Continuous mill control method and device, electronic equipment and storage medium

Abstract

The invention provides a continuous mill control method and device, electronic equipment and a storage medium, and belongs to the technical field of continuous mill control. The method comprises the steps that nonlinear transformation is conducted on rolled strip steel parameters, and intermediate variables corresponding to strip steel waveform curve coefficients are obtained; the rolled strip steel parameters comprise strip steel quality parameters, strip steel initial specification parameters, rolling process parameters and strip shape defect parameters; performing fuzzy processing based on the plate shape defect parameter and the intermediate variable to obtain an adjustment amount corresponding to the intermediate variable; obtaining a strip steel waveform curve coefficient based on the initial strip steel waveform curve coefficient and the adjustment amount corresponding to the intermediate variable; generating a target strip steel waveform curve based on the strip steel waveform curve coefficient; and the reduction rate and the roll shifting amount of the continuous mill are controlled based on the target strip steel waveform curve. According to the continuous mill control method and device, the electronic equipment and the storage medium, the qualified rate of strip steel can be increased.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of continuous rolling mill control, and more specifically, relates to a continuous rolling mill control method and device, an electronic device, and a storage medium. Background Art

[0002] During the production process of ultra-low carbon steel (such as IF steel) thin gauge products (thickness ≤ 0.5 mm, width ≥ 1200 mm) in an acid rolling continuous rolling mill unit, due to the low deformation resistance of the material and large fluctuations in rolling tension, periodic or continuous single-sided wave defects are likely to form at the strip edges. In traditional processes, the roll reduction distribution and shift roll amount setting of each stand do not fully consider the coordinated control of the extension difference between the edge and the middle, resulting in the difficulty of completely eradicating the single-sided wave problem. Existing technologies usually adopt the roll reduction and shift roll amount automatically issued by the model, which cannot meet the requirements of the dynamically changing stress distribution in ultra-thin gauge rolling, affecting subsequent processing and the qualified product rate. Summary of the Invention

[0003] The purpose of the present disclosure is to provide a continuous rolling mill control method and device, an electronic device, and a storage medium to improve the qualified product rate of the strip.

[0004] In the first aspect of the embodiments of the present disclosure, a continuous rolling mill control method is provided, including: Performing a non-linear transformation on the rolling strip parameters to obtain an intermediate variable corresponding to the strip waveform curve coefficient, where the rolling strip parameters include strip material parameters, strip initial specification parameters, rolling process parameters, and shape defect parameters, and the strip waveform curve coefficient includes an edge shape correction coefficient, a middle shape control coefficient, and a shape tilt correction coefficient; Performing a fuzzification process based on the shape defect parameters and the intermediate variable to obtain an adjustment amount corresponding to the intermediate variable; Obtaining the strip waveform curve coefficient based on the initial strip waveform curve coefficient and the adjustment amount corresponding to the intermediate variable; Generating a target strip waveform curve based on the strip waveform curve coefficient; Controlling the roll reduction and shift roll amount of the continuous rolling mill based on the target strip waveform curve.

[0005] In the second aspect of the embodiments of the present disclosure, a continuous rolling mill control device is provided, including: A non-linear transformation module for performing a non-linear transformation on the rolling strip parameters to obtain an intermediate variable corresponding to the strip waveform curve coefficient, where the rolling strip parameters include strip material parameters, strip initial specification parameters, rolling process parameters, and shape defect parameters, and the strip waveform curve coefficient includes an edge shape correction coefficient, a middle shape control coefficient, and a shape tilt correction coefficient; The blurring processing module is configured to perform blurring processing based on the shape defect parameters and intermediate variables to obtain an adjustment amount corresponding to the intermediate variables; The coefficient generation module is configured to obtain strip waveform curve coefficients based on the initial strip waveform curve coefficients and the adjustment amount corresponding to the intermediate variables; The curve generation module is configured to generate a target strip waveform curve based on the strip waveform curve coefficients; The control module is configured to control the reduction rate and roll shifting amount of the continuous rolling mill based on the target strip waveform curve.

[0006] In a third aspect of the embodiments of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above continuous rolling mill control method are implemented.

[0007] In a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above continuous rolling mill control method are implemented.

[0008] The beneficial effects of the continuous rolling mill control method, device, electronic device, and storage medium provided by the embodiments of the present disclosure are as follows: In the embodiments of the present disclosure, first, non-linear transformation is performed on various rolling strip parameters including material, specifications, etc., deeply mining complex relationships, and accurately obtaining intermediate variables. The blurring processing can flexibly handle the uncertainties of shape defects and intermediate variables. Dynamically adjusting the strip waveform curve coefficients and generating a target curve make the control of the strip shape more in line with actual requirements. Controlling the reduction rate and roll shifting amount based on the target curve can consider the differences in edge and middle elongation in a coordinated manner, adapt to dynamic stress changes, effectively eliminate the single-sided wave defect, improve the strip size accuracy and strip shape quality, and further improve the subsequent processing performance and the qualified product rate of the product. Description of the Drawings

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0010] Figure 1 It is a schematic flowchart of a continuous rolling mill control method provided by an embodiment of the present disclosure; Figure 2 It is a structural block diagram of a continuous rolling mill control device provided by an embodiment of the present disclosure; Figure 3 It is a schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Detailed implementation manners

[0011] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present disclosure. However, those skilled in the art should clearly understand that the present disclosure can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present disclosure.

