A method for monitoring and early warning of strip deviation of a cold rolling continuous annealing furnace

Abstract

A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace includes the following steps: S1: Acquire strip shape data from three levels of computers (L1, L2, and L3) through a multi-source data acquisition module; S2: Extract features of strip shape information based on the acquired strip shape data; S3: Determine the extracted features according to the set feature extraction rules, and provide an early warning result based on the determination result. This method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace, based on cross-process information integration, achieves early warning of strip deviation and integrated monitoring of cross-process data at the same time and location. This is beneficial for improving the stability of strip operation in the cold-rolled annealing furnace and minimizing the occurrence of strip edge rubbing and scratching within the furnace.

Description

Technical Field

[0001] This invention belongs to the field of cold rolling continuous annealing technology, specifically relating to a method for monitoring and early warning of strip deviation in a cold rolling continuous annealing furnace. Background Technology

[0002] Strip deviation within a cold rolling continuous annealing furnace can cause anything from minor production line slowdowns to severe strip breakage and production shutdowns. Therefore, researching strip deviation within the annealing furnace is crucial for improving the stable production capacity of the unit. Strip deviation within the annealing furnace involves numerous factors and is a complex nonlinear problem. Despite significant research findings, continuous innovation is needed to minimize the impact of strip deviation and continuously improve the production level of cold rolling continuous annealing units.

[0003] Solving the problem of strip steel deviation requires in-depth utilization of data from different processes, physical spaces, and time ranges. Current issues include: separation of raw strip shape data from the cold rolling mill from the annealing mill's production process control; separation of information outside and inside the annealing furnace; and a lack of automation of expert knowledge. Therefore, there is an urgent need to research a method and system for monitoring and early warning of strip steel deviation in cold rolling continuous annealing furnaces based on big data and internet technology to support the improvement of unit operation.

[0004] The invention application with application number CN 201710433843.4 discloses "a method and device for early warning of strip steel deviation on a cold-rolled galvanizing line", which includes: during the operation of the strip steel on the cold-rolled galvanizing line, obtaining the strip steel position deviation value from each of the N furnace front correction sensors out of M furnace front correction sensors, so as to obtain N strip steel position deviation values ​​for N positions in front of the annealing furnace, determining whether each of the N strip steel position deviation values ​​belongs to the corresponding preset deviation range, and if it is determined that any one of the N strip steel position deviation values ​​does not belong to the corresponding preset deviation range, then issuing a strip steel deviation warning message, wherein preset deviation ranges are set for each of the N furnace front correction sensors.

[0005] The invention application with application number CN 201910934118.4 discloses "a method for predicting strip deviation and breakage at the inlet of a continuous annealing unit", which includes: obtaining the process parameters of the main equipment in the inlet section of the continuous annealing unit; determining the transverse distribution function of internal tensile stress in the strip based on the process parameters; determining the strip deviation amount based on the process parameters; and determining the strip breakage risk result based on the transverse distribution function of internal tensile stress in the strip and the strip deviation amount. Summary of the Invention

[0006] To address the above problems, this invention provides a method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace, the specific technical solution of which is as follows:

[0007] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace, characterized by the following steps:

[0008] S1: Obtain strip shape data from L1, L2, and L3 computers through the configured multi-source data acquisition module.

[0009] S2: Extract features of strip shape information based on the acquired strip shape data;

[0010] S3: Determine the extracted features according to the set feature extraction rules, and give an early warning result based on the determination result.

[0011] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0012] Tracking blocks are also set at equal intervals along the length of the strip. By determining the furnace position of each tracking block at each time point in the time series, the furnace information at each time point in the time series is assigned to each tracking block. This establishes the monitoring of strip information at each tracking block position in the time series and the monitoring of strip information at each tracking block position along the length of the strip at the same time.

[0013] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0014] The characteristics of the strip shape information consist of two dimensions: the flatness measurement value of the strip surface and the tension difference value between the two sides of the strip corresponding to the measurement position.

[0015] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0016] Establish an overall plate shape channel consisting of a plate shape data channel, a strip length channel, and a plate shape tension difference channel;

[0017] The plate shape data channel is used to store the collected flatness time series data;

[0018] Plate shape data, strip length data, and tension difference data are collected and stored simultaneously.

[0019] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0020] The strip shape data is distributed symmetrically in six regions along the width direction of the strip, from the driving side to the operating side. The six regions are: the driving side edge, the driving side 1 / 4, the driving side middle, the operating side middle, the operating side 1 / 4, and the operating side edge.

[0021] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0022] Based on the actual occupancy of the channel, the channel ID value of each area is determined by calculating and measuring the correspondence between the channel ID and the actual area. Based on this, the dictionary of wave-shaped areas and channel IDs is established. Based on the established dictionary of wave-shaped areas and channel IDs, six characteristic values ​​at each meter position are calculated along the length of the strip to form a characteristic value matrix.

[0023] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0024] The establishment of the wave-shaped region and channel ID dictionary is completed according to the following steps:

[0025] SS1: Determine the number of valid channels;

[0026] SS2: Calculate the strip width based on the total number of channels, the number of effective channels, the width of the edge channels, and the width of the middle channels;

[0027] SS3: Calculate the calculated width of the middle wave based on the calculated width of the strip, the width of the edge channel, and the number of edge channels;

[0028] SS4: Calculate the number of mid-wave channels based on the width of the mid-wave, the width of the mid-channel, and the total number of channels;

[0029] SS5: Determine the number of quarter-wave channels based on the overall number of channels, the number of edge channels, and the number of mid-wave channels.

[0030] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0031] In step SS1, the determination of whether a channel is a valid channel is established by comparing the count of non-zero values ​​at the corresponding channel acquisition points with a set threshold.

[0032] When the count of non-zero values ​​at the corresponding channel acquisition point is greater than or equal to the set threshold, it is determined to be a valid channel; otherwise, it is an invalid channel.

[0033] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0034] Step SS2 is specifically determined according to the following formula:

[0035]

[0036] in,

[0037] width cal: Calculated width of strip steel, unit: mm;

[0038] N: Total number of channels;

[0039] Width of the edge passage, unit: mm;

[0040] Width of the central passage, unit: mm;

[0041] L effective Number of valid channels.

[0042] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0043] Step SS3 is determined specifically according to the following formula:

[0044]

[0045] in,

[0046] midwidth cal : Width of wave in the middle section, unit: mm;

[0047] ceil: rounds up to the nearest integer.

[0048] width cal : Calculated width of strip steel, unit: mm;

[0049] Width of the edge passage, unit: mm;

[0050] N edge Number of side passages.

[0051] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0052] When the ratio of the calculated mid-wave width to the total number of channels is greater than the set mid-wave channel width, step SS4 is performed according to the following formula:

[0053]

[0054] Otherwise, proceed as follows:

[0055]

[0056] in,

[0057] N mid Number of central passageways;

[0058] midwidth cal: Width of wave in the middle section, unit: mm;

[0059] Width of the central passage, unit: mm;

[0060] N: Total number of channels.

[0061] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0062] Step SS5 is determined specifically according to the following formula:

[0063] N qua =NN edge -N mid ,

[0064] in,

[0065] N qua : 1 / 4 of the number of channels;

[0066] N: Total number of channels;

[0067] N edge Number of side passages;

[0068] N mid Number of central passageways.

[0069] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0070] The strip is divided into three regions along its length: head, middle, and tail.

