METHOD FOR PRODUCING COLD ROLLED STEEL STRIP

Field of the invention

This invention relates to a method for producing cold-rolled steel strip.

Background of the invention

For some applications the mechanical properties are paramount. For instance, for structural applications in constructions, it is particularly important that the mechanical properties such as tensile strength, yield strength and elongation values are at a certain level to allow the application of the material to function reliably and safely in a construction such as in a building or in a vehicle.

In other cases, the surface properties are paramount. For instance, for applications where the material is subjected to press forming, the surface of the substrate slides along the press tools where high contact pressure and poor lubrication may result in surface damage such as galling of the substrate and in case of an applied zinc coating, may result in zinc pollution of the press tools. To reduce these effects, the state-of-the- art solution is to use extremely smooth tool surfaces and rather rough strip surfaces as well as applying oil for lubrication. One hypothesis is that high roughness on the strip helps to capture oil for good lubrication properties. This strategy of improving press performance by increasing strip roughness in critical forming conditions has proven to work in press shops. However, it also has a negative effect on the painting process and appearance as high roughness generally increases waviness which is detrimental for the paint appearance. So, for these applications, such as the outer parts of a vehicle, the surface properties must be at a certain level to allow the application of the material to function reliably and safely in a construction and still be visually attractive after press forming and painting.

In many cases both the mechanical properties and the surface properties must meet certain minimum levels. To increase the mechanical properties the temper rolling reduction after recrystallisation annealing or hot-dip galvanising is increased, whereas control of the surface texture may require a reduction in temper rolling reduction in a temper rolling mill (TRM).

Thin gauge flat steel products are typically made in a process of hot rolling, pickling, and cold rolling. Annealing is required after cold rolling to compensate the strain hardening during cold rolling and obtain the desired mechanical properties, e.g. yield strength (Rpo.2) or elongation. Optionally (hot dip) galvanizing can be applied to add a corrosion protection coating to the steel substrate.

After annealing the steel has softened but became sensitive to so called "yield point elongation". This yield point elongation results in uncontrolled local strain and localized bands of plastic deformation also known as slip-banding or Liiders bands, when tensile strain is applied in a tensile test or during deformation of a flat steel sheet into a 3-dimensional shape, such as a chassis part or an outer panel for a vehicle. This uncontrolled and visible plastic deformation is highly undesirable. A known solution to the problem is temper rolling, or skin-pass rolling, where a small amount of bulk elongation, typically 0.5-3% is applied in a temper rolling mill (TRM). This process induces microscopic Luders bands in the steel substrate at such a scale that subsequent plastic deformation in regular sheet metal forming appears homogeneous. An alternative solution is applying tension levelling, where controlled bulk strain is applied by stretching and bending in a tension leveller (TL).

Both skin-pass rolling and tension levelling have to fulfil additional objectives. Skin-pass rolling is also used to correct strip shape, by varying elongation over the strip width, and to apply a surface texture, also called surface roughness, by means of using work rolls with relatively high roughness. The roughness peaks on the work roll induce locally high contact pressures and local plastic deformation of the steel substrate, and/or metallic coating. In tension levelling the main objective is to flatten the steel strip by inducing a small amount of bulk elongation as well. Tension levelling may also be used for the purpose of strip bulk elongation in the case the temper mill has insufficient rolling force capability to apply the required strip bulk elongation for solving the yield point elongation issue.

The problem is that both the skin-pass rolling and tension levelling have conflicting objectives. Skin-pass rolling has to balance the bulk elongation to solve the yield point elongation with the local plastic strain at the surface to create roughness. For example: hard steel substrates require small diameter work rolls to prevent the rolling forces becoming too high. Softer steel substrates can be rolled with large diameter work rolls to realize the desired roughness transfer at a defined bulk elongation. Consequently, once a work roll type has been installed in the skin-pass mill the rolling force depends on the required bulk elongation to solve the yield point elongation (and define the yield strength Rpo.2) and roughness transfer is also fixed. Only small variation in strip roughness is possible by adjusting the line tension, and thus required rolling force, while maintaining a constant bulk elongation. Improved roughness control is highly desired to enable more narrow strip roughness specifications, compensate for work roll surface wear, and react to variations in steel substrate properties.