[0012] To make the objectives, technical solutions, and advantages of the present disclosure clearer, the following will be described through specific embodiments with reference to the accompanying drawings.

[0013] Please refer to Figure 1 , Figure 1 , which is a schematic flowchart of a continuous rolling mill control method provided by an embodiment of the present disclosure. The method includes: S101: Perform a non-linear transformation on the rolling strip parameters to obtain intermediate variables corresponding to the strip waveform curve coefficients. The rolling strip parameters include strip material parameters, initial strip specification parameters, rolling process parameters, and shape defect parameters. The strip waveform curve coefficients include edge shape correction coefficients, middle shape control coefficients, and shape tilt correction coefficients.

[0014] In this embodiment, the rolling strip parameters of the strip to be rolled can be collected as input data, which may include: Strip material parameters: such as elastic modulus, yield strength; Initial strip specification parameters: such as initial strip thickness, strip width; Rolling process parameter P: such as rolling force F i (i = 1, 2,..., n, where n is the number of stands), rolling speed; Shape defect parameters: height of single-sided wave, amplitude of middle wave, shape tilt.

[0015] Perform a normalization process on the rolling strip parameters to obtain standard rolling strip parameters and map them to the [0, 1] interval. For example, for a certain parameter x, the normalization formula is:

[0016] Among them, and are the maximum and minimum values of this parameter respectively.

[0017] The standard rolling strip parameters can be input into a neural network to perform a non-linear transformation on the standard rolling strip parameters to obtain intermediate variables corresponding to the strip waveform curve coefficients.

[0018] The purpose of the non - linear transformation is to explore the complex non - linear relationship between the input parameters and the strip waveform curve coefficients.

[0019] Since there is no simple linear correlation between the standard rolling strip parameters and the strip waveform curve coefficients, non - linear transformation is needed to handle this. The purpose of the non - linear transformation is to obtain the non - linear relationship between the standard rolling strip parameters and the strip waveform curve coefficients.

[0020] For example, through the calculation of multiple - layer neurons and non - linear activation functions in a neural network, it can automatically learn the characteristics and patterns of the input standard rolling strip parameters and convert them into intermediate variables related to the strip waveform curve coefficients. The intermediate variables contain valuable information for strip shape control.

[0021] S102: Based on the shape defect parameters and the intermediate variables, perform fuzzification to obtain the adjustment amount corresponding to the intermediate variables.

[0022] In this embodiment, the values of the shape defect parameters and the intermediate variables have a certain degree of fuzziness and cannot be defined by precise numerical values. For example, the severity of the single - side wave height can be described by fuzzy concepts such as "slight", "medium", "severe", etc. Therefore, fuzzification can better handle this uncertainty.

[0023] The fuzzification process includes: Divide the fuzzy sets, and divide the shape defect parameters and the intermediate variables into different fuzzy sets.

[0024] For example, divide the single - side wave height into three fuzzy sets: "low", "medium", and "high". Each set corresponds to a membership function, which is used to determine the degree to which the input value belongs to this set.

[0025] Formulate fuzzy rules, and fuzzy rules can be formulated according to expert experience.

[0026] For example, if the single - side wave height is "high" and a certain intermediate variable is within a specific range, the adjustment amount of this intermediate variable can be "a large positive value".

[0027] Fuzzy inference: According to the input shape defect parameters and intermediate variables, calculate the membership degrees of the shape defect parameters and intermediate variables belonging to each fuzzy set through the membership function, and then perform inference based on the fuzzy rules to obtain the fuzzy output of the adjustment amount corresponding to the intermediate variables.

[0028] Defuzzification: Convert the fuzzy output into an exact adjustment amount.

[0029] S103: Based on the initial strip waveform curve coefficients and the adjustment amount corresponding to the intermediate variables, obtain the strip waveform curve coefficients.

[0030] In this embodiment, the initial strip waveform curve coefficient can be obtained based on experience or preliminary setting. The initial coefficient is corrected by the obtained adjustment amount of the intermediate variable.

[0031] For example, the edge shape correction coefficient , where is the initial edge shape correction coefficient, is the adjustment amount corresponding to the intermediate variable. In this way, the waveform curve coefficient can be dynamically adjusted according to the actual rolling situation and shape defects.

[0032] S104: Generate the target strip waveform curve based on the strip waveform curve coefficient.

[0033] In this embodiment, the strip waveform curve coefficient determines the thickness deviation or shape change law of the strip at different lateral positions. Using the strip waveform curve coefficient and combining with the cubic polynomial model , the target strip waveform curve can be generated. The target strip waveform curve can visually represent the desired strip shape state.

[0034] where a, b, and c are the edge shape correction coefficient, the middle shape control coefficient, and the shape tilt correction coefficient respectively, x is the lateral position coordinate of the target strip, and y is the thickness deviation of the target strip.

[0035] S105: Control the reduction rate and the roll shift amount of the continuous rolling mill based on the target strip waveform curve.

[0036] In this embodiment, the reduction rate refers to the ratio of the reduction amount of the strip thickness during rolling to the strip thickness before rolling, and it is an important parameter for controlling the strip thickness and shape. The roll shift amount refers to the movement amount of the roll in the axial direction.

[0037] If the target strip waveform curve shows that the overall thickness deviation of the strip is large, the reduction rate of the overall stand needs to be adjusted; if only the thickness of a local area does not meet the requirements, the reduction rate of the local stand is adjusted.

[0038] For example, if the edge thickness of the strip is too thick, the reduction rate of the corresponding stand at the edge can be increased to further compress and thin the edge.

