Cord steel rolling temperature trend early warning method and device based on trend accumulation and sum

Abstract

The present invention provides a trend warning method and device for the rolling temperature of cord steel based on trend accumulation and relates to the technical field of process industry quality control. The method comprises: collecting time series data of the rolling temperature during the rolling process of cord steel; preprocessing the time series data to form an analysis data set; smoothing the analysis data set based on a double exponential weighted average algorithm to extract the trend item of the rolling temperature between different batches; calculating the trend accumulation and the rolling temperature based on an improved trend accumulation and control chart; optimizing the smoothing coefficient in the double exponential weighted average algorithm and the allowable drift parameter in the trend accumulation and control chart to establish an online trend warning model; using the online trend warning model to monitor the abnormal trend of the rolling temperature online, and quantitatively evaluate the degree of trend deviation. The present invention can improve the accuracy and comprehensiveness of the trend warning of the rolling temperature of cord steel.

Description

Technical Field

[0001] The present invention relates to the technical field of process industry quality control, and in particular to a cord steel rolling temperature trend early warning method and device based on trend accumulation and sum. Background Art

[0002] As a typical representative of high-end wire rods, cord steel has a complex manufacturing process and has the characteristics of high strength, high toughness, and uniform internal structure. It has been recognized by the industry as a cutting-edge product in wire rods. Among them, the rolling temperature has a significant impact on the mechanical properties, metallographic structure, carbide and other product qualities of cord steel, and is a key control variable in the rolling process. When the rolling temperature is too high, it will accelerate the grain growth and bubble diffusion of the cord steel, reduce the mechanical properties and plasticity of the cord steel, and lead to a decrease in the strength and toughness of the product. When the rolling temperature is too low, the steel is prone to problems such as fracture and folding. Too low a temperature will also lead to insufficient precipitation of carbides in the cord steel, thereby reducing the strength and hardness of the product and affecting the mechanical properties of the product. Therefore, it is of great significance to develop process monitoring technology for the rolling temperature of cord steel.

[0003] An effective trend warning method can improve the process control level of rolling temperature and optimize the production process. In actual production, trend anomalies caused by equipment aging or process control deviations are often not discovered in time. In serious cases, they will cause batch quality accidents and bring huge losses to the company. Production managers hope to monitor the changing trends of rolling temperatures between different batches to pay close attention to the production status. However, objective reasons such as sensor noise, ambient temperature changes, differences in operating habits of different teams, frequent steel grade and process switching, etc., bring huge challenges to the trend warning of rolling temperature. Therefore, the existing conventional methods cannot comprehensively and accurately warn of abnormal trends in the rolling temperature of cord steel. Summary of the invention

[0004] In view of the above problems, the purpose of the present invention is to provide a method and device for warning the trend of the cord steel rolling temperature based on the cumulative sum of trends, so as to solve the problem of inaccurate warning of abnormal trend of the cord steel rolling temperature by conventional methods in the prior art, and to improve the accuracy and comprehensiveness of warning of the trend of the cord steel rolling temperature.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] On the one hand, a cord steel rolling temperature trend early warning method based on trend accumulation and sum is provided, the method comprising the following steps:

[0007] S1. Collecting time series data of rolling temperature during the rolling process of cord steel;

[0008] S2, preprocessing the time series data to form an analysis data set;

[0009] S3, smoothing the analysis data set based on the double exponential weighted average algorithm to extract the trend item of the rolling temperature between different batches;

[0010] S4. Calculating the trend cumulative sum of the rolling start temperature based on the improved trend cumulative sum control chart;

[0011] S5. Optimize the smoothing coefficient in the double exponential weighted average algorithm and the allowed drift parameter in the trend accumulation and control chart to establish an online trend warning model;

[0012] S6. Use the online trend warning model to monitor the abnormal trend of the rolling temperature online and quantitatively evaluate the degree of trend deviation.

[0013] Optionally, the step S1 includes:

[0014] The high-frequency data acquisition interface is used to collect time series data from the production site in real time. The sampling frequency is set according to the rolling speed to perform equal-interval sampling and store it in the time series database.

[0015] The time series data is interpolated in combination with the rolling speed to eliminate missing values ​​and outliers.

[0016] Optionally, step S2 includes:

[0017] According to the billet material number obtained by the spray number recognition system at the entrance of the heating furnace and combined with the material tracking algorithm, a unique material number is matched for each temperature curve in the time series data to achieve material tracking and batch segmentation of the rolling temperature;

[0018] In the process of extracting the rolling temperature features, after removing the temperature of 5% of the head length, the maximum value of the temperature curve is taken to represent the rolling temperature of this batch.