[0071] Feature extraction rules for plate shape characteristics are established in each region of the head and tail based on five factors: tension difference, mid-wave area ratio, average straightness, edge wave straightness difference, and mid-wave straightness difference. The plate shape characteristics are then determined based on the established feature extraction rules.

[0072] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0073] The aforementioned "calculating 6 characteristic values ​​at each meter position along the length of the strip to form an characteristic value matrix" specifically involves calculating the weighted average of the straightness of each wavy region within each meter length, and using the weighted average result of each region to characterize the straightness of the corresponding region.

[0074] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0075] The first rule is set based on the tension difference: the average tension difference between the drive side and the operating side is greater than or equal to 0.

[0076] The second rule is set based on the mid-wave area ratio: the ratio of the number of straightness values ​​greater than or equal to 0 in the mid-wave cross section to the total number of straightness values ​​in the mid-wave cross section is less than 1 / 2;

[0077] The third rule is set based on the average straightness value: the average straightness value of the cross section is less than or equal to the set straightness threshold value.

[0078] Based on the difference in straightness between the edge waves and the difference in straightness between the middle waves, the following rules, from fourth to sixth, are established:

[0079] Rule 4: The difference between the side waves on the driving side and the operating side is less than or equal to the set threshold for the side wave difference, and the difference between the middle waves on the driving side and the operating side is less than or equal to the set threshold for the middle wave difference.

[0080] Rule 5: The difference between the side wave on the driving side and the operating side is greater than or equal to the set threshold for the side wave difference, and the difference between the middle wave on the driving side and the operating side is greater than or equal to the set threshold for the middle wave difference, and the average straightness of the cross section is less than or equal to the set average straightness of the cross section.

[0081] Rule 6: The absolute value of the side wave difference between the driving side and the operating side is less than or equal to the set threshold for the absolute value of the side wave difference, and the mid-wave difference between the driving side and the operating side is less than or equal to the set threshold for the mid-wave difference.

[0082] If a paper is satisfied with any one of the following rules, namely, Rule 1, Rule 2, Rule 3, and any one of the following rules, then the paper is deemed a qualified paper; otherwise, it is deemed an unqualified paper and a warning is issued.

[0083] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0084] The phrase "by determining the furnace position of each tracking block at each time point in the time series, thereby assigning furnace information at each time point in the time series to each tracking block" specifically refers to:

[0085] First, calculate the real-time furnace position of each tracking block. Then, calculate the relative position of each tracking block to the functional area. Finally, store the real-time furnace information of the area to which the tracking block belongs in the tracking block.

[0086] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0087] The real-time furnace position of each tracking block is calculated as follows:

[0088] First, generate the sequence number of tracking blocks along the length of each roll;

[0089] Next, the inlet position of the annealing furnace is set as the origin, and the distance between each tracking block and the origin is determined according to the sequence number of the tracking blocks, i.e. the length of the tracking blocks;

[0090] Finally, the real-time furnace position of each tracking block is determined by subtracting the distance between each tracking block and the origin from the distance from the strip head to the origin.

[0091] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0092] Calculate the relative positional relationship between each tracking block and the functional area, specifically including the following steps:

[0093] SA1: Set the annealing furnace inlet position as the origin, and calculate the difference between the real-time furnace position of each tracking block and the distance of each functional area from the origin.

[0094] SA2: Determine whether the difference between the real-time furnace position of each tracking block and the distance between each functional area and the origin is greater than or equal to 0 and less than or equal to the sum of the lengths of each functional area and the lengths of each tracking block; if true, determine that the tracking block is in the functional area, otherwise determine that it is not in the functional area.

[0095] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0096] The real-time furnace position of each tracking block is calculated as follows:

[0097] First, generate the sequence number of tracking blocks along the length of each roll;

[0098] Secondly, the distance between each tracking block and the origin is determined based on the sequence number of the tracking blocks, i.e., the length of the tracking block;

[0099] Finally, the real-time furnace position of each tracking block is determined by subtracting the distance between each tracking block and the origin from the distance from the strip head to the origin.

[0100] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0101] Each tracking block also contains the planning information and the shape information of the strip to which it belongs.

[0102] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention is characterized in that:

[0103] The strip shape information contained in each tracking block consists of two dimensions: the flatness measurement value of the strip surface and the tension difference value between the two sides of the strip at the measurement position.

[0104] The tension difference value on both sides of the strip corresponding to the measurement position is directly assigned;

[0105] The flatness measurement value of the strip surface is assigned according to the following steps:

[0106] First, a strip shape data channel is established to store the collected flatness time series data; and along the strip width direction, from the driving side to the operating side, a symmetrical strip shape data distribution is formed in six regions: the driving side edge, the driving side 1 / 4, the driving side middle, the operating side middle, the operating side 1 / 4, and the operating side edge.

[0107] Then, based on the actual occupancy of the channel, the channel ID value of each area is determined by calculating the correspondence between the channel ID and the actual area. Based on this, the dictionary of wave-shaped areas and channel IDs is established. Based on the established dictionary of wave-shaped areas and channel IDs, six characteristic values ​​at each meter position are calculated along the length of the strip to form a characteristic value matrix.

[0108] Finally, the feature values ​​of each region within the corresponding length of each tracking block are calculated by weighted averaging within each region, and the result represents the feature values ​​of each region.

[0109] This invention discloses a method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace. Through a multi-source data acquisition and processing module, it acquires L3 plan data, cold-rolling mill strip shape data, annealing mill external data, and annealing furnace internal data in real time via network interconnection, completing the static and dynamic positioning of data in different areas of the strip. Then, according to the production plan sequence, it analyzes the strip shape data and provides early warnings for planned strip deviation based on strip shape characteristics. Finally, it tracks the operation of the strip in the furnace, displaying information such as strip shape, external CPC (Cost Per Scale), and internal tension, speed, and internal CPC in the strip zones, enabling all strip zones to be in the same position and at the same time. The invention integrates and monitors key information; the strip shape characteristics, in addition to flatness, also take into account the tension difference between the two sides of the strip, fully considering influencing factors; when calculating multiple characteristic values, a weighted average is used to make the values ​​more accurate; in summary, the present invention provides a method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace. Based on cross-process information integration, it realizes early warning of strip deviation and integrated monitoring of strip cross-process data at the same time and location, which is conducive to improving the stability of strip operation in cold-rolled annealing furnaces, minimizing the occurrence of strip edge rubbing and scratching in the furnace, and improving the operational stability of the unit. Attached Figure Description

[0110] Figure 1 This is a schematic diagram of the deviation warning steps of the present invention;

[0111] Figure 2 This is a schematic block diagram illustrating the working principle of the present invention.

[0112] Figure 3 This is a schematic diagram showing the division of the strip width into six regions in the working principle section of this invention.

[0113] Figure 4 This is a schematic diagram showing the distance between each working area and the set origin in the working principle section of this invention;

[0114] Figure 5 This is a schematic diagram showing the division of the tracking blocks along the length of the strip in the working principle section of this invention;

[0115] Figure 6 This is a schematic diagram illustrating the data of a single strip block at different times throughout the entire process at time t1 in this embodiment of the invention.

[0116] Figure 7 This is a schematic diagram showing the data of all blocks of the strip at the same time t2 in an embodiment of the present invention. Detailed Implementation

[0117] The following is a further detailed description of a method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace according to the present invention, based on the accompanying drawings and specific embodiments.

[0118] A method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace includes the following steps:

[0119] S1: Obtain strip shape data from L1, L2, and L3 computers through the set multi-source data acquisition module;

[0120] S2: Extract features of strip shape information based on the acquired strip shape data;

[0121] S3: Determine the extracted features according to the set feature extraction rules, and give an early warning result based on the determination result.