[0039] The roll shift amount is used to adjust the pressure distribution in the contact area between the roll and the strip and improve the shape. When the target strip waveform curve shows the existence of single-sided waves or local shape defects, by adjusting the roll shift amount and changing the axial position of the roll, the concentrated stress at the edge is diffused towards the middle to eliminate the shape defects.

[0040] It can be concluded from the above that in this embodiment, non-linear transformation is first performed on various rolling strip parameters including material, specifications, etc., deeply mining complex relationships, and accurately obtaining intermediate variables. The fuzzy processing can flexibly handle the uncertainties of strip shape defects and intermediate variables. Dynamically adjusting the strip waveform curve coefficients and generating a target curve can make the control of strip shape more in line with actual requirements. Based on the target curve, adjusting the reduction rate and the roll shift amount can consider the differences in elongation between the edge and the middle in a coordinated manner, adapt to the dynamic changes of stress, effectively eliminate the single-sided wave defect, improve the dimensional accuracy and strip shape quality of the strip, and further improve the subsequent processing performance and the qualified rate of products.

[0041] In an embodiment of the present disclosure, non-linear transformation is performed on rolling strip parameters to obtain intermediate variables corresponding to strip waveform curve coefficients, including: Performing non-linear transformation on rolling strip parameters to obtain parameter characteristics corresponding to the rolling strip parameters; Mapping the parameter characteristics corresponding to the rolling strip parameters to intermediate variables corresponding to the strip waveform curve coefficients.

[0042] In this embodiment, non-linear transformation can be performed on rolling strip parameters through a neural network to obtain intermediate variables corresponding to strip waveform curve coefficients.

[0043] Taking the neural network as an example, the neural network can automatically learn the complex patterns and characteristics of input parameters through the calculations of multiple layers of neurons and non-linear activation functions.

[0044] For example, when processing rolling force and strip material parameters, the neural network can discover the unique deformation characteristics of strips with different materials under different rolling forces, and convert these original rolling strip parameters into more representative and distinguishable parameter characteristics. The parameter characteristics contain information that affects the strip waveform curve coefficients.

[0045] In this embodiment, after obtaining the parameter characteristics corresponding to the rolling strip parameters, it is necessary to establish a connection between them and the strip waveform curve coefficients. The strip waveform curve coefficients include an edge shape correction coefficient, a middle shape control coefficient, and a shape tilt correction coefficient, and these coefficients directly affect the final strip shape. Through a specific mapping algorithm or model, the parameter characteristics are converted into intermediate variables corresponding to these coefficients. This mapping process can be regarded as a conversion and transmission of information, converting the feature information extracted from the original parameters into intermediate information related to shape control.

[0046] The intermediate variables synthesize the information related to shape control in the parameter characteristics.

[0047] For example, a certain intermediate variable reflects the degree of association between the edge deformation of the strip and the target edge shape. By adjusting this intermediate variable, the edge shape correction coefficient can be accurately corrected, thereby achieving precise control of the strip shape.

[0048] In an embodiment of the present disclosure, a non-linear transformation is performed on the rolling strip parameters to obtain parameter features corresponding to the rolling strip parameters, including: Performing a non-linear transformation on the rolling strip parameters based on an adaptive neural network; The non-linear transformation is expressed as:

[0049]

[0050] where, represents the input value of the th neuron in the th layer, represents the number of layers of the neural network, represents the th neuron in , represents the output value of the th neuron in the th layer, represents the weight of the connection between the th neuron in the th layer and the th neuron in the th layer, represents the bias of the th neuron in the th layer, represents the number of neurons in the th layer, represents the output value of the th neuron in the th layer.

[0051] In this embodiment, the adaptive neural network is a multi-layer structure, and each layer contains multiple neurons. The number of layers is represented by , and there are multiple neurons in the th layer, and the th neuron is identified by ; there are also multiple neurons in the th layer, and the th neuron is identified by . The rolling strip parameters enter the first layer (input layer) of the neural network as input data, and after multi-layer calculations and processing, the parameter features corresponding to the rolling strip parameters are obtained from the output layer.

[0052] For the th neuron in the th layer, its input value is obtained by The outputs of all neurons in the layer are weighted summed, plus the bias of the neuron The calculation formula is:

[0053] Weight Indicates Tier The neuron pairs Tier Different weight values ​​determine the intensity and direction of information transmission. By adjusting the weights, the neural network can learn the complex relationship between different features in the input data.

[0054] Bias An adjustable threshold is provided for neurons. Even if the input is zero, the neuron may produce output due to the existence of bias, which increases the flexibility and expressiveness of the model.

[0055] In this embodiment, a nonlinear transformation can be performed by using a ReLU activation function, namely:

[0056] The relationship between the parameters of rolled strip is nonlinear, and linear models cannot accurately describe these complex relationships. The ReLU activation function sets input values ​​less than zero to zero and keeps input values ​​greater than zero unchanged, introducing nonlinear characteristics to the neural network, allowing the neural network to learn more complex patterns and features.

[0057] Through nonlinear transformation, neurons can filter and transform input information, highlight important features, and suppress irrelevant information. As the number of neural network layers increases, neurons in each layer continuously extract and abstract features, and finally obtain parameter features that can reflect the essential characteristics of rolled strip parameters.

[0058] It can be concluded from the above that the adaptive neural network in this embodiment can effectively mine the complex relationship between parameters by performing multi-layer weighted summation, nonlinear transformation and adaptive adjustment on the rolled strip parameters, thereby obtaining representative parameter characteristics.