Objectives of the invention

Therefore, it is an objective of the invention to provide a method that allows obtaining good surface texture and good mechanical properties.

It is another objective of the invention to provide a cold-rolled strip with a good surface texture and good mechanical properties for press forming.

Description of the invention

One or more of the objectives is reached with the method in accordance with claim 1. Preferred embodiments are provided in the dependent claims.

The invention solves the problem of poor strip surface roughness control by an integrated system including both the skin-pass mill and the tension leveller using a control scheme which combines predictive models, optimization routine and conventional controllers to gain accurate control of surface roughness while maintaining the yield strength properties.

The setpoint generation is shown in Figure 1 and is comprises five elements: Input:

1. Coil: coil of certain chemical composition and dimensions (thickness, width, length) which has been cold rolled, annealed, and optionally coated.

Process:

2. Skin-pass rolling: rolling process in which a small thickness reduction is applied, a surface texture is created in the strip surface, and optionally strip shape is adjusted. The internal control system targets an elongation setpoint and adjusts the rolling force continuously.

3. Tension levelling: process of applying bulk strain by stretching and bending. The internal control systems target an elongation setpoint and adjust the tension and roll positions continuously.

4. Finishing: process where quality inspection is performed, surface roughness is measured and oil is applied.

Output:

5. Finished coil: finished product with properties according to the order.

The desired output is reached based on the given input and the conduct of the process steps in between. The input consists of the physical coil that is to be processed with its incoming mechanical properties and surface properties and of the target requirements that the final finished coil must meet according to the customers demand and any applicable standards. Usually, a final finished coil has to meet one or more target mechanical properties (prop-tar), which includes at least the yield strength. The yield strength is usually expressed as the Rpo.2, the 0.2% offset yield strength which is defined as the stress that has to be imposed on a material in a tensile test that will result in a plastic strain of 0.2% (ISO 6892-1 :2019). It should be noted that the system as claimed is also for yield strengths that are defined differently, e.g Rpo.s. Another important prop-tar could be the yield point extension (A _e or YPE, often referred to as yield point elongation). This is defined as the extension between the start of yielding and the start of uniform work-hardening in a tensile test, expressed as a percentage of the extensometer gauge length, L _e (ISO 6892-1 :2019). The effect of the postprocessing of a metal strip in a TRM and/or TL is that the YPE is reduced, sometimes to 0 depending on the steel grade and the requirements of the customer and/or standard. The final finished coil also has to meet one or more target surface texture specifications (texttar) such as the roughness values Ra (ISO 21920-2:2021), which is the average, or arithmetic average of profile height deviations from the mean line. Setting these targets is the task for step a: the order section.

Based on these targets a work roll is selected where the prop-tar defines the work roll diameter and the text-tar determine the work roll roughness and peak count. This task is performed in step b: the work roll selection. The relevant data from step a and b are fed into an initial setpoint generator to generate initial setpoints for the temper rolling mill ,TRM and tension leveller, TL in step c using tables or models based on historical data. The historical data corresponds to data that are collected from processes. The main initial setpoints are the line tension in the TRM, the elongation in the TRM (e_TRM), the work roll bending force during TRM and the total elongation (e_tot) of the processed metal strip in the TRM + TL. The total elongation e_tot is sum of the elongation in the TRM and the elongation in the TL (e_TRM + e_TL The line tension in the TRM is important to control the flatness of the entering strip and combined with the work roll bending force in the TRM also for control of the flatness of the strip exiting the skin-pass mill. The e_TRM+e_TL has to be controlled to obtain the required yield strength (Rp) and obtain the suppression of the yield point elongation (YPE) where e_TRM is the elongation in the TRM and e_TL is the elongation in the TL. e_YPE is defined as the minimal elongation required to suppress the YPE effect. The e_tot according to the invention must be distributed over the TRM and the TL and the distribution of the elongation over TRM and TL is defined according to Eqs. 1 and 2:

Rp = a* e_TRM + b * e_TL + c (Eq. 1) e_YPE = d* e_TRM + e * e_TL (Eq. 2)

Thus, e_TL follows from an initial setpoint e_TRM given a target Rp and minimal e_YPE which must be satisfied.