[0019] Optionally, step S3 includes:

[0020] Suppose the data sequence in the analysis data set is , the double exponential weighted average algorithm DEWMA is defined as shown in formula (1):

[0021] Formula (1)

[0022] Where EWMA represents the moving exponential weighted average algorithm, DEWMA represents the double exponential weighted average algorithm; xi represents the i-th data, 1≤i≤n, and n is the total number of data; λ∈(0,1] represents the smoothing coefficient. The closer the data is to the current moment, the greater the weight it has, and the farther the data is from the current moment, the smaller the weight it has.

[0023] Optionally, in step S4, improving the trend accumulation and control chart includes:

[0024] In the traditional trend accumulation and control chart, is a data sequence, and two assumptions are made for this data sequence: the process is controlled and the process is not controlled, as shown in equations (2) and (3):

[0025] Formula (2)

[0026] Formula (3)

[0027] Among them, H0 represents the assumption that the process is controlled, H1 represents the assumption that the process is not controlled; xi represents the i-th data, n is the total number of data; N(μ0,σ) and N(μ1,σ) represent normal distributions with mean μ0 and μ1 and variance σ respectively; ξ represents the number of samples that obey the N(μ0,σ) distribution, and δ represents the degree of mean shift;

[0028] Let f(x) represent the probability density function of the normal distribution, and the likelihood functions corresponding to the two hypotheses are shown in equations (4) and (5):

[0029] Formula (4)

[0030] Formula (5)

[0031] Among them, L(n,μ0) is the likelihood function of hypothesis H0, and L(n,μ1) is the likelihood function of hypothesis H1; ∏ represents the multiplication operation, f0(xi) and f1(xi) represent the probability density functions corresponding to the normal distribution in hypothesis H0 and hypothesis H1 respectively;

[0032] The likelihood ratio Ln of hypothesis H1 to hypothesis H0 is shown in formula (6):

[0033]

[0034] Formula (6)

[0035] Maximizing Ln and taking its logarithm yields:

[0036] Formula (7)

[0037] Formula (8)

[0038] in, and Respectively represent the cumulative sum of upper deviation and lower deviation of the i-th data; and They represent the cumulative sum of upper deviation and lower deviation of the i-1th data respectively; k is the allowable drift parameter, which represents the allowable offset of the data;

[0039] By changing μ0 in the traditional trend cumulative sum control chart to the smoothed data xi-1 at the previous moment, we can get the improved trend cumulative sum control chart, as shown in equations (9) and (10):

[0040] Formula (9)

[0041] Formula (10)

[0042] The monitoring control limit CL and the allowable drift parameter k are set according to the actual situation of the cord steel rolling temperature;

[0043] like , then it is considered that there is an upward shift trend; if , then it is considered that there is a downward trend; if and , it is considered that no significant deviation has occurred.

[0044] Optionally, the step S4 further comprises: a relative standard deviation calculation method based on process specifications;

[0045] The relative standard deviation yi of the rolling temperature and process specification of each billet is calculated by equation (11) and equation (12);

[0046] Formula (11)

[0047] Formula (12)

[0048] Where yi represents the relative standard deviation of the ith billet, Ti represents the characteristic value of the rolling temperature of the ith billet; Taim represents the control target of the rolling temperature; TUCL represents the upper control limit of the rolling temperature; TLCL represents the lower control limit of the rolling temperature, and .

[0049] Optionally, step S5 includes:

[0050] The smoothing coefficient λ in the double exponential weighted average algorithm and the allowable drift parameter k in the trend accumulation and control chart are optimized; the smoothing coefficient λ is used to control the smoothing intensity of the rolling temperature, and the allowable drift parameter k is used to control the allowable offset of the monitoring data.

[0051] Optionally, the step S5 further includes: an online monitoring process for early warning of rolling temperature trend;

[0052] Firstly, the relative standard deviation of the rolling temperature of each billet is calculated according to the upper and lower control limits of the rolling temperature specified in the process specification; secondly, it is smoothed by the double exponential weighted average method, and the cumulative trend sum of the rolling temperature is further calculated; then, according to the set monitoring control limits, it is judged whether there is an abnormal trend and its deviation direction, and the end position of the trend is determined during the cyclic monitoring process; finally, after the process returns to stability, the parameters are reset to continue the new online monitoring process.

[0053] Optionally, step S6 includes:

[0054] When the upward trend or downward trend ends, the cumulative trend sum at that position serves as a quantitative evaluation indicator of the degree of trend deviation.

