Steel rolling parameter optimization method and device, electronic equipment and storage medium
By screening and optimizing steel rolling parameters, the problems of low efficiency and high cost of steel flatness in the prior art have been solved, and efficient and low-cost steel flatness are achieved.
Patent Information
- Application Number
- CN202510301996.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-14
- Publication Date
- 2025-05-06
AI Technical Summary
The prior art is inefficient and expensive in improving the flatness of steel, and relies on increasing the preset distance of roll joints and reverse rolling to explore the best rolling parameters.
By receiving multiple rolling parameters to be screened, their sensitivity is determined, and the target rolling parameters are screened in the order of sensitivity from large to small, and the target rolling parameters are optimized based on the parameter optimization model to obtain the target rolling parameter value.
It realizes efficiently obtaining steel with higher flatness without repeatedly setting rolling parameters and conducting multiple tests, reducing production costs.
Smart Images
Figure CN119940146A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of steel manufacturing, and in particular to a method, device, electronic equipment and storage medium for optimizing steel rolling parameters. Background Art
[0002] Steel flatness is an important quality parameter of steel plates during the rolling process. If the steel flatness is too low, it will lead to uneven stress distribution on the steel surface, stress concentration in local areas, reducing the overall bearing capacity and possibly causing the risk of structural deformation or fracture.
[0003] However, the inventors have found that the existing technologies mostly rely on industrial tests such as increasing the preset distance of the roll gap and controlling the reverse rolling of the plate to explore the optimal rolling parameters in order to obtain higher steel flatness. This method is not only inefficient but also costly. Summary of the invention
[0004] The objectives of the present invention include, for example, providing a method, device, electronic device and storage medium for optimizing steel rolling parameters, which can at least partially solve the above-mentioned technical problems.
[0005] The embodiments of the present invention can be implemented as follows:
[0006] In a first aspect, an embodiment of the present invention provides a method for optimizing steel rolling parameters, the method comprising:
[0007] receiving a plurality of rolling parameters to be screened, wherein the types of the rolling parameters to be screened are different;
[0008] Determining the sensitivity of each rolling parameter to be screened, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the influence of the rolling parameter to be screened on the flatness of the steel;
[0009] Selecting at least one target rolling parameter from the rolling parameters to be selected in descending order of sensitivity;
[0010] Based on the parameter optimization model, the target rolling parameters are optimized to obtain target rolling parameter values corresponding to the target rolling parameters.
[0011] Optionally, determining the sensitivity of each rolling parameter to be screened includes:
[0012] Constructing a function polynomial of each rolling parameter to be screened and the flatness of the steel;
[0013] Acquire a plurality of rolling parameter value groups, and determine target coefficients of each term in the function polynomial based on a genetic algorithm and the rolling parameter value groups;
[0014] The partial derivative of the function polynomial with the determined target coefficients is obtained to obtain the sensitivity of each rolling parameter to be screened.
[0015] Optionally, the number of the rolling parameter value groups is the same as the number of the rolling parameters to be screened, and each of the rolling parameter value groups includes a rolling parameter value corresponding to each of the rolling parameters to be screened, and an actual value of steel flatness; the target coefficients of each term in the function polynomial are determined based on the genetic algorithm and the rolling parameter value groups, including:
[0016] Assigning powers of each item in the function polynomial respectively by means of random number assignment, wherein the random number is within a preset range;
[0017] Substituting the rolling parameter values of each rolling parameter value group into the function polynomial assigned with power numbers to obtain a polynomial group;
[0018] Solving the polynomial group to obtain the first coefficient of each term of the function polynomial;
[0019] Based on the function polynomial after determining the first coefficient and assigning the power number, a plurality of steel flatness prediction values are obtained;
[0020] An adaptive function of the predicted value of the steel flatness, the actual value of the steel flatness and the average value of the steel flatness error is constructed, and iteration is performed with the goal of minimizing the average value of the steel flatness error to obtain the target coefficients of each term of the function polynomial.
[0021] Optionally, the function polynomial is:
[0022]
[0023] Among them, Y is the flatness of the steel, a is the target coefficient, b is the power, x is the rolling parameter to be screened, and n is the number of the rolling parameters to be screened.
