Hot rolling mill vibration data processing method, system, terminal and storage medium

Abstract

The application discloses a hot rolling mill vibration data processing method, system, terminal and storage medium, and the method comprises the following steps: collecting original vibration data; performing data segmentation to obtain a plurality of data subblocks and index labeling, performing peak value calculation on the data subblocks to obtain data subblock peak values; performing descending order sorting, extracting the data subblock peak values in the first few digits of the sequence, performing mean value calculation on the remaining data subblock peak values to obtain data subblock mean peak values; comparing the extracted data subblock peak values with the data subblock mean peak values respectively to obtain vibration data segmentation boundaries; and segmenting the original vibration data according to the vibration data segmentation boundaries. The application collects original vibration data in the whole process of the operation of the hot rolling mill, obtains vibration data segmentation boundaries through algorithm analysis and deduction, and then segments the original vibration data into steel biting process vibration data and normal rolling process vibration data, so that the segmentation process is intuitive and accurate, and the segmentation result has high reliability.

Description

Technical Field

[0001] This application relates to a vibration data processing scheme, specifically a method, system, terminal, and storage medium for processing vibration data of a hot rolling mill during the hot rolling process, belonging to the field of predictive maintenance technology for mechanical equipment. Background Technology

[0002] Vibration analysis is a widely used and relatively mature technology in predictive maintenance of mechanical equipment. This technology collects vibration data during the operation of mechanical equipment and combines it with methods such as spectrum conversion to complete data analysis, thereby effectively determining whether the mechanical equipment has a fault.

[0003] Hot rolling is a common production process in the metallurgical industry. It primarily uses slabs (mainly continuously cast slabs) as raw materials, heating them before hot rolling mills produce bars, strips, and other products. Hot rolling mills in hot rolling production lines are heavy-duty equipment, operating under high loads and impact loads for extended periods. Generally speaking, compared to ordinary, stable rotary equipment, the rotating components in hot rolling mills experience greater and more complex stresses, making them more prone to various failures. Therefore, more and more technicians are beginning to apply vibration analysis technology to hot rolling mills.

[0004] The normal rolling process of a hot rolling mill is as follows: Figure 1 As shown, the hot rolling mill is in a relatively stable state at this time, and vibration analysis can be performed using commonly used methods in the industry, such as amplitude spectrum and envelope spectrum. However, once a vibration occurs... Figure 2 The hot rolling mill bite phenomenon shown is caused by the speed difference between the slab and the rolls of the hot rolling mill. A strong rigid impact occurs at the moment of contact between the slab and the rolls, resulting in vibrations with amplitudes far exceeding those of the normal rolling process. Therefore, analyzing vibration data from the bite process and the normal rolling process together leads to significantly larger values ​​for the calculated vibration time-domain parameters compared to the normal rolling process, often exceeding the standard vibration threshold and generating false alarms. Conversely, increasing the vibration threshold exceeds the fault threshold in the normal rolling process, resulting in missed fault detections. Furthermore, processing data from both processes together introduces frequency mixing during spectrum transformation, increasing the difficulty of vibration analysis and making it hard to determine the true fault. Therefore, filtering out hot rolling mill bite data and preventing its influence on subsequent vibration analysis has become a major challenge in equipment fault diagnosis in the metallurgical industry.

[0005] To address the aforementioned issues, several solutions have emerged in the industry. For example, Chinese patent CN202010144161.3 proposes a method for filtering bite impact data based on operating condition signals. This method reads the mill's process parameters, compares these parameters with pre-configured threshold values ​​to determine if bite has occurred, and then delays the data acquisition module to collect data, thus achieving data filtering. While feasible, this solution requires hardware access to the hot rolling mill to obtain process parameters and input the bite threshold values. This increases the workload and burden on production personnel, as incorrect or missing threshold values ​​can lead to incorrect operating condition judgments and data filtering failure. Furthermore, the vibration data acquisition terminal used in this solution needs to be linked with the automation system to obtain hot rolling mill operating condition data before data acquisition via control. This undoubtedly increases the complexity and coupling of the system operation. Moreover, many metallurgical plants, for safety reasons, do not open the interfaces of their automation systems, thus limiting the practicality of this solution.

[0006] In summary, how to propose a vibration data processing scheme for hot rolling mills that differs from the aforementioned related technologies, efficiently and accurately extract vibration data during the normal rolling process of hot rolling mills, avoid the impact of vibration data from the steel biting process on subsequent vibration analysis, and ensure the accuracy of vibration analysis results has become a common concern for those skilled in the art. Summary of the Invention

[0007] In view of the above-mentioned defects in the existing technology, the purpose of this application is to propose a method, system, terminal and storage medium for processing vibration data of hot rolling mills in the hot rolling process, as follows.