[0055] On the other hand, a cord steel rolling temperature trend early warning device based on trend accumulation and is provided, which is used to implement any of the above methods, and the device comprises:

[0056] The data acquisition module is used to collect the time series data of the rolling temperature during the rolling process of the cord steel;

[0057] The preprocessing module is used to preprocess the time series data to form an analysis data set;

[0058] The first calculation module is used to smooth the analysis data set based on a double exponential weighted average algorithm to extract the trend item of the rolling temperature between different batches;

[0059] A second calculation module is used to calculate the trend cumulative sum of the rolling start temperature based on the improved trend cumulative sum control chart;

[0060] The optimization module is used to optimize the smoothing coefficient in the double exponential weighted average algorithm and the allowed drift parameters in the trend accumulation and control chart, and establish an online trend warning model;

[0061] The monitoring module is used to monitor the abnormal trend of the rolling temperature online using the online trend warning model, and quantitatively evaluate the degree of trend deviation.

[0062] In another aspect, an electronic device is provided, the electronic device comprising:

[0063] processor;

[0064] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are loaded and executed by the processor, the steps of the above-mentioned cord steel rolling temperature trend early warning method are implemented.

[0065] On the other hand, a computer-readable storage medium is provided, in which a program code is stored. The program code can be called by a processor to execute the steps of the above-mentioned cord steel rolling temperature trend early warning method.

[0066] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0067] The method of the present invention solves the problem of insufficient accuracy and versatility of conventional trend warning in the prior art. It warns of abnormal trends in the rolling temperature of cord steel under the premise of considering the influence of noise and steel type and process switching, and gives a quantitative evaluation of abnormal trends, which helps to find abnormalities in equipment or processes such as heating furnaces and descaling water that affect the rolling temperature. In addition, the method of the present invention can also be applied to other metallurgical processes and even different scenarios of process industries through retraining and parameter adjustment, such as process parameters such as superheat in the smelting process, and quality indicators such as tensile strength, to provide technical support for improving process control stability and preventing batch quality accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0069] Figure 1 It is a flow chart of a cord steel rolling temperature trend early warning method based on trend accumulation and provided by an embodiment of the present invention;

[0070] Figure 2 It is a schematic diagram of the equipment layout of a production line provided by an embodiment of the present invention;

[0071] Figure 3 It is a schematic diagram of the original time series data trend of the rolling temperature provided by an embodiment of the present invention;

[0072] Figure 4 is a schematic diagram of characteristic values ​​of the steel billet rolling temperature to be analyzed provided by an embodiment of the present invention;

[0073] Figure 5 is a trend graph after DEWMA smoothing provided by an embodiment of the present invention;

[0074] Figure 6 is a schematic diagram of judgment logic provided by an embodiment of the present invention;

[0075] Figure 7 It is a schematic diagram of an online monitoring process provided by an embodiment of the present invention;

[0076] Figure 8is a TCUSUM control diagram provided by an embodiment of the present invention;

[0077] Fig. 9 is a temperature trend graph with abnormal trend start and end position marks provided by an embodiment of the present invention;

[0078] Fig.10 It is a structural schematic diagram of a cord steel rolling temperature trend early warning device based on trend accumulation and provided by an embodiment of the present invention;

[0079] Fig.11 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0081] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0082] The embodiment of the present invention provides a cord steel rolling temperature trend early warning method based on trend accumulation and the method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 As shown, the method comprises the following steps:

[0083] S1. Collecting time series data of rolling temperature during the rolling process of cord steel;

[0084] S2, preprocessing the time series data to form an analysis data set;

[0085] S3, smoothing the analysis data set based on the double exponential weighted average algorithm to extract the trend item of the rolling temperature between different batches;

[0086] S4. Calculating the trend cumulative sum of the rolling start temperature based on the improved trend cumulative sum control chart;

[0087] S5. Optimize the smoothing coefficient in the double exponential weighted average algorithm and the allowed drift parameter in the trend accumulation and control chart to establish an online trend warning model;

[0088] S6. Use the online trend warning model to monitor the abnormal trend of the rolling temperature online and quantitatively evaluate the degree of trend deviation.

[0089] The method of the present invention can solve the problems of insufficient accuracy and versatility of conventional trend warnings in the prior art, and warns of abnormal trends in the cord steel rolling temperature under the premise of considering noise influence and steel type and process switching, and gives a quantitative evaluation of the abnormal trend, thereby improving the temperature control stability of the cord steel rolling process.

