Non-oriented silicon steel plate shape control method

Abstract

The invention discloses a non-oriented silicon steel plate shape control method, and belongs to the technical field of cold rolling. According to the method, the core control factors with significant influence are screened out, data acquisition and statistical analysis are carried out based on the six-sigma management idea, a mathematical model between the finished strip steel unevenness and the core control factors is constructed, and the unevenness of the finished strip steel is determined according to the differentiated requirements for the unevenness of the strip steel. Reversely deducing and optimizing the parameter combination of the core control factors through the mathematical model, and carrying out the coordinated regulation and control of the cold rolling process and the continuous annealing process; by combining six-sigma data analysis with multiple regression modeling, nonlinear influences of parameters can be dynamically captured, after different clients are classified according to unevenness requirements, unevenness values required by the clients are determined, the taper of a first middle roller, the convexity of a second middle roller and the opening degree of an air door are changed, and then the maximum running speed of a finished steel coil is obtained. On the premise of meeting the use requirements of customers, the production efficiency is improved, the production cost is reduced, and the product competitiveness is improved.

Description

Technical Field

[0001] This invention relates to the field of cold rolling technology, and in particular to a method for controlling the shape of non-oriented silicon steel sheets. Background Technology

[0002] Cold-rolled non-oriented silicon steel is widely used as the core material for industrial motors, compressors, and drive motors in new energy vehicles. Currently, the stamping speed of high-speed presses is increasing, and the stacking height of stator and rotor cores is also increasing, which places increasingly higher demands on the sheet shape of cold-rolled non-oriented silicon steel. Currently, users evaluate the sheet shape of non-oriented silicon steel primarily based on strip flatness. Strips with significant flatness will have gaps between the stator and rotor laminations after stamping, affecting the dimensional accuracy of the stator and rotor. Significant flatness can even cause the press to jam and stop.

[0003] Currently, the main processing steps for cold-rolled non-oriented silicon steel are steelmaking, hot rolling, normalizing, cold rolling, annealing, and rewinding. The main processes affecting the flatness of the finished strip are cold rolling and annealing. Existing flatness improvement technologies mainly focus on optimizing the shape control technology during cold rolling, with fewer technologies for improving flatness during the annealing process. Furthermore, research on the improvement of flatness through cold rolling and continuous annealing has been conducted independently. Summary of the Invention

[0004] The purpose of this invention is to provide a method for controlling the shape of non-oriented silicon steel sheets to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for controlling the shape of non-oriented silicon steel sheets, comprising the following steps:

[0006] With the goal of precisely controlling the flatness of cold-rolled non-oriented silicon steel strip, we systematically analyzed and sorted out the key factors affecting the flatness of the strip throughout the entire process, and screened out the core control factors with significant influence.

[0007] Based on the Six Sigma management concept, data collection and statistical analysis were carried out to construct a mathematical model between the flatness of finished strip steel and core control factors.

[0008] Based on the differentiated requirements for strip flatness, the parameter combination of core control factors is derived and optimized through mathematical modeling to achieve coordinated control of cold rolling and continuous annealing processes.

[0009] Furthermore, the key factors affecting the flatness of the strip steel throughout the entire process include steelmaking, hot rolling, normalizing, cold rolling, annealing, and rewinding processes.

[0010] The core control factors selected include cold rolling process-related factors and continuous annealing process-related factors. Among them, the cold rolling process-related factors are the taper of the first intermediate roll and the convexity of the second intermediate roll, and the continuous annealing process-related factors are the opening degree of the cooling system damper and the running speed of the strip in the annealing furnace.

[0011] The core control factors work together to affect the plastic elongation, thermal stress distribution, and residual stress state of the strip.

[0012] Furthermore, the first intermediate roll is a tapered roll, which changes the effective support length of the work roll at the edge by axial movement, thereby adjusting the elongation rate of the strip edge region; the second intermediate roll is equipped with macroscopic crown, which sets the crown value to shape the overall profile of the bearing roll gap, affecting the elongation rate of the strip center region; the damper opening is used to control the uniformity of the cooling rate in the continuous annealing cooling section; the strip running speed affects the overall cooling efficiency and longitudinal cooling uniformity.