[0059] In one embodiment of the present disclosure, fuzzy processing is performed based on the plate shape defect parameter and the intermediate variable to obtain the adjustment amount corresponding to the intermediate variable, including: Determine the membership degree corresponding to a plurality of fuzzy sets, wherein the plurality of fuzzy sets are fuzzy sets divided based on the plate shape defect parameters and the intermediate variables respectively; The fuzzy reasoning is performed on the membership degrees and fuzzy rules corresponding to multiple fuzzy sets to obtain the adjustment amount corresponding to the intermediate variable.

[0060] In this embodiment, the shape defect parameters include the height of one-sided wave, the amplitude of center wave, and the shape inclination.

[0061] Since it is difficult to define the values of the shape defect parameters and the intermediate variables with precise numerical values, it is necessary to divide the shape defect parameters and the intermediate variables into multiple fuzzy sets respectively.

[0062] For example, the height of one-sided wave can be divided into three fuzzy sets: "low", "medium", and "high"; the intermediate variables are also divided into different fuzzy sets according to their value ranges and impacts on shape control. This can better describe the uncertainty and fuzziness of the parameters.

[0063] In this embodiment, for each fuzzy set, it is necessary to determine the degree to which the shape defect parameter and the intermediate variable belong to this set, that is, the membership degree.

[0064] Exemplarily, the input variables are divided into different fuzzy sets, for example: Height of one-sided wave : Divided into three fuzzy sets: "low" (L), "medium" (M), and "high" (H); Amplitude of center wave : Divided into three fuzzy sets: "small" (S), "medium" (M), and "large" (B); Shape inclination : Divided into three fuzzy sets: "slight" (S), "medium" (M), and "severe" (H). Similarly, the intermediate variables are also divided into different fuzzy sets respectively. The intermediate variables can be represented by , and respectively.

[0065] The membership function (such as the triangular membership function) can be used to determine the membership degrees of the input variables belonging to each fuzzy set. For example, for the height of one-sided wave The membership degree belonging to the "low" fuzzy set can be expressed as:

[0066] where and are the boundary values of the "low" fuzzy set.

[0067] In this embodiment, a series of fuzzy rules can be formulated according to expert experience. For example: Rule 1: If is "high" and is "small", then is "a large negative value"; Rule 2: If is "large" and is "medium", then is a "large positive value"; Rule 3: If is "severe" and is "large", then is a "large negative value".

[0068] According to the membership degrees of the input variables and the fuzzy rules, the fuzzy output is obtained through fuzzy inference. For example, for Rule 1, its activation degree is:

[0069] wherein, represents the minimum operation, represents the height of the single-sided wave belongs to the membership degree of the "high" fuzzy set, represents the intermediate variable belongs to the membership degree of the fuzzy set S.

[0070] Then, according to the conclusion part of the rule, the membership function of the fuzzy output is obtained. Aggregate the fuzzy outputs of all rules to obtain the membership function of the final fuzzy output , , .

[0071] Use a defuzzification method (such as the centroid method) to convert the fuzzy output into an exact value.

[0072] For example, for , its exact value is:

[0073] wherein, is 's membership function.

[0074] In this embodiment, the initial strip waveform curve coefficients can be defined as , and .

[0075] Then the strip waveform curve coefficient is expressed as:

[0076]

[0077]

[0078] It can be concluded from the above that this embodiment can effectively cope with the uncertainties of the shape defect parameters and the values of intermediate variables. By dividing the fuzzy sets and determining the membership degrees, the parameter characteristics can be obtained. Reasoning based on the fuzzy rules formulated according to expert experience can comprehensively consider the complex relationships among various factors. The result obtained by fuzzy reasoning is converted into an exact value through defuzzification, which can dynamically adjust the strip waveform curve coefficient. This helps the continuous rolling mill to adjust the control parameters in a timely and accurate manner according to the shape condition of the strip, thereby improving the shape quality of the strip, reducing shape defects, and enhancing the qualified rate and production efficiency of the product.

[0079] In an embodiment of the present disclosure, the continuous rolling mill control method further includes: Based on the thickness change rate of the target strip waveform curve along the length direction of the target strip, determine the roll reduction adjustment strategy of the continuous rolling mill, where the target strip is the strip after being rolled by the continuous rolling mill; The roll reduction adjustment strategy of the continuous rolling mill includes overall stand roll reduction adjustment and local stand roll reduction adjustment.

[0080] In this embodiment, the target strip waveform curve reflects the ideal shape of the strip after being rolled by the continuous rolling mill, and the thickness change rate along its length direction reflects the change trend and degree of the strip thickness in the length direction. Different thickness change rates mean different deformation requirements at different positions of the strip during the rolling process.

[0081] When the target strip waveform curve shows that the thickness change along the length direction is relatively uniform, that is, the thickness change rate is relatively stable throughout the length of the strip and there is an overall thickness deviation, it is suitable to adopt the overall stand roll reduction adjustment strategy.

[0082] For example, if the overall thickness of the strip is thicker than the target value and the thickness change rate is relatively consistent, the roll reduction rates of all stands can be increased, so that the strip is subjected to a greater compressive force at each stand, thereby reducing the overall thickness of the strip and making it close to the target thickness.