[0090] Furthermore, in step S1, time series data is collected from the production site in real time through a high-frequency data acquisition interface, and sampling is performed at equal intervals according to the sampling frequency set according to the rolling speed. The obtained time series data records rich temperature change information and is stored in a time series database.

[0091] In the embodiment of the present invention, the equipment layout of the high-quality wire production line of a steel plant is as follows: Figure 2 As shown in the figure, the rolling temperature measurement point indicated by the large triangle icon is located at the entrance of the first rolling mill, which is used to measure the surface temperature of the steel billet after the heating furnace is heated and the high-pressure water is used to remove the iron oxide scale. Since the temperature sensor is fixed and the steel billet moves along the roller, the rolling temperature data obtained is a continuous time series data, which records the complete temperature change information along the length direction of the steel billet, and is collected by the IBA data acquisition system and stored in the time series database.

[0092] In addition, problems such as unplanned downtime and data acquisition network fluctuations that occur occasionally on site may cause missing values ​​or outliers in the high-frequency time series data collection. Therefore, it is necessary to interpolate the collected time series data in combination with the billet rolling speed to eliminate missing values ​​and outliers and improve the accuracy of trend warning results.

[0093] Furthermore, in step S2, the collected raw data is the time series data recorded by the pyrometer during the continuous production process, including invalid data recorded when the steel is not passed, and the rolling temperature in the batch also shows a temperature drop trend from the head to the tail, which cannot be directly used for trend warning. Therefore, it is necessary to use material tracking, batch segmentation, and feature extraction methods to pre-process the data to obtain a data set that can be used to analyze the trend change of the rolling temperature between cord steel batches.

[0094] Figure 3It is a schematic diagram of the original time series data trend of the rolling temperature. The temperature curve intercepted by the two cursor lines represents the temperature change trend from the head to the tail of a billet. The temperature shows a downward trend because the cooling time is different due to the different distances between the head and the tail from the temperature measurement point. The billet material number obtained by the spray number recognition system at the entrance of the heating furnace is combined with the material tracking algorithm to match a unique material number for each temperature curve in the time series data, so as to realize the material tracking and batch segmentation of the rolling temperature.

[0095] The method proposed by the present invention is to provide early warning of trends between different batches of cord steel. Figure 3 The decreasing trend of the rolling temperature in the batch will undoubtedly affect the accuracy of the early warning algorithm. Therefore, considering the unevenness of the head structure and the head cutting process during the rolling process, the present invention removes the temperature of 5% of the head length during the rolling temperature feature extraction process, and takes the maximum value of the temperature curve to represent the rolling temperature of this batch for trend early warning. The characteristic values ​​of the rolling temperature of the 320 steel billets to be analyzed are as follows: Figure 4 As shown, it can be seen that the arrow part of the figure has a significant trend of first decreasing and then increasing, and there are also two abnormal points with lower temperatures.

[0096] Furthermore, in step S3, the double exponential weighted average method (DEWMA) is used to solve the problem of noise influence on the trend change of the rolling temperature caused by the system's own fluctuations and sensor measurement errors. Compared with the traditional moving exponential weighted average method (EWMA), the method in the present invention has better performance in monitoring the mean shift of the process. The smoothed rolling temperature eliminates the influence of random noise and short-term fluctuations, and can better reflect the trend change of the data.

[0097] The moving exponential weighted average method (EWMA) can detect smaller shifts in historical data, and can effectively reduce the impact of random fluctuations on monitoring results by adjusting the weights. However, as the observation time shortens, the weight attached to the observation decreases exponentially. To solve this problem, the present invention introduces the double exponential weighted average method (DEWMA), which is better than the control chart based on EWMA in monitoring process mean shifts. DEWMA performs two exponential smoothings to characterize the weighted combination of current and past observations.

[0098] Specifically, suppose the data sequence in the analysis data set is , the double exponential weighted average algorithm DEWMA is defined as shown in formula (1):

[0099] Formula (1)

[0100] Where EWMA represents the moving exponential weighted average algorithm, DEWMA represents the double exponential weighted average algorithm; xi represents the i-th data, 1≤i≤n, and n is the total number of data; λ∈(0,1] represents the smoothing coefficient. The closer the data is to the current moment, the greater the weight it has, and the farther the data is from the current moment, the smaller the weight it has.

[0101] Therefore, DEWMA can eliminate random noise and short-term fluctuations and better reflect the trend changes of data, especially for time series data with relatively stable long-term trends. Figure 5 As shown, it can be seen that the temperature anomalies of a single batch have disappeared, and only the characteristics of trend changes are retained.