[0013] The taper of the first intermediate roll ranges from 0.3 to 0.9 mm, the crown of the second intermediate roll ranges from 200 to 1200 μm, the damper opening ranges from 30% to 100%, and the strip running speed ranges from 80 to 150 m / min.

[0014] Furthermore, the cold rolling process also includes a real-time monitoring step for the first and second intermediate rolls: monitoring the roll surface temperature, wear amount, and roundness deviation. When the roll surface temperature exceeds 80°C, the wear amount exceeds 0.1mm, or the roundness deviation exceeds 0.05mm, the rolls are cooled, ground, or replaced.

[0015] Furthermore, the continuous annealing process also includes a cooling medium status control step: monitoring the temperature, pressure, and cleanliness of the cooling medium, maintaining the temperature fluctuation range within ±5℃ and the pressure fluctuation range within ±0.1MPa, periodically filtering the cooling medium, and ensuring that the cleanliness level of the cooling medium is not lower than NAS8.

[0016] Furthermore, the data collection and statistical analysis based on Six Sigma management principles includes the following specific processes:

[0017] The sample size, collection frequency, and collection accuracy of data collection should be clearly defined. At least 30 sets of valid samples should be collected for each core control factor parameter combination. The collection frequency should be 1 set of data for every 10 coils of strip steel produced. The collection accuracy should be matched with the adjustment accuracy of each core control factor.

[0018] Among them, the acquisition accuracy of the taper of the first intermediate roll is 0.001mm, the acquisition accuracy of the crown of the second intermediate roll is 1µm, the acquisition accuracy of the damper opening is 0.5%, the acquisition accuracy of the strip running speed is 0.1m / min, and the acquisition accuracy of the flatness of the finished strip is 0.01%.

[0019] Conduct full-condition data collection: covering the main product specifications of non-oriented silicon steel, collecting the actual parameter values ​​of core control factors and the corresponding flatness detection values ​​of finished strip steel under different production conditions, including different raw material compositions, different rolling forces and different annealing temperatures;

[0020] During the data collection process, production environment parameters need to be recorded simultaneously, including workshop temperature, humidity, and power grid voltage fluctuations.

[0021] After preprocessing the collected raw data, valid data were obtained. The effective data were analyzed using the analysis of variance method to calculate the contribution rate of each core control factor to the flatness of the finished strip steel. Factors with a contribution rate of ≥8% were selected as the final modeling variables.

[0022] Furthermore, a mathematical model is constructed to connect the flatness of finished strip steel with the core control factors. Specifically, a combination of multiple linear regression and quadratic term fitting is used to construct the model based on the effective data. The specific process includes:

[0023] The dependent variable Y is set as the flatness of the cold-rolled finished strip, and the independent variables are the taper of the first intermediate roll, the crown of the second intermediate roll, the opening degree of the damper, and the running speed of the strip.

[0024] Construct a multiple linear regression model to analyze the linear relationship between the independent variables and the dependent variable;

[0025] Based on the residual analysis results, a significant quadratic term is introduced to construct a mixed regression model containing linear and quadratic terms;

[0026] The constructed model is validated using cross-validation. The valid data is randomly divided into a training set and a validation set. The training set data is used to fit the model, and the validation set data is used to test the model's prediction accuracy.

[0027] The final mathematical model was determined through fitting and verification.

[0028] Furthermore, the process for detecting and calculating the flatness of the cold-rolled finished strip steel specifically includes:

[0029] For sample selection, three test sections are selected from the head, middle and tail of each coil of finished strip steel, and five test points are evenly selected from each section along the width of the strip steel.

[0030] Unevenness is detected using a laser strip shaper. During the detection, the laser strip shaper scans at a constant speed along the length of the strip and records the actual height value of each detection point.

[0031] Based on the height value obtained from the detection, identify the wavy defects on the surface of the strip steel, determine the position of the wavy crests and troughs, and measure the vertical distance between the wavy crests and troughs as well as the distance between adjacent crests.