[0024] Optionally, the adaptive function is:
[0025]
[0026] Among them, Y 误差 is the average flatness error of the steel, Y 实际值i is the actual value of the steel flatness, Y 预测值i is the predicted value of the steel flatness, and n is the number of rolling parameters to be screened.
[0027] Optionally, the method further comprises a step of obtaining the parameter optimization model, the step comprising:
[0028] Acquire a preset number of target rolling parameter value groups, each of the target rolling parameter values comprising a target rolling parameter value corresponding to each target rolling parameter and an actual value of steel flatness;
[0029] Based on the roulette method, a preset number of target rolling parameter value groups are divided into a training sample group and a test sample group according to a preset ratio;
[0030] Constructing a parameter optimization pre-model based on the training sample group and the test sample group;
[0031] The number of hidden layers and the number of nodes of the parameter optimization pre-model are determined based on the particle swarm algorithm to obtain the parameter optimization model.
[0032] Optionally, determining the number of hidden layers and the number of nodes of the parameter optimization pre-model based on a particle swarm algorithm includes:
[0033] Respectively set the range of the number of hidden layers and the range of the number of nodes;
[0034] Determine the number of initialized particle swarms and the number of iterations;
[0035] The maximum steel flatness is taken as the goal to optimize and obtain the number of hidden layers and the number of nodes.
[0036] In a second aspect, an embodiment of the present invention provides a device for optimizing steel rolling parameters, the device comprising:
[0037] A rolling parameter receiving unit to be screened, used for receiving a plurality of rolling parameters to be screened, each of which has a different type;
[0038] A sensitivity determination unit, used to determine the sensitivity of each rolling parameter to be screened, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the influence of the rolling parameter to be screened on the flatness of the steel;
[0039] A target rolling parameter screening unit, used for screening at least one target rolling parameter from the rolling parameters to be screened in descending order of sensitivity;
[0040] The target rolling parameter value determination unit is used to optimize the target rolling parameter based on the parameter optimization model to obtain the target rolling parameter value corresponding to the target rolling parameter.
[0041] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the program.
[0042] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a computer program, and when the computer program is executed, the server where the computer-readable storage medium is located is controlled to implement the steps of any one of the above methods.
[0043] The beneficial effects of the embodiments of the present invention include, for example:
[0044] The target rolling parameters are screened out by determining the sensitivity of the received rolling parameters to be screened, and then the target rolling parameters are optimized by the parameter optimization model to obtain the target rolling parameter values corresponding to the target rolling parameters. Users can directly use the target rolling parameter values to roll steel plates on the production line to obtain steel with higher flatness. The entire process does not require repeated setting of rolling parameters on the production line, nor does it require multiple tests, which is efficient and low-cost. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0046] Figure 1 A block diagram of an electronic device provided by an embodiment of the present invention;
[0047] Figure 2 A flow chart of the steps of a method for optimizing steel rolling parameters provided by an embodiment of the present invention;
[0048] Figure 3 A flow chart of a steel rolling parameter optimization method provided by an embodiment of the present invention;
[0049] Figure 4 An architectural diagram of a steel rolling parameter optimization device provided in an embodiment of the present invention.
[0050] Icons: 100 - electronic device; 110 - memory; 120 - processor; 130 - communication module; 300 - steel rolling parameter optimization device; 301 - rolling parameter receiving unit to be screened; 302 - sensitivity determination unit; 303 - target rolling parameter screening unit; 304 - target rolling parameter value determination unit. DETAILED DESCRIPTION
[0051] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.
[0052] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0053] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0054] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.
[0055] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.
[0056] The flatness of steel is very important for steel plates. Insufficient flatness of steel will lead to uneven stress distribution on the steel surface, stress concentration in local areas, reducing the overall bearing capacity, and may cause structural deformation or fracture risks. In terms of fatigue strength and service life, uneven steel surface will accelerate crack initiation and expansion, significantly shortening the fatigue life of steel under cyclic loads.