[0102] Furthermore, in step S4, the trend of the rolling temperature after smoothing is monitored based on the improved trend cumulative sum control chart (TCUSUM). The relative standard deviation of the rolling temperature after smoothing at the previous moment is used as the target value to calculate the trend cumulative sum. If there is a significant difference between the relative standard deviation of the current data and the data at the previous moment, it means that the data has an abnormal trend. This monitoring method can detect potential problems or abnormalities earlier, so that appropriate measures can be taken to intervene and adjust.

[0103] In an embodiment of the present invention, the improved trend cumulative sum control chart is improved from the traditional cumulative sum control chart (CUSUM). The principle of CUSUM is to calculate the difference between the data point and the target value and add these differences. If the cumulative sum reaches the predetermined control limit, it means that a change has occurred in the production process. The target value is usually the control target of the production process, or it can be the average value of the previous time window, and the control limit is the threshold that triggers an alarm or corrective action when the cumulative sum exceeds the limit.

[0104] Improvements to trend accumulation and control charts include:

[0105] In the traditional trend accumulation and control chart, is a data sequence, and two assumptions are made for this data sequence: the process is controlled and the process is not controlled, as shown in equations (2) and (3):

[0106] Formula (2)

[0107] Formula (3)

[0108] Among them, H0 represents the hypothesis that the process is controlled, H1 represents the hypothesis that the process is not controlled; xi represents the i-th data, n is the total number of data; N(μ0,σ) and N(μ1,σ) represent normal distributions with mean μ0 and μ1 and variance σ respectively; ξ represents the number of samples that obey the N(μ0,σ) distribution, and δ represents the degree of mean shift.

[0109] Let f(x) represent the probability density function of the normal distribution, and the likelihood functions corresponding to the two hypotheses are shown in equations (4) and (5): Formula (4)

[0110] Formula (5)

[0111] Among them, L(n,μ0) is the likelihood function of hypothesis H0, and L(n,μ1) is the likelihood function of hypothesis H1; ∏ represents a multiplication operation, and f0(xi) and f1(xi) represent the probability density functions corresponding to the normal distribution in hypothesis H0 and hypothesis H1, respectively.

[0112] The likelihood ratio Ln of hypothesis H1 to hypothesis H0 is shown in formula (6):

[0113] Formula (6)

[0114] Maximizing Ln and taking its logarithm yields:

[0115] Formula (7)

[0116] Formula (8)

[0117] in, and Respectively represent the cumulative sum of upper deviation and lower deviation of the i-th data; and They represent the upper deviation cumulative sum and the lower deviation cumulative sum of the i-1th data respectively; k is the allowable drift parameter, which represents the allowable offset of the data.

[0118] CUSUM monitors the deviation between the observed data and the target value. In order to respond to the changing trend more promptly, the target mean μ0 in the traditional trend cumulative sum control chart is changed to the smoothed data xi-1 at the previous moment, and an improved trend cumulative sum control chart is obtained, as shown in equations (9) and (10):

[0119] Formula (9)

[0120] Formula (10)

[0121] The monitoring control limit CL and the allowable drift parameter k are set according to the actual situation of the cord steel rolling temperature;

[0122] like , then it is considered that there is an upward shift trend; if , then it is considered that there is a downward trend; if and , it is considered that no obvious deviation has occurred. The judgment logic is as follows Figure 6 shown.

[0123] TCUSUM can detect continuous small deviations, thereby discovering trend changes in the production process at an earlier stage and monitoring abnormal trends.

[0124] Furthermore, step S4 also includes: a relative standard deviation calculation method based on process specifications. In the actual continuous production process of cord steel, there is a switch between different steel grades and specifications, and the corresponding process control requirements are also different, which will lead to frequent trend warnings based on the observed value of the rolling temperature, and it is impossible to accurately capture the trend deviation caused by the weakening of the production line capacity or abnormal process operation. The present invention calculates the relative standard deviation of the rolling temperature through the rolling temperature observation data and the process control limit and uses it for trend warning, which solves the problem that the traditional trend warning method is not suitable for multiple steel grades and specifications.

[0125] Cross-steel or cross-order production will lead to process specification switching. The difference in control targets in different process specifications will cause the start rolling temperature to show an upward or downward trend. In order to eliminate the dimensional impact of different process specification switching on trend warning, the relative standard deviation yi of the start rolling temperature of each billet and the process specification is calculated by equations (11) and (12);

[0126] Formula (11)

[0127] Formula (12)

[0128] Where yi represents the relative standard deviation of the ith billet, Ti represents the characteristic value of the rolling temperature of the ith billet; Taim represents the control target of the rolling temperature; TUCL represents the upper control limit of the rolling temperature; TLCL represents the lower control limit of the rolling temperature, and .