[0032] The unevenness value at each detection point is calculated, and the average value of all unevenness values ​​at all detection points is taken as the final unevenness of the coiled steel.

[0033] Furthermore, it also includes dynamic adjustment steps in the production process: real-time detection of the actual unevenness of the strip during production. If the deviation between the actual unevenness and the target value exceeds ±0.02%, the key influencing factors are fine-tuned based on the mathematical model, and the fine-tuning range does not exceed 5% of the value range of each factor.

[0034] Furthermore, based on the differentiated requirements for strip unevenness, the parameter combination of the core control factors is derived and optimized through mathematical modeling, specifically including:

[0035] Based on different users' requirements for strip flatness, precision requirements are divided into high precision requirements, medium precision requirements, and ordinary precision requirements. The flatness of finished strip steel with high precision requirements is ≤0.65%; the flatness range of finished strip steel with medium precision requirements is (0.65%, 1.2%); and the flatness range of finished strip steel with ordinary precision requirements is (1.2%, 1.65%).

[0036] Establish requirement files for different accuracy needs and clarify the target unevenness requirements for different users;

[0037] Substitute the target value of unevenness into the mathematical model, and combine the value range of each core control factor to construct a multivariate constrained optimization problem;

[0038] The particle swarm optimization algorithm is used to solve multivariable constrained optimization problems;

[0039] The optimal parameter combination obtained by the solution is verified by an experiment, and the unevenness of the experimental product is detected.

[0040] The verified optimal parameter combination is transformed into specific production operation instructions, specifying the taper setting value of the first intermediate roll and the convex setting value of the second intermediate roll in the cold rolling process, as well as the adjustment value of the damper opening and the control value of the strip running speed in the continuous annealing process.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] 1. By focusing on the key processes of cold rolling and continuous annealing, locking in core parameters such as the taper of the first intermediate roll, and combining real-time monitoring of the rolls with precise control of the cooling medium, this invention can effectively avoid strip shape defects caused by roll system abnormalities and uneven cooling, reduce strip flatness fluctuations, improve roll gap stiffness stability, and solve the limitations of single-process adjustment.

[0043] 2. This invention combines Six Sigma data analysis with multiple regression modeling, collects full-condition data according to scientific standards and screens high-contribution factors, and then constructs a hybrid regression model with quadratic terms using MINITAB. This reduces model prediction bias, improves real-time fine-tuning response, and enhances fitting accuracy. It can dynamically capture the nonlinear effects of parameters, avoid the problem of lag in manual adjustment, and achieve accurate prediction and dynamic control of unevenness, thus enabling more effective control of strip unevenness.

[0044] 3. This invention classifies users by combining user demand classification with particle swarm optimization. Users are divided into three categories according to their flatness requirements. Different customers have different flatness requirements for cold-rolled finished strip steel. After clarifying the flatness value required by the customer, the maximum running speed of the finished steel coil is obtained by changing the taper of the first intermediate roll, the convexity of the second intermediate roll, and the opening of the damper. Under the premise of meeting the customer's usage requirements, this invention improves production efficiency, reduces production costs, and enhances product competitiveness. Attached Figure Description

[0045] Figure 1 This is a schematic diagram illustrating the model optimization of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Please see Figure 1 The present invention provides the following technical solutions:

[0048] A method for controlling the shape of non-oriented silicon steel sheets includes the following steps:

[0049] With the goal of precisely controlling the flatness of cold-rolled non-oriented silicon steel strip, we systematically analyzed and sorted out the key factors affecting the flatness of the strip throughout the entire process, and screened out the core control factors with significant influence.

[0050] Based on the Six Sigma management concept, data collection and statistical analysis were carried out to construct a mathematical model between the flatness of finished strip steel and core control factors.

[0051] Based on the differentiated requirements for strip flatness, the parameter combination of core control factors is derived and optimized through mathematical modeling to achieve coordinated control of cold rolling and continuous annealing processes.

[0052] The key factors affecting the flatness of strip steel throughout the entire process include steelmaking, hot rolling, normalizing, cold rolling, annealing, and recoiling.

