Simulated fixed-weight cutting data processing method and system based on cognitive intelligent process model, and device and medium

Abstract

A simulated fixed-weight cutting data processing method based on a cognitive intelligent process model, the method comprising: mounting one hydraulic lifting platform scale per strand at a cooling bed transfer run-out table of a multi-strand continuous casting machine; each time after the platform scale performs weighing, feeding back weighing data to a fixed-weight cutting model; on the basis of the fixed-weight cutting model, using a mathematical calculation formula to calculate the length of a steel billet; issuing steel billet length information to a flame cutting machine for cutting, so as to obtain a cut steel billet; weighing the cut steel billet again by means of the platform scale, and calculating a length required for the next instance of steel billet cutting; ensuring that a process is continuously and circularly performed, and collecting a series of parameters in the above process; and studying dynamic relationships between the parameters and the mathematical calculation formula. The method improves the prediction accuracy of the weighing data. Further comprised are a simulated fixed-weight cutting data processing system based on a cognitive intelligent process model, and an electronic device and a computer-readable storage medium. The effect of simulating fixed-weight cutting is achieved, thereby ensuring the cutting accuracy.

Description

Simulation fixed weight cutting data processing method, system, equipment and medium based on cognitive intelligent process model

[0001] The present application claims priority from the Chinese patent application No. 2024106434898, filed on May 22, 2024, and entitled "Simulation fixed weight cutting data processing method, system, equipment and medium based on cognitive intelligent process model", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application belongs to the field of data modeling, and particularly relates to a simulation fixed weight cutting data processing method based on a cognitive intelligent process model. BACKGROUND

[0003] At present, there are mainly the following fixed weight cutting methods:

[0004] The Chinese invention patent No. CN201910336630.9 proposes a continuous casting machine fixed weight cutting method and system, which comprises the following steps: first, determining the standard weight of the billet, calculating and setting the standard length of the billet according to the standard weight, weighing the billet cut according to the set standard length of the billet, comparing the weighed weight with the standard weight, and adjusting the cutting length of the next billet according to the comparison result; the system comprises a fixed-length cutting module, a weighing track control module, a fixed weight control module and a number and weight statistics module, the fixed-length cutting module is used for cutting the billet, the weighing roller control module is used for obtaining the weighed weight of the billet, the fixed weight control module is used for comparing the weighed weight with the standard weight, calculating the most reasonable cutting position, and transmitting the cutting position information to the fixed-length cutting module for cutting the billet. The present application effectively improves the length fluctuation of the rolling steel, improves the rolling yield of the steel, and reduces the labor intensity of the workers. However, the technical features of the fixed weight cutting model mainly adopt the method of entering the fixed weight control module with the weight value for fixed weight cutting, and the method of calculating the cognitive neural network model is not involved.

[0005] The Chinese patent with the application number CN202011165738.5 provides a continuous casting billet weight control cutting system and method, which includes: a continuous casting information acquisition device for acquiring continuous casting information related to the continuous casting machine; a temperature measuring device for acquiring temperature distribution information of the current billet being cut; a weighing device for measuring the weight of the current billet; a control module for determining the target cutting length of the next billet according to the continuous casting information, the temperature distribution information of the current billet, and the weight of the current billet; and a cutting module for cutting the next billet according to the target cutting length. The present application combines billet temperature distribution information when cutting the billet to weight, thereby improving the accuracy of setting the cutting length of the next billet, and effectively improving the accuracy of continuous casting billet weight cutting. However, the invention patent uses a weighing machine as the target cutting method for corresponding weight cutting, but it is not suitable for simulating weight cutting. SUMMARY

[0006] To solve the above problems, a weight cutting method is used to improve the yield of square billets. This method uses a weighing machine to accurately weigh the square billet after each casting flow, and adjusts the cutting length of the next time according to the result to ensure that the weight of the square billet cut each time is maintained within a stable range. However, over a long period of operation, the weighing machine needs frequent maintenance, and its accuracy and the accuracy of the weighing instrument may decrease over time, which often leads to inaccurate cutting results. To solve this problem, the present invention innovatively proposes a simulation weight cutting data processing method based on a cognitive intelligent process model, which realizes the effect of simulation weight cutting and ensures the accuracy of cutting, specifically including:

[0007] S1: First, install one hydraulic lifting platform scale at the cold bed conveying roller of the 8-machine 8-flow continuous casting machine per flow. To ensure production rhythm, the lifting time of the hydraulic lifting platform scale is controlled within 20S.