[0122] in,

[0123] Tracking blocks are also set at equal intervals along the length of the strip. By determining the furnace position of each tracking block at each time point in the time series, the furnace information at each time point in the time series is assigned to each tracking block. This establishes the monitoring of strip information at each tracking block position in the time series and the monitoring of strip information at each tracking block position along the length of the strip at the same time.

[0124] in,

[0125] The phrase "by determining the furnace position of each tracking block at each time point in the time series, thereby assigning furnace information at each time point in the time series to each tracking block" specifically refers to:

[0126] First, calculate the real-time furnace position of each tracking block. Then, calculate the relative position of each tracking block to the functional area. Finally, store the real-time furnace information of the area to which the tracking block belongs in the tracking block.

[0127] in,

[0128] The real-time furnace position of each tracking block is calculated as follows:

[0129] First, generate the sequence number of tracking blocks along the length of each roll;

[0130] Next, the inlet position of the annealing furnace is set as the origin, and the distance between each tracking block and the origin is determined according to the sequence number of the tracking blocks, i.e. the length of the tracking blocks;

[0131] Finally, the real-time furnace position of each tracking block is determined by subtracting the distance between each tracking block and the origin from the distance from the strip head to the origin.

[0132] in,

[0133] Calculate the relative positional relationship between each tracking block and the functional area, specifically including the following steps:

[0134] SA1: Set the annealing furnace inlet position as the origin, and calculate the difference between the real-time furnace position of each tracking block and the distance of each functional area from the origin.

[0135] SA2: Determine whether the difference between the real-time furnace position of each tracking block and the distance between each functional area and the origin is greater than or equal to 0 and less than or equal to the sum of the lengths of each functional area and the lengths of each tracking block; if true, determine that the tracking block is in the functional area, otherwise determine that it is not in the functional area.

[0136] in,

[0137] The real-time furnace position of each tracking block is calculated as follows:

[0138] First, generate the sequence number of tracking blocks along the length of each roll;

[0139] Secondly, the distance between each tracking block and the origin is determined based on the sequence number of the tracking blocks, i.e., the length of the tracking block;

[0140] Finally, the real-time furnace position of each tracking block is determined by subtracting the distance between each tracking block and the origin from the distance from the strip head to the origin.

[0141] In the above,

[0142] Each tracking block also contains the planning information and the shape information of the strip to which it belongs.

[0143] in,

[0144] The strip shape information contained in each tracking block consists of two dimensions: the flatness measurement value of the strip surface and the tension difference value between the two sides of the strip at the measurement position.

[0145] The tension difference value on both sides of the strip corresponding to the measurement position is directly assigned;

[0146] The flatness measurement value of the strip surface is assigned according to the following steps:

[0147] First, a strip shape data channel is established to store the collected flatness time series data; and along the strip width direction, from the driving side to the operating side, a symmetrical strip shape data distribution is formed in six regions: the driving side edge, the driving side 1 / 4, the driving side middle, the operating side middle, the operating side 1 / 4, and the operating side edge.

[0148] Then, based on the actual occupancy of the channel, the channel ID value of each area is determined by calculating the correspondence between the channel ID and the actual area. Based on this, the dictionary of wave-shaped areas and channel IDs is established. Based on the established dictionary of wave-shaped areas and channel IDs, six characteristic values ​​at each meter position are calculated along the length of the strip to form a characteristic value matrix.

[0149] Finally, the feature values ​​of each region within the corresponding length of each tracking block are calculated by weighted averaging within each region, and the result represents the feature values ​​of each region.

[0150] in,

[0151] The characteristics of the strip shape information consist of two dimensions: the flatness measurement value of the strip surface and the tension difference value between the two sides of the strip corresponding to the measurement position.

[0152] in,

[0153] Establish an overall plate shape channel consisting of a plate shape data channel, a strip length channel, and a plate shape tension difference channel;

[0154] The plate shape data channel is used to store the collected flatness time series data;

[0155] Plate shape data, strip length data, and tension difference data are collected and stored simultaneously.

[0156] in,

[0157] The strip shape data is distributed symmetrically in six regions along the width direction of the strip, from the driving side to the operating side. The six regions are: the driving side edge, the driving side 1 / 4, the driving side middle, the operating side middle, the operating side 1 / 4, and the operating side edge.

[0158] in,

[0159] Based on the actual occupancy of the channel, the channel ID value of each area is determined by calculating and measuring the correspondence between the channel ID and the actual area. Based on this, the dictionary of wave-shaped areas and channel IDs is established. Based on the established dictionary of wave-shaped areas and channel IDs, six characteristic values ​​at each meter position are calculated along the length of the strip to form a characteristic value matrix.

[0160] in,

[0161] The establishment of the wave-shaped region and channel ID dictionary is completed according to the following steps:

[0162] SS1: Determine the number of valid channels;

[0163] SS2: Calculate the strip width based on the total number of channels, the number of effective channels, the width of the edge channels, and the width of the middle channels;

[0164] SS3: Calculate the calculated width of the middle wave based on the calculated width of the strip, the width of the edge channel, and the number of edge channels;

[0165] SS4: Calculate the number of mid-wave channels based on the width of the mid-wave, the width of the mid-channel, and the total number of channels;

[0166] SS5: Determine the number of quarter-wave channels based on the overall number of channels, the number of edge channels, and the number of mid-wave channels.

[0167] in,

[0168] In step SS1, the determination of whether a channel is a valid channel is established by comparing the count of non-zero values ​​at the corresponding channel acquisition points with a set threshold.

[0169] When the count of non-zero values ​​at the corresponding channel acquisition point is greater than or equal to the set threshold, it is determined to be a valid channel; otherwise, it is an invalid channel.

[0170] in,

[0171] Step SS2 is specifically determined according to the following formula:

[0172]

[0173] in,

[0174] width cal : Calculated width of strip steel, unit: mm;

[0175] N: Total number of channels;

[0176] Width of the edge passage, unit: mm;

[0177] Width of the central passage, unit: mm;

[0178] L effective Number of valid channels.

[0179] in,

[0180] Step SS3 is determined specifically according to the following formula:

[0181]

[0182] in,

[0183] midwidth cal : Width of wave in the middle section, unit: mm;

[0184] ceil: rounds up to the nearest integer.

[0185] width cal : Calculated width of strip steel, unit: mm;

[0186] Width of the edge passage, unit: mm;

[0187] N edge Number of side passages.

[0188] in,

[0189] When the ratio of the calculated mid-wave width to the total number of channels is greater than the set mid-wave channel width, step SS4 is performed according to the following formula:

[0190]

[0191] Otherwise, proceed as follows:

[0192]

[0193] in,

[0194] N mid Number of central passageways;

[0195] midwidth cal : Width of wave in the middle section, unit: mm;

[0196] Width of the central passage, unit: mm;

[0197] N: Total number of channels.

[0198] in,

[0199] Step SS5 is determined specifically according to the following formula:

[0200] Nqua =NN edge -N mid ,

[0201] in,

[0202] N qua : 1 / 4 of the number of channels;

[0203] N: Total number of channels;

[0204] N edge Number of side passages;

[0205] N mid Number of central passageways.

[0206] in,

[0207] The strip is divided into three regions along its length: head, middle, and tail.

[0208] Feature extraction rules for plate shape characteristics are established in each region of the head and tail based on five factors: tension difference, mid-wave area ratio, average straightness, edge wave straightness difference, and mid-wave straightness difference. The plate shape characteristics are then determined based on the established feature extraction rules.

