Cold rolling force calculation method, electronic equipment and program product

Abstract

The invention provides a cold rolling force calculation method, electronic equipment and a program product, and the method comprises the steps: S1, obtaining the basic information of a steel coil, the basic information comprising the steel type of the steel coil, the thickness of a strip steel raw material, the width of the strip steel, the roller diameter, chemical components, raw material information and a friction coefficient calculated according to a preset rolling speed; s2, judging whether the second-class deformation resistance model coefficient of the steel type does not exist in a preset database or not; if yes, the first-class deformation resistance model coefficient of the steel type is calculated through a preset first algorithm, and the rolling force is calculated through the first-class deformation resistance model coefficient by adopting a preset first rolling mathematical model; otherwise, the rolling force is calculated by adopting a preset second rolling mathematical model through the second-class deformation resistance model coefficient; by means of the technical scheme, high-precision prediction of the cold rolling force of the new steel grade can be achieved, the calculation error at the initial stage of production is remarkably reduced, and the product quality and the production stability are further effectively improved.

Description

Technical Field

[0001] The invention relates to the field of steel cold rolling production, and in particular to a calculation method for cold rolling force, electronic equipment and program products. Background Art

[0002] In the field of steel cold rolling production, the construction of the cold rolling force model is crucial. It is not only related to the quality and production efficiency of the product, but also directly affects the production cost. The cold rolling force is affected by many factors, including but not limited to the geometric size, tension, deformation resistance and friction coefficient of the material. Among them, the deformation resistance model coefficient and the friction coefficient are difficult to measure directly, which becomes the key difficulty of the model accuracy. The cold rolling deformation resistance model coefficient is related to the material and the cumulative reduction rate, and the friction coefficient is related to the speed and lubrication state. For steel grades that have been stably produced, these parameters can be estimated more accurately through long-term experience accumulation and data feedback; however, for new steel grades, due to the lack of historical data support, it is particularly difficult to obtain the initial deformation resistance model coefficient, which often leads to large errors in the calculation of the rolling force in the initial production, thus affecting product quality and production stability. Summary of the invention

[0003] The embodiments of the present invention provide a method for calculating the cold rolling force, an electronic device and a program product to achieve high-precision prediction of the cold rolling force of new steel grades, significantly reduce calculation errors in the early stages of production, and further effectively improve product quality and production stability.

[0004] In order to achieve the above object, on the one hand, a method for calculating cold rolling force is provided, the method comprising:

[0005] S1, obtaining basic information of the steel coil, the basic information including: steel type, strip steel raw material thickness, strip steel width, roll diameter, chemical composition, raw material information and friction coefficient calculated according to a predetermined rolling speed;

[0006] S2, judging whether there is no second-class deformation resistance model coefficient of the steel type in the predetermined database; if so, calculating the first-class deformation resistance model coefficient of the steel coil by a predetermined first algorithm, and calculating the rolling force by using the first-class deformation resistance model coefficient and a predetermined first rolling mathematical model; otherwise, calculating the rolling force by using the second-class deformation resistance model coefficient and a predetermined second rolling mathematical model;

[0007] The first algorithm is as follows: inputting the chemical composition and raw material information of the steel coil into a model pre-trained according to historical data to obtain a type of deformation resistance model coefficient of the steel coil;

[0008] The first rolling mathematical model is:

[0009]

[0010] P represents the rolling force, B represents the strip width, R represents the roller radius, Δh represents the reduction, ε represents the reduction rate, μ represents the friction coefficient, H represents the strip raw material thickness, the strip raw material thickness is the entrance thickness, K represents the deformation resistance, μ t represents the tension friction coefficient, t b Indicates the inlet tension, t f Indicates outlet tension;

[0011] The second rolling mathematical model is:

[0012]

[0013] α3=1.08-1.02·ε

[0014]

[0015] α7=α1·α3+α2

[0016]

[0017]

[0018] h2, h1, Dd0 and α1~α9 are predetermined temporary variables; μ t ,ν,E,v R and E R is a predetermined constant, k fm represents the average deformation resistance calculated by the predetermined deformation resistance model, k h represents the deformation resistance at the rolling exit calculated by the deformation resistance model; h represents the exit thickness; R0 represents the original working roll radius, and R' represents the working roll radius after flattening;

[0019] With respect to the first type of deformation resistance model coefficient and the second type of deformation resistance model coefficient, the deformation resistance model is:

[0020]

[0021] k0, k1, k e Respectively represent predetermined constants.