[0129] Furthermore, step S5 includes the optimization of two hyperparameters, namely the smoothing coefficient and the allowed drift parameter in the trend warning method. The smoothing coefficient λ is used to control the smoothing strength of the rolling temperature in the double exponential weighted average algorithm, so that the data change trend is retained under the premise of eliminating the influence of noise. The allowed drift parameter k is used to control the allowed offset of the monitoring data in the trend accumulation and control chart, and can be adjusted according to actual needs to achieve the purpose of tightening or relaxing the warning strength.

[0130] The step S5 also includes an online monitoring process for early warning of rolling temperature trend. Figure 7 As shown in the figure, first, the relative standard deviation of the rolling temperature of each billet is calculated according to the upper and lower control limits of the rolling temperature specified in the process specification; secondly, it is smoothed by the double exponential weighted average method, and the cumulative trend sum of the rolling temperature is further calculated; then, according to the set monitoring control limits, it is judged whether there is an abnormal trend and its offset direction, and the end position of the trend is determined in the process of cyclic monitoring; finally, after the process returns to stability, the various parameters are reset to continue the new online monitoring process.

[0131] The improved TCUSUM control chart is used to monitor the abnormal trend of the rolling temperature of the cord steel produced by the same team, and CL=0.3, λ=0.25, k=10-3 are set. The TCUSUM control chart is as follows: Figure 8 As shown, the start and end positions of the abnormal trend are marked in the temperature trend graph as shown in Fig. 9 As shown, it can be seen that there are two significant abnormal trends, which are consistent with the manual judgment results. This shows that the trend warning method in the present invention can be unaffected by a small number of abnormal batches and normal fluctuations, and can effectively detect abnormal trend conditions in the production process.

[0132] Furthermore, step S6 includes a method for quantitatively evaluating the degree of trend deviation. The more significant the trend change of the rolling temperature is, the greater the trend cumulative sum in the control chart is. When the upward trend or the downward trend ends, the trend cumulative sum at this position can be used as a quantitative evaluation index of the degree of trend deviation. The quantitative results of the three abnormal trends in the example of the present invention are shown in Table 1.

[0133] Table 1

[0134] The method of the present invention can solve the problems of insufficient accuracy and versatility of conventional trend warnings in the prior art, and warns of abnormal trends in the cord steel rolling temperature under the premise of considering noise influence and steel type and process switching, and gives a quantitative evaluation of the abnormal trend, thereby improving the temperature control stability of the cord steel rolling process.

[0135] Accordingly, an embodiment of the present invention further provides a cord steel rolling temperature trend early warning device based on trend accumulation and sum, Fig.10 1 is a structural block diagram of a cord steel rolling temperature trend early warning device based on trend accumulation and sum according to an exemplary embodiment. Fig.10 As shown, the device comprises:

[0136] The data acquisition module 201 is used to collect the time series data of the rolling temperature during the rolling process of the cord steel;

[0137] A preprocessing module 202 is used to preprocess the time series data to form an analysis data set;

[0138] The first calculation module 203 is used to perform smoothing processing on the analysis data set based on a double exponential weighted average algorithm to extract trend items of the rolling temperature between different batches;

[0139] A second calculation module 204 is used to calculate the trend cumulative sum of the rolling start temperature based on the improved trend cumulative sum control chart;

[0140] An optimization module 205 is used to optimize the smoothing coefficient in the double exponential weighted average algorithm and the allowed drift parameter in the trend accumulation and control chart to establish an online trend warning model;

[0141] The monitoring module 206 is used to monitor the abnormal trend of the start rolling temperature online by using the online trend warning model, and quantitatively evaluate the degree of trend deviation.

[0142] For ease of explanation, Fig.10 Only the main components of the device are shown. The device of this embodiment can be used to perform Figure 1 The technical solution of the method embodiment shown has similar implementation principles and technical effects, which will not be repeated here.

[0143] In an exemplary embodiment, the present invention further provides an electronic device, the electronic device comprising:

[0144] processor;

[0145] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are loaded and executed by the processor, the steps of the above-mentioned method for early warning of the trend of the rolling temperature of the cord steel based on the trend accumulation sum are implemented.

[0146] Fig.11 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Fig.11As shown, the electronic device 300 may include a processor 3001 and a memory 3002. Optionally, the electronic device 300 may further include a transceiver 3003. The processor 3001, the memory 3002 and the transceiver 3003 may be connected, for example, via a communication bus. The memory 3002 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 3001, the steps of the cord steel rolling temperature trend early warning method based on the trend accumulation sum are implemented.