[0209] in,

[0210] The aforementioned "calculating 6 characteristic values ​​at each meter position along the length of the strip to form an characteristic value matrix" specifically involves calculating the weighted average of the straightness of each wavy region within each meter length, and using the weighted average result of each region to characterize the straightness of the corresponding region.

[0211] in,

[0212] The first rule is set based on the tension difference: the average tension difference between the drive side and the operating side is greater than or equal to 0.

[0213] The second rule is set based on the mid-wave area ratio: the ratio of the number of straightness values ​​greater than or equal to 0 in the mid-wave cross section to the total number of straightness values ​​in the mid-wave cross section is less than 1 / 2;

[0214] The third rule is set based on the average straightness value: the average straightness value of the cross section is less than or equal to the set straightness threshold value.

[0215] Based on the difference in straightness between the edge waves and the difference in straightness between the middle waves, the following rules, from fourth to sixth, are established:

[0216] Rule 4: The difference between the side waves on the driving side and the operating side is less than or equal to the set threshold for the side wave difference, and the difference between the middle waves on the driving side and the operating side is less than or equal to the set threshold for the middle wave difference.

[0217] Rule 5: The difference between the side wave on the driving side and the operating side is greater than or equal to the set threshold for the side wave difference, and the difference between the middle wave on the driving side and the operating side is greater than or equal to the set threshold for the middle wave difference, and the average straightness of the cross section is less than or equal to the set average straightness of the cross section.

[0218] Rule 6: The absolute value of the side wave difference between the driving side and the operating side is less than or equal to the set threshold for the absolute value of the side wave difference, and the mid-wave difference between the driving side and the operating side is less than or equal to the set threshold for the mid-wave difference.

[0219] If a paper is satisfied with any one of the following rules, namely, Rule 1, Rule 2, Rule 3, and any one of the following rules, then the paper is deemed a qualified paper; otherwise, it is deemed an unqualified paper and a warning is issued.

[0220] Working process and principle

[0221] For the sake of explanation, the following will take the form of first constructing a computer system for monitoring and early warning of deviation.

[0222] [1] Establish a computer model system for monitoring and early warning of strip deviation. The system includes: a communication module that interacts with the cold rolling mill's strip shape server and L2 process control computer, as well as with the annealing furnace's L1 system; a strip shape analysis module; a rule knowledge module; a deviation early warning module; and a data integration module.

[0223] [2] Using the communication module of step 1, communicate with the L2 process control computer to obtain the annealing furnace production plan data in a timely manner; then, using the communication module, communicate with the cold rolling mill plate shape server to obtain the plate shape data of the strip in the plan, and parse the plate shape data to extract the plate shape features;

[0224] [3] Using the plate shape features obtained in step 2, the deviation warning module is called to make inferences and judgments according to the given knowledge rule base, and deviation warning information is given to each steel coil in the production plan before production.

[0225] [4] During the production process, the strip is divided into tracking blocks along the length of the strip. The communication module of step 1 is used to establish real-time tracking of the strip blocks from the entrance to the exit of the annealing furnace. The strip plan information, plate shape information and the real-time furnace information corresponding to the tracking blocks are linked with the blocks in real time. [5] Based on the block position tracking and data integration established in step [4], the data related to deviation are integrated and monitored, including the data display of a single strip block at different times in the whole process and the data display of all strip blocks at the same time.

[0226] The steps are described in detail below:

[0227] I. Establish a computer model system for monitoring and early warning of strip steel deviation. This system includes: a communication module for data interaction with the cold rolling mill's strip shape server and L2 process control computer, as well as with the annealing furnace's L1 system; a strip shape analysis module; a rule knowledge module; a deviation early warning module; and a data integration module.

[0228] The strip misalignment data integration and monitoring system involved in step 1 is as follows: Figure 2 As shown, it involves a communication model module, a strip shape analysis module, a rule knowledge module, a deviation warning module, a data integration and monitoring module, and a main control module that includes data processing, strip tracking, and logic processing.

[0229] Second, using the communication module from step 1, communicate with the L2 process control computer to obtain the annealing furnace production plan data in a timely manner; then, using the communication module, communicate with the cold rolling mill strip shape server to obtain the strip shape information in the plan, and extract the strip shape features based on the strip shape information.

[0230] The strip shape information includes the flatness measurement value of the strip surface, the measurement position corresponding to the measurement value, and the tension difference between the two sides of the strip corresponding to the measurement position.

[0231] The main data for annealing furnace production planning includes: strip tracking number, process annealing temperature, process tension, process speed, and strip length. strip strip width w strip strip thickness t strip , strip steel tapping mark g, etc.

[0232] After the annealing furnace production plan is updated, the L2 process control computer sends the updated plan to the strip misalignment monitoring and early warning computer model system in real time. In addition, the strip misalignment monitoring and early warning computer model system can also proactively request a plan from the L2 process control computer. After receiving the request information, the L2 process control computer sends the latest plan information to the strip misalignment monitoring and early warning computer model system.

[0233] A communication module is used to communicate with the cold rolling mill strip shape server to obtain strip shape data from the annealing furnace production plan and extract features from the strip shape information.

[0234] The cold rolling mill's plate shape server reserves N+2 channels with fixed IDs as channels for overall plate shape. The channel set is as follows:

[0235] S ID =[channel1,channel2,…,channel i ,…,channel N ,channel length ,channel tendiff ],

[0236] 1≤i≤N

[0237] Among them, channel i This is a plate-shaped data channel used to store the collected flatness time series data. For t s Time Channel i The collected straightness values ​​reflect the shape characteristics of the strip; strip length channel. length and plate tension difference channel tendiff (The strip tension difference is the tension difference between the driving side and the operating side), which are used to store the strip position information and tension difference information corresponding to the strip shape time series, respectively:

[0238] For t s The position of the strip corresponding to the flatness measurement point at any given time.

[0239] For t s The tension difference of the plate shape at each position is measured at all times.

[0240] Strip shape data, strip length data, and strip tension difference data are collected and stored simultaneously. s Time, channel length The value obtained from the collection point is The channel corresponding to that moment i The value of the collection point The channel corresponding to that moment tendiff The value of the collection point

[0241] After obtaining the strip shape data, strip shape features are extracted based on the strip shape information. Distinguishing between the drive side and the operating side, the strip is divided into six regions along its width direction, symmetrically: drive side edge (area1), drive 1 / 4 section (area2), drive side middle section (area3), operating side middle section (area4), operating side 1 / 4 section (area5), and operating side edge section (area6). Along the length direction, six feature values ​​corresponding to each meter position in the width direction are calculated, a feature value index is established, and a strip shape feature matrix is ​​generated, as shown below. Figure 3 As shown.

[0242] After the cold rolling mill's strip shape server collects and stores flatness data, it allocates the data according to the actual width of the strip. Therefore, when generating strip shape features, it is necessary to calculate the correspondence between the measurement channel ID and the actual area based on the actual channel occupancy, determine the channel ID values ​​corresponding to the edge wave area, quarter wave area, and middle wave area on the drive side and operation side, and form a wave area and channel ID dictionary. Where area i For the area name, For area i The corresponding set of channels.

[0243] Overall Channel S ID Includes edge channel egde and central channel mid The width of the side passage is denoted as The width of the central passage is denoted as The total number of channels is N, and the number of edge channels is N. edge The number of central passages is N mid The number of channels in 1 / 4 section is N qua When allocating channels, the number of edge channels is a known quantity, predefined. The number of middle channels is calculated using a formula.