[0022] Preferably, the method for calculating the cold rolling force further comprises:

[0023] Collecting production data of the steel coil during the production process according to a predetermined cycle, the production data including: actually measured rolling force, inlet thickness, outlet thickness, the rolling speed, reduction rate, inlet tension and outlet tension;

[0024] After the production is completed, the production data of N groups are selected through T verification, and the production data of N groups are respectively combined with the basic information of the steel coil to obtain the processing data of N groups;

[0025] The first deformation resistance of each set of processed data is calculated by a predetermined third rolling mathematical model, and a type of deformation resistance model coefficient of the steel coil is calculated by a predetermined second intelligent algorithm, and the basic information of the steel coil and the type of deformation resistance model coefficient of the steel coil are stored in the database.

[0026] Preferably, in the method for calculating the cold rolling force, the third rolling mathematical model is a mathematical model obtained by inverse operation of the first rolling mathematical model, wherein the third rolling mathematical model is:

[0027]

[0028] K * represents the first deformation resistance, and P1 represents the actually measured rolling force.

[0029] Preferably, the method for calculating the cold rolling force, wherein calculating a type of deformation resistance model coefficient of the steel coil by a predetermined second intelligent algorithm comprises the following steps:

[0030] S401, define the coefficient C related to the deformation resistance, where C = (k0, k1, k e ), determine the four initial values ​​of the coefficient C;

[0031] S402, in each of the N groups of processed data, respectively calculate the deviation between the deformation resistance obtained by the four initial values ​​of the coefficient C and the first deformation resistance, and sort the deviations from small to large, and the values ​​of the coefficient C corresponding to the deviations after the sorting are C1, C2, C3, and C4, respectively; wherein the deviations are calculated by the following formula:

[0032]

[0033] err represents the deviation, f(C i ) represents the i-th deviation, K n (C i ) represents the reduction rate of the nth group of processed data among the N groups of processed data and the i-th coefficient C i Calculated deformation resistance, K n * represents the first deformation resistance of the nth group of processed data, i=[1,4];

[0034] Through the calculation of the above deviation, we can obtain: f(C1)≤f(C2)≤f(C3)≤f(C4);

[0035] S403. If the Euclidean distance between C1 and C4 in the multi-dimensional space is less than a predetermined value, output C1 as the first type of deformation resistance model coefficient of the steel coil and end the process;

[0036] S404. Calculate the first type of deformation resistance model coefficient of the steel coil through a predetermined optimization algorithm, where:

[0037] Calculate the centroid C of other points except C4 o :

[0038]

[0039] And calculate the reflection point C in the search space r , where the search space is the set of all first type of deformation resistance model coefficients,

[0040] C r = C o + α·(C o - C4)

[0041] where α is a predetermined constant, α > 0;

[0042] When f(C1) ≤ f(C r ) < f(C3), replace C4 with C r and return to step S402;

[0043] When f(C r ) < f(C1), calculate the expansion point C e in the search space;

[0044] C e = C o + γ·(C r - C o )

[0045] where γ is a predetermined constant, γ > 1; if f(C r ) < f(C e ), then replace C4 with C r and return to step S402; otherwise, replace C4 with C e and return to step S401;

[0046] When f(C3) ≤ f(C r ) < f(C4) or f(C r ) ≥ f(C4), calculate the contraction point C c in the search space;

[0047] C c = C o+ρ·(C r -C o )

[0048] Where ρ is a predetermined constant, ρ<1; if f(C c ) <f(C r ), using C c Replace C4 and return to step S402; otherwise, except for C1, other parameters C are rolled back using the following formula:

[0049] C i =C1+σ·(C i -C1)

[0050] Wherein, σ is a predetermined constant, σ<1; and return to step S402.

[0051] Preferably, in the cold rolling force calculation method, in step S401, the initial coefficient C is randomly generated within a predetermined empirical range or randomly selected from historical data.