[0147] In a specific implementation, as an embodiment, the processor 3001 may include one or more CPUs, such as Fig.11 CPU0 and CPU1 are shown in FIG.

[0148] In a specific implementation, as an embodiment, the electronic device 300 may also include multiple processors, such as Fig.11 3001 and processor 3004 are shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0149] The memory 3002 is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor 3001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0150] The transceiver 3003 is used to communicate with a network device or a terminal device.

[0151] Optionally, the transceiver 3003 may include a receiver and a transmitter, wherein the receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0152] Optionally, the transceiver 3003 may be integrated with the processor 3001 , or may exist independently and be coupled to the processor 3001 via an interface circuit of the electronic device 300 , which is not specifically limited in the embodiment of the present invention.

[0153] It should be noted that Fig.11 The structure of the electronic device 300 shown in the figure does not constitute a limitation on the electronic device. The actual electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In addition, the technical effects of the electronic device 300 can refer to the technical effects of the above-mentioned method embodiment, which will not be repeated here.

[0154] In an exemplary embodiment, the present invention further provides a computer-readable storage medium, wherein at least one instruction is stored in the computer-readable storage medium, and the at least one instruction is loaded and executed by a processor to implement the steps of the above-mentioned method for early warning of the trend of the cord steel rolling temperature based on the trend accumulation sum. For example, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0155] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0156] References in the specification to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments", etc. indicate that the described embodiments may include a particular feature, structure, or characteristic, but not every embodiment may include the particular feature, structure, or characteristic. In addition, when a particular feature, structure, or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the relevant art to implement such feature, structure, or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0157] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0158] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0159] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0161] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0162] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0163] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A cord steel rolling temperature trend early warning method based on trend accumulation and sum, characterized in that: The following steps are involved: S1. Collecting time series data of rolling temperature during the rolling process of cord steel; S2. Preprocessing the time series data to form an analysis data set; S3, smoothing the analysis data set based on a double exponential weighted average algorithm to extract trend items of rolling temperature between different batches; S4. Calculating the trend cumulative sum of the rolling start temperature based on the improved trend cumulative sum control chart; In step S4, improving the trend accumulation and control chart includes: In the traditional trend accumulation and control chart, let x1, x2, ..., x i , ..., x n is a data sequence, and two assumptions are made for this data sequence: process controlled and process uncontrolled, as shown in equations (2) and (3): H0: x i ~N(μ0, σ), i=1, 2, 3,..., n Equation (2) H1:x i ~N(μ0,σ),i=1,2,3,…,ξ;μ1=μ0+δσ Equation (3) x i ~N(μ1,σ),i=ξ+1,ξ+2,…,n;1≤i≤n Among them, H0 represents the hypothesis that the process is under control, and H1 represents the hypothesis that the process is not under control; x i represents the i-th data, n is the total number of data; N(μ0, σ) and N(μ1, σ) respectively represent normal distributions with mean μ0 and μ1 and variance σ; ξ represents the number of samples that obey the N(μ0, σ) distribution, and δ represents the degree of mean shift; Let f(x) represent the probability density function of the normal distribution, and the likelihood functions corresponding to the two hypotheses are shown in equations (4) and (5): Where L(n, μ0) is the likelihood function of hypothesis H0, L(n, μ1) is the likelihood function of hypothesis H1; Π represents the multiplication operation, f0(x i ) and f1(x i ) represent the probability density functions corresponding to the normal distribution in hypothesis H0 and hypothesis H1 respectively; The likelihood ratio L of hypothesis H1 to hypothesis H0 n As shown in formula (6): For L n Finding the maximum value and taking the logarithm yields: in, and Respectively represent the cumulative sum of upper deviation and lower deviation of the i-th data; and They represent the cumulative sum of upper deviation and lower deviation of the i-1th data respectively; k is the allowable drift parameter, which represents the allowable offset of the data; Change μ0 in the traditional trend accumulation and control chart to the smoothed data x at the previous moment. i-1 , we get the improved trend accumulation and control chart, as shown in equations (9) and (10): The monitoring control limit CL and the allowable drift parameter k are set according to the actual situation of the cord steel rolling temperature; like It is considered that there is an upward shift trend; if It is considered that there is a downward trend; if and It is considered that no significant deviation has occurred; The step S4 also includes: a relative standard deviation calculation method based on process specifications; The relative standard deviation y of the rolling temperature and process specifications of each billet is calculated by equation (11) and equation (12): i ; Among them, y i represents the relative standard deviation of the ith billet, T i represents the characteristic value of the rolling temperature of the i-th billet; T aim Indicates the control target of rolling temperature; T UCL Indicates the upper limit of rolling temperature; T LCL represents the lower control limit of rolling temperature, and T alm =(T UCL +T LCL ) / 2; S5. Optimize the smoothing coefficient in the double exponential weighted average algorithm and the allowed drift parameter in the trend accumulation and control chart to establish an online trend warning model; The step S5 also includes: an online monitoring process for early warning of rolling temperature trend; First, the relative standard deviation of the rolling temperature of each billet is calculated according to the upper and lower control limits of the rolling temperature specified in the process specification; second, the double exponential weighted average method is used for smoothing, and the trend cumulative sum of the rolling temperature is further calculated; then, according to the set monitoring control limits, it is determined whether there is an abnormal trend and its deviation direction, and the position where the trend ends is determined during the cyclic monitoring process; finally, after the process returns to stability, the various parameters are reset to continue the new online monitoring process; S6. Using the online trend warning model to monitor the abnormal trend of the start rolling temperature online, and quantitatively evaluate the degree of trend deviation.