[0057] The flattening process (such as rolling with a small reduction rate) can eliminate the yield platform of the steel, optimize its mechanical property curve, and avoid the "slip line" defect during stamping.
[0058] The existing method of improving the flatness of steel plates is often to change a rolling parameter in the rolling process, then roll the steel plate, and compare the flatness of the new steel plate with the previous steel plate. After the comparison result is obtained, the rolling parameters are adjusted for rolling...Since there are many rolling parameters in the steel plate rolling process, this method is not only huge in workload, but also inefficient and costly.
[0059] Based on the above situation, an embodiment of the present invention provides a steel rolling parameter optimization method, device, electronic device and storage medium, which can effectively alleviate the above technical problems.
[0060] Please refer to Figure 1, is a block diagram of an electronic device 100 provided by the present application. The electronic device 100 may be a device capable of performing data processing, which is not limited in this embodiment. The electronic device 100 includes a memory 110, a processor 120, and a communication module 130. The memory 110, the processor 120, and the communication module 130. Each component is electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines.
[0061] The memory 110 is used to store programs or data. The memory 110 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.
[0062] The processor 120 is used to read / write data or programs stored in the memory and execute corresponding functions.
[0063] The communication module 130 is used to establish a communication connection between the server and other communication terminals through the network, and to send and receive data through the network.
[0064] It should be understood that Figure 1 The structure shown is only a schematic diagram of the structure of the electronic device 100. The electronic device 100 may also include Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown. Figure 1 The components shown in the figure can be implemented by hardware, software or a combination thereof. The electronic device 100 can be arranged in other devices or as an independent device.
[0065] The embodiment of the present invention provides a method for optimizing steel rolling parameters, which can be applied to an electronic device 100. The method includes: Figure 2 The following steps are shown:
[0066] Step S110: receiving a plurality of rolling parameters to be screened, wherein the types of the rolling parameters to be screened are different.
[0067] Step S120: determining the sensitivity of each rolling parameter to be screened, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the influence of the rolling parameter to be screened on the flatness of the steel.
[0068] Step S130: selecting at least one target rolling parameter from the rolling parameters to be screened in descending order of sensitivity.
[0069] Step S140: Based on the parameter optimization model, the target rolling parameters are optimized to obtain target rolling parameter values corresponding to the target rolling parameters.
[0070] In step S110, a plurality of rolling parameters to be screened are received, and the types of the rolling parameters to be screened are different.
[0071] The rolling parameters to be screened may be rolling parameters that the user inputs through the terminal and that the user believes have an impact on the flatness of the steel, such as rough rolling temperature, waiting thickness, final rolling temperature, and number of rapid cooling water groups, etc. After the user inputs these rolling parameters to be screened, the controller receives these rolling parameters to be screened and prepares for subsequent processing of the rolling parameters to be screened.
[0072] In step S120, the sensitivity of each rolling parameter to be screened is determined, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the influence of the rolling parameter to be screened on the flatness of the steel.
[0073] Sensitivity can be a parameter that characterizes the effect of the rolling parameters to be screened on the flatness of steel. The rolling parameters to be screened input by the user are the rolling parameters that the user believes may have an effect on the flatness of steel, but the user does not know which rolling parameters have a large effect on the flatness of steel and which rolling parameters have a very small or even negligible effect on the flatness of steel. Therefore, by determining the sensitivity of the rolling parameters to be screened, the rolling parameters with greater sensitivity can be screened out for optimization, which not only reduces the amount of calculation, but also improves the efficiency of rolling parameter optimization.
[0074] Optionally, determining the sensitivity of each rolling parameter to be screened includes:
[0075] Construct a function polynomial of each rolling parameter to be screened and the flatness of the steel.
[0076] A plurality of rolling parameter value groups are obtained, and target coefficients of each term in the function polynomial are determined based on a genetic algorithm and the rolling parameter value groups.
[0077] The partial derivative of the function polynomial with the determined target coefficients is obtained to obtain the sensitivity of each rolling parameter to be screened.