[0052] Preferably, the cold rolling force calculation method further includes: calculating the second type of deformation resistance model coefficient of the steel grade by inputting N groups of processed data into a predetermined long-term self-learning model, and storing the second type of deformation resistance model coefficient of the steel grade into a database.

[0053] Preferably, in the method for calculating the cold rolling force, the pre-trained model in step S2 is established and trained by the following steps:

[0054] S501, obtaining historical processing data stored in a database and corresponding coefficients of a type of deformation resistance model;

[0055] S502, establishing a model according to the preset initial parameters, and randomly generating a training set and a test set of a predetermined proportion from the data obtained in step S501;

[0056] S503, use a predetermined function to train the model and use the test set to evaluate the model. If the accuracy of the model is greater than a predetermined threshold, return to step S502 to optimize the parameters and perform training again.

[0057] Preferably, in the method for calculating the cold rolling force, the pre-trained model in step S2 is an XGBoost model.

[0058] On the other hand, an embodiment of the present invention provides an electronic device, which includes a memory and a processor, wherein the memory stores at least one program, and the at least one program is executed by the processor to implement any of the above-mentioned methods for calculating the cold rolling force.

[0059] On the other hand, an embodiment of the present invention provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the cold rolling force calculation method as described above is implemented.

[0060] The above technical solution has the following technical effects:

[0061] The technical solution of the embodiment of the present invention obtains the basic information of the steel coil, including the steel type, strip raw material thickness, strip width, roll diameter, chemical composition, raw material information and the friction coefficient calculated according to the predetermined rolling speed; according to this information, if there is no second-class deformation resistance model coefficient of the steel type in the predetermined database, a pre-trained model is used to calculate a first-class deformation resistance model coefficient based on the chemical composition and raw material information of the steel coil, and the rolling force is calculated using the first rolling mathematical model; if there is a second-class deformation resistance model coefficient in the database, the second rolling mathematical model is directly used to calculate the rolling force; these two models respectively consider different influencing factors and parameters, ensure the accuracy and reliability of the calculation results, can effectively improve the control accuracy in the rolling process, optimize production efficiency, reduce scrap rate, and can quickly generate the required model coefficients through the pre-trained model in the absence of ready-made model coefficients, thereby realizing rapid adaptation and processing of new or unknown steel types, further improving the automation and intelligence level in the cold rolling process, and improving product quality and production stability.

[0062] In a further embodiment, actual measured data of the steel coil during the production process are collected according to a predetermined cycle, including rolling force, entrance thickness, exit thickness, rolling speed, reduction rate, entrance tension and exit tension; after the production is completed, N groups of representative data are selected from these data through T verification, and these data are combined with the basic information of the steel coil to generate processed data; the first deformation resistance of each group of processed data is calculated using a third rolling mathematical model, and a type of deformation resistance model coefficient of the steel coil is calculated through a second intelligent algorithm; the basic information of the steel coil and the calculated type of deformation resistance model coefficient are stored in a database, which not only improves the accuracy and reliability of the deformation resistance model, but also can continuously update and optimize the model parameters, so as to better adapt to changes in different steel grades and production conditions, and ultimately improve the control accuracy of the cold rolling process and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic flow chart of a method for calculating cold rolling force according to an embodiment of the present invention;

[0064] Figure 2 A flow chart of a method for calculating cold rolling force according to an embodiment of the present invention;

[0065] Figure 3It is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] To further illustrate the various embodiments, the present invention provides drawings. These drawings are part of the disclosure of the present invention, which are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, a person of ordinary skill in the art should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0067] The present invention will now be further described with reference to the accompanying drawings and specific implementation methods.