2. The cord steel rolling temperature trend early warning method based on trend accumulation and sum according to claim 1, characterized in that: The step S1 comprises: The high-frequency data acquisition interface is used to collect time series data from the production site in real time. The sampling frequency is set according to the rolling speed to perform equal-interval sampling and store it in the time series database. The time series data is interpolated in combination with the rolling speed to eliminate missing values ​​and outliers.

3. The cord steel rolling temperature trend early warning method based on trend accumulation and sum according to claim 1, characterized in that: The step S2 comprises: According to the billet material number obtained by the spray number recognition system at the entrance of the heating furnace and combined with the material tracking algorithm, a unique material number is matched for each temperature curve in the time series data to achieve material tracking and batch segmentation of the rolling temperature; In the process of extracting the rolling temperature characteristics, after removing the temperature of 5% of the head length, the maximum value of the temperature curve is taken to represent the rolling temperature of this batch.

4. The cord steel rolling temperature trend early warning method based on trend accumulation and sum according to claim 1, characterized in that: The step S3 comprises: Suppose the data sequence in the analysis data set is x1, x2, ..., x i , ..., x n , the double exponential weighted average algorithm DEWMA is defined as shown in formula (1): CONTINUE i =WHAT? i +(1-λ)CONTINUE i-1 formula(1) EWMA i =λx i +(1-λ)EWMA i-1 Among them, EWMA represents the moving exponential weighted average algorithm, DEWMA represents the double exponential weighted average algorithm; i represents the i-th data, 1≤i≤n, n is the total number of data; λ∈(0,1] represents the smoothing coefficient, the closer the data is to the current moment, the greater the weight it has, and the farther the data is from the current moment, the smaller the weight it has.

5. The cord steel rolling temperature trend early warning method based on trend accumulation and sum according to claim 1, characterized in that: The step S5 comprises: The smoothing coefficient λ in the double exponential weighted average algorithm and the allowable drift parameter k in the trend accumulation and control chart are optimized; the smoothing coefficient λ is used to control the smoothing intensity of the rolling temperature, and the allowable drift parameter k is used to control the allowable offset of the monitoring data.

6. The cord steel rolling temperature trend early warning method based on trend accumulation and sum according to claim 1, characterized in that: The step S6 comprises: When the upward trend or downward trend ends, the cumulative trend sum at that position serves as a quantitative evaluation indicator of the degree of trend deviation.

7. A cord steel rolling temperature trend early warning device based on trend accumulation and sum, the device is used to implement the method according to any one of claims 1 to 6, characterized in that: The device comprises: The data acquisition module is used to collect the time series data of the rolling temperature during the rolling process of the cord steel; A preprocessing module, used for preprocessing the time series data to form an analysis data set; A first calculation module is used to perform smoothing processing on the analysis data set based on a double exponential weighted average algorithm to extract trend items of the rolling temperature between different batches; A second calculation module is used to calculate the trend cumulative sum of the rolling start temperature based on the improved trend cumulative sum control chart; The optimization module is used to optimize the smoothing coefficient in the double exponential weighted average algorithm and the allowed drift parameters in the trend accumulation and control chart, and establish an online trend warning model; The monitoring module is used to monitor the abnormal trend of the rolling temperature online by using the online trend warning model, and quantitatively evaluate the degree of trend deviation.

Images