[0008] Firstly, a method for processing vibration data from a hot rolling mill, comprising:

[0009] Collect raw vibration data of the hot rolling mill;

[0010] The original vibration data is segmented to obtain multiple data sub-blocks, and each data sub-block is indexed and labeled according to time sequence. Peak values ​​are calculated for each data sub-block, and the peak values ​​of multiple data sub-blocks are summed.

[0011] Sort the peak values ​​of multiple data sub-blocks in descending order, extract the first few peak values ​​of the data sub-blocks in the sequence and record the index number corresponding to the extracted peak values ​​of the data sub-blocks, and calculate the average peak value of the remaining peak values ​​of the data sub-blocks after extraction.

[0012] The peak values ​​of the extracted data sub-blocks are compared with the average peak value of the data sub-blocks. Based on the comparison results, combined with the preset threshold and the index number, the vibration data segmentation boundary is obtained.

[0013] The original vibration data is segmented according to the vibration data segmentation boundary to obtain vibration data of the steel biting process and vibration data of the normal rolling process.

[0014] Preferably, the acquisition of raw vibration data from the hot rolling mill includes:

[0015] Vibration data of the hot rolling mill is collected using a vibration acceleration sensor, and the raw vibration data is obtained and saved.

[0016] Preferably, the original vibration data is segmented to obtain multiple data sub-blocks, and each data sub-block is indexed and labeled according to time sequence. Peak values ​​are calculated for each data sub-block, and the peak values ​​of the multiple data sub-blocks are summarized, including:

[0017] The original vibration data is segmented at a preset time interval to obtain multiple data sub-blocks, and each data sub-block is indexed and labeled in time sequence. Each data sub-block contains several data points.

[0018] Peak values ​​are calculated for each data sub-block, and the peak values ​​of multiple data sub-blocks and their corresponding index numbers are obtained by summing them up.

[0019] Preferably, the step of sorting the peak values ​​of multiple data sub-blocks in descending order, extracting the first few peak values ​​of the data sub-blocks in the sequence and recording the index number corresponding to the extracted peak values, and calculating the average peak value of the remaining peak values ​​of the data sub-blocks after extraction, includes:

[0020] The peak values ​​of the multiple data sub-blocks are sorted in descending order, and the top three peak values ​​of the data sub-blocks in the sequence are extracted and denoted as the first maximum peak value, the second maximum peak value, and the third maximum peak value, respectively.

[0021] Record the index numbers corresponding to the peak values ​​of the extracted data sub-blocks, and denote them as the first maximum peak index number, the second maximum peak index number, and the third maximum peak index number, respectively.

[0022] The mean value of the remaining data sub-blocks after extraction from the sequence is calculated to obtain the mean value of the data sub-blocks.

[0023] Preferably, the step of comparing the peak values ​​of the extracted data sub-blocks with the average peak value of the data sub-blocks, and determining the vibration data segmentation boundary based on the comparison results, a preset threshold, and the index number, includes:

[0024] Calculate the ratio of the first maximum peak value to the average peak value of the data sub-block to obtain the first peak value comparison value. If the first peak value comparison value is less than the preset first peak value comparison threshold, it is considered that vibration data segmentation is not required. If the first peak value comparison value is not less than the first peak value comparison threshold, the ratio of the second maximum peak value to the average peak value of the data sub-block is calculated in sequence.

[0025] Calculate the ratio of the second maximum peak value to the average peak value of the data sub-block to obtain the second peak comparison value. If the second peak comparison value is less than the preset second peak comparison threshold, the data sub-block corresponding to the first maximum peak index number is considered to be the steel biting process occurrence sub-block. If the second peak comparison value is greater than the preset third peak comparison threshold and the difference between the second maximum peak index number and the first maximum peak index number is greater than the preset first index difference threshold, vibration data segmentation is not required. If the second peak comparison value is greater than the second peak comparison threshold and the difference between the second maximum peak index number and the first maximum peak index number is equal to the preset second index difference threshold, the data sub-blocks corresponding to the first maximum peak index number to the second maximum peak index number are all included in the steel biting process occurrence sub-block. Calculate the ratio of the third maximum peak value to the average peak value of the data sub-block in sequence.

[0026] Calculate the ratio of the third maximum peak value to the average peak value of the data sub-blocks to obtain a third peak comparison value. If the third peak comparison value is less than a preset fourth peak comparison threshold, then the data sub-blocks corresponding to the first maximum peak index number to the second maximum peak index number are considered to be steel biting process occurrence sub-blocks. If the third peak comparison value is not less than the fourth peak comparison threshold and the difference between the third maximum peak index number and the second maximum peak index number is equal to a preset third index difference threshold, then the data sub-block corresponding to the third maximum peak index number is considered to be a steel biting process occurrence sub-block, and the data sub-blocks corresponding to the first maximum peak index number to the third maximum peak index number are considered to be steel biting process occurrence sub-blocks.