[0008] S2: After each weighing by the hydraulic platform scale, the weighing data is fed back to the weight cutting model. The weight cutting model uses the following mathematical calculation method to calculate the length: ρ*L*S=W

[0009] Where W is the weight of the hydraulic platform scale; ρ is the density of the billet; L is the length of the billet; and S is the cross-sectional area of the billet.

[0010] W and S are known, and ρ is the theoretical density, which is calculated to obtain the size of L. W is used to iteratively calculate the length of the billet each time.

[0011] The size of ρ is related to the square billet drawing speed, square billet nozzle size, square billet cooling water volume, square billet cooling temperature, etc. The calculation method is described in the subsequent steps.

[0012] S3: After the length of the billet is calculated, the length information is sent to the flame cutting machine for cutting. After each cutting, the billet scale is used for re-weighing, and the cutting length of the next billet is calculated.

[0013] Through this cycle, the weight of the square billet after cutting is stabilized within a very small range, realizing fixed weight cutting.

[0014] S4: After realizing fixed weight cutting and continuous program running for 3-4 months, the billet weight, billet number, square billet speed, square billet nozzle size, square billet cooling water quantity, square billet cooling temperature, and secondary cooling water quantity are collected during the process to obtain process data about square billet production within 3-4 months.

[0015] S5: Through the study of the dynamic relationship between the above variables and p, artificial intelligence data modeling is performed between the above process parameters and the actual weighing to accurately predict the billet weight after each production to replace the data of the weighing machine.

[0016] The specific steps include:

[0017] (1) For the analysis of 1 flow billets, the data of each billet of 1 flow is aligned in time, the square billet speed is granulated in time, and the square billet speed is granulated into 100 parts through the time required for the production of 1 flow billet (about 10 minutes), the list value of the speed [X, X, X, X, …, X] is obtained, and then the speed list value is used as the input of the neural network.

[0018] (2) For the square billet nozzle size, square billet cooling water quantity, square billet cooling temperature, and secondary cooling water quantity, the same method is used, the production time of each billet is granulated, and then the list value is obtained for the input of the neural network. Time granulation refers to sampling and time discretization of continuous time.

[0019] (3) The neural network uses a multi-dimensional feedforward neural network, which is different from the ordinary feedforward neural network in that it uses multi-dimensional numerical input. The matrix operation input expression f(WX+B)+δY is used for neural network reasoning inside the neural network. The input takes the speed list of each flow of each billet (100), the square billet nozzle size list (100), the square billet cooling water quantity list (100), the square billet cooling temperature list (100), and the four groups of numerical values as the input of a single node of the neural network for calculation.

[0020] Each input value adjusts the proportion of input data by adjusting the weight w, and the setting of the weight w is calculated by the cognitive computing engine. The calculation mode mainly adopts the steel grade process weight calculation library to calculate the corresponding calculation, and the correlation between the steel grade list and the size of the nozzle, the size of the cooling water, and the size of the cooling temperature is taken as the basis for calculation. The cognitive computing engine calculates the list to perform corresponding analysis and calculation, and obtains the process parameter weight value under the accurate steel grade, and then enters the corresponding neural network model calculation.

[0021] (4) The output of the neuron adopts the weighing value obtained by comparison and weighing as the output of the neural network.

[0022] (5) The weighing data AY11, AY21, AY31 obtained by weighing is fed back to the neuron, and delta Y is used as a compensation item to modify the model.

[0023] (6) The data of each billet in each flow is used for training, and the model is trained for about 3 months of production time, a total of 12000 billets, to obtain the neural network parameter value under the fitting effect, to obtain the value of W and B, and to obtain the simulation neural network data model under the condition of the flow.

[0024] (7) In the training process, the weighing data of the weighing body is compared with the simulated weighing data after the simulation of the fixed weight cutting data, and the model is dynamically refined and adjusted, and W and B values are stabilized through continuous training. The model is stabilized by determining the size of W and B.

[0025] (8) The weighing data of the weighing body is separated from the feedback link, and then the actual production data is connected, the model output value of the simulation neural network data is obtained by reasoning, the weight data of the simulation billet is obtained, and the corresponding cutting optimization information is fed back, so as to achieve the effect of simulation fixed weight cutting optimization.

[0026] Based on the simulation fixed weight cutting data processing method based on the cognitive intelligent process model proposed above, in order to better realize the present application, a simulation fixed weight cutting data processing system based on the cognitive intelligent process model is further proposed, comprising: a data collection module, a length calculation module, a billet cutting module, a billet weighing module, and a data prediction module.