[0244] The effective channel set is The number of valid channels is M, where M ≤ N. M is determined by the valid channel judgment criteria. The condition for judging whether a channel is a valid channel is that the number of non-zero values ​​at the channel's acquisition points exceeds a set threshold.

[0245]

[0246] in channel i The set of collection points that are not equal to 0, threshold effective The conditional threshold for an effective channel.

[0247] The area is allocated according to the formula, and the calculated width of the strip is calculated first.

[0248]

[0249] Among them, width cal For calculating the width of the strip, L effective This represents the number of valid channels.

[0250] Then, the width of the wave (midwidth) is calculated using the formula. cal

[0251]

[0252] Where midwidth cal The width of the wave is calculated using ceil, which rounds up to the nearest integer.

[0253] Then use midwidth cal Calculate the number N of wave channels according to the formula. mid ,

[0254]

[0255] After obtaining the number of intermediate wave channels, calculate the number of quarter wave channels:

[0256] N qua =NN edge -N mid

[0257] After obtaining the number of edge waves, mid waves, and quarter waves, the channels can be assigned according to the symmetry of the board shape. Along the length direction, six eigenvalues ​​are calculated at each meter position, forming an eigenvalue matrix M. feature .

[0258] Third, based on the plate shape features obtained in step 2, and according to the given knowledge rule base, the deviation warning module is called to perform reasoning and judgment, and deviation warning information is given to each steel coil in the production plan before production.

[0259] Retrieve rule set S from the knowledge rule base role The rule set is used to issue early warnings for deviation of each coil of strip steel in the production plan. The rule set includes the following rules:

[0260] Feature extraction rule 1: The mean tension difference between the driving side and the operating side is ≥0, and is denoted as role1;

[0261] Feature extraction rule 2: The ratio of the number of straightness values ​​greater than or equal to 0 in the mid-wave cross section to the total number of straightness values ​​in the mid-wave cross section is less than 1 / 2, denoted as role2.

[0262] Feature extraction rule 3: Mean straightness of cross-section ≤ threshold cross , denoted as role3;

[0263] Feature extraction rule 4.1: The difference between the driving side and the operating side wave edge ≤ threshold1 and the difference between the driving side and the operating side wave mid-wave edge ≤ threshold2 are denoted as role. 4-1 ;

[0264] Feature extraction rule 4.2: The difference between the driving side and the operating side wave margin is ≥threshold1, and the difference between the driving side and the operating side wave margin is ≥threshold2, and the I value is ≤threshold.I , denoted as role 4-2 ;

[0265] Feature extraction rule 4.3: The absolute value of the difference between the driving side and the operating side wave difference ≤ threshold3 and the difference between the driving side and the operating side wave difference ≤ threshold2 is denoted as role. 4-3 ;

[0266] For each coil of strip steel, using the shape feature matrix and tension difference information from step 2, as well as the strip steel position information, the strip steel deviation warning is calculated according to the rules. The method is as follows:

[0267] Along its length, the strip is divided into three sections: head, middle, and tail. The head region is denoted as area. head area body area tail Calculate the rule compliance status for each of the three regions.

[0268] Obtain the average tension difference between the head, middle, and tail, and determine whether it conforms to rule 1.

[0269] Obtain the number of straightness measurement points at the head, middle, and tail that are greater than or equal to 0, and the number of straightness measurement points in the middle wave area. Calculate the ratio between the two and determine whether it meets rule 2.

[0270] Obtain the mean values ​​of the head, middle, and tail cross sections, and compare them with the threshold. cross Compare and determine whether it conforms to rule 3.

[0271] Obtain the edge wave difference and middle wave difference values ​​on the head, middle, and tail operation side driving sides. Compare the edge wave difference with threshold1 and the middle wave difference with threshold2 to determine whether they meet rule 4.1.

[0272] Obtain the edge wave difference, middle wave difference, and average straightness value on the head, middle, and tail operation side drive sides. Compare the edge wave difference with threshold1, the middle wave difference with threshold2, and the straightness value with threshold. I Determine whether it conforms to rule 4.2.

[0273] Obtain the edge wave difference and middle wave difference values ​​on the head, middle, and tail operation sides. Compare the absolute value of the edge wave difference with ≤threshold3 and the middle wave difference with threshold2 to determine if they meet rule 4.3.

[0274] 4. During the production process, the strip is divided into tracking blocks along the length of the strip. The communication module of step 1 is used to establish real-time tracking of the strip blocks from the entrance to the exit of the annealing furnace. The strip planning information, strip shape information and the real-time furnace information corresponding to the tracking blocks are linked to the blocks in real time.

[0275] First, establish the coordinate system of the continuous annealing furnace. weld The origin of the coordinate system is the inlet of the annealing furnace. The length l of each functional area is obtained through production line configuration information. area The distance l between the functional area and the origin is calculated using production line configuration information. area-weld ,like Figure 4 As shown.

[0276] Establish a coordinate system for the strip steel strip The origin of this coordinate system is the coil head. Based on the length information of each coil of strip steel and the length setting information of the tracking block, the tracking blocks are divided along the length direction to establish real-time tracking of the strip steel from the inlet position of the annealing furnace to the outlet position of the annealing furnace.

[0277] Let the set of tracking blocks be S. block = [block1, block2, ... block l ], block refers to the tracking blocks divided along the length of the strip, such as Figure 5 As shown.

[0278] The tracking block includes its distance from the coil head, its position within the furnace, its associated strip plan information, its strip shape information, and the corresponding real-time furnace information. The distance between the tracking block and the strip head, the associated strip plan information, and the associated strip shape information are calculated and assigned values ​​during the tracking block's generation. The calculation and assignment method is as follows:

[0279] Based on the length of each coil of strip steel l strip With tracking block length l block Generate tracking blocks along the length of each roll, and generate sequence number IDs for the tracking blocks in sequence. block The generated tracking blocks are stored in the tracking block sequence. The strip steel planning information to which each tracking block belongs is stored in the corresponding tracking block. The distance l between each tracking block and the origin of the strip steel coordinate system is calculated. block-strip And store it in the corresponding tracking block.

[0280] l block-strip =ID block ×l block

[0281] Based on the distance l between each tracking block and the origin of the strip steel coordinate system block-stripThe features are then matched with the plate shape features from step 2, and six plate shape feature values ​​corresponding to each tracking block are calculated and stored in the tracking block. The matching method is as follows:

[0282] The distance l between each tracking block and the origin of the strip steel coordinate system block-strip To track the starting index of the block's starting position along the length direction in the plate-shaped feature matrix, l block-strip +I block To track the end position of the block in the length direction of the plate feature matrix, the average value of the six region indices is calculated based on the values ​​matched by the start and end indices.

[0283] The furnace position value of the tracking block is calculated, and the real-time furnace information corresponding to the tracking block is assigned through real-time tracking calculation. The real-time tracking construction method is as follows:

[0284] Calculate the real-time furnace position of each tracking block (distance from the origin of the continuous annealing furnace coordinate system). The calculation method is as follows:

[0285] l block-weld =l strip-weld -l block-strip ,

[0286] Calculate the real-time positional relationship between each tracking block and the functional area, and store the real-time furnace information of the area to which the tracking block belongs in that tracking block. First, calculate the relative positional relationship between each tracking block and each functional area. block-area .

[0287] l block-area =l block-weld -l area-weld

[0288] According to l block-area The value is used to calculate the position status of the block to determine whether the tracked block is in the corresponding functional area.