[0083] The overall stand roll reduction adjustment can comprehensively control the thickness of the strip and ensure that the overall thickness of the strip meets the target requirements. By synchronously adjusting the roll reduction rates of all stands, the stability of the strip during the rolling process can be maintained, and the shape problems caused by excessive local adjustment can be avoided.

[0084] If the target strip waveform curve shows that there are obvious differences in the thickness change rate in the length direction of the strip, that is, the thickness changes abnormally in some local areas, such as sudden increases or decreases in thickness, then the local stand roll reduction adjustment strategy needs to be adopted. For example, if the thickness change rate is large within a certain length of the strip and it is thicker than the target thickness, the roll reduction rate of the local stand corresponding to this area can be increased specifically, so that this part of the strip is compressed more, thereby correcting the local thickness deviation.

[0085] The adjustment of the local mill reduction rate can accurately handle the local thickness problems of the strip steel, improving the thickness accuracy and shape quality of the strip steel. By adjusting the reduction rate of specific stands, it is possible to more flexibly adapt to the deformation requirements of the strip steel at different positions, avoiding unnecessary impacts on the overall strip steel quality.

[0086] From the above, it can be concluded that in this embodiment, by analyzing the thickness change rate of the target strip steel waveform curve along the length direction, the uniformity and local characteristics of the strip steel thickness change are judged, and then a suitable overall or local mill reduction rate adjustment strategy is selected to achieve precise control of the strip steel thickness and improve the rolling quality of the strip steel.

[0087] In an embodiment of the present disclosure, the continuous rolling mill control method further includes: In response to the reduction rate adjustment strategy of the continuous rolling mill being the overall stand reduction rate adjustment, determining the average thickness of the target strip steel based on the thickness change rate; Determining the adjustment amount of the overall stand reduction rate based on the deviation between the average thickness of the target strip steel and the ideal thickness; In response to the reduction rate adjustment strategy of the continuous rolling mill being the local stand reduction rate adjustment, determining the adjustment amount of each local stand reduction rate based on the correlation model; the correlation model is the correlation model between the local area thickness deviation of the target strip steel and the reduction rate adjustment amount.

[0088] In this embodiment, when the reduction rate adjustment strategy of the continuous rolling mill is determined to be the overall stand reduction rate adjustment, it indicates that the thickness change of the target strip steel in the length direction is relatively uniform, and there is an overall thickness deviation. Therefore, based on the thickness change rate of the target strip steel waveform curve along the length direction, the thickness of the entire target strip steel is comprehensively calculated to obtain the average thickness of the target strip steel.

[0089] For example, assume the length of the target strip steel is L, and a coordinate axis is established along the length direction of the strip steel , at the position the thickness corresponding to the target strip steel waveform curve is .

[0090] The thickness data of the target strip steel waveform curve in the length direction can be integrated and then divided by the strip steel length to obtain the average thickness . This can eliminate the influence of local thickness fluctuations and obtain the overall thickness information. Its calculation formula can be expressed as:

[0091] In practical applications, since a continuous function cannot be obtained, at this time, the strip steel length can be discretized. Assume that the strip steel length is equally divided into n small segments, each segment length is , and the midpoint position of each small segment is , the corresponding thickness measurement value is , and the average thickness can be calculated by the rectangle method. The calculation formula of the rectangle method is:

[0092] Compare the average thickness with the ideal thickness and calculate the deviation between the two. The ideal thickness is the strip thickness value that meets the product quality requirements and is preset.

[0093] Determine the adjustment amount of the overall mill reduction ratio according to this deviation. The greater the deviation, the greater the reduction ratio that needs to be adjusted, so that the average thickness of the strip is closer to the ideal thickness.

[0094] For example, if the average thickness is greater than the ideal thickness, it is necessary to increase the reduction ratio of the overall mill to further compress and thin the strip; conversely, if the average thickness is less than the ideal thickness, it is necessary to decrease the reduction ratio of the overall mill.

[0095] In this embodiment, the correlation model is the correlation model between the thickness deviation of the local area of the target strip and the adjustment amount of the reduction ratio. In actual production, through a large amount of experimental data and historical rolling records, statistical analysis can be carried out on different local area thickness deviation situations and the corresponding adjustment amounts of the reduction ratio, and a relationship model between the two can be established using mathematical methods (such as regression analysis, machine learning algorithms, etc.). The correlation model reflects the magnitude of the reduction ratio adjustment required to make the thickness of the area reach the ideal state under different local area thickness deviations.

[0096] When the reduction ratio adjustment strategy of the tandem mill is the adjustment of the local mill reduction ratio, it indicates that the thickness change of the target strip in the length direction is uneven and there are local thickness deviations. Therefore, the target strip is divided into multiple local areas, and the thickness deviation of each local area (that is, the difference between the actual thickness and the ideal thickness of the local area) is calculated.

[0097] Input the thickness deviation of each local area into the established correlation model, and determine the corresponding adjustment amount of the reduction ratio of each local mill according to the output of the model.

[0098] For example, if the thickness of a certain local area is greater than the ideal thickness, the correlation model will output a corresponding adjustment amount of the reduction ratio, indicating that the local mill corresponding to this area increases the reduction ratio to reduce the thickness of this area.

[0099] It can be concluded from the above that in this embodiment, by distinguishing the overall and local mill reduction ratio adjustment strategies and using different methods to determine the adjustment amounts respectively, the precise control of the reduction ratio of the tandem mill is realized, thus meeting the requirements for thickness accuracy and strip shape quality in the strip rolling process.