[0027] The data collection module feeds the weighing data back to the fixed weight cutting model after each weighing of the platform scale.

[0028] The length calculation module is a fixed weight cutting model, which calculates the length of the billet by using a mathematical calculation formula.

[0029] The billet cutting module: the billet length information is sent to the flame cutting machine, the billet is cut, and the excess billet after cutting is obtained.

[0030] Billet weighing module: the remaining billet after cutting is weighed on the platform scale to calculate the length required for the next billet cutting;

[0031] Data prediction module: study the dynamic relationship between parameters and mathematical calculation formula to improve the prediction accuracy of weighing data.

[0032] Based on the above-mentioned simulation fixed weight cutting data processing method based on the cognitive intelligent process model, in order to better realize the present application, further put forward an electronic equipment, including memory and processor;The memory stores a computer program;When the computer program is executed on the processor, the simulation fixed weight cutting data processing method based on the cognitive intelligent process model is realized.

[0033] Based on the above-mentioned simulation fixed weight cutting data processing method based on the cognitive intelligent process model, in order to better realize the present application, further put forward a computer readable storage medium, the computer readable storage medium stores computer instructions;When the computer instructions are executed on the above-mentioned electronic equipment, the simulation fixed weight cutting data processing method based on the cognitive intelligent process model is realized. Advantages:

[0034] At the cold bed conveying roller of the casting machine, a hydraulic lifting platform scale is installed for each flow to accurately measure the weight of the billet. However, the weighing of the feedback scale is cancelled, and the simulation model is used for fixed weight cutting. After the hydraulic platform scale completes the weighing each time, the relevant data will be quickly fed back to the fixed weight cutting model, which uses a unique mathematical calculation method to accurately calculate the required cutting length. Further, by studying the dynamic relationship between the process parameters and the actual weighing, an accurate artificial intelligence data model is established using these relationships. This model can accurately predict the weight of the billet after each production, realizing a more efficient and intelligent production process. The present application not only eliminates the influence of the weighing machine on the production process, but also realizes the effect of simulation fixed weight and cutting by using virtual weighing data as signal feedback, significantly improving the production efficiency and product quality. BRIEF DESCRIPTION OF DRAWINGS

[0035] Fig. 1 is a multi-dimensional feedforward neural network in a simulation fixed weight cutting data processing method based on a cognitive intelligent process model. DETAILED DESCRIPTION

[0036] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0037] The present application will be described in further detail below with reference to the drawings:

[0038] At present, the cutting mode of the cutting model in the continuous casting industry production line mainly has two cutting modes of fixed size and fixed weight. The fixed size mode often causes the influence of the range size, resulting in the low product yield of the rolling steel. Therefore, in order to improve the yield of the square billet, the fixed weight cutting mode is adopted, that is, after each casting flow, the weighing machine is used to weigh the weight of the square billet to correct the cutting length of the next time, so that the weight of the square billet cut each time is within a stable range. However, since the weighing machine needs to be maintained for a long time, and the weighing machine precision and weighing instrument will decrease after a long time, the hardware often causes inaccurate factors in each cutting. Therefore, the present application adopts a simulated fixed weight cutting mode to perform data modeling, which can eliminate the weighing influence of the weighing machine, achieve the effect of virtual weighing data as signal feedback, and realize the effect of simulated fixed weight and cutting.

[0039] The method is as follows:

[0040] S1: First, one hydraulic lifting platform scale is installed at the cold bed conveying roller of the 8-machine 8-flow continuous casting machine. In order to ensure the production rhythm, the lifting time of the hydraulic lifting platform scale is controlled within 20S.

[0041] S2: After the hydraulic platform scale is lifted and weighed each time, the weighing data is fed back to the fixed weight cutting model. The fixed weight cutting model adopts the following mathematical calculation method to calculate the length: ρ*L*S=W

[0042] Wherein, W is the weight of the hydraulic platform scale; ρ is the density of the billet; L is the length of the billet; and S is the cross-sectional area of the billet.

[0043] W and S are known, and ρ is the theoretical density. The size of L is obtained by calculation. The length of the billet is iteratively calculated by W each time.

[0044] The size of ρ is related to the square billet drawing speed, the square billet nozzle size, the square billet cooling water quantity, the square billet cooling temperature and the like. The calculation method is shown in the subsequent steps.

[0045] S3: After the length of the billet is calculated, the length information is sent to the flame cutting machine for cutting. After each cutting, the billet scale is used for re-weighing, and the cutting length of the next billet is calculated.