[0078] As an optional implementation, a function polynomial of each rolling parameter to be screened and the flatness of the steel can be constructed. A plurality of rolling parameter value groups pre-stored in the database or a plurality of rolling parameter value groups input by the user through the terminal are obtained, and a genetic algorithm is used to substitute the obtained plurality of rolling parameter value groups into the function polynomial to obtain the target coefficients of each term in the function polynomial.
[0079] After determining the target coefficient, the partial derivative of the function polynomial is calculated to obtain the partial derivative of each term. This partial derivative is the sensitivity of each rolling parameter to be screened.
[0080] Optionally, the number of the rolling parameter value groups is the same as the number of the rolling parameters to be screened, and each of the rolling parameter value groups includes a rolling parameter value corresponding to each of the rolling parameters to be screened, and an actual value of steel flatness. The step of determining the target coefficients of each term in the function polynomial based on the genetic algorithm and the rolling parameter value groups includes:
[0081] The powers of the various items in the function polynomial are assigned values respectively by means of random number assignment, wherein the random numbers are within a preset range.
[0082] Substitute the rolling parameter values of each rolling parameter value group into the function polynomial with the assigned power respectively to obtain a polynomial group.
[0083] The polynomial group is solved to obtain the first coefficient of each term of the function polynomial.
[0084] Based on the function polynomial after determining the first coefficient and assigning the power number, a plurality of steel flatness prediction values are obtained.
[0085] An adaptive function of the predicted value of the steel flatness, the actual value of the steel flatness and the average value of the steel flatness error is constructed, and iteration is performed with the goal of minimizing the average value of the steel flatness error to obtain the target coefficients of each term of the function polynomial.
[0086] The number of rolling parameter value groups is the same as the number of rolling parameters to be screened, and each rolling parameter value group includes a rolling parameter value corresponding to each rolling parameter to be screened, and an actual value of steel flatness. For example, there are 9 rolling parameters to be screened, namely, rough rolling temperature, waiting thickness, final rolling temperature, number of rapid cooling water groups, rapid cooling water volume, red-returning temperature, straightening reduction, straightening tilting amount, and upper cooling bed temperature. Then the number of rolling parameter value groups is 9 groups, and each rolling parameter value group includes a value corresponding to each of the rough rolling temperature, waiting thickness, final rolling temperature, number of rapid cooling water groups, rapid cooling water volume, red-returning temperature, straightening reduction, straightening tilting amount, and upper cooling bed temperature, and an actual value of steel flatness.
[0087] After constructing the function polynomial, the power of each term of the function polynomial can be assigned a value by random number assignment, and the assignment range is within a preset range. For example, if the preset range is 1 to 10, then the power of each term is randomly selected from 1 to 10 as the power value of the term.
[0088] After the power is assigned, each rolling parameter in the rolling parameter value group can be substituted into the function polynomial to obtain a polynomial group. If there are 9 rolling parameter value groups, there are 9 polynomial groups.
[0089] In the polynomial group at this time, each rolling parameter to be screened is an independent variable, the flatness of the steel is a dependent variable, and the coefficients of each term are unknown numbers. By solving the polynomial, the first coefficients of each term of the function polynomial can be obtained.
[0090] Substitute the rolling parameters in each rolling parameter value group into a function polynomial with randomly assigned powers and a determined first coefficient to obtain multiple predicted values of steel flatness. Then, an adaptive function of the predicted value of steel flatness, the actual value of steel flatness and the average value of steel flatness error is constructed, and it is iterated with the goal of minimizing the average value of steel flatness error, so that the target coefficients of each term of the function polynomial can be obtained.
[0091] Optionally, the function polynomial is:
[0092]
[0093] Among them, Y is the flatness of the steel, a is the target coefficient, b is the power, x is the rolling parameter to be screened, and n is the number of the rolling parameters to be screened.
[0094] For example, if there are 9 rolling parameters to be screened, namely rough rolling temperature, waiting thickness, final rolling temperature, number of quick cooling water groups, quick cooling water volume, red return temperature, straightening reduction, straightening tilt momentum, and upper cooling bed temperature, then n is equal to 9, x1, x2,…, x9 represent rough rolling temperature, waiting thickness, final rolling temperature, number of quick cooling water groups, quick cooling water volume, red return temperature, straightening reduction, straightening tilt momentum, and upper cooling bed temperature, respectively, a1, a2,…, a9 are the coefficients corresponding to each variable, respectively, and b1, b2,…, b9 are the powers corresponding to each variable, respectively.