[0053] Through significance analysis and practical verification, the core control factors selected include cold rolling process-related factors and continuous annealing process-related factors. Among them, the cold rolling process-related factors are the taper of the first intermediate roll and the convexity of the second intermediate roll, and the continuous annealing process-related factors are the opening degree of the cooling system damper and the running speed of the strip in the annealing furnace.

[0054] The core control factors work together to affect the plastic elongation, thermal stress distribution, and residual stress state of the strip steel;

[0055] The first intermediate roll is a tapered roll, which changes the effective support length of the work roll at the edge by axial movement, thereby adjusting the elongation of the strip edge region; the second intermediate roll is equipped with macroscopic crown, which shapes the overall profile of the bearing roll gap by setting the crown value, affecting the elongation of the strip center region; the damper opening determines the cooling intensity distribution along the width direction of the strip, and is used to control the uniformity of the cooling rate in the continuous annealing cooling section; the strip running speed determines the heat exposure time of the strip in the cooling section, affecting the overall cooling efficiency and longitudinal cooling uniformity;

[0056] The taper of the first intermediate roll ranges from 0.3 to 0.9 mm, the crown of the second intermediate roll ranges from 200 to 1200 μm, the damper opening ranges from 30% to 100%, and the strip running speed ranges from 80 to 150 m / min.

[0057] The cold rolling process also includes a real-time monitoring step for the first and second intermediate rolls: monitoring the roll surface temperature, wear amount, and roundness deviation. When the roll surface temperature exceeds 80°C, the wear amount exceeds 0.1mm, or the roundness deviation exceeds 0.05mm, the rolls are cooled, ground, or replaced to ensure the stability of the transverse stiffness distribution of the roll gap.

[0058] The continuous annealing process also includes a cooling medium condition control step: monitoring the temperature, pressure and cleanliness of the cooling medium, keeping the temperature fluctuation range within ±5℃ and the pressure fluctuation range within ±0.1MPa, filtering the cooling medium regularly, and ensuring that the cleanliness level of the cooling medium is not lower than NAS8.

[0059] In the above embodiments, by systematically sorting out the key factors of the entire process and screening the core control factors, the two key processes of cold rolling and continuous annealing can be targeted and focused on. This avoids the limitations of strip shape control caused by neglecting the synergistic effect between processes. By taking parameters such as the taper of the first intermediate roll and the crown of the second intermediate roll as core factors, they can be accurately applied to the key links of strip elongation and thermal stress distribution. At the same time, combined with real-time monitoring of the rolls and control of the cooling medium, strip shape defects caused by roll system abnormalities and uneven cooling can be effectively avoided, the range of strip unevenness fluctuation can be reduced, the stability of the roll gap lateral stiffness can be improved, and the cooling uniformity satisfaction rate can reach more than 98%. This lays a stable process foundation for subsequent mathematical modeling and parameter optimization, which is significantly better than the control effect of traditional single-process adjustment.

[0060] Data collection and statistical analysis based on Six Sigma management principles include the following specific processes:

[0061] The sample size, collection frequency, and collection accuracy of data collection should be clearly defined. At least 30 sets of valid samples should be collected for each core control factor parameter combination. The collection frequency should be 1 set of data for every 10 coils of strip steel produced. The collection accuracy should be matched with the adjustment accuracy of each core control factor.

[0062] Among them, the acquisition accuracy of the taper of the first intermediate roll is 0.001mm, the acquisition accuracy of the crown of the second intermediate roll is 1µm, the acquisition accuracy of the damper opening is 0.5%, the acquisition accuracy of the strip running speed is 0.1m / min, and the acquisition accuracy of the flatness of the finished strip is 0.01%.

[0063] Conduct full-condition data collection: covering the main product specifications of non-oriented silicon steel (thickness 0.15~0.5mm, width 800~1250mm), collecting the actual parameter values ​​of core control factors and the corresponding flatness detection values ​​of finished strip steel under different production conditions, including different raw material compositions, different rolling forces and different annealing temperatures;

[0064] During the data collection process, production environment parameters need to be recorded simultaneously, including workshop temperature, humidity, and power grid voltage fluctuations, for subsequent data anomaly investigation.