[0100] In one embodiment of the present disclosure, the continuous rolling mill control method further includes: Determine the wave type of the target strip based on the target strip waveform curve, where the wave type includes middle waves and single-sided waves; Obtain the adjustment amount of the roll shifting amount based on the wave type.

[0101] In this embodiment, the target strip waveform curve visually presents the shape characteristics of the strip. By analyzing the curve, the wave type of the strip can be judged.

[0102] If the target strip waveform curve shows that there are periodic or continuous fluctuations in the thickness or shape of the middle area of the strip, deviating from the ideal flat state, it can be determined as middle waves. Middle waves are caused by inconsistent elongation between the middle and edges of the strip, or uneven stress in the middle during the rolling process.

[0103] When the target strip waveform curve indicates that there are wavy thickness or shape changes only on one side of the strip while the other side is relatively flat, it can be determined as single-sided waves. The generation of single-sided waves is related to factors such as uneven distribution of rolling force and tension on both sides during the rolling process, or inconsistent wear conditions of the rolls.

[0104] In this embodiment, the adjustment of the roll shifting amount can change the distribution of the contact area between the roll and the strip, and further adjust the lateral pressure distribution of the strip. When it is determined as middle waves, if the middle thickness increases (the middle bulges), it means that the elongation in the middle is less than that on both sides. The roll can be shifted to one side to reduce the contact area between the roll and the middle of the strip, make the pressure distribution more uniform, thereby reducing the middle thickness and eliminating the middle waves. The specific adjustment amount of the roll shifting amount needs to be determined according to the severity of the middle waves (i.e., the amplitude of the middle bulge or depression of the target strip waveform curve). The greater the severity, the greater the adjustment amplitude of the roll shifting amount.

[0105] For single-sided waves, if the thickness on one side of the strip increases (one side bulges), the roll is shifted to that side to increase the contact pressure between the roll on that side and the strip, so that the strip on that side is further elongated, thereby reducing the thickness difference with the other side and eliminating the single-sided waves. Similarly, the adjustment amount of the roll shifting amount should be determined according to the severity of the single-sided waves. The higher the severity, the more the roll shifting amount is adjusted.

[0106] It can be concluded from the above that in this embodiment, the wave type of the strip is identified by analyzing the target strip waveform curve, and then the corresponding adjustment amount of the roll shifting amount is determined according to different wave types and their severities to effectively control the wave defects of the strip and improve the shape quality of the strip.

[0107] Exemplarily, for the rolling of ultra-low carbon steel on a five-stand six-high rolling mill in a certain steel plant, in the rolling process of ultra-low carbon steel (C < 0.01%), width 1250 mm, and thickness 0.35 mm, the following key process points were optimized to effectively solve the problem of single-sided wave: I. Reduction ratio distribution As shown in Table 1, the reduction ratio of the F5 stand was increased by 5%, and the reduction ratios of the F1-F4 stands were recalculated and distributed according to the reduction ratio of the five stands. The rolling force of the F5 stand increased by 1000 KN, and the leveling effect was good, effectively eliminating the single-sided wave.

[0108] Table 1 Reduction ratios before and after optimization

[0109] II. Adjusting the shifting amount of intermediate rolls The shifting amount distribution is shown in Table 2. The shifting amount of the F1-F4 stands is 0, that is, in the middle position, reducing the stress at the strip edges in the first 4 stands, thus weakening the edge wave. The F5 stand is automatically adjusted according to the strip shape mill model.

[0110] Table 2 Shifting amounts before and after optimization

[0111] III. Adding an edge compensation mechanism By correcting the strip waveform curve coefficients (values of a, b, c), edge compensation was increased to further achieve the purpose of eliminating the single-sided wave.

[0112] Reducing the value of a can increase the strip edge tension, making the strip edge shape tighter; increasing the value of b can reduce the strip middle tension, making the strip middle shape looser; reducing the value of c can make the strip edge shape on the operating side tighter and the edge on the drive side looser, achieving the purpose of eliminating the single-sided wave on the operating side. The comparison of the values of a, b, c is shown in Table 3, and the specific strip shape curve is as Figure 1 shown.

[0113] Table 3 Waveform curve coefficients before and after optimization

[0114] By increasing the reduction ratio of the F5 stand by 5%, the rolling force increased by 1000 KN, the shifting amount of the F1-F4 stands was 0, the coefficient a: -13, b: -3, c: +3, and the edge wave was greatly optimized, with the edge wave less than 3 mm.

[0115] By redistributing the reduction ratio, adjusting the shifting amount of the intermediate rolls, and adding a strip edge compensation mechanism, the leveling effect of the F5 stand was enhanced, the strip edge wave generation in the F1-F4 stands was weakened, and the strip shape curve adopted the method of being tight on the operating side - loose in the middle - relatively tight on the drive side to jointly optimize the single-sided wave and reduce the wave height to below 3 mm.

[0116] The continuous rolling mill control method corresponding to the above embodiments Figure 2 is a structural block diagram of a continuous rolling mill control device provided by an embodiment of the present disclosure. For ease of illustration, only parts related to the embodiments of the present disclosure are shown. Refer to Figure 2 , the continuous rolling mill control device 20 includes: a non-linear transformation module 21, a fuzzy processing module 22, a coefficient generation module 23, a curve generation module 24, and a control module 25.