[0095] Optionally, the adaptive function is:
[0096]
[0097] Among them, Y 误差 is the average flatness error of the steel, Y 实际值i is the actual value of the steel flatness, Y 预测值iis the predicted value of the steel flatness, and n is the number of rolling parameters to be screened.
[0098] In an optional implementation, the above adaptive function can be constructed, and iterated with the goal of minimizing the average value of the steel flatness error to obtain the target coefficients of each term of the function polynomial. 误差 Minimization is the goal, and the iteration is performed for 2000 times. Finally, the specific values of a1, a2, ..., a9 and b1, b2, ..., b9 are determined. Then, the above values are substituted into the function polynomial to obtain
[0099]
[0100] At this time, Y 误差 is 0.12mm. By taking the partial derivative of the above formula, we can get the partial derivatives of x1, x2, …, x9 respectively.
[0101] In step S130, at least one target rolling parameter is selected from the rolling parameters to be selected in descending order of sensitivity.
[0102] After obtaining the partial derivatives of each item, a set number of partial derivatives can be selected from each partial derivative in descending order of sensitivity, and the rolling parameters to be screened corresponding to these partial derivatives are the target rolling parameters. Alternatively, a sensitivity threshold is determined, and the rolling parameters to be screened corresponding to the partial derivatives whose partial derivatives are greater than the sensitivity threshold are determined as the target rolling parameters.
[0103] In step S140, the target rolling parameters are optimized based on the parameter optimization model to obtain target rolling parameter values corresponding to the target rolling parameters.
[0104] After the target rolling parameters are determined, they are input into the parameter optimization model to optimize the target rolling parameters, that is, the target rolling parameter values corresponding to the target rolling parameters can be obtained.
[0105] Optionally, the method further comprises a step of obtaining the parameter optimization model, the step comprising:
[0106] A preset number of target rolling parameter value groups are obtained, each of the target rolling parameter values comprising a target rolling parameter value corresponding to each target rolling parameter and an actual value of steel flatness.
[0107] Based on the roulette method, a preset number of target rolling parameter value groups are divided into a training sample group and a test sample group according to a preset ratio.
[0108] A parameter optimization pre-model is constructed based on the training sample group and the test sample group.
[0109] The number of hidden layers and the number of nodes of the parameter optimization pre-model are determined based on the particle swarm algorithm to obtain the parameter optimization model.
[0110] In an optional implementation, the parameter optimization model can be obtained in the following manner:
[0111] First, a preset number of target rolling parameter value groups are obtained, including a target rolling parameter value corresponding to each of the target rolling parameters and an actual value of the steel flatness. Through the roulette method, a part of them is randomly selected as the training sample group and the other part is randomly selected as the test sample group.
[0112] Then, a parameter optimization pre-model (such as a BP neural network model) is constructed through the training sample group and the test sample group, and the number of hidden layers and nodes of the parameter optimization pre-model is determined by the particle swarm algorithm to obtain the parameter optimization model.
[0113] Optionally, determining the number of hidden layers and the number of nodes of the parameter optimization pre-model based on a particle swarm algorithm includes:
[0114] The range of the number of hidden layers and the range of the number of nodes are set respectively. The number of initialized particle swarms and the number of iterations are determined.
[0115] The maximum steel flatness is taken as the goal to optimize and obtain the number of hidden layers and the number of nodes.
[0116] As an optional implementation, the range of the number of hidden layers and the range of the number of nodes can be set respectively, and the number of initialized particle swarms and the number of iterations can be determined. After setting, the number of hidden layers and the number of nodes can be obtained by optimizing with the maximum steel flatness as the goal.
[0117] For example, the number of hidden layers is set to 1 to 10, the number of hidden layer nodes is set to 10 to 100, the number of initialized particle swarms is set to 20, and the number of iterations is set to 5000. The maximum steel flatness is used as the goal for automatic optimization, and the optimal number of hidden layers and nodes is finally obtained. The fitness value of each particle is the unevenness value predicted by the BP neural network corresponding to the particle.