The values for the coefficients a to e can be based on historical results or based on models that determine the parameters based on relevant product and process data. The determination of the initial setpoints is performed in step c.

The initial setpoints are subsequently checked, validated, against the limits imposed by the TRM - and TL process and installations. A rolling force model calculates the rolling force based on the e_TRM, line tension, work roll properties, strip properties. A roughness model calculates the strip roughness based on the rolling force, work roll properties and strip properties and also a validation whether the calculated YPE, Rpo.2 and Ra meets the prop-tar and text-tar values. The validation of the initial setpoints is performed in step d. Thus, as described above a setpoint validator, using predictive models, to validate if the initial setpoints will meet the one or more target mechanical properties and the one or more target surface texture specifications of the outgoing strip is obtained.

When the initial setpoints have been validated, i.e. the target values are obtainable in principle on the basis of the input coil and the chosen process, then as last step the setpoints may be optimised, for instance to adjust the setpoints so as to steer the target values towards the centre of their respective specification windows for instance to avoid the risk of a realised property value to be too close to a reject value and to make the process conditions as stable and reproducible as possible. In the case that the target values appear not to be obtainable with the initial setpoints, then the setpoint optimisation will adjust the setpoints until these optimised setpoints do enable to obtain the target values. The setpoint optimisation is performed in step e.

By means of non-limiting examples: In case the setpoint validation of step d indicates that the text-tar specification will not be met, the setpoint optimizer will adjust the initial setpoint according to one or more of the following strategies:

I. Adjust line tension (TRM) to influence the rolling force, which subsequently influences the surface texture, while maintaining a constant e_TRM and observing the limits of the installation and the process;

II. Adjust the e_TRM to influence the rolling force, which subsequently influences the surface texture and correcting the e_TL to maintain a constant Rpo.2;

III. Adjust e_TRM to influence the rolling force, which subsequently influences the surface texture, without adjusting e_TL.

The last step is to implement the final setpoints in the TRM and TL and execute the process to produce a strip with the desired prop-tar and text-tar. This is step f. During the execution of the process measured rolling forces, observations of strip shape and measured surface texture values can be fed back into the control system in a feedback loop to adapt the setpoints of the running process or for storage and later use as historical results or as improvement of the models that determine the parameters based on relevant product and process data.

The control system of the present invention is different from conventional closed- loop control systems such as defined in the Chinese patent application CN 107008758B. In CN 107008758B a real-time measurement on the finished product is used to do realtime correction of the process settings such as e_TRM and e_TL. The control system of the present invention is an open loop control system or a feed forward control system. The disadvantage of a closed-loop control such as described in CN107008758B is that the system is re-active to a difference between the real state and the desired state of the system; meaning that it reacts on an error in target properties, such as roughness as in CN 107008758B. On the other hand, the present invention is based on an openloop control or a feed forward control system where knowledge, which is captured from models using Eqs. 1 and 2, is used to define optimal settings, which prevent any error or difference between the desired state and the real state (prop-tar). For generation of set-point also models are used to predict surface properties (text-tar). To further improve accuracy of the models for small effects, which are not included in the model, a correction factor or an adaptation factor is used based on the difference between properties of the final product and the model prediction.

In an embodiment the actual rolling force value as measured in the TRM is used as input into the setpoint validator for the next coil. This feed-back adaptation improves the predictive capability of the method according to the invention.