[0117] Among them, the non-linear transformation module 21 is used to perform non-linear transformation on the rolling strip parameters to obtain an intermediate variable corresponding to the strip waveform curve coefficient. The rolling strip parameters include strip material parameters, strip initial specification parameters, rolling process parameters, and shape defect parameters. The strip waveform curve coefficients include edge shape correction coefficients, middle shape control coefficients, and shape tilt correction coefficients; The fuzzy processing module 22 is used to perform fuzzy processing based on the shape defect parameters and the intermediate variable to obtain an adjustment amount corresponding to the intermediate variable; The coefficient generation module 23 is used to obtain the strip waveform curve coefficient based on the initial strip waveform curve coefficient and the adjustment amount corresponding to the intermediate variable; The curve generation module 24 is used to generate a target strip waveform curve based on the strip waveform curve coefficient; The control module 25 is used to control the reduction rate and roll shifting amount of the continuous rolling mill based on the target strip waveform curve.

[0118] In an embodiment of the present disclosure, the non-linear transformation module 21 specifically is used for: Performing non-linear transformation on the rolling strip parameters to obtain parameter characteristics corresponding to the rolling strip parameters; Mapping the parameter characteristics corresponding to the rolling strip parameters to an intermediate variable corresponding to the strip waveform curve coefficient.

[0119] In an embodiment of the present disclosure, the non-linear transformation module 21 is specifically further used for: Performing non-linear transformation on the rolling strip parameters based on an adaptive neural network; The non-linear transformation is expressed as:

[0120]

[0121] Among them, represents the input value of the th neuron in the th layer, represents the number of layers of the neural network, represents the th neuron in , represents the The output value of the neuron in the layer, from the neuron in the layer to the weight of the connection between the neuron in the layer, and the bias of the neuron in the layer, where represents the output value of the

[0122] In one embodiment of the present disclosure, the fuzzy processing module 22 is specifically configured to: Determine the membership degrees corresponding to multiple fuzzy sets, where the multiple fuzzy sets are fuzzy sets divided respectively based on strip shape defect parameters and intermediate variables; Perform fuzzy inference on the membership degrees corresponding to the multiple fuzzy sets and the fuzzy rules to obtain the adjustment amount corresponding to the intermediate variable.

[0123] In one embodiment of the present disclosure, the tandem mill control device 20 further includes: a roll reduction rate adjustment module, which is specifically configured to: Based on the thickness change rate of the target strip waveform curve along the length direction of the target strip, determine the roll reduction rate adjustment strategy of the tandem mill, where the target strip is the strip after being rolled by the tandem mill; The roll reduction rate adjustment strategy of the tandem mill includes overall stand roll reduction rate adjustment and local stand roll reduction rate adjustment.

[0124] In one embodiment of the present disclosure, the roll reduction rate adjustment module is further specifically configured to: In response to the roll reduction rate adjustment strategy of the tandem mill being the overall stand roll reduction rate adjustment, determine the average thickness of the target strip based on the thickness change rate; Determine the adjustment amount of the overall stand roll reduction rate based on the deviation between the average thickness of the target strip and the ideal thickness; In response to the roll reduction rate adjustment strategy of the tandem mill being the local stand roll reduction rate adjustment, determine the adjustment amount of the roll reduction rate of each local stand based on the correlation model; the correlation model is the correlation model between the thickness deviation of the local area of the target strip and the adjustment amount of the roll reduction rate.

[0125] In one embodiment of the present disclosure, the tandem mill control device 20 further includes: a roll shift amount adjustment module, which is specifically configured to: Based on the target strip waveform curve, determine the wave type of the target strip, where the wave type includes middle wave and single-sided wave; Obtain the adjustment amount of the roll shift amount based on the wave type.​

[0126] See Figure 3 , Figure 3 which is a schematic block diagram of an electronic device provided by an embodiment of the present disclosure. As Figure 3 shown, the electronic device 300 in this embodiment may include: one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The above-mentioned processors 301, input devices 302, output devices 303, and memories 304 communicate with each other through a communication bus 305. The memory 304 is used to store computer programs, and the computer programs include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. Among them, the processor 301 is configured to call the program instructions to execute the functions of each module in the above device embodiments, for example Figure 2 the functions of the non-linear transformation module 21, the fuzzy processing module 22, the coefficient generation module 23, the curve generation module 24, and the control module 25 shown

[0127] It should be understood that in the embodiments of the present disclosure, the so-called processor 301 may be a central processing unit (CPU), and this processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or this processor may also be any conventional processor, etc.

[0128] The input device 302 may include a touchpad, a fingerprint acquisition sensor (for acquiring the fingerprint information and the direction information of the fingerprint of the user), a microphone, etc., and the output device 303 may include a display (such as an LCD), a speaker, etc.

[0129] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A part of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0130] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure may execute the implementation manners described in the first and second embodiments of the continuous rolling mill control method provided by the embodiments of the present disclosure, and may also execute the implementation manner of the electronic device described in the embodiments of the present disclosure, which will not be elaborated herein.

[0131] In another embodiment of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, all or part of the processes in the method of the foregoing embodiments are implemented. It may also be completed by instructing relevant hardware through the computer program. The computer program may be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the foregoing method embodiments may be implemented. Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0132] The computer-readable storage medium may be an internal storage unit of the electronic device in any of the foregoing embodiments, such as the hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the electronic device. Further, the computer-readable storage medium may also include both the internal storage unit and the external storage device of the electronic device. The computer-readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store the data that has been output or will be output.

[0133] Those of ordinary skill in the art will realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of the examples have been generally described in terms of function in the above description. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this disclosure.