[0027] The index number corresponding to the sub-block where the steel biting process occurs is used as the boundary for vibration data segmentation.

[0028] Preferably, the first peak comparison threshold is 8, the second peak comparison threshold is 3, the third peak comparison threshold is 4, and the fourth peak comparison threshold is 2;

[0029] The first sequence number difference threshold is 2, the second sequence number difference threshold is 1 or 2, and the third sequence number difference threshold is 1.

[0030] Preferably, the step of segmenting the original vibration data according to the vibration data segmentation boundary to obtain vibration data of the steel biting process and vibration data of the normal rolling process includes:

[0031] The original vibration data is segmented according to the vibration data segmentation boundary. All data sub-blocks in the original data before the vibration data segmentation boundary are taken as vibration data of the steel biting process, and all data sub-blocks in the original data after the vibration data segmentation boundary are taken as vibration data of the normal rolling process.

[0032] The vibration data processing method for hot rolling mill proposed in this application collects the original vibration data of the entire operation process of the hot rolling mill, and combines algorithm analysis and derivation to obtain the vibration data segmentation boundary. This enables the data segmentation process to divide the original vibration data into vibration data of the steel biting process and vibration data of the normal rolling process. The data segmentation process is intuitive and accurate, and the data segmentation results are highly reliable.

[0033] Meanwhile, the method in this application does not require additional hardware equipment during implementation, and the solution has a wide range of applications, strong practicality, and low implementation cost.

[0034] Secondly, a vibration data processing system for a hot rolling mill, comprising:

[0035] The raw data acquisition module is configured to acquire raw vibration data from the hot rolling mill;

[0036] The preprocessing and peak calculation module is configured to segment the original vibration data to obtain multiple data sub-blocks, index and label each data sub-block in time sequence, calculate the peak value of each data sub-block, and summarize the peak values ​​of multiple data sub-blocks.

[0037] The peak extraction and average peak value calculation module is configured to sort the peak values ​​of multiple data sub-blocks in descending order, extract the first few peak values ​​of the data sub-blocks in the sequence and record the index number corresponding to the extracted peak values ​​of the data sub-blocks, and calculate the average peak value of the remaining peak values ​​of the data sub-blocks after extraction to obtain the average peak value of the data sub-blocks.

[0038] The segmentation boundary calculation module is configured to compare the peak values ​​of the extracted data sub-blocks with the average peak value of the data sub-blocks respectively, and to determine the vibration data segmentation boundary based on the comparison results, combined with a preset threshold and the index number.

[0039] The vibration data segmentation module is configured to segment the original vibration data according to the vibration data segmentation boundary to obtain vibration data of the steel biting process and vibration data of the normal rolling process.

[0040] Third, an intelligent terminal includes a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the hot rolling mill vibration data processing method as described above.

[0041] Fourth, a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the hot rolling mill vibration data processing method as described above.

[0042] Corresponding to the aforementioned method, the hot rolling mill vibration data processing system, terminal, and storage medium proposed in this application realizes a data segmentation process that divides the original vibration data into vibration data of the steel biting process and vibration data of the normal rolling process through a systematic and standardized processing flow. This provides technical support for subsequent vibration analysis and monitoring operations. The hardware solution has high adaptability and compatibility and can be effectively applied to the vibration analysis scenario of hot rolling mills.

[0043] This application also provides a reference for other technical solutions related to predictive maintenance of mechanical equipment, which can be used as a basis for further development and in-depth research, and has a very broad application prospect.

[0044] The specific implementation methods of this application will be further described in detail below with reference to the accompanying drawings, so as to make the technical solution of this application easier to understand and master. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings:

[0046] Figure 1 This is a schematic diagram of the normal rolling process of a hot rolling mill;

[0047] Figure 2 This is a schematic diagram of the steel biting process in a hot rolling mill;

[0048] Figure 3 This is a flowchart illustrating the method for processing vibration data from a hot rolling mill according to an embodiment of this application.

[0049] Figure 4 This is a time-domain plot of a segment of raw vibration data from a specific operational example of this application;

[0050] Figure 5 This is a frequency domain plot of a segment of raw vibration data from a specific operational example of this application;

[0051] Figure 6 This is a time-domain plot after filtering out vibration data during the steel-gripping process in a specific operational example of this application;

[0052] Figure 7 This is a frequency domain diagram after filtering out vibration data during the steel biting process in a specific operational example of this application;

[0053] Figure 8 This is a schematic diagram of the architecture of the hot rolling mill vibration data processing system according to an embodiment of this application. Detailed Implementation

[0054] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0055] On the one hand, this application relates to a method for processing vibration data of a hot rolling mill. By collecting the original vibration data of the entire operation process of the hot rolling mill, the method uses an algorithm to analyze and derive the vibration data segmentation boundary for data division, and then divides the original vibration data into vibration data of the steel biting process and vibration data of the normal rolling process. Users only need to perform spectrum analysis and diagnosis on the segmented vibration data of the normal rolling process to achieve the purpose of vibration monitoring and fault diagnosis of the hot rolling mill.