[0289] 0 indicates that the tracking block does not belong to the corresponding functional area, and 1 indicates that the tracking block belongs to the corresponding functional area.

[0290] If according to l block-area The calculated position state is 0, indicating that the tracking block is not in this functional region. If l block-area A value of 1 indicates that the tracking block is in this functional region, and the real-time measurement values ​​of this functional region are stored in the tracking block.

[0291] Fifth, based on the Block position tracking and data integration established in step [4], conduct integrated monitoring of deviation-related data, including data display of a single Block of the strip at different times throughout the process and data display of all Blocks of the strip at the same time.

[0292] Example

[0293] Step 1: According to Figure 1 Establish a deviation monitoring and early warning system, which includes: a communication module that interacts with the cold rolling mill's shape server and L2 process control computer, as well as with the annealing furnace's L1 system; a shape analysis module; a rule knowledge module; a deviation early warning module; and a data integration module.

[0294] Step 2: Periodically obtain strip steel information, for example, for a specific update, obtain the annealing furnace production plan M. plan It contains 20 volumes. Volume information:

[0295] coil1=[1700,1450,0.6,GN3848]

[0296] The cold rolling mill's plate shape server reserves 52 channels with fixed IDs as channels for overall plate shape.

[0297] S ID =[channel1,channel2,…,channel 52 The number of data collection points per channel is 25457, and the data for the 52nd channel is as follows:

[0298] channel 52 = [0.05, -0.08, ..., 0.6]

[0299] strip length channel length and a plate tension difference channel tendiff The number of collection points is also 25457.

[0300] Calculate the wave region and channel ID dictionary: Width of the middle side passage Width of the central passage The total number of channels is 52, the number of edge channels is 4, and the number of central channels is N. mid To be determined, the number N of channels in a 1 / 4 unit. qua The number of valid channel sets is 47, which needs to be determined.

[0301] The area is allocated according to the formula, and the calculated width of the strip is calculated first.

[0302] width cal = 52 × 12 + 26 × (47 - 12)

[0303] width cal =1586

[0304] Then, the width of the wave (midwidth) is calculated using the formula.cal

[0305]

[0306] midwidth cal =769

[0307] Then use midwidth cal The number of waves N is calculated using the formula. mid .

[0308]

[0309] N mid =32,

[0310] N qua =52-4-32, N qua =16.

[0311] After obtaining the number of edge waves, middle waves, and quarter waves, the channels can be assigned according to the symmetry of the wave pattern. The assignment results are shown in Table 1.

[0312] Table 1

[0313]

[0314]

[0315] Along the length direction, six characteristic values ​​are calculated at each meter position. The length of the coiled steel is 1700 meters. Taking the characteristic value at 20m as an example, the calculation method for the other positions is the same.

[0316] Calculate the characteristic value of the driving side wave at this location, and obtain the average flatness value of channels 1 and 2 at this location based on the driving side wave ID and position index.

[0317] Calculate the characteristic value of the driving side 1 / 4 wave at this location. Based on the driving side 1 / 4 wave ID and position index, obtain the average flatness value of channels 3, 4, 5, 6, 7, 8, 9, and 10 at this location.

[0318] Calculate the characteristic value of the mid-wave on the driving side at this location. Based on the mid-wave ID and position index on the driving side, obtain the average flatness value of channels 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26 at this location.

[0319] Calculate the wave characteristic value on the operation side at this location. Based on the wave ID and position index on the operation side, obtain the average flatness value of channels 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42 at this location.

[0320] Calculate the characteristic value of the 1 / 4 wave on the operation side at this location. Based on the ID of the 1 / 4 wave on the operation side and the position index, obtain the average flatness value of channels 43, 44, 45, 46, 47, 48, 49, and 50 at this location.

[0321] Calculate the characteristic value of the side wave at this operation point, and obtain the average flatness value of channels 51 and 52 at this point based on the wave ID and position index of the operation.

[0322] After calculation, the characteristic value at 20m is [1 1 1 1 1 1], and the characteristic values ​​at other locations are calculated in the same way.

[0323] Step 3: For each coil of strip steel, using the shape feature matrix, tension difference information, and strip position information from Step 2, calculate the strip deviation warning according to the rules. Divide the strip steel into head, middle, and tail sections along its length. Let the head region be denoted as area. head area body area tail For the strip shape feature matrix obtained in step 2, feature values ​​are obtained according to the region and applied to the rules to determine the deviation risk of the strip in different regions. In step 2, a total of 4 coils were obtained, as shown in Table 2:

[0324] Table 2

[0325]

[0326]

[0327] The analysis is performed on the first volume, 512136210. The analysis process for the remaining volumes is the same. The start and end position ranges for the head region are: [0, 20], the middle region is: [20, 1680], and the tail region is:

[0328] [1680:1700]. Based on the position index in the eigenvalue matrix in step two, obtain the eigenvalue matrix of the head region, the eigenvalue matrix of the middle region, the eigenvalue matrix of the tail region, the tension difference of the head region, the tension difference of the middle region, and the tension difference of the tail region. As shown in Table 3.

[0329] Table 3

[0330]

[0331] The difference in wave area and flatness between the head, middle and tail regions is calculated according to the formula, as shown in Table 4.

[0332] Table 4

[0333]

[0334] The compliance of each rule in the head, middle, and tail regions is calculated based on the rule formula, as shown in Table 5.

[0335] Table 5

[0336]

[0337]

[0338] Considering the comprehensiveness of the rule evaluation, the rules are combined to evaluate the conformity of the board shape. The comprehensive evaluation rule is: a roll that simultaneously meets the rules 1, 2, and 3, and also meets any of the rules 4.1, 4.2, and 4.3, is considered a qualified roll in terms of board shape conformity. The table above shows the quantitative values ​​of the board shape for the beginning, middle, and end of the roll; the remaining rolls are displayed in the same way.

[0339] Step 4: Using the communication module from Step 1, communicate with the L2 process control computer to acquire high-frequency measurement data inside the furnace, strip position tracking data, and annealing furnace production plan data to obtain strip information:

[0340] coil1=[1700,1450,0.6,GN3848]

[0341] Calculate the length l of the strip in this coil. strip The length is 1770 meters, and the tracking block length is l. block For a length of 40 meters, tracking blocks are generated along the length of each roll, and the number of tracking blocks generated is... There are 45 in total. The set of tracking blocks is S. block = [block0, block1, ... block 44 ].

[0342] Establish the coordinate system of the continuous annealing furnace enter The origin of the coordinate system is the entrance position of the annealing furnace. The distance l between the functional area and the origin is calculated using the production line configuration information. area-enter The key functional areas involved in the embodiment include: the position of the hydraulic cylinder at CPC8, the strip offset at CPC8, the furnace temperature of RTF1 section, the speed of the furnace rollers in RTF1 section, the load of the furnace rollers in RTF1 section, and the furnace temperature of RTF2 section.

[0343] Based on the production line configuration information, obtain the length l of each functional area. area Cpc8 measures position length l cpc8 = 50 meters, RTF1 zone length l RTF =200 meters, RTF2 zone length l RTF =250 meters

[0344] Based on the production line configuration information, calculate the distance between each functional area and the annealing furnace inlet. area-enter .

[0345] l jpf-enter =200 meters, l RTF1-enter = 400 meters, l RTF2-enter =700 meters

[0346] The calculation is performed on block0, and the calculation method for the other blocks is the same.

[0347] block0 contains the distance value between itself and the coil head, its position in the furnace, the strip plan information, the strip shape information, and the real-time furnace information corresponding to the tracking block.