[0068] Embodiment 1:

[0069] In order to achieve high-precision prediction of the cold rolling force of new steel grades, significantly reduce the calculation error in the early stage of production, and further effectively improve product quality and production stability, an embodiment of the present invention provides a method for calculating the cold rolling force. Figure 1 FIG. 1 is a flow chart of a method for calculating cold rolling force according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0070] S1, obtaining basic information of the steel coil, including: steel type of the steel coil, strip steel raw material thickness, strip steel width, roll diameter, chemical composition, raw material information and friction coefficient calculated according to a predetermined rolling speed;

[0071] S2, judging whether there is no second-class deformation resistance model coefficient of the steel type in the predetermined database; if so, calculating the first-class deformation resistance model coefficient of the steel coil by a predetermined first algorithm, and calculating the rolling force by using the first-class deformation resistance model coefficient and a predetermined first rolling mathematical model; otherwise, calculating the rolling force by using the second-class deformation resistance model coefficient and a predetermined second rolling mathematical model;

[0072] The first algorithm is: inputting the chemical composition and raw material information of the steel coil into a model pre-trained according to historical data to obtain a type of deformation resistance model coefficient of the steel coil;

[0073] The first rolling mathematical model is:

[0074]

[0075] P represents rolling force, B represents strip width, R represents roller radius, Δh represents reduction, ε represents reduction rate, μ represents friction coefficient, H represents strip raw material thickness, strip raw material thickness is the inlet thickness, K represents deformation resistance, μ t represents the tension friction coefficient, t b Indicates the inlet tension, tf Indicates outlet tension;

[0076] The second rolling mathematical model is:

[0077]

[0078] h1=Δh+h2

[0079]

[0080]

[0081] α3=1.08-1.02·ε

[0082]

[0083] α7=α1·α3+α2

[0084]

[0085] h2, h1, D0 and α1~α9 are predetermined temporary variables; μ t ,ν,E,v R and E R is a predetermined constant, k fm represents the average deformation resistance calculated by the predetermined deformation resistance model, k h represents the deformation resistance at the rolling exit calculated by the deformation resistance model; h represents the exit thickness; R0 represents the original working roll radius, and R' represents the working roll radius after flattening;

[0086] For the first type of deformation resistance model coefficient and the second type of deformation resistance model coefficient, the deformation resistance model is:

[0087]

[0088] k0, k1, k e Respectively represent predetermined constants.

[0089] Embodiment 2:

[0090] In order to improve the product quality of cold rolling mills, especially to solve the problem of poor accuracy of calculated rolling force due to lack of initial deformation resistance model coefficient when new steel grades are produced for the first time, the present invention proposes a method for calculating cold rolling force, which can save deformation resistance testing costs and industrial production trial and error costs, and further improve the stability of the rolling process through high-precision rolling force calculation results. Figure 2 FIG. 1 is a flow chart of a method for calculating cold rolling force according to an embodiment of the present invention. Figure 2 As shown, the method includes:

[0091] S101, when the steel coil reaches the entrance walking beam or entrance saddle and is ready for subsequent production, basic information of the steel coil is obtained;

[0092] Preferably, the basic information includes: steel type of the steel coil, strip steel raw material thickness, strip steel width, roll diameter, chemical composition, raw material information and friction coefficient calculated according to a predetermined rolling speed.

[0093] S201, judging whether there is a second type of deformation resistance model coefficient for the steel type; specifically, judging whether there is no second type of deformation resistance model coefficient for the steel type in a predetermined database according to the steel type of the steel coil; if so, it indicates that the steel type is a new steel type, and a first type of deformation resistance model coefficient of the steel coil is calculated by a predetermined first algorithm, and a predetermined first rolling mathematical model is used to calculate the rolling force by using the first type of deformation resistance model coefficient; otherwise, it indicates that the steel type is a produced steel type, and a predetermined second rolling mathematical model is used to calculate the rolling force by using the second type of deformation resistance model coefficient;

[0094] Preferably, the first algorithm is: inputting the chemical composition and raw material information of the steel coil into a model pre-trained according to historical data to obtain a type of deformation resistance model coefficient of the steel coil;

[0095] Preferably, the pre-trained model is an XGBoost model created based on the xgboost library.

[0096] In a specific embodiment, the pre-trained model is established and trained through the following steps:

[0097] S201-1, obtaining historical processing data and corresponding historical first-class deformation resistance model coefficients stored in a database;

[0098] Preferably, the chemical composition includes: aluminum Al, arsenic As, boron B, carbon C, calcium Ca, chromium Cr, copper Cu, manganese Mn, molybdenum Mo, nitrogen N, niobium Nb, nickel Ni, phosphorus P, sulfur S, silicon Si, tin Sn, titanium Ti, vanadium V, zirconium Zr, etc.;

[0099] Preferably, the raw material information includes: hot-rolled strip outlet thickness, hot-rolling final rolling temperature, coiling temperature, previous process code, etc.