[0065] After preprocessing the collected raw data to obtain valid data, the effective data were analyzed using the analysis of variance method. The contribution rate of each core control factor to the flatness of the finished strip steel was calculated, and factors with a contribution rate of ≥8% were selected as the final modeling variables to ensure the effectiveness and simplicity of the model.

[0066] In the above embodiments, the data analysis process based on Six Sigma management principles ensures the scientific validity and effectiveness of the collected data, clarifies the requirements for sample size, frequency, and accuracy, avoids random data bias, covers all working conditions and environmental parameter records, eliminates interference from raw materials, working conditions, and the environment, provides reliable data support for model construction, and selects high contribution factors through preprocessing and variance analysis, which can eliminate interference from irrelevant variables, making the modeling variables more accurate, increasing the proportion of effective data to over 95%, and achieving an accuracy rate of 92% in identifying the contribution rate of core factors. This ensures the high fit and practicality of the subsequent mathematical model and significantly reduces the risk of model bias caused by data quality issues.

[0067] A mathematical model was constructed to connect the unevenness of finished steel strip with core control factors. Specifically, MINITAB statistical analysis software was used, employing a combination of multiple linear regression and quadratic term fitting to build the model on valid data. The specific process included:

[0068] The dependent variable Y is set as the flatness (%) of the cold-rolled finished strip, and the independent variables are the taper of the first intermediate roll (X1), the crown of the second intermediate roll (X2), the opening degree of the damper (X3), and the running speed of the strip (X4).

[0069] Construct a multiple linear regression model to analyze the linear relationship between the independent variables and the dependent variable;

[0070] Based on the residual analysis results, a significant quadratic term (X2) is then introduced. 2 X3 2 To improve the model's fitting accuracy, a mixed regression model containing linear and quadratic terms is constructed.

[0071] The constructed model was validated using cross-validation. The valid data was randomly divided into a training set (70%) and a validation set (30%). The training set data was used to fit the model, and the validation set data was used to test the model's predictive accuracy.

[0072] After fitting and verification, the final mathematical model is as follows:

[0073] Y=1.881-1.758X1-0.002773X2-0.0301X3+0.01816X4+0.000003X2 2 +0.000299X3 2 The coefficients are obtained by solving the least squares method, and the determination coefficient R of the model is... 2 A value of ≥0.92 indicates that the model can explain more than 92% of the unevenness variation of finished strip steel, and has high practical value.

[0074] The actual unevenness of the strip steel is detected in real time during the production process. If the deviation between the actual unevenness and the target value exceeds ±0.02%, the key influencing factors are fine-tuned based on the mathematical model. The fine-tuning range does not exceed 5% of the value range of each factor.

[0075] The process for detecting and calculating the flatness of cold-rolled finished strip steel includes:

[0076] For sample selection, three test sections are selected from the head, middle and tail of each coil of finished strip steel, and five test points are evenly selected from each section along the width of the strip steel.

[0077] Unevenness is detected using a laser strip shaper. During the detection, the laser strip shaper scans at a constant speed along the length of the strip and records the actual height value of each detection point.

[0078] Based on the height value obtained from the detection, identify the wavy defects on the surface of the strip steel, determine the position of the wavy crests and troughs, and measure the vertical distance between the wavy crests and troughs as well as the distance between adjacent crests.

[0079] The unevenness value at each detection point is calculated, and the average value of all unevenness values ​​at all detection points is taken as the final unevenness of the coiled steel.

[0080] In the above embodiments, the modeling method combining multiple linear regression and quadratic term fitting can fully fit the complex relationship between core factors and unevenness. The introduction of quadratic terms can accurately capture the nonlinear influence of convexity and damper opening on unevenness. Compared with a single linear model, the fitting accuracy is improved. Cross-validation and high coefficient of determination can ensure the accuracy of model prediction. The real-time fine-tuning mechanism can quickly respond to production fluctuations and avoid unevenness exceeding tolerance. The standardized unevenness detection process can stably control the detection error, and the data reliability is significantly improved. After the model is applied, the average deviation between the predicted and actual unevenness values ​​is ≤3%, and the real-time adjustment response time is shortened to within 1 minute, effectively solving the problems of lag and low accuracy of traditional manual adjustment, and realizing dynamic and precise control of plate shape.