[0118] Finally, the number of hidden layers and nodes obtained by optimization are substituted into the parameter optimization pre-model to obtain the parameter optimization model. The determined target rolling parameters (for example, the target rolling parameters are the number of rapid cooling water groups, the amount of rapid cooling water, the amount of straightening reduction, the amount of straightening tilting, and the upper cooling bed temperature) are optimized, and the corresponding target rolling parameter values can be obtained, which are: 2 groups of rapid cooling water groups, 180m3 / h of rapid cooling water, 1.15mm of straightening reduction, 0.54 of straightening tilting, and 400℃ of upper cooling bed temperature. When users are rolling steel, they only need to set them according to the obtained target rolling parameter values.
[0119] For a better explanation of the present invention, see Figure 3 , an embodiment of the present invention provides a flow chart of a method for optimizing steel rolling parameters to illustrate the solution of the present invention.
[0120] like Figure 3 As shown, the controller receives the rolling parameters to be screened, constructs a function polynomial according to the number of rolling parameters to be screened, and uses a genetic algorithm to obtain the target coefficients of each item. The partial derivative of the function polynomial with the target coefficient is obtained to obtain the sensitivity of each rolling parameter to be screened. According to the sensitivity, at least one target rolling parameter is determined from each rolling parameter to be screened, and the target rolling parameter value corresponding to the target rolling parameter is obtained by inputting the target rolling parameter into the parameter optimization model.
[0121] The construction steps of the parameter optimization model are as follows: obtaining multiple target rolling parameter value groups, training sample groups and test sample groups for the multiple target rolling parameter value groups, and constructing a parameter optimization pre-model. The particle swarm algorithm is used to determine the number of hidden layers and the number of hidden layer nodes of the parameter optimization pre-model, and then the parameter optimization model is obtained.
[0122] Based on the same inventive concept, Figure 4 As shown, the embodiment of the present invention provides a steel rolling parameter optimization device 300, comprising:
[0123] The rolling parameter receiving unit 301 is used to receive a plurality of rolling parameters to be screened, and the types of the rolling parameters to be screened are different.
[0124] The sensitivity determination unit 302 is used to determine the sensitivity of each rolling parameter to be screened, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the impact of the rolling parameter to be screened on the flatness of the steel.
[0125] The target rolling parameter screening unit 303 is used to screen out at least one target rolling parameter from the rolling parameters to be screened in descending order of sensitivity.
[0126] The target rolling parameter value determination unit 304 is used to optimize the target rolling parameter based on the parameter optimization model to obtain the target rolling parameter value corresponding to the target rolling parameter.
[0127] Regarding the above-mentioned steel rolling parameter optimization device 300, the specific functions of each unit therein have been described in detail in the embodiment of the steel rolling parameter optimization method provided in this specification, and will not be elaborated here.
[0128] Based on the same inventive concept, an embodiment of the present invention specification provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of any method of the steel rolling parameter optimization method described above are implemented.
[0129] The present invention has at least the following beneficial effects:
[0130] The target rolling parameters are screened out by determining the sensitivity of the received rolling parameters to be screened, and then the target rolling parameters are optimized by the parameter optimization model to obtain the target rolling parameter values corresponding to the target rolling parameters. Users can directly use the target rolling parameter values to roll steel plates on the production line to obtain steel with higher flatness. The entire process does not require repeated setting of rolling parameters on the production line, nor does it require multiple tests, which is efficient and low-cost.
[0131] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0132] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.
[0133] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.
[0134] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for optimizing steel rolling parameters, characterized in that: The method comprises: receiving a plurality of rolling parameters to be screened, wherein the types of the rolling parameters to be screened are different; Determining the sensitivity of each rolling parameter to be screened, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the influence of the rolling parameter to be screened on the flatness of the steel; Selecting at least one target rolling parameter from the rolling parameters to be selected in descending order of sensitivity; Based on the parameter optimization model, the target rolling parameters are optimized to obtain target rolling parameter values corresponding to the target rolling parameters.