[0100] S201-1, establishing a model according to preset initial parameters, and randomly generating a training set and a test set of predetermined proportions from the data obtained in step S201-1;

[0101] Preferably, the preset initial parameters include: training objective objective = 'reg:squarederror'; learning rate eta = 0.3; maximum tree depth max_depth = 5; evaluation model eval_metric = 'rmse'; other parameters are default;

[0102] Preferably, a training set and a test set of predetermined proportions are randomly generated, wherein the training set accounts for 80% of the samples and the test set accounts for the other 20% of the samples.

[0103] S201-3, train the model using a predetermined function and evaluate the model using a test set. If the accuracy of the model is greater than a predetermined threshold, return to step S201-2 to optimize the parameters and perform training again;

[0104] Preferably, the predetermined function is a train function.

[0105] Preferably, the first rolling mathematical model is:

[0106]

[0107] P represents rolling force, B represents strip width, R represents roller radius, Δh represents reduction, ε represents reduction rate, μ represents friction coefficient, H represents strip raw material thickness, strip raw material thickness is the inlet thickness, K represents deformation resistance, μ t represents the tension friction coefficient, t b Indicates the inlet tension, t f Indicates outlet tension;

[0108] Preferably, the second rolling mathematical model is:

[0109]

[0110] h1=Δh+h2

[0111]

[0112] α3=1.08-1.02·ε

[0113]

[0114] h2, h1, D0 and α1~α9 are predetermined temporary variables; μ t ,ν,E,v R and E R is a predetermined constant, k fm represents the average deformation resistance calculated by the predetermined deformation resistance model, k h represents the deformation resistance at the rolling exit calculated by the deformation resistance model; h represents the exit thickness; R0 represents the original working roll radius, and R' represents the working roll radius after flattening;

[0115] Preferably, for the first type of deformation resistance model coefficient and the second type of deformation resistance model coefficient, the deformation resistance model is:

[0116]

[0117] k0, k1, k e Respectively represent predetermined constants;

[0118] S301, collecting measured data during the production process;

[0119] Specifically, the production data of the steel coil during the production process, i.e., measured data, is collected according to a predetermined cycle. The production data includes: actually measured rolling force, entry thickness, exit thickness, the rolling speed, reduction rate, entry tension, and exit tension.

[0120] S401, using T verification to screen the measured data;

[0121] Specifically, after the production is completed, N groups of production data are selected through T verification, and the N groups of production data are respectively combined with the basic information of the steel coil to obtain N groups of processing data;

[0122] Preferably, N=10.

[0123] S501, calculating the actual deformation resistance through the third rolling mathematical model, and calculating a type of deformation resistance model coefficient through the second intelligent algorithm;

[0124] Specifically, the first deformation resistance, i.e., the actual deformation resistance, of each set of processed data is calculated by a predetermined third rolling mathematical model, and a type of deformation resistance model coefficient of the steel coil is calculated by a predetermined second intelligent algorithm, and the basic information of the steel coil and the type of deformation resistance model coefficient of the steel coil are stored in a database for subsequent calculation of a type of deformation resistance model coefficient of a new steel grade by the first intelligent algorithm;

[0125] Preferably, the third rolling mathematical model is:

[0126]

[0127] K * represents the first deformation resistance, and P1 represents the actually measured rolling force.

[0128] In a specific embodiment, calculating a type of deformation resistance model coefficient of a steel coil by a predetermined second intelligent algorithm includes the following steps:

[0129] S501-1, define the coefficient C related to deformation resistance, where C = (k0, k1, k e ), determine the four initial values ​​of the coefficient C;

[0130] Preferably, the initial value of the coefficient C is randomly generated within a predetermined empirical range or randomly selected from historical data.