[0081] Based on the differentiated requirements for strip unevenness, the parameter combination of core control factors is derived and optimized through mathematical modeling, specifically including:

[0082] Based on different users' requirements for strip flatness, precision requirements are divided into high precision, medium precision, and ordinary precision requirements. High precision requirements require a finished strip flatness of ≤0.65%; medium precision requirements require a finished strip flatness range of (0.65%, 1.2%); and ordinary precision requirements require a finished strip flatness range of (1.2%, 1.65%).

[0083] Establish requirement files for different accuracy needs and clarify the target unevenness requirements for different users;

[0084] Substitute the target value of unevenness into the mathematical model, and combine the value range of each core control factor to construct a multivariate constrained optimization problem;

[0085] The particle swarm optimization algorithm is used to solve the multivariate constrained optimization problem. The optimization objective is to maximize the strip running speed while ensuring that the flatness of the finished strip is less than or equal to the target flatness value, so as to improve the production efficiency of the continuous annealing unit. The number of particles in the algorithm is set to 50, the number of iterations is set to 100, and the inertia weight is set to a value range of 0.4 to 0.9 to ensure that the algorithm can converge to the optimal solution quickly.

[0086] The optimal parameter combination obtained by the solution is verified by an experiment, and the unevenness of the experimental product is detected.

[0087] The validated optimal parameter combinations are translated into specific production operation instructions, specifying the taper setting value of the first intermediate roll and the crown setting value of the second intermediate roll in the cold rolling process, as well as the adjustment value of the damper opening and the control value of the strip running speed in the continuous annealing process. These instructions are then issued to the production workshop to ensure that operators can execute them accurately.

[0088] In the above embodiments, classifying users according to their precision requirements and optimizing parameter combinations accordingly enables differentiated and precise services. Establishing files for different needs clarifies the target orientation. Combined with particle swarm optimization algorithms, production speed is maximized while meeting unevenness requirements. Compared with fixed parameter production mode, this improves the production efficiency of continuous annealing units. Verification tests and operation command conversion processes ensure the accuracy of parameter implementation, avoiding the disconnect between theoretical optimization and actual production. The unevenness compliance rate of products with different precision requirements is consistently above 98%. This not only guarantees the high precision requirements of high-end customers (such as in the field of new energy vehicle motors) but also reduces production costs for ordinary customers, significantly improving product market competitiveness and customer satisfaction.

[0089] The method will be further illustrated below through two examples:

[0090] Example 1:

[0091] For the production of WN series silicon steel products, the customer uses WN series silicon steel to manufacture large motors using oscillating stamping, which has relatively low requirements for strip flatness control. After confirming that the customer requires an flatness of less than 1.65%, the taper of the first intermediate roll, the crown of the second intermediate roll, and the damper opening were adjusted using a model. It was determined that to meet the 1.65% strip flatness requirement, the following settings were used: taper of the first intermediate roll X1 = 0.8mm, crown of the second intermediate roll X2 = 500um, and damper opening 48%. This resulted in a maximum strip running speed of 145m / min. In actual production, the following settings were used: taper of the first intermediate roll 0.8mm, crown of the second intermediate roll 500mm, damper opening 48%, and strip running speed 145m / min. The final measured flatness of the finished coil was 1.63%, which is basically consistent with the theoretical value calculated by the model. While meeting the customer's requirements, the continuous annealing unit's production speed was increased by 25m / min, production efficiency was improved, and the overall yield was increased by 20.8%, significantly reducing manufacturing costs.