[0046] Through this cycle, the weight of the square billet after cutting is stabilized within a very small range, realizing fixed weight cutting.

[0047] S4: After realizing fixed weight cutting and continuous running of the program for 3-4 months, the billet weight, billet number, square billet speed, square billet nozzle size, square billet cooling water quantity, square billet cooling temperature, and secondary cooling water quantity are collected during the process to obtain process data about square billet production for 3-4 months.

[0048] S5: The dynamic relationship between the above variables and p is studied to perform artificial intelligence data modeling between the above process parameters and the actual weighing to accurately predict the billet weight after each production to replace the data of the weighing machine.

[0049] The specific steps include:

[0050] (1) For the analysis of 1 flow billets, the data of each billet of 1 flow is aligned in time, the square billet speed is granulated in time, and the square billet speed is granulated into 100 parts through the time required for the production of 1 flow billet (about 10 minutes), the list value of the speed [X, X, X, X, …, X] is obtained, and then the speed list value is used as the input of the neural network.

[0051] (2) For the square billet nozzle size, square billet cooling water quantity, square billet cooling temperature, and secondary cooling water quantity, the same method is used, the production time of each billet is granulated, and then the list value is obtained for the input of the neural network. The time granulation refers to sampling the continuous time and then discretizing the time.

[0052] (3) As shown in FIG. 1, the neural network uses a multi-dimensional feedforward neural network, which is different from the ordinary feedforward neural network in that the neural network uses multi-dimensional numerical input. The matrix operation input expression f(WX+B)+δY is used inside the neural network to perform neural network reasoning. The input takes the speed list of each flow of each billet (100), the square billet nozzle size list (100), the square billet cooling water quantity list (100), the square billet cooling temperature list (100), and the four groups of numerical values as the input of a single node of the neural network for calculation.

[0053] Each input value adjusts the proportion of input data by adjusting the weight w, and the setting of the weight w is calculated by the cognitive computing engine. The calculation mode mainly adopts the steel grade process weight calculation library to calculate the corresponding calculation, and the correlation between the steel grade list and the size of the nozzle, the size of the cooling water, and the size of the cooling temperature is taken as the basis for calculation. Through the cognitive computing engine calculation list, the corresponding analysis calculation is carried out, and the process parameter weight value under the accurate steel grade is obtained, and then the corresponding neural network model calculation is entered.

[0054] (4) The output of the neuron adopts the weighing value obtained by comparison and weighing as the output of the neural network.

[0055] (5) The weighing data AY11, AY21, AY31 obtained by weighing is fed back to the neuron, and δY is used as a compensation item to modify the model.

[0056] (6) The data of each billet in each flow is used for training, and the model is trained for about 3 months of production time, a total of 12000 billets, to obtain the neural network parameter value under the fitting effect, obtain the value of W and B, and obtain the simulation neural network data model under the condition of the flow.

[0057] (7) In the training process, the weighing data of the weighing body is compared with the simulation weighing data after the simulation fixed weight cutting data, and the model is dynamically refined and adjusted, and W and B values are stabilized through continuous training. The size of W and B values is finally determined to stabilize the model.

[0058] (8) The weighing data of the weighing body is separated from the feedback link, and then the actual production data is connected, the model output value of the simulation neural network data is obtained by reasoning, the weight data of the simulation billet is obtained, and the corresponding cutting optimization information is fed back, so as to achieve the effect of simulation fixed weight cutting optimization.

[0059] The embodiment also provides a simulation fixed weight cutting data processing system based on a cognitive intelligent process model, which comprises a data collection module, a length calculation module, a billet cutting module, a billet weighing module and a data prediction module.

[0060] The data collection module feeds the weighing data to the fixed weight cutting model after each weighing of the platform scale.

[0061] The length calculation module is a fixed weight cutting model, which calculates the length of the billet by using a mathematical calculation formula.

[0062] The billet cutting module: the billet length information is sent to the flame cutting machine, the billet is cut, and the excess billet after cutting is obtained.

[0063] The billet weighing module: the excess billet after cutting is weighed by the platform scale, and the length required for the next billet cutting is calculated.

[0064] Data prediction module: study the dynamic relationship between parameters and mathematical calculation formula, improve the prediction accuracy of weighing data.

[0065] The embodiment further provides an electronic device, including a memory and a processor; the memory stores a computer program; when the computer program is executed on the processor, the simulation fixed weight cutting data processing method based on the cognitive intelligent process model is realized.

[0066] The embodiment further provides a computer readable storage medium, and the computer readable storage medium stores computer instructions; when the computer instructions are executed on the electronic device, the simulation fixed weight cutting data processing method based on the cognitive intelligent process model is realized.