[0131] S501-2. For each set of processing data among the N sets of processing data, calculate the deviation between the deformation resistance obtained from the four initial values of the passing coefficient C and the first deformation resistance, and sort the magnitudes of the deviations in ascending order of label. The values of the coefficient C corresponding to the deviations after label sorting are C1, C2, C3, and C4 in sequence. Among them, the deviation is calculated by the following formula:

[0132]

[0133] err represents the deviation, f(C i ) represents the i-th deviation, K n (C i ) represents the deformation resistance calculated through the reduction ratio of the n-th set of processing data among the N sets of processing data and the i-th coefficient C i K n * represents the first deformation resistance of the n-th set of processing data, i = [1, 4];

[0134] Through the calculation of the above deviation, we get: f(C1) ≤ f(C2) ≤ f(C3) ≤ f(C4);

[0135] S501-3. When the Euclidean distance between C1 and C4 in the multi-dimensional space is less than a predetermined value, output C1 as the coefficient of the first type of deformation resistance model of the steel coil and end the process;

[0136] S501-4. Calculate the coefficient of the first type of deformation resistance model of the steel coil through a predetermined optimization algorithm, where:

[0137] Calculate the centroid C of other points except C4 o :

[0138]

[0139] And calculate the reflection point C r in the search space. The search space is the set of all possible coefficients of the first type of deformation resistance model,

[0140] C r = C o + α·(C o - C4)

[0141] where α is a predetermined constant, α > 0; preferably, the value of α is 1;

[0142] When f(C1) ≤ f(C r ) < f(C3), that is, the reflection point is average. Replace C4 with C r and return to step S501-2;

[0143] When f(Cr ) When f(C1), that is, the reflection point is optimal, expand on the basis of the current solution to explore a farther solution space in the hope of finding a better solution, and calculate the expansion point C in the search space e ;

[0144] C e = C o + γ·(C r - C o )

[0145] where γ is a predetermined constant, γ > 1; if f(C r ) < f(C e ), then replace C4 with C r and return to step S501-2; otherwise, replace C4 with C e and return to step S501-1; preferably, the value of γ is 2;

[0146] When f(C3) ≤ f(C r ) < f(C4) or f(C r ) ≥ f(C4), that is, the reflection point is poor or the worst, then choose to shrink the search range and return to a point closer to the current solution in the hope of finding a better solution in a smaller search range, and calculate the contraction point C in the search space c ;

[0147] C c = C o + ρ·(C r - C o )

[0148] where ρ is a predetermined constant, ρ < 1; preferably, the value of ρ is 0.5; if f(C c ) < f(C r ), replace C4 with C c and return to step S501-2; otherwise, except for C1, other parameters C are rolled back through the following formula:

[0149] C i = C1 + σ·(C i - C1)

[0150] where σ is a predetermined constant, σ < 1; and return to step S501-2; preferably, the value of σ is 0.5.

[0151] S601, calculate the coefficient of the secondary deformation resistance model through the long-term self-learning model;

[0152] Specifically, the second type of deformation resistance model coefficient of the steel grade is calculated by inputting N groups of processed data into a predetermined long-term self-learning model, and the second type of deformation resistance model coefficient of the steel grade is stored in a database;

[0153] Preferably, the predetermined long-term self-learning model includes: Newton-Raphson algorithm, Powell optimization algorithm, etc.; or, the predetermined long-term self-learning model is based on an intelligent algorithm, such as: particle swarm algorithm, genetic algorithm, etc.

[0154] Among them, the first type of deformation resistance model coefficient is for steel coils, and the second type of deformation resistance model coefficient is for steel grades. The first type of deformation resistance model coefficient is used for the calculation of the first rolling mathematical model, and the second type of deformation resistance model coefficient is used for the calculation of the second rolling mathematical model; there is a certain difference in the values ​​of the two deformation resistance model coefficients;

[0155] The first rolling mathematical model is a simplified version of the second rolling mathematical model, and its main feature is that it is simple and easy to inversely calculate; the second rolling mathematical model takes into account the flattening of the rolls and the elastic deformation zone at the inlet and outlet, so it has higher accuracy and is suitable for use after the steel grade is stably produced. Both algorithms are based on the classic Bland-Ford-Hill cold rolling force model; the third rolling mathematical model is the inverse operation of the first rolling mathematical model, and its method is simple and easy to implement, and the calculation results are used for the second intelligent algorithm calculation.