[0092] Example 2:

[0093] For the production of new energy silicon steel series, customers have high requirements for the level of strip flatness control. To ensure the flatness control level of the finished steel coil meets customer requirements (<0.65%) while maintaining a continuous annealing unit production speed of 100m / min, the model was used to optimize and adjust the taper of the first intermediate roll, the crown of the second intermediate roll, and the damper opening. It was found that increasing the taper of the first intermediate roll can maintain the production efficiency of the continuous annealing unit. The optimal production scheme is: first intermediate roll taper X1 = 0.90mm, second intermediate roll crown X2 = 500um, damper opening X3 = 48%, and continuous annealing production speed X4 = 100m / min. The calculated strip flatness under this scheme is 0.63%, which meets customer requirements. After actual production using this scheme, the measured thickness deviation of the finished coil was 0.61%, which is basically consistent with the model calculation.

[0094] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for controlling the shape of non-oriented silicon steel sheets, characterized in that, Includes the following steps: With the goal of precisely controlling the flatness of cold-rolled non-oriented silicon steel strip, we systematically analyzed and sorted out the key factors affecting the flatness of the strip throughout the entire process, and screened out the core control factors with significant influence. Based on the Six Sigma management concept, data collection and statistical analysis were carried out to construct a mathematical model between the flatness of finished strip steel and core control factors. Based on the differentiated requirements for strip flatness, the parameter combination of core control factors is derived and optimized through mathematical modeling to achieve coordinated control of cold rolling and continuous annealing processes.

2. The non-oriented silicon steel sheet shape control method as described in claim 1, characterized in that, The key factors affecting the flatness of strip steel throughout the entire process include steelmaking, hot rolling, normalizing, cold rolling, annealing, and recoiling processes. The core control factors selected include cold rolling process-related factors and continuous annealing process-related factors. Among them, the cold rolling process-related factors are the taper of the first intermediate roll and the convexity of the second intermediate roll, and the continuous annealing process-related factors are the opening degree of the cooling system damper and the running speed of the strip in the annealing furnace. The core control factors work together to affect the plastic elongation, thermal stress distribution, and residual stress state of the strip.

3. The non-oriented silicon steel sheet shape control method as described in claim 2, characterized in that, The first intermediate roll is a tapered roll, which changes the effective support length of the work roll at the edge by axial movement, thereby adjusting the elongation of the strip edge region; the second intermediate roll is equipped with macroscopic crown, which shapes the overall profile of the bearing roll gap by setting the crown value, affecting the elongation of the strip center region; the damper opening is used to control the uniformity of the cooling rate in the continuous annealing cooling section; the strip running speed affects the overall cooling efficiency and longitudinal cooling uniformity. The taper of the first intermediate roll ranges from 0.3 to 0.9 mm, the crown of the second intermediate roll ranges from 200 to 1200 μm, the damper opening ranges from 30% to 100%, and the strip running speed ranges from 80 to 150 m / min.

4. The non-oriented silicon steel sheet shape control method as described in claim 2, characterized in that, The cold rolling process also includes a real-time monitoring step for the first and second intermediate rolls: monitoring the roll surface temperature, wear amount, and roundness deviation. When the roll surface temperature exceeds 80°C, the wear amount exceeds 0.1mm, or the roundness deviation exceeds 0.05mm, the rolls are cooled, ground, or replaced.

5. The method for controlling the shape of non-oriented silicon steel sheets as described in claim 2, characterized in that, The continuous annealing process also includes a cooling medium condition control step: monitoring the temperature, pressure and cleanliness of the cooling medium, keeping the temperature fluctuation range within ±5℃ and the pressure fluctuation range within ±0.1MPa, filtering the cooling medium regularly, and ensuring that the cleanliness level of the cooling medium is not lower than NAS8.