[0348] The distance between the tracking block and the head of its corresponding strip, the strip plan information to which the tracking block belongs, and the strip shape information to which the tracking block belongs are calculated and assigned values ​​when the tracking block is generated. The calculation and assignment method is as follows:

[0349] At time t1, communication with the L2 process control computer will be established, and the acquired production plan data for the annealing furnace will be assigned to tracking block 0. The assignment method for the remaining blocks is the same. After assignment, the strip information for block 0 is: coil1 = [1700, 1450, 0.6, GN3848]. The strip information for the remaining blocks is assigned in the same way. The distance between block 0 and the coil head (the distance from the origin of the strip coordinate system) is calculated. block-strip =0*40=0, and the calculation method for the remaining blocks is the same. The strip shape information of the tracking block block0 is obtained from the strip shape feature matrix in step 2. The distance between block0 and the roll head is 0, the length of block0 is 40 meters, the starting index in the strip shape feature matrix is ​​0, and the ending index is 40, resulting in the strip shape feature matrix belonging to this tracking block, with a dimension of 6×40. The acquisition method for the remaining blocks is the same.

[0350] The furnace position value of the tracking block (block0) is calculated and assigned using real-time tracking information. The real-time tracking construction method is as follows:

[0351] At time t2, the position tracking data of the coiled steel was received: This indicates that the distance between the head of the steel coil and the welding machine is 220 meters. At time t2, high-frequency data from inside the furnace was received, including measurement information for the key area: Calculate the real-time positional relationship between tracking block 0 and each functional area, and store the real-time furnace information of the area to which the tracking block belongs in the tracking block. The calculation method for the other blocks is the same.

[0352] Calculate the real-time furnace position of tracking block0 (distance from the origin of the continuous annealing furnace coordinate system). The remaining tracking blocks are calculated using the same formula.

[0353] Calculate the relative positional relationship between block0 and each functional region at time t2.

[0354]

[0355]

[0356]

[0357] According to l block-area Calculate the position state, based on Calculation results:

[0358] 0≤20≤90, -180 < 0, -480 < 0.

[0359] Based on the calculation results, block0 is located at CPC8. Therefore, the values ​​of the hydraulic cylinder position at CPC8 and the strip offset at CPC8 are assigned to block0.

[0360] At time t3, the position tracking data of the coiled steel was received: This indicates that the distance between the head of the steel coil and the welding machine is 460 meters. At time t2, high-frequency data from inside the furnace was received, including measurement information for the key area: Calculate the real-time positional relationship between tracking block 0 and each functional area, and store the real-time furnace information of the area to which the tracking block belongs in the tracking block. The calculation method for the other blocks is the same.

[0361] Calculate the real-time furnace position of tracking block0 (distance from the origin of the continuous annealing furnace coordinate system). The remaining tracking blocks are calculated using the same formula.

[0362] Calculate the relative positional relationship between block0 and each functional region at time t2.

[0363]

[0364]

[0365]

[0366] According to l block-area Calculate the position state, based on Calculation results:

[0367] 260 > 240 0 < 60 < 240, -240 < 0.

[0368] Based on the calculation results, block0 is located at RTF1. Therefore, the values ​​of RTF1 furnace temperature, RTF1 furnace roller speed, and RTF1 furnace roller load are assigned to block0. The other tracking blocks are calculated using the same formula.

[0369] Step 5: Based on the tracking in Step 4, perform integrated monitoring of deviation-related data, including data display of a single strip block at different times throughout the process and data display of all strip blocks at the same time.

[0370] Data display of a single strip block at different times throughout the entire process at time t1: Taking the first tracking block as an example, the data display of block0 in RTF1 segment and CPC8 segment (e.g.) Figure 6 (as shown);

[0371] Data display of all blocks of the strip at time t2 (e.g., Figure 7 As shown in the image, the current strip steel information inside the furnace is: 512186800, and the inlet width is: 1231mm. The monitoring and display method is the same for other times.

[0372] This invention discloses a method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace. Through a multi-source data acquisition and processing module, it acquires L3 plan data, cold-rolling mill strip shape data, annealing mill external data, and annealing furnace internal data in real time via network interconnection, completing the static and dynamic positioning of data in different areas of the strip. Then, according to the production plan sequence, it analyzes the strip shape data and provides early warnings for planned strip deviation based on strip shape characteristics. Finally, it tracks the operation of the strip in the furnace, displaying information such as strip shape, external CPC (Cost Per Scale), and internal tension, speed, and internal CPC in the strip zones, enabling all strip zones to be in the same position and at the same time. The invention integrates and monitors key information; the strip shape characteristics, in addition to flatness, also take into account the tension difference between the two sides of the strip, fully considering influencing factors; when calculating multiple characteristic values, a weighted average is used to make the values ​​more accurate; in summary, the present invention provides a method for monitoring and early warning of strip deviation in a cold-rolled continuous annealing furnace. Based on cross-process information integration, it realizes early warning of strip deviation and integrated monitoring of strip cross-process data at the same time and location, which is conducive to improving the stability of strip operation in cold-rolled annealing furnaces, minimizing the occurrence of strip edge rubbing and scratching in the furnace, and improving the operational stability of the unit.

Claims

1. A method for monitoring and early warning of strip deviation in a cold rolling continuous annealing furnace, characterized in that It comprises the following steps: S1: Obtain the strip shape data of L1, L2 and L3 computers through the set multi-source data acquisition module; S2: Extract the characteristics of the strip shape information according to the obtained strip shape data; S3: Determine the extracted characteristics according to the set characteristic extraction rule, and give the early warning result according to the determination result, The characteristics of the strip shape information are composed of two dimensions of the flatness measurement value of the strip surface and the strip tension difference value corresponding to the measurement position; An overall strip shape channel is established, which is composed of a strip shape data channel, a strip length channel and a strip shape tension difference channel; The strip shape data channel is used to store the collected flatness time series data; The strip shape data, strip length data and tension difference data are collected and stored at the same time.

2. The cold rolling continuous annealing furnace strip deviation monitoring and early warning method according to claim 1, characterized in that: Along the length direction of the strip, equidistantly scattered tracking blocks are also arranged, and the furnace position of each tracking block at each time point on the time sequence is determined, so that the furnace information of each time point on the time sequence is assigned to each tracking block, thereby establishing the monitoring of the strip information on the time sequence at each tracking block position and the monitoring of the strip information at the same time on the length direction of the strip at each tracking block position.

3. The cold rolling continuous annealing furnace strip deviation monitoring and early warning method according to claim 1, characterized in that: The strip shape data is distributed symmetrically along the width direction of the strip from the driving side to the operating side to form six regions, which are the driving side edge, the driving side 1 / 4 part, the driving side middle part, the operating side middle part, the operating side 1 / 4 part and the operating side edge.

4. The cold rolling continuous annealing furnace strip deviation monitoring and early warning method according to claim 3, characterized in that: According to the actual occupation of the channel, the correspondence between the measurement channel ID and the actual regions is calculated to determine the channel ID value of each region, and the establishment of the wave shape region and channel ID dictionary is completed accordingly; and according to the established wave shape region and channel ID dictionary, the six characteristic values at each meter position in the length direction of the strip are calculated to form a characteristic value matrix.