[0056] like Figure 3 As shown in the embodiments of this application, the hot rolling mill vibration data processing method includes the following steps:

[0057] S1. Collect raw vibration data of the hot rolling mill. This step can be specified as follows.

[0058] Vibration data of the hot rolling mill is collected using a vibration acceleration sensor, and the raw vibration data is obtained and saved.

[0059] S2. The original vibration data is segmented to obtain multiple data sub-blocks, and each data sub-block is indexed and labeled according to time sequence. Peak values ​​are calculated for each data sub-block, and the peak values ​​of the multiple data sub-blocks are summed. This step can be specified as the following process.

[0060] S21. The original vibration data is segmented according to a preset time interval to obtain multiple data sub-blocks, and each data sub-block is indexed and labeled according to time sequence. Each data sub-block contains several data points.

[0061] In a specific operational example, the preset time interval is 200ms, and each data sub-block contains n data points.

[0062] S22. Calculate the peak value for each data sub-block and summarize the peak values ​​of multiple data sub-blocks and their corresponding index numbers.

[0063] In a specific operational example, peak calculation for each data sub-block can be directly transformed into finding the absolute average X of the 10 data points with the largest absolute values ​​among n data points. p The calculation formula is as follows:

[0064]

[0065] In the formula, j is the data point number in each of the data sub-blocks, and X... pj The data is the one with the largest absolute value when the data sub-blocks are sorted in descending order.

[0066] S3. Sort the peak values ​​of multiple data sub-blocks in descending order, extract the top few peak values ​​of the data sub-blocks in the sequence and record the index number corresponding to the extracted peak values ​​of the data sub-blocks, and calculate the average peak value of the remaining peak values ​​of the data sub-blocks after extraction. This step can be specified as the following process.

[0067] S31. Sort the peak values ​​of the multiple data sub-blocks in descending order, and extract the top three peak values ​​of the data sub-blocks in the sequence, which are respectively denoted as the first maximum peak value, the second maximum peak value, and the third maximum peak value.

[0068] In a specific operational example, the first maximum peak value is denoted as X. p1 The second maximum peak value is denoted as X. p2 The third maximum peak value is denoted as X. p3 .

[0069] S32. Record the index numbers corresponding to the peak values ​​of the extracted data sub-blocks, and denot them as the first maximum peak index number, the second maximum peak index number, and the third maximum peak index number, respectively.

[0070] In a specific operational example, the index number of the first maximum peak value is denoted as I. p1 The second maximum peak index number is denoted as I. p2 The index number of the third maximum peak value is denoted as I. p3 .

[0071] S33. Calculate the mean of the peak values ​​of the remaining data sub-blocks after extraction from the sequence to obtain the average peak value of the data sub-blocks.

[0072] In a specific operational example, the average peak value of the data sub-block is denoted as X. pm .

[0073] S4. Compare the peak values ​​of the extracted data sub-blocks with the average peak value of the data sub-blocks respectively. Based on the comparison results, combined with the preset threshold and the index number, a judgment is made to obtain the vibration data segmentation boundary. This step can be specified as the following process.

[0074] S41. Calculate the ratio of the first maximum peak value to the average peak value of the data sub-block to obtain the first peak value comparison value;

[0075] If the first peak comparison value is less than the preset first peak comparison threshold, then vibration data segmentation is not required.

[0076] If the first peak comparison value is not less than the first peak comparison threshold, then the ratio of the second maximum peak value to the average peak value of the data sub-block is calculated sequentially.

[0077] In a specific operational example, the first peak comparison threshold is set to 8, and the above judgment process can be further specified as follows:

[0078] If X p1 / X pm If the value is less than 8, it is considered that there is no vibration data for the steel biting process and vibration data segmentation is not required; otherwise, proceed to the subsequent S42 step.

[0079] S42. Calculate the ratio of the second maximum peak value to the average peak value of the data sub-block to obtain the second peak value comparison value;

[0080] If the second peak comparison value is less than the preset second peak comparison threshold, then the data sub-block corresponding to the first maximum peak index number is considered to be the steel biting process occurrence sub-block;

[0081] If the second peak comparison value is greater than the preset third peak comparison threshold and the difference between the second maximum peak index number and the first maximum peak index number is greater than the preset first index difference threshold, then vibration data segmentation is not required.