[0156] Embodiment three:

[0157] The present invention also provides an electronic device, such as Figure 3 As shown, the electronic device includes a processor 301, a memory 302, a bus 303, and a computer program stored in the memory 302 and executable on the processor 301. The processor 301 includes one or more processing cores. The memory 302 is connected to the processor 301 via the bus 303. The memory 302 is used to store program instructions. When the processor executes the computer program, the steps in the above method embodiment of the first embodiment of the present invention are implemented.

[0158] Further, as an executable solution, the electronic device may be a computer unit, and the computer unit may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The computer unit may include, but is not limited to, a processor and a memory. Those skilled in the art may understand that the composition structure of the above-mentioned computer unit is only an example of a computer unit and does not constitute a limitation on the computer unit, and may include more or less components than the above-mentioned, or a combination of certain components, or different components. For example, the computer unit may also include input and output devices, network access devices, buses, etc., which are not limited in the embodiments of the present invention.

[0159] Further, as an executable solution, the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the computer unit, and various interfaces and lines are used to connect the various parts of the entire computer unit.

[0160] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the computer unit by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required for a function; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0161] Embodiment 4:

[0162] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the method described above are implemented.

[0163] Although the present invention has been specifically shown and described in conjunction with the preferred embodiments, it should be understood by those skilled in the art that various changes may be made to the present invention in form and details without departing from the spirit and scope of the present invention as defined by the appended claims, all of which are within the scope of protection of the present invention.

Claims

1. A method for calculating cold rolling force, characterized in that: include: S1, obtaining basic information of the steel coil, the basic information including: steel type, strip steel raw material thickness, strip steel width, roll diameter, chemical composition, raw material information and friction coefficient calculated according to a predetermined rolling speed; S2, judging whether there is no second-class deformation resistance model coefficient of the steel type in the predetermined database; if so, calculating the first-class deformation resistance model coefficient of the steel coil by a predetermined first algorithm, and calculating the rolling force by using the first-class deformation resistance model coefficient and a predetermined first rolling mathematical model; otherwise, calculating the rolling force by using the second-class deformation resistance model coefficient and a predetermined second rolling mathematical model; The first algorithm is as follows: inputting the chemical composition and raw material information of the steel coil into a model pre-trained according to historical data to obtain a type of deformation resistance model coefficient of the steel coil; The first rolling mathematical model is: P represents the rolling force, B represents the strip width, R represents the roller radius, Δh represents the reduction, ε represents the reduction rate, μ represents the friction coefficient, H represents the strip raw material thickness, the strip raw material thickness is the entrance thickness, K represents the deformation resistance, μ t represents the tension friction coefficient, t b Indicates the inlet tension, t f Indicates outlet tension; The second rolling mathematical model is: h1=Δh+h2 α3=1.08-1.02·ε α7＝α1·α3+α2 R'＝α9 2 h2, h1, D0 and α1~α9 are predetermined temporary variables; μ t ,ν,E,v R and E R is a predetermined constant, k fm represents the average deformation resistance calculated by the predetermined deformation resistance model, k h represents the deformation resistance at the rolling exit calculated by the deformation resistance model; h represents the exit thickness; R0 represents the original working roll radius, and R' represents the working roll radius after flattening; With respect to the first type of deformation resistance model coefficient and the second type of deformation resistance model coefficient, the deformation resistance model is: k0, k1, k e Respectively represent predetermined constants.

2. The method for calculating the cold rolling force according to claim 1, characterized in that: Also includes: Collecting production data of the steel coil during the production process according to a predetermined cycle, the production data including: actually measured rolling force, inlet thickness, outlet thickness, the rolling speed, reduction rate, inlet tension and outlet tension; After the production is completed, the production data of N groups are selected through T verification, and the production data of N groups are respectively combined with the basic information of the steel coil to obtain the processing data of N groups; The first deformation resistance of each set of processed data is calculated by a predetermined third rolling mathematical model, and a type of deformation resistance model coefficient of the steel coil is calculated by a predetermined second intelligent algorithm, and the basic information of the steel coil and the type of deformation resistance model coefficient of the steel coil are stored in the database.