5. The cold rolling continuous annealing furnace strip deviation monitoring and early warning method according to claim 4, characterized in that: The establishment of the wave shape region and channel ID dictionary is completed according to the following steps: SS1: Determine the number of effective channels; SS2: Calculate the strip calculation width according to the total number of channels, the number of effective channels, the width of the edge channel and the width of the middle channel; SS3: Calculate the middle wave calculation width according to the strip calculation width, the width of the edge channel and the number of edge channels; SS4: Calculate the number of middle wave channels according to the middle wave calculation width, the width of the middle channel and the total number of channels; SS5: Determine the number of 1 / 4 wave channels according to the total number of channels, the number of edge channels and the number of middle wave channels.

6. The cold rolling continuous annealing furnace strip deviation monitoring and early warning method according to claim 5, characterized in that: In step SS1, a determination of whether a channel is a valid channel is made by comparing the count of values other than 0 at the corresponding channel collection point with a set threshold value; When the count of values other than 0 at the corresponding channel collection point is greater than or equal to the set threshold value, the channel is determined to be a valid channel, otherwise, the channel is determined to be an invalid channel.

7. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 5, characterized in that: Step SS2 is specifically determined according to the following formula: Wherein, width cal : strip calculated width, unit: mm; N: the total number of channels; width of the edge channel, in mm; width of the middle passage, unit: mm; L effective : number of valid channels.

8. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 5, characterized in that: Step SS3 is specifically determined according to the following formula: Wherein, midwidth cal : midwidth, unit: mm ceil: rounding up; width cal : strip calculated width, unit: mm; width of the edge channel, in mm; N edge : Number of edge channels.

9. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 5, characterized in that: When the ratio of the middle wave calculation width to the total number of channels is greater than the set width of the middle channel, step SS4 is performed according to the following formula: Otherwise, according to the following formula: N mid : number of middle passages; midwidth cal : midwidth, unit: mm width of the middle passage, unit: mm; N: the total number of channels.

10. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 5, characterized in that: Step SS5 is specifically determined according to the following formula: N qua = N - N edge - N mid , Wherein, N qua : 1 / 4 of the number of passages; N: the total number of channels; N edge : number of edge channels N mid : number of middle passages.

11. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 4, characterized in that: The strip is divided into a head, a middle and a tail along the length direction of the strip, In each of the head, tail and middle regions, feature extraction rules for plate shape characteristics are established according to five factors of tension difference, middle wave area ratio, flatness average, edge wave flatness difference and middle wave flatness difference, and the determination of the plate shape characteristics is completed according to the established feature extraction rules.

12. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 4, characterized in that: The "calculating 6 characteristic values at each meter position in the length direction of the strip to form a characteristic value matrix" is specifically to calculate the weighted average of the flatness of each wave region in each region, and the weighted average result in each region represents the flatness of the corresponding region.

13. The strip deviation monitoring and early warning method of the cold rolling continuous annealing furnace according to claim 11, characterized in that: A first rule is set according to the tension difference: the average of the tension difference between the driving side and the operating side is greater than or equal to 0; A second rule is set according to the middle wave area ratio: the ratio of the number of flatness values greater than or equal to 0 in the middle wave cross section to the total number of flatness values in the middle wave cross section is less than 1 / 2; A third rule is set according to the flatness average: the average of the flatness of the cross section is less than or equal to a set flatness threshold value; The following fourth to sixth rules are set according to the edge wave flatness difference and the middle wave flatness difference: Rule four, the edge wave difference between the driving side and the operating side is less than or equal to a set edge wave difference threshold value, and the middle wave difference between the driving side and the operating side is less than or equal to a set middle wave difference threshold value; Rule five, the edge wave difference between the driving side and the operating side is greater than or equal to a set threshold value of the edge wave difference, the middle wave difference between the driving side and the operating side is greater than or equal to a set threshold value of the middle wave difference, and the average value of the flatness of the cross section is less than or equal to a set average value of the flatness of the cross section; Rule six, the absolute value of the edge wave difference between the driving side and the operating side is less than or equal to a set threshold value of the absolute value of the edge wave difference, and the middle wave difference between the driving side and the operating side is less than or equal to a set threshold value of the middle wave difference; When the rule one, the rule two and the rule three are simultaneously satisfied, and any one of the rule four to the rule six is satisfied, it is determined that the roll is a qualified roll, otherwise it is determined to be an unqualified roll, and a warning is issued.

14. The cold rolling continuous annealing furnace strip deviation monitoring and warning method according to claim 2, characterized in that: The "assigning the furnace information of each time point in the time sequence to each tracking block by determining the position of each tracking block in the furnace at each time point in the time sequence" is specifically: First, calculate the real-time furnace position of each tracking block, then calculate the relative position relationship between each tracking block and the functional area, and finally store the real-time information of the tracking block in the area to which the tracking block belongs in the tracking block.

15. The cold rolling continuous annealing furnace strip deviation monitoring and warning method according to claim 14, characterized in that: The real-time furnace position of each tracking block is calculated, specifically: First, generate the sequence number of the tracking block along the length direction of each roll; Second, set the annealing furnace entrance position as the origin, and determine the distance between each tracking block and the origin according to the sequence number of the tracking block, i.e. the length of the tracking block; Finally, determine the real-time furnace position of each tracking block by subtracting the distance between each tracking block and the origin from the distance between the strip head and the origin.

16. The cold rolling continuous annealing furnace strip deviation monitoring and warning method according to claim 14, characterized in that: The relative position relationship between each tracking block and the functional area is calculated, specifically including the following steps: SA1: Set the annealing furnace entrance position as the origin, and calculate the difference between the real-time furnace position of each tracking block and the distance of each functional area from the origin; SA2: Determine whether the difference between the real-time furnace position of each tracking block and the distance of each functional area from the origin is greater than or equal to 0 and less than or equal to the sum of the length of each functional area and the length of each tracking block; if it is true, it is determined that the tracking block is in the functional area, otherwise it is determined that it is not.

17. The cold rolling continuous annealing furnace strip deviation monitoring and warning method according to claim 16, characterized in that: The real-time furnace position of each tracking block is calculated, specifically: First, generate the sequence number of the tracking block along the length direction of each roll; Second, determine the distance between each tracking block and the origin according to the sequence number of the tracking block, i.e. the length of the tracking block; Finally, determine the real-time furnace position of each tracking block by subtracting the distance between each tracking block and the origin from the distance between the strip head and the origin.

18. The cold rolling continuous annealing furnace strip deviation monitoring and warning method according to claim 2, characterized in that: Each tracking block further contains the planned information of the corresponding strip and the shape information of the corresponding strip.

19. The strip deviation monitoring and early warning method for a cold rolling continuous annealing furnace according to claim 18, characterized in that: the flatness measurement value of the strip surface and the strip tension difference value corresponding to the measurement position of the strip on both sides of the two-dimensional features constitute the flatness information of the strip included in each tracking block, the strip tension difference value corresponding to the measurement position is directly assigned; the flatness measurement value of the strip surface is assigned according to the following steps: first, establish a flatness data channel for storing the collected flatness time series data; and along the width direction of the strip, from the drive side to the operation side, form a symmetric distribution of the flatness data of the six regions of the drive side edge, the drive side 1 / 4, the drive side middle, the operation side middle, the operation side 1 / 4 and the operation side edge; then, according to the actual occupation of the channel, the correspondence between the measurement channel ID and the actual regions is calculated to determine the channel ID value of each region, and the establishment of the waviness region and channel ID dictionary is completed accordingly; and according to the established waviness region and channel ID dictionary, the six characteristic values at each meter position in the length direction of the strip are calculated to form a characteristic value matrix; finally, the weighted average calculation of the characteristic values of each region in the corresponding length of each tracking block is carried out, and the calculation results represent the characteristic values in each region.

Images