[0082] If the second peak comparison value is greater than the second peak comparison threshold and the difference between the second maximum peak index number and the first maximum peak index number is equal to the preset second index difference threshold, then it is considered that the data sub-blocks corresponding to the first maximum peak index number to the second maximum peak index number are all included in the steel biting process occurrence sub-block, and the ratio of the third maximum peak value to the average peak value of the data sub-block is calculated in sequence.

[0083] In a specific operational example, the second peak comparison threshold is set to 3, the third peak comparison threshold is set to 4, the first sequence number difference threshold is set to 2, and the second sequence number difference threshold is set to 1 or 2. The above judgment process can be further specified as follows:

[0084] If X p2 / X pm If <3, then I is considered to be... p1 The corresponding data sub-block is the steel biting process occurrence sub-block;

[0085] If X p2 / X pm >4 and I p2 -I p1 If the value is greater than 2, it is considered that there is no vibration data for the steel biting process, or that the original vibration data itself is erroneous / faulty data, and there are multiple impact waveforms, so vibration data segmentation is not required.

[0086] If X p2 / X pm >3 and I p2 -I p1 =1 or I p2 -I p1 =2, then I is considered to be p1 to I p2 The corresponding data sub-blocks are all included in the steel biting process generation sub-block, and then proceed to the subsequent S43 step process.

[0087] S43. Calculate the ratio of the third maximum peak value to the average peak value of the data sub-block to obtain the third peak value comparison value;

[0088] If the third peak comparison value is less than the preset fourth peak comparison threshold, then the data sub-blocks corresponding to the first maximum peak index number to the second maximum peak index number are considered to be steel biting process occurrence sub-blocks.

[0089] If the third peak comparison value is not less than the fourth peak comparison threshold and the difference between the third maximum peak index number and the second maximum peak index number is equal to the preset third index difference threshold, then the data sub-block corresponding to the third maximum peak index number is considered to be the steel biting process occurrence sub-block, and the data sub-blocks corresponding to the first maximum peak index number to the third maximum peak index number are all steel biting process occurrence sub-blocks.

[0090] In a specific operational example, the fourth peak comparison threshold is set to 2, and the third sequence number difference threshold is set to 1. The above judgment process can be further specified as follows:

[0091] If X p3 / X pmIf <2, then I is considered to be p1 to I p2 The corresponding data sub-blocks are all sub-blocks of the steel biting process, and proceed to the subsequent S44 step process;

[0092] If X p3 / X pm ≥2 and I p3 -I p2 =1, then I is considered to be p3 The corresponding data sub-block is the steel biting process generation sub-block, I p1 to I p3 The corresponding data sub-blocks are all sub-blocks of the steel biting process, and proceed to the subsequent S44 step process.

[0093] It should be noted that in the specific operational examples above, the settings for each comparison threshold and difference threshold were calculated based on 650 strip steel as the standard. However, in actual application of the solution, the settings for each comparison threshold and difference threshold will vary depending on the type of steel being processed and the shape and specifications of the processed products.

[0094] S44. The index number corresponding to the sub-block in the steel biting process is used as the boundary for vibration data segmentation.

[0095] S5. The original vibration data is segmented according to the vibration data segmentation boundary to obtain vibration data of the steel biting process and vibration data of the normal rolling process. This step can be specified as the following process.

[0096] The original vibration data is segmented according to the vibration data segmentation boundary. All data sub-blocks in the original data before the vibration data segmentation boundary are taken as vibration data of the steel biting process, and all data sub-blocks in the original data after the vibration data segmentation boundary are taken as vibration data of the normal rolling process.

[0097] For example, if we consider I p1 to I p2 If the corresponding data sub-blocks are all sub-blocks where the steel biting process occurs, then the original data, I p2 All previously mentioned data sub-blocks are used as vibration data for the steel-gripping process. The original data, specifically the I-th sub-block... p2+1 All subsequent data sub-blocks are used as vibration data during normal rolling processes.

[0098] Subsequently, the obtained vibration data of the steel biting process and the vibration data of the normal rolling process are saved. Under normal circumstances, the vibration data of the normal rolling process can be used for vibration analysis and fault diagnosis using conventional methods; the vibration data of the steel biting process can also be used to determine the vibration changes of the equipment during the steel biting phenomenon through peak value, spectrum and other analysis methods.

[0099] Figures 4-7 This is an example of segmenting and filtering a piece of original vibration data according to the method of this application. Table 1 shows the calculation results of key vibration time-domain parameters before and after filtering. As can be seen from the figure, the method of this application can effectively determine whether there is steel biting process vibration data in a piece of original vibration data, and segment and filter the steel biting process vibration data.

[0100] Peak acceleration RMS speed cliff Before filtration 214.46 13.65 45.73 After filtration 37.27 7.49 2.91

[0101] Table 1 Comparison of key time-domain parameters in vibration data before and after filtration

[0102] In summary, the vibration data processing method for hot rolling mills proposed in this application has an intuitive and accurate data segmentation process and highly reliable data segmentation results.