In an embodiment the actual surface texture parameter as measured after TRM and TL is used as input into the setpoint validator for the next coil. This feed-back adaptation improves the predictive capability of the method according to the invention and enables to achieve the desired surface texture of the future coils with a better accuracy, thereby reducing the risk of producing sub-standard products.

It is preferable that the adjustment by the setpoint optimizer of the line tension in the TRM is such that the adjusted line tension does not exceed a lower or an upper limit. A line operates most flexibly when the initial line settings are in the middle of the operational window.

In another embodiment a method wherein the adjustment by the setpoint validator of the e_TRM is such that the rolling force is adjusted to bring at least one target surface texture specification within a specification window.

In an embodiment the adjustment by the setpoint optimizer of the elongation in the TRM is such that the at least one target surface texture parameter or target surface texture specification is brought within a specification window as a result of the modification of the rolling forces in the TRM due to the adjusted elongation in the TRM.

In an embodiment the e_TL is adjusted according to Eq. 1 to maintain the one or more target mechanical properties in the specification window, preferably without affecting the surface texture properties. By using the equation that describes the relation between e_TRM and e-TL on the one hand and a mechanical or microstructural property on the other hand the optimal combination of e_TRM and e-TL can be selected and thereby reduce the risk of producing sub-standard products.

In an embodiment the e_TRM elongation and the e_TL elongation are adjusted according to Eq. 1 to maintain the one or more target mechanical properties in the specification window and maintain one or more target surface texture parameters in the specification window. By using the equation that describes the relation between e_TRM and e-TL on the one hand and the equation that describes the relation between e_TRM and the surface texture parameter on the other hand the optimal combination of e_TRM and e-TL can be selected and thereby reduce the risk of producing sub-standard products.

In an embodiment the e_TRM elongation is adjusted according to eq. 1 to maintain the one or more target mechanical properties in the specification window and maintain one or more target surface texture parameters in the specification window. By using the equation that describes the relation between e_TRM and e-TL on the one hand and the equation that describes the relation between e_TRM and the surface texture parameter on the other hand the optimal e_TRM can be selected and thereby reduce the risk of producing sub-standard products.

In an embodiment wherein the setpoint optimiser is set up such as to meet the one or more target mechanical properties and/or the one or more target surface texture specifications in the centre of the respective specification window. To give the control system sufficient room for manoeuvre it is preferable that the targets are in the middle of the specification window and not near one end or the other of the specification. This also reduces the risk of producing sub-standard products.

In an embodiment the models predicting the skin-pass rolling force and surface roughness are statistical models based on historical process data. The more process data are available the more reliable these models are and the more likely it is that the desired properties can be obtained with a small chance of process outliers or the production of substandard products.

In a preferred embodiment the method is used in the line set-up in which the skinpass rolling step precedes the tension levelling step although it is equally possible in the situation in which the skin-pass rolling step follows the tension levelling step.

In a preferable embodiment the method is as described below wherein a. the skin-pass reduction e_TRM is between 0 (excluding 0) and 3.0 %, or wherein b. the tension levelling reduction e_TL is between 0 (excluding 0) and 3.0 %, preferably e_TL is at least 0.20 %, more preferably at least 0.50%, or wherein c. the sum of the e_TL and the e_TRM ( X (e_TL + e_TRM)) is 0.20 to 6.0 %, wherein the tension levelling reduction is at least 0.20%.

The inventors found that the ranges allow the operator and the method according to the invention sufficient leeway to control the process reproducibly and produce the desired products to the required specifications.

According to a second aspect the invention is also embodied in a computerised process automation for a continuous processing line, wherein the process automation is embodied such that, during operation, it carries out a method according to the invention. The method according to the invention is perfectly suited to control the TRM and TL process in an automated fashion and control the process reproducibly and produce the desired products to the required specifications.

According to a third aspect the invention is also embodied in a continuous processing line which is controlled by a computerised process automation according to the invention.

In another aspect a continous processing line which is controlled by a computerised process automation wherein the process automation is embodied such that, during operation, it carries out a method as described in the above mentioned embodiments.