[0067] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent substitution for part or all of the technical features; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the description of the present application.

Claims

1. A simulation of weight cut data processing method based on cognitive intelligence process model, characterized in that, The application relates to a hydraulic lifting platform scale installed at a cold bed conveying roller of a multi-flow continuous casting machine. The platform scale feeds weighing data to a fixed weight cutting model after each weighing; The fixed weight cutting model calculates the length of a steel billet by using a mathematical formula; The length information of the steel billet is fed to a flame cutting machine to cut the steel billet and obtain a residual steel billet after cutting; The residual steel billet after cutting is weighed by the platform scale to calculate the length required for the next steel billet cutting; The program is ensured to be continuous and cyclic, and process parameters are collected; The dynamic relationship between the parameters and the mathematical formula is studied to improve the prediction accuracy of the weighing data. The lifting time of the hydraulic lifting platform scale is controlled within 20s.

2. The method of claim 1, wherein the method is based on a cognitive intelligence process model. The mathematical formula is: rho*L*S=W; 3. The method of claim 1, wherein the method further comprises: W is the weight of the hydraulic platform scale, rho is the density of the steel billet, L is the length of the steel billet, and S is the cross-sectional area of the steel billet; The size of L is calculated, and the length of the steel billet is iteratively calculated by W each time. The parameters include the weight of the steel billet each time, the steel billet number, the square billet drawing speed, the square billet nozzle size, the square billet cooling water volume, the square billet cooling temperature and the two-cooling water volume of the square billet production process data information.

4. The method of claim 1, wherein the method further comprises: The parameters and the artificial intelligence data modeling between the actual weighing are used to accurately predict the weight of the steel billet after each production to replace the data of the platform scale.

5. The method of claim 1, wherein the method further comprises: The parameters of each flow are analyzed, the data of the parameters of each flow are aligned in time, the square billet drawing speed is granulated in time, the list value of the drawing speed is obtained, a neural network model is constructed, and the list value of the drawing speed is used as the input of the neural network.

6. The method of claim 5, wherein the method further comprises: The neural network model adopts a multi-dimensional feedforward neural network; 7. The method of claim 6, wherein the method further comprises: The neural network model uses a matrix operation input expression of f(WX+B)+delta Y to perform neural network reasoning; The delta Y is used as a compensation item to correct the neural network model. The output of the neural network model is a weighing value obtained by comparing neurons.

8. The simulation of re-cutting data processing method based on the cognitive intelligent process model according to claim 6 or 7, characterized in that, The neural network model is trained by using the data of each steel billet to obtain the neural network parameter value under the fitting effect, the values of W and B, and a simulated neural network data model is constructed.

9. The method of claim 6, wherein the simulation of the data processing of the re-cutting process based on the cognitive intelligence process model is characterized by, During the neural network model training process, the weighing data of the weighing body are compared with the simulated weighing data after the simulated fixed weight cutting data, the dynamic adjustment of the neural network model is performed, and the model tends to be stable; 10. The method of claim 6, wherein the simulation of the data processing of the re-cutting process based on the cognitive intelligence process model is characterized by, The weighing data of the weighing body are separated from the feedback link and connected to the actual production data to obtain the model output value of the simulated neural network data; The weight data of the simulated steel billet are obtained, and optimization information feedback is performed. The application relates to a hydraulic lifting platform scale installed at a cold bed conveying roller of a multi-flow continuous casting machine.

11. A simulated fixed weight cutting data processing system based on a cognitive intelligence process model, characterized by The platform scale feeds weighing data to a fixed weight cutting model after each weighing; The fixed weight cutting model calculates the length of a steel billet by using a mathematical formula; The length information of the steel billet is fed to a flame cutting machine to cut the steel billet and obtain a residual steel billet after cutting; ​ ​ The billet weighing module: the cut billet is weighed on the platform scale, and the length required for the next billet cutting is calculated; The data prediction module: the dynamic relationship between the parameters and the mathematical calculation formula is studied, and the prediction accuracy of the weighing data is improved.

12. An electronic device, comprising: The computer readable storage medium stores computer instructions; when the computer instructions are executed on the electronic device of claim 12, the simulation cutting data processing method based on the cognitive intelligent process model of any one of claims 1-10 is realized.

13. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions; when the computer instructions are executed on the electronic device of claim 12, the simulation cutting data processing method based on the cognitive intelligent process model of any one of claims 1-10 is realized.

Images