[0103] Meanwhile, the method in this application does not require additional hardware equipment during implementation, and the solution has a wide range of applications, strong practicality, and low implementation cost.

[0104] In addition, the vibration data of the steel biting process in this application is also collected and stored. Since the steel biting phenomenon occurs when the hot rolling mill is subjected to the greatest impact force, if the equipment itself is faulty, the impact on it will be greater when the steel biting phenomenon occurs. Saving the vibration data of the steel biting process and analyzing it can help users to judge whether there is a fault in the hot rolling mill.

[0105] On the other hand, this application also relates to a vibration data processing system for a hot rolling mill, the system architecture of which is as follows: Figure 8 As shown, it includes:

[0106] The raw data acquisition module is configured to acquire raw vibration data from the hot rolling mill;

[0107] The preprocessing and peak calculation module is configured to segment the original vibration data to obtain multiple data sub-blocks, index and label each data sub-block in time sequence, calculate the peak value of each data sub-block, and summarize the peak values ​​of multiple data sub-blocks.

[0108] The peak extraction and average peak value calculation module is configured to sort the peak values ​​of multiple data sub-blocks in descending order, extract the first few peak values ​​of the data sub-blocks in the sequence and record the index number corresponding to the extracted peak values ​​of the data sub-blocks, and calculate the average peak value of the remaining peak values ​​of the data sub-blocks after extraction to obtain the average peak value of the data sub-blocks.

[0109] The segmentation boundary calculation module is configured to compare the peak values ​​of the extracted data sub-blocks with the average peak value of the data sub-blocks respectively, and to determine the vibration data segmentation boundary based on the comparison results, combined with a preset threshold and the index number.

[0110] The vibration data segmentation module is configured to segment the original vibration data according to the vibration data segmentation boundary to obtain vibration data of the steel biting process and vibration data of the normal rolling process.

[0111] In one possible implementation, the raw data acquisition module includes:

[0112] The raw data acquisition unit is configured to collect vibration data of the hot rolling mill using a vibration acceleration sensor, obtain raw vibration data, and save it.

[0113] In one possible implementation, the preprocessing and peak calculation module includes:

[0114] The data preprocessing unit is configured to segment the original vibration data at preset time intervals to obtain multiple data sub-blocks and index and label each data sub-block in time sequence, with each data sub-block containing several data points.

[0115] The peak calculation unit is configured to perform peak calculation on each of the data sub-blocks and summarize the peak values ​​of multiple data sub-blocks and their respective index numbers.

[0116] In one possible implementation, the peak extraction and mean peak value calculation module includes:

[0117] The peak extraction unit is configured to sort the peak values ​​of multiple data sub-blocks in descending order, and extract the top three peak values ​​of the data sub-blocks in the sequence, which are respectively denoted as the first maximum peak value, the second maximum peak value, and the third maximum peak value.

[0118] The index sequence number recording unit is configured to record the index sequence number corresponding to the peak value of the extracted data sub-block, which are respectively recorded as the first maximum peak index number, the second maximum peak index number, and the third maximum peak index number;

[0119] The average peak value calculation unit is configured to calculate the average peak value of the remaining data sub-blocks after extraction from the sequence to obtain the average peak value of the data sub-blocks.

[0120] In one possible implementation, the segmentation boundary calculation module includes:

[0121] The first maximum peak value comparison unit is configured to calculate the ratio of the first maximum peak value to the average peak value of the data sub-block to obtain a first peak value comparison value. If the first peak value comparison value is less than a preset first peak value comparison threshold, it is considered that vibration data segmentation is not required. If the first peak value comparison value is not less than the first peak value comparison threshold, the ratio of the second maximum peak value to the average peak value of the data sub-block is calculated in sequence.

[0122] The second maximum peak value comparison unit is configured to calculate the ratio of the second maximum peak value to the average peak value of the data sub-block to obtain a second peak value comparison value. If the second peak value comparison value is less than a preset second peak value comparison threshold, the data sub-block corresponding to the first maximum peak value index number is considered to be a steel biting process occurrence sub-block. If the second peak value comparison value is greater than a preset third peak value comparison threshold and the difference between the second maximum peak value index number and the first maximum peak value index number is greater than a preset first sequence number difference threshold, vibration data segmentation is not required. If the second peak value comparison value is greater than the second peak value comparison threshold and the difference between the second maximum peak value index number and the first maximum peak value index number is equal to a preset second sequence number difference threshold, the data sub-blocks corresponding to the first maximum peak value index number to the second maximum peak value index number are all included in the steel biting process occurrence sub-block, and the ratio of the third maximum peak value to the average peak value of the data sub-block is calculated sequentially.

