current target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled. Furthermore, based on the optimal index values ​​of the historical sample seamless steel pipe set, the optimal index values ​​of the new sample seamless steel pipe set under the current target billet blanking production process stage, and the optimal index values ​​of each subset, the specific process for monitoring and controlling the current target billet blanking production process of the seamless steel pipe to be monitored is as follows:

[0046] The ratio of the optimal index value under the historical sample seamless steel pipe set to the optimal index value under the new sample seamless steel pipe set under the current target billet blanking production process stage is denoted as the judgment index value under the current target billet blanking production process stage; the set of ratios of the optimal index values ​​under each subset in the subset sequence to the optimal index values ​​under the historical sample seamless steel pipe set is denoted as the characteristic ratio set; the ratio of the number of characteristic ratios less than the judgment index value to the total number of historical sample seamless steel pipes is denoted as the abnormal characterization value of the seamless steel pipe production to be monitored under the current target billet blanking production process stage; the expression for the abnormal characterization value of the seamless steel pipe production to be monitored under the current target billet blanking production process stage is:

[0047]

[0048] Where Y represents the abnormal characterization value of the seamless steel pipe production to be monitored at the current target billet blanking production stage, and H represents the number of subsets in the subset sequence, which is also the total number of historical sample seamless steel pipes. These are the optimal index values ​​based on the historical sample set of seamless steel pipes. These are the optimal index values ​​for the new sample of seamless steel pipes under the current target billet blanking production process stage. Let R be the optimal index value for the a-th subset in the subset sequence, and R() be the discriminant function. When the condition is met, that is... The function outputs 1 when the condition is not met. The function outputs 0. This indicates the degree of deviation of the new sample from the optimal normal state constituted by the historical seamless steel pipe set. The larger the ratio, the more significantly the addition of the new sample reduces the stability or monotonicity of the overall data, indicating that the sample has disrupted the original normal pattern and is very likely an outlier. It is a key indicator for measuring whether a new sample deviates from normal operating conditions; It characterizes the degree to which the existence of the a-th historical sample supports or disrupts the overall data distribution pattern; Greater than or equal to The more frequently the occurrences, the more damage the current seamless steel pipe under monitoring has caused to the model, reaching or exceeding the fluctuation range caused by multiple historical samples. This indicates that the fluctuation of the current seamless steel pipe under monitoring cannot be explained by normal specification differences, and is therefore highly likely caused by abnormal operating conditions. Greater than or equal to The more frequently it occurs, the larger Y becomes. Therefore, the larger Y is, the greater the probability that the state of the seamless steel pipe production under the current target billet blanking production process is abnormal, or the more likely the current target billet blanking production process of the seamless steel pipe under the monitoring is an abnormal condition. Conversely, the smaller Y is, the greater the probability that the state of the seamless steel pipe production under the current target billet blanking production process is normal, or the more likely the current target billet blanking production process of the seamless steel pipe under the monitoring is a normal condition.

[0049] Finally, based on the abnormal characterization value of the seamless steel pipe production under the current target billet blanking production process stage, abnormal monitoring and control are performed on the current target billet blanking production process of the seamless steel pipe to be monitored. Specifically, it is determined whether the abnormal characterization value of the seamless steel pipe production under the current target billet blanking production process stage is greater than the preset abnormal threshold. If it is greater, it is determined that the seamless steel pipe production under the current target billet blanking production process is in an abnormal processing state or that the current target billet blanking production process of the seamless steel pipe to be monitored is in an abnormal operating condition. At this time, production control must be implemented immediately, including but not limited to issuing alarm prompts, reducing processing speed, or stopping processing for maintenance and inspection, in order to avoid If an abnormal state or abnormal operating condition escalates and causes equipment damage or material waste, such as when the target billet blanking production process is a sawing process, it can be determined that the sawing process of the seamless steel pipe under monitoring is in an abnormal processing state or abnormal operating condition. In this case, an early warning can be issued immediately, and relevant personnel can be notified to carry out maintenance immediately. If it is determined that the abnormal characteristic value of the seamless steel pipe under monitoring at the current target billet blanking production stage is not greater than the preset abnormal threshold, it is determined that the seamless steel pipe under monitoring is in a normal processing state or that the target billet blanking production process of the seamless steel pipe under monitoring is in a normal operating condition. The existing operating state can be maintained or production and processing can continue. In specific applications, the implementer can set the preset abnormal threshold according to the actual situation. For example, in this embodiment, the preset abnormal threshold can be set to 0.75.