3. The method for calculating the cold rolling force according to claim 2, characterized in that: The third rolling mathematical model is a mathematical model obtained by inverse operation of the first rolling mathematical model, wherein the third rolling mathematical model is: K * represents the first deformation resistance, and P1 represents the actually measured rolling force.

4. The method for calculating the cold rolling force according to claim 2, characterized in that: Calculating a type of deformation resistance model coefficient of the steel coil by a predetermined second intelligent algorithm comprises the following steps: S401, define the coefficient C related to deformation resistance, where C = (k0, k1, k e ), determine the four initial values ​​of the coefficient C; S402, in each of the N groups of processed data, respectively calculate the deviation between the deformation resistance obtained by the four initial values ​​of the coefficient C and the first deformation resistance, and sort the deviations from small to large, and the values ​​of the coefficient C corresponding to the deviations after the sorting are C1, C2, C3, and C4, respectively; wherein the deviations are calculated by the following formula: err represents the deviation, f(C i ) represents the i-th deviation, K n (C i ) represents the reduction rate of the nth group of processed data among the N groups of processed data and the i-th coefficient C i Calculated deformation resistance, K n * represents the first deformation resistance of the nth group of processed data, i=[1, 4]; Through the calculation of the above deviation, we can obtain: f(C1)≤f(C2)≤f(C3)≤f(C4); S403, when the Euclidean distance between C1 and C4 in the multidimensional space is less than a predetermined value, outputting C1 as a type of deformation resistance model coefficient of the steel coil, and terminating the process; S404, calculating a type of deformation resistance model coefficient of the steel coil by a predetermined optimization algorithm, wherein: Calculate the centroid C of all points except C4 o : And calculate the reflection point C in the search space r , the search space is the set of all coefficients of a type of deformation resistance model, C r =C o +α·(C o -C4) Wherein, α is a predetermined constant, α>0; When f(C1)≤f(C r )<f(C3), use C r Replace C4 and return to step S402; When f(C r )<f(C1), calculate the expansion point C in the search space e ; C e =C o +γ·(C r -C o ) Where γ is a predetermined constant, γ>1; if f(C r )<f(C e ), then use C r Replace C4 and return to step S402; otherwise, replace C e Replace C4 and return to step S401; When f(C3)≤f(C r )<f(C4) or f(C r )≥f(C4), calculate the contraction point C in the search space c ; C c =C o +ρ·(C r -C o ) Where ρ is a predetermined constant, ρ<1; if f(C c )<f(C r ), using C c Replace C4 and return to step S402; otherwise, except for C1, other parameters C are rolled back using the following formula: C i =C1+σ·(C i -C1) Wherein, σ is a predetermined constant, σ<1; and return to step S402.

5. The method for calculating the cold rolling force according to claim 4, characterized in that: In step S401, the initial value of the coefficient C is randomly generated within a predetermined empirical range or randomly selected from historical data.

6. The method for calculating the cold rolling force according to claim 2, characterized in that: Also includes: The second type of deformation resistance model coefficient of the steel grade is calculated by inputting N groups of processed data into a predetermined long-term self-learning model, and the second type of deformation resistance model coefficient of the steel grade is stored in a database.

7. The method for calculating the cold rolling force according to claim 2, characterized in that: The pre-trained model in step S2 is established and trained by the following steps: S501, obtaining historical processing data stored in a database and corresponding coefficients of a type of deformation resistance model; S502, establishing a model according to the preset initial parameters, and randomly generating a training set and a test set of a predetermined proportion from the data obtained in step S501; S503, use a predetermined function to train the model and use the test set to evaluate the model. If the accuracy of the model is greater than a predetermined threshold, return to step S502 to optimize the parameters and perform training again.

8. The method for calculating the cold rolling force according to claim 1, characterized in that: The pre-trained model in step S2 is an XGBoost model.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores at least one program, and the at least one program is executed by the processor to implement the method for calculating the cold rolling force according to any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for calculating the cold rolling force according to any one of claims 1 to 8 is implemented.

Images