[0123] The third maximum peak value comparison unit is configured to calculate the ratio of the third maximum peak value to the average peak value of the data sub-blocks to obtain a third peak value comparison value. If the third peak value comparison value is less than a preset fourth peak value comparison threshold, then the data sub-blocks corresponding to the first maximum peak value index number to the second maximum peak value index number are considered to be steel biting process occurrence sub-blocks. If the third peak value comparison value is not less than the fourth peak value comparison threshold and the difference between the third maximum peak value index number and the second maximum peak value index number is equal to a preset third index difference threshold, then the data sub-block corresponding to the third maximum peak value index number is considered to be a steel biting process occurrence sub-block, and the data sub-blocks corresponding to the first maximum peak value index number to the third maximum peak value index number are all steel biting process occurrence sub-blocks.

[0124] The segmentation boundary determination unit is configured to use the index number corresponding to the obtained steel biting process sub-block as the vibration data segmentation boundary.

[0125] In one possible implementation, the vibration data segmentation module includes:

[0126] The vibration data segmentation unit is configured to segment the original vibration data according to the vibration data segmentation boundary, taking all the data sub-blocks in the original data before the vibration data segmentation boundary as the steel biting process vibration data, and taking all the data sub-blocks in the original data after the vibration data segmentation boundary as the normal rolling process vibration data.

[0127] Furthermore, this application also relates to a smart terminal, including a memory and a processor. The memory stores at least one instruction, at least one program, code set, or instruction set, which is loaded and executed by the processor to implement the steps of the hot rolling mill vibration data processing method described above, for example... Figure 3 The steps shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 8 The functions of each module / unit are shown.

[0128] In another aspect, this application also relates to a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the hot rolling mill vibration data processing method as described above.

[0129] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user device. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0130] Corresponding to the aforementioned method, the hot rolling mill vibration data processing system, terminal, and storage medium proposed in this application realizes a data segmentation process that divides the original vibration data into vibration data of the steel biting process and vibration data of the normal rolling process through a systematic and standardized processing flow. This provides technical support for subsequent vibration analysis and monitoring operations. The hardware solution has high adaptability and compatibility and can be effectively applied to the vibration analysis scenario of hot rolling mills.

[0131] Furthermore, this application provides a reference for other technical solutions related to predictive maintenance of mechanical equipment, which can be used as a basis for further development and in-depth research, and has a very broad application prospect.

[0132] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit and essential characteristics of this application. Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within this application.

[0133] Finally, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method of processing vibration data of a hot rolling mill, characterized in that, The method comprises the following steps: Collecting original vibration data of a hot rolling mill; Data segmentation is performed on the original vibration data to obtain a plurality of data sub-blocks, each of which is indexed and labeled in time sequence, and peak values of each data sub-block are calculated respectively, and a plurality of data sub-block peak values are obtained by summarizing; The plurality of data sub-block peak values are sorted in descending order, the data sub-block peak values in the first few digits in the sequence are extracted, the index sequence numbers corresponding to the extracted data sub-block peak values are recorded, and the mean value of the remaining data sub-block peak values after extraction is calculated to obtain a data sub-block mean peak value; The extracted data sub-block peak values are compared with the data sub-block mean peak value respectively, and a vibration data segmentation boundary is obtained by judging according to the comparison result, combining a preset threshold value and the index sequence number, wherein, a first peak value comparison value is obtained by calculating the ratio of the first maximum peak value in the plurality of data sub-block peak values to the data sub-block mean peak value, if the first peak value comparison value is less than a preset first peak value comparison threshold value, it is considered that vibration data segmentation is not needed, if the first peak value comparison value is not less than the first peak value comparison threshold value, a second maximum peak value in the plurality of data sub-block peak values is calculated in sequence, a second peak value comparison value is obtained by calculating the ratio of the second maximum peak value to the data sub-block mean peak value, if the second peak value comparison value is less than a preset second peak value comparison threshold value, it is considered that the data sub-block corresponding to the first maximum peak value index sequence number is a steel biting process occurring sub-block, if the second peak value comparison value is greater than a preset third peak value comparison threshold value and the difference between the second maximum peak value index sequence number and the first maximum peak value index sequence number is greater than a preset first sequence number difference threshold value, it is considered that vibration data segmentation is not needed, if the second peak value comparison value is greater than the second peak value comparison threshold value and the difference between the second maximum peak value index sequence number and the first maximum peak value index sequence number is equal to a preset second sequence number difference threshold value, it is considered that the data sub-blocks corresponding to the first maximum peak value index sequence number to the second maximum peak value index sequence number are all included in the steel biting process occurring sub-block, a third maximum peak value in the plurality of data sub-block peak values is calculated in sequence, a third peak value comparison value is obtained by calculating the ratio of the third maximum peak value to the data sub-block mean peak value, if the third peak value comparison value is less than a preset fourth peak value comparison threshold value, it is considered that the data sub-blocks corresponding to the first maximum peak value index sequence number to the second maximum peak value index sequence number are all steel biting process occurring sub-blocks, if the third peak value comparison value is not less than the fourth peak value comparison threshold value and the difference between the third maximum peak value index sequence number and the second maximum peak value index sequence number is equal to a preset third sequence number difference threshold value, it is considered that the data sub-block corresponding to the third maximum peak value index sequence number is a steel biting process occurring sub-block, and the data sub-blocks corresponding to the first maximum peak value index sequence number to the third maximum peak value index sequence number are all steel biting process occurring sub-blocks. Taking the index number corresponding to the obtained biting process occurring sub-block as a vibration data segmentation boundary; Segmenting the original vibration data according to the vibration data segmentation boundary to obtain biting process vibration data and normal rolling process vibration data.