2. The method for optimizing steel rolling parameters according to claim 1, characterized in that: Determining the sensitivity of each rolling parameter to be screened includes: Constructing a function polynomial of each rolling parameter to be screened and the flatness of the steel; Acquire a plurality of rolling parameter value groups, and determine target coefficients of each term in the function polynomial based on a genetic algorithm and the rolling parameter value groups; The partial derivative of the function polynomial with the determined target coefficients is obtained to obtain the sensitivity of each rolling parameter to be screened.
3. The method for optimizing steel rolling parameters according to claim 2, characterized in that: The number of the rolling parameter value groups is the same as the number of the rolling parameters to be screened, and each of the rolling parameter value groups includes a rolling parameter value corresponding to each of the rolling parameters to be screened, and an actual value of steel flatness; The step of determining the target coefficients of each item in the function polynomial based on the genetic algorithm and the rolling parameter value group includes: Assigning powers of each item in the function polynomial respectively by random number assignment, wherein the random number is within a preset range; Substituting the rolling parameter values of each rolling parameter value group into the function polynomial assigned with power numbers to obtain a polynomial group; Solving the polynomial group to obtain the first coefficient of each term of the function polynomial; Based on the function polynomial after determining the first coefficient and assigning the power number, a plurality of steel flatness prediction values are obtained; An adaptive function of the predicted value of the steel flatness, the actual value of the steel flatness and the average value of the steel flatness error is constructed, and iteration is performed with the goal of minimizing the average value of the steel flatness error to obtain the target coefficients of each term of the function polynomial.
4. The method for optimizing steel rolling parameters according to claim 3, characterized in that: The function polynomial is: Among them, Y is the flatness of the steel, a is the target coefficient, b is the power, x is the rolling parameter to be screened, and n is the number of the rolling parameters to be screened.
5. The method for optimizing steel rolling parameters according to claim 3, characterized in that: The adaptive function is: Among them, Y 误差 is the average flatness error of the steel, Y 实际值i is the actual value of the steel flatness, Y 预测值i is the predicted value of the steel flatness, and n is the number of rolling parameters to be screened.
6. The method for optimizing steel rolling parameters according to claim 1, characterized in that: The method further comprises the step of obtaining the parameter optimization model, the step comprising: Acquire a preset number of target rolling parameter value groups, each of the target rolling parameter values comprising a target rolling parameter value corresponding to each target rolling parameter and an actual value of steel flatness; Based on the roulette method, a preset number of target rolling parameter value groups are divided into a training sample group and a test sample group according to a preset ratio; Constructing a parameter optimization pre-model based on the training sample group and the test sample group; The number of hidden layers and the number of nodes of the parameter optimization pre-model are determined based on the particle swarm algorithm to obtain the parameter optimization model.
7. The method for optimizing steel rolling parameters according to claim 1, characterized in that: The method of determining the number of hidden layers and the number of nodes of the parameter optimization pre-model based on the particle swarm algorithm includes: Respectively set the range of the number of hidden layers and the range of the number of nodes; Determine the number of initialized particle swarms and the number of iterations; The maximum steel flatness is taken as the goal to optimize and obtain the number of hidden layers and the number of nodes.
8. A steel rolling parameter optimization device, characterized in that: The steel rolling parameter optimization device comprises: A rolling parameter receiving unit to be screened, used for receiving a plurality of rolling parameters to be screened, wherein the types of the rolling parameters to be screened are different; A sensitivity determination unit, used to determine the sensitivity of each rolling parameter to be screened, wherein the greater the sensitivity of the rolling parameter to be screened, the greater the influence of the rolling parameter to be screened on the flatness of the steel; A target rolling parameter screening unit, used for screening at least one target rolling parameter from the rolling parameters to be screened in descending order of sensitivity; The target rolling parameter value determination unit is used to optimize the target rolling parameter based on the parameter optimization model to obtain the target rolling parameter value corresponding to the target rolling parameter.
9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the program.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a computer program, and when the computer program is executed, the server where the computer-readable storage medium is located is controlled to implement the steps of the method according to any one of claims 1 to 7.