[0050] Thus, this embodiment completes the monitoring and control of abnormal conditions in the seamless steel pipe billet cutting process. Furthermore, this embodiment improves the monitoring and control effect by combining the billet specification parameters of historical sample seamless steel pipes with the current data sequence corresponding to the target billet cutting production process of historical sample seamless steel pipes, thereby enhancing the accuracy and sensitivity of abnormal condition detection and identification.

[0051] In summary, this embodiment first obtains the historical sample seamless steel pipe set, the billet specification parameters of the historical sample seamless steel pipes in the historical sample seamless steel pipe set, and the current data sequence corresponding to the target billet blanking production process of the historical sample seamless steel pipes. Then, based on the initial weight factor combination, the differences between the billet specification parameters of different historical sample seamless steel pipes in the historical sample seamless steel pipe set, and the differences between the current data sequences corresponding to the target billet blanking production process of different historical sample seamless steel pipes, the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The initial weight factor combination is iterated using the particle swarm optimization algorithm, and the fitness index value corresponding to the new weight factor combination obtained in each iteration is calculated until the fitness index value reaches the elbow inflection point, at which point the iteration stops. The maximum fitness index value is selected as the optimal index value under the historical sample seamless steel pipe set, and the weight factor combination corresponding to the maximum fitness index value is selected as the optimal weight factor combination. Finally, based on the optimal index value and the optimal weight factor combination under the historical sample seamless steel pipe set, the target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled. Furthermore, this embodiment uses the optimal index value and optimal weight factor combination obtained by jointly analyzing the billet specification parameters of historical seamless steel pipes and the current data sequence corresponding to the target billet blanking production process of historical seamless steel pipes as the basis for monitoring and controlling the target billet blanking production process of the seamless steel pipe to be monitored. This can improve the effect of abnormal monitoring and control of the seamless steel pipe billet blanking production process, such as improving the ability to identify abnormal billet blanking conditions.

[0052] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 this application, and should all be included within the protection scope of this application.

Claims

1. A monitoring and control system for abnormal conditions during seamless steel pipe billet cutting, comprising a processor and a memory, characterized in that, The processor executes the computer program stored in the memory to perform the following steps: Obtain the historical sample seamless steel pipe set, the billet specification parameters of the historical sample seamless steel pipe in the historical sample seamless steel pipe set, and the current data sequence corresponding to the target billet blanking production process of the historical sample seamless steel pipe; Based on the initial weight factor combination, the differences in billet specification parameters between different historical seamless steel pipes in the historical sample seamless steel pipe set, and the differences in current data sequences corresponding to the target billet blanking production process of different historical seamless steel pipes, the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The initial weight factor combination is iterated using the particle swarm optimization algorithm, and the fitness index value corresponding to the new weight factor combination obtained in each iteration is calculated until the fitness index value reaches the elbow inflection point and the iteration stops. The maximum fitness index value is selected as the optimal index value under the historical sample seamless steel pipe set, and the weight factor combination corresponding to the maximum fitness index value is selected as the optimal weight factor combination. Based on the optimal index values ​​and optimal weight factor combinations of the historical sample seamless steel pipe set, the target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled. The initial weighting factor combination consists of the initial weighting factors of all billet specification parameter types. The initial weighting factors of all billet specification parameter types are the same and their sum is 1. The method for obtaining the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set includes: Based on the initial weight factor combination and the differences in billet specification parameters between seamless steel pipes from different historical samples, the comprehensive difference characterization value of specification parameters of seamless steel pipes from different historical samples under the initial weight factor combination is obtained. Based on the differences between the current data sequences corresponding to the target billet blanking production process of seamless steel pipes from different historical samples, the current difference characterization value of seamless steel pipes from different historical samples under the initial weight factor combination is obtained. A two-dimensional mapping space is constructed, wherein the horizontal axis of the two-dimensional mapping space is the comprehensive difference characterization value of specification parameters and the vertical axis is the characterization value of current difference. The comprehensive difference characterization values ​​of specification parameters and current difference characterization values ​​of different historical sample seamless steel pipes in the historical sample seamless steel pipe set under the initial weight factor combination are all mapped to the two-dimensional mapping space, and the points obtained by mapping are all recorded as the points to be analyzed. Based on the upper and lower envelopes of the points to be analyzed in the two-dimensional mapping space, the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The points to be analyzed are sorted in ascending order of their horizontal coordinates to obtain a sequence of points to be analyzed. Based on the slope between adjacent points to be analyzed in the sequence, the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set is obtained. The product of the stability assessment value and the monotonicity assessment value is recorded as the fitness index value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set.

2. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 1, characterized in that, The specifications of the tube blank include its length, diameter, wall thickness, and weight.

3. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 1, characterized in that, The method for obtaining the comprehensive difference in specification parameters between the i-th and j-th historical seamless steel pipe samples in the historical sample seamless steel pipe set under the initial weight factor combination includes: The normalized result of the absolute value of the difference between the billet specification parameters of the i-th historical seamless steel pipe and the j-th historical seamless steel pipe under each billet specification parameter type is denoted as the specification parameter difference between the i-th and j-th historical seamless steel pipes under the corresponding specification parameter type. The product of the initial weight factor of each specification parameter type and the specification parameter difference between the i-th and j-th historical seamless steel pipes under the corresponding specification parameter type is denoted as the weighted specification parameter difference between the i-th and j-th historical seamless steel pipes under the corresponding specification parameter type. The sum of the weighted differences in specification parameters between the i-th and j-th historical seamless steel pipes under all specification parameter types is denoted as the comprehensive difference in specification parameters between the i-th and j-th historical seamless steel pipes under the initial weight factor combination.

4. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 1, characterized in that, The current difference representation value between the i-th and j-th historical seamless steel pipes in the historical sample seamless steel pipe set under the initial weight factor combination is the normalized result of the DTW distance between the current data sequences corresponding to the target billet blanking production process of the i-th and j-th historical seamless steel pipes.

5. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 1, characterized in that, The method for obtaining the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set includes: Arrange all the horizontal coordinates of the points to be analyzed in the two-dimensional mapping space in ascending order and record them as the horizontal coordinate sequence; calculate the vertical distance between the upper and lower envelopes of each horizontal coordinate in the horizontal coordinate sequence and record it as the envelope width of the corresponding horizontal coordinate; record the average envelope width of all horizontal coordinates in the horizontal coordinate sequence as the average width between the upper and lower envelopes; record the negative correlation mapping result of the average width as the first feature value; record the negative correlation mapping result of the average absolute value of the difference between the envelope width of each horizontal coordinate in the horizontal coordinate sequence and the average width as the second feature value; record the product of the first feature value and the second feature value as the stability evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set.

6. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 1, characterized in that, The method for obtaining the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set includes: The sequence of slopes between adjacent points in the analysis point sequence is denoted as the slope sequence. The frequency of occurrence of a previous slope value being less than a subsequent slope value in the slope sequence is counted and recorded as the monotonicity evaluation value corresponding to the initial weight factor combination under the historical sample seamless steel pipe set.

7. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 1, characterized in that, A method for monitoring and controlling the billet blanking production process of the seamless steel pipe to be monitored, based on the optimal index values ​​and optimal weight factor combinations of the historical sample seamless steel pipe set, includes: The new set consisting of the historical sample seamless steel pipe set and the seamless steel pipe to be monitored is denoted as the new sample seamless steel pipe set, and the optimal index value under the new sample seamless steel pipe set is obtained. Obtain the fitness index value corresponding to the optimal weight factor combination under each subset in the subset sequence corresponding to the historical sample seamless steel pipe set, and record it as the optimal index value under the corresponding subset. The a-th subset in the subset sequence corresponding to the historical sample seamless steel pipe set is the set of seamless steel pipes remaining after removing the a-th historical sample seamless steel pipe from the historical sample seamless steel pipe set. Based on the optimal index values ​​of the historical sample seamless steel pipe set, the optimal index values ​​of the new sample seamless steel pipe set, and the optimal index values ​​of each subset, the target billet blanking production process of the seamless steel pipe to be monitored is monitored and controlled.

8. The seamless steel pipe billet abnormal condition monitoring and control system as described in claim 7, characterized in that, Based on the optimal index values ​​of the historical seamless steel pipe set, the optimal index values ​​of the new seamless steel pipe set, and the optimal index values ​​of each subset, a method for monitoring and controlling the target billet blanking production process of the seamless steel pipe to be monitored includes: The ratio of the optimal index value under the historical sample seamless steel pipe set to the optimal index value under the new sample seamless steel pipe set is denoted as the judgment index value; the set of ratios of the optimal index values ​​under each subset to the optimal index values ​​under the historical sample seamless steel pipe set is denoted as the feature ratio set; and the ratio of the number of feature ratios less than the judgment index value to the total number of historical sample seamless steel pipes is denoted as the abnormal characterization value of the seamless steel pipe production to be monitored under the current target billet blanking production process stage. Based on the aforementioned abnormal characterization values, anomaly monitoring and control are performed on the current target billet blanking production process of the seamless steel pipe to be monitored.

Images