2. The hot rolling mill vibration data processing method of claim 1, wherein, The original vibration data of the hot rolling mill is collected, including: The vibration data of the hot rolling mill is collected by using a vibration acceleration sensor to obtain original vibration data and save the original vibration data.

3. The hot rolling mill vibration data processing method of claim 1, wherein, The original vibration data is segmented to obtain a plurality of data sub-blocks, and each data sub-block is indexed and labeled in time sequence, and the peak value of each data sub-block is calculated to obtain a plurality of data sub-block peak values, including: The original vibration data is segmented at a predetermined time interval to obtain a plurality of data sub-blocks, and each data sub-block is indexed and labeled in time sequence, and each data sub-block contains a plurality of data points; The peak value of each data sub-block is calculated, and a plurality of data sub-block peak values and their respective index numbers are obtained.

4. The hot rolling mill vibration data processing method of claim 3, wherein, The plurality of data sub-block peak values are sorted in descending order, the first few data sub-block peak values in the sequence are extracted, and the index numbers corresponding to the extracted data sub-block peak values are recorded, and the remaining data sub-block peak values after extraction are calculated to obtain a data sub-block average peak value, including: The plurality of data sub-block peak values are sorted in descending order, and the first three data sub-block peak values in the sequence are extracted and recorded as a first maximum peak value, a second maximum peak value and a third maximum peak value, respectively. The index numbers corresponding to the extracted data sub-block peak values are recorded and recorded as a first maximum peak value index number, a second maximum peak value index number and a third maximum peak value index number, respectively. The remaining data sub-block peak values in the sequence after extraction are calculated to obtain a data sub-block average peak value.

5. The hot rolling mill vibration data processing method of claim 1, wherein: The first peak value comparison threshold is 8, the second peak value comparison threshold is 3, the third peak value comparison threshold is 4, and the fourth peak value comparison threshold is 2. The first sequence number difference threshold is 2, the second sequence number difference threshold is 1 or 2, and the third sequence number difference threshold is 1.

6. The hot rolling mill vibration data processing method of claim 1, wherein, The original vibration data is segmented according to the vibration data segmentation boundary to obtain biting process vibration data and normal rolling process vibration data, including: The original data before the vibration data segmentation boundary is segmented as biting process vibration data, and the original data after the vibration data segmentation boundary is segmented as normal rolling process vibration data.

7. A hot rolling mill vibration data processing system characterized by, including: An original data collection module configured to collect original vibration data of a hot rolling mill; A preprocessing and peak value calculation module configured to segment the original vibration data to obtain a plurality of data sub-blocks, and index and label each data sub-block in time sequence, and calculate the peak value of each data sub-block to obtain a plurality of data sub-block peak values. The peak extraction and average peak calculation module is configured to sort the data sub-block peaks in descending order, extract the data sub-block peaks in the first few digits in the sequence, record the index sequence number corresponding to the extracted data sub-block peaks, and record the index sequence number corresponding to the extracted data sub-block peaks, and calculate the average of the data sub-block peaks remaining after extraction to obtain the data sub-block average peak; The segmentation limit calculation module is configured to compare the extracted data sub-block peaks with the data sub-block average peak respectively, judge according to the comparison result, combine the preset threshold and the index sequence number, and obtain the vibration data segmentation limit; The vibration data segmentation module is configured to segment the original vibration data according to the vibration data segmentation limit to obtain the steel biting process vibration data and the normal rolling process vibration data.

8. A smart terminal, characterized by The memory stores at least one instruction, at least one program, a code set or an instruction set, and the processor loads and executes the at least one instruction, at least one program, code set or instruction set to implement the hot rolling mill vibration data processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the processor loads and executes the at least one instruction, at least one program, code set or instruction set to implement the hot rolling mill vibration data processing method according to any one of claims 1 to 6.

Images