[0134] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described electronic devices and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0135] In several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces or units, or can also be electrical, mechanical, or other forms of connection.

[0136] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of this disclosure.

[0137] In addition, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0138] The above is only the specific implementation manner of this disclosure, but the protection scope of this disclosure is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or substitutions within the technical scope disclosed by this disclosure, and these modifications or substitutions should all be covered by the protection scope of this disclosure. Therefore, the protection scope of this disclosure should be subject to the protection scope of the claims.

Claims

1. A continuous rolling mill control method, characterized in that: include: Performing nonlinear transformation on the rolled steel strip parameters to obtain intermediate variables corresponding to the strip steel waveform curve coefficients, wherein the rolled steel strip parameters include strip steel material parameters, strip steel initial specification parameters, rolling process parameters and plate shape defect parameters, and the strip steel waveform curve coefficients include edge plate shape correction coefficients, middle plate shape control coefficients and plate shape tilt correction coefficients; Performing fuzzy processing based on the plate shape defect parameters and the intermediate variables to obtain an adjustment amount corresponding to the intermediate variables; The strip steel waveform curve coefficient is obtained based on the initial strip steel waveform curve coefficient and the adjustment amount corresponding to the intermediate variable; Generate a target strip steel waveform curve based on the strip steel waveform curve coefficient; The reduction rate and roll shifting amount of the continuous rolling mill are controlled based on the target strip waveform curve.

2. The method for controlling a continuous rolling mill according to claim 1, characterized in that: The nonlinear transformation of the rolled strip parameters to obtain the intermediate variables corresponding to the strip waveform curve coefficients includes: Perform nonlinear transformation on the rolled strip parameters to obtain parameter characteristics corresponding to the rolled strip parameters; The parameter characteristics corresponding to the rolled steel strip parameters are mapped into intermediate variables corresponding to the steel strip waveform curve coefficients.

3. The method for controlling a continuous rolling mill according to claim 2, characterized in that: The nonlinear transformation of the rolled strip parameters to obtain parameter characteristics corresponding to the rolled strip parameters includes: Performing nonlinear transformation on the rolled strip parameters based on an adaptive neural network; The nonlinear transformation is expressed as: in, Indicates Tier The input value of a neuron, represents the number of layers of the neural network, Indicates The neurons, Indicates Tier The output value of a neuron, Indicates Tier Neuron to Tier The weights of the connections between neurons, Indicates Tier The bias of a neuron, Indicates The number of neurons in the layer, Indicates Tier The output value of a neuron.

4. The method for controlling a continuous rolling mill according to claim 1, wherein: The fuzzy processing is performed based on the plate shape defect parameter and the intermediate variable to obtain the adjustment amount corresponding to the intermediate variable, including: Determining membership degrees corresponding to a plurality of fuzzy sets, wherein the plurality of fuzzy sets are fuzzy sets divided based on the plate shape defect parameters and the intermediate variables respectively; The fuzzy reasoning is performed on the membership degrees and fuzzy rules corresponding to multiple fuzzy sets to obtain the adjustment amount corresponding to the intermediate variable.

5. The method for controlling a continuous rolling mill according to claim 1, characterized in that: Also includes: Determining a rolling reduction rate adjustment strategy of a continuous rolling mill based on a thickness change rate of a target steel strip waveform curve along a length direction of the target steel strip, wherein the target steel strip is a steel strip rolled by the continuous rolling mill; The reduction rate adjustment strategy of the continuous rolling mill includes overall stand reduction rate adjustment and local stand reduction rate adjustment.

6. The method for controlling a continuous rolling mill according to claim 5, characterized in that: Also includes: In response to the rolling reduction rate adjustment strategy of the tandem rolling mill being overall stand rolling reduction rate adjustment, determining an average thickness of a target strip based on the thickness variation rate; Determine the adjustment amount of the overall stand reduction rate based on the deviation between the average thickness of the target strip and the ideal thickness; In response to the reduction rate adjustment strategy of the continuous rolling mill being local stand reduction rate adjustment, the adjustment amount of the reduction rate of each local stand is determined based on an association model; the association model is an association model between the local area thickness deviation of the target strip and the reduction rate adjustment amount.

7. The method for controlling a continuous rolling mill according to claim 1, characterized in that: Also includes: Determining a wave type of the target steel strip based on the target steel strip waveform curve, wherein the wave type includes a middle wave and a single-side wave; An adjustment amount of the roll shifting amount is obtained based on the wave type.

8. A continuous rolling mill control device, characterized in that: include: A nonlinear transformation module is used to perform nonlinear transformation on the parameters of the rolled steel strip to obtain intermediate variables corresponding to the strip waveform curve coefficients, wherein the rolled steel strip parameters include the strip material parameters, the strip initial specification parameters, the rolling process parameters and the plate shape defect parameters, and the strip waveform curve coefficients include the edge plate shape correction coefficient, the middle plate shape control coefficient and the plate shape tilt correction coefficient; A fuzzy processing module, used for performing fuzzy processing based on the plate shape defect parameters and the intermediate variables to obtain an adjustment amount corresponding to the intermediate variables; A coefficient generation module, used for obtaining the strip steel waveform curve coefficient based on the initial strip steel waveform curve coefficient and the adjustment amount corresponding to the intermediate variable; A curve generation module, used for generating a target strip steel waveform curve based on a strip steel waveform curve coefficient; A control module is used to control the reduction rate and roll shifting amount of the continuous rolling mill based on the target strip steel waveform curve.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images