6. The method for controlling the shape of non-oriented silicon steel sheets as described in claim 1, characterized in that, The data collection and statistical analysis based on Six Sigma management principles includes the following specific processes: The sample size, collection frequency, and collection accuracy of data collection should be clearly defined. At least 30 sets of valid samples should be collected for each core control factor parameter combination. The collection frequency should be 1 set of data for every 10 coils of strip steel produced. The collection accuracy should be matched with the adjustment accuracy of each core control factor. Among them, the acquisition accuracy of the taper of the first intermediate roll is 0.001mm, the acquisition accuracy of the crown of the second intermediate roll is 1µm, the acquisition accuracy of the damper opening is 0.5%, the acquisition accuracy of the strip running speed is 0.1m / min, and the acquisition accuracy of the flatness of the finished strip is 0.01%. Conduct full-condition data collection: covering the main product specifications of non-oriented silicon steel, collecting the actual parameter values ​​of core control factors and the corresponding flatness detection values ​​of finished strip steel under different production conditions, including different raw material compositions, different rolling forces and different annealing temperatures; During the data collection process, production environment parameters need to be recorded simultaneously, including workshop temperature, humidity, and power grid voltage fluctuations. After preprocessing the collected raw data, valid data were obtained. The effective data were analyzed using the analysis of variance method to calculate the contribution rate of each core control factor to the flatness of the finished strip steel. Factors with a contribution rate of ≥8% were selected as the final modeling variables.

7. The method for controlling the shape of non-oriented silicon steel sheets as described in claim 1, characterized in that, A mathematical model is constructed to connect the unevenness of finished strip steel with core control factors. Specifically, a combination of multiple linear regression and quadratic term fitting is used to build the model on effective data. The specific process includes: The dependent variable Y is set as the flatness of the cold-rolled finished strip, and the independent variables are the taper of the first intermediate roll, the crown of the second intermediate roll, the opening degree of the damper, and the running speed of the strip. Construct a multiple linear regression model to analyze the linear relationship between the independent variables and the dependent variable; Based on the residual analysis results, a significant quadratic term is introduced to construct a mixed regression model containing linear and quadratic terms; The model was validated using cross-validation. The effective data was randomly divided into training and validation sets. The training set data was used to fit the model, and the validation set data was used to test the model's prediction accuracy. The final mathematical model was determined through fitting and verification.

8. The non-oriented silicon steel sheet shape control method as described in claim 7, characterized in that, The process for detecting and calculating the flatness of cold-rolled finished strip steel specifically includes: For sample selection, three test sections are selected from the head, middle and tail of each coil of finished strip steel, and five test points are evenly selected from each section along the width of the strip steel. Unevenness is detected using a laser strip shaper. During the detection, the laser strip shaper scans at a constant speed along the length of the strip and records the actual height value of each detection point. Based on the height value obtained from the detection, identify the wavy defects on the surface of the strip steel, determine the position of the wavy crests and troughs, and measure the vertical distance between the wavy crests and troughs as well as the distance between adjacent crests. The unevenness value at each detection point is calculated, and the average value of all unevenness values ​​at all detection points is taken as the final unevenness of the coiled steel.

9. The non-oriented silicon steel sheet shape control method as described in claim 7, characterized in that, It also includes dynamic adjustment steps in the production process: real-time detection of the actual flatness of the strip during production. If the deviation between the actual flatness and the target value exceeds ±0.02%, the key influencing factors are fine-tuned based on the mathematical model, and the fine-tuning range does not exceed 5% of the value range of each factor.

10. The method for controlling the shape of non-oriented silicon steel sheets as described in claim 1, characterized in that, Based on the differentiated requirements for strip unevenness, the parameter combination of core control factors is derived and optimized through mathematical modeling, specifically including: Based on different users' requirements for strip flatness, precision requirements are divided into high precision requirements, medium precision requirements, and ordinary precision requirements. The flatness of finished strip steel with high precision requirements is ≤0.65%; the flatness range of finished strip steel with medium precision requirements is (0.65%, 1.2%); and the flatness range of finished strip steel with ordinary precision requirements is (1.2%, 1.65%). Establish requirement files for different accuracy needs and clarify the target unevenness requirements for different users; Substitute the target value of unevenness into the mathematical model, and combine the value range of each core control factor to construct a multivariate constrained optimization problem; The particle swarm optimization algorithm is used to solve multivariable constrained optimization problems; The optimal parameter combination obtained by the solution is verified by an experiment, and the unevenness of the experimental product is detected. The verified optimal parameter combination is transformed into specific production operation instructions, specifying the taper setting value of the first intermediate roll and the convex setting value of the second intermediate roll in the cold rolling process, as well as the adjustment value of the damper opening and the control value of the strip running speed in the continuous annealing process.

Images