Examples

The example demonstrate how the invention can be used to adjust setpoints to enable both Rpo.2 and Ra within the required specification windows and how the invention can be used to suppress the YPE and enable both Rpo.2 and Ra within the required specification windows. The steel making process, annealing process and skinpass rolling process always contain some natural spread. Therefore, the target values for Rp0.2, and Ra are actively steered toward the centre of the windows using the invention.

Example 1 is based on the following order information:

Example 1, 2 and 3 relate to an IF grade which does not show a YPE. For this grade both values for a and b are 1 for eq.l.

The initial setpoint generation defines:

• a line tension setpoint,

• a work roll bending force setpoint,

• a Rpo.2 = 190+/-2 MPa based on e_TRM = 1.7% and e_TL = 0.2%

• a rolling force

Comparative case Cl

The initial setpoints defined in example 1 are validated using the models (Figure 2) which leads to a predicted roughness Ra = 1.4 pm which is exactly on the upper limit of the Ra window. Considering natural spread in the process and roughness measurement, it has a high risk to get out of specification and is not desired.

Inventive case II

With use of the invention case Cl can be adjusted. Following the setpoint optimization, the initial setpoint for the line tension is increased to the maximum limit (these limit values depend on the processing line and therefore the exact values are not relevant for the example) which decreases the roughness to Ra = 1.3. In a second setpoint optimization step the initial setpoint for e_TRM is reduced to 1.4% and the roughness model predicts Ra = 1.2. At the same time the e_TL is adjusted to 0.5% to target the Rpo.2 = 190 MPa. The setup optimization according to the invention brings both the predicted Rpo.2 (Eq. 1) and Ra to the target values.

Example 2 is based on the following order information:

The initial setpoint generation defines:

• a line tension setpoint,

• a work roll bending force setpoint,

• a Rpo,2 = 175+/-2 MPa based on e_TRM = 0.9% and e_TL = 0.4%

• a rolling force

Comparative case C2

The initial setpoints defined in example 2 are validated using the models (Figure 2) which leads to a roughness Ra = 1.05 pm which is within the Ra window but not meets the Ra target of 1.2 pm. Inventive case 12

With use of the invention case C2 can be adjusted. Following the setpoint optimization, the initial setpoint for the line tension is required for a stable rolling process. Therefore, the initial setpoints for e_TRM and e_TL are considered. Adjusting the e_TRM to 1.2% and the e_TL to 0.1% will enable a target Ra = 1.2 pm.

Example 3 is based on the following order information:

The initial setpoint generation defines:

• a line tension setpoint,

• a work roll bending force setpoint,

• a Rp0.2 = 180+/-2 MPa based on e_TRM = 0.9% and e_TL = 0.6%

• a rolling force

Comparative case C3

The initial setpoints defined in example 3 are validated using the models (Figure 2) which leads to a roughness Ra = 1.05 pm which is outside the Ra window.

Inventive case 13

With use of the invention case C3 can be adjusted. Following the setpoint optimization, the initial setpoint for the line tension is considered first but may not be reduced as the strip has a poor shape and high tension is required for a stable rolling process. Therefore, the initial setpoints for e_TRM and e_TL are considered. Adjusting the e_TRM to 1.6% and the e_TL to 0% will enable a target Ra = 1.35 pm which is close to the target. However, during rolling the first few meters of the coil with these setpoints, it becomes apparent that the strip shape is still poor, and operators adjust the e_TL elongation to 0.3% to correct the shape issue. Automatically, the setpoint validation now indicates that the Rpo.2 will become 190 MPa (equation 1) which is on the border of the Rpo.2 window. The setup optimizer will adjust e_TRM to 1.4% such that the Rpo.2 = 185 MPa and the Ra = 1.26, thus both well within their windows.

Example 4 is based on the following order information:

This steel is a HSLA grade which shows a high YPE. For this grade d = l and e=l.l in eq. 2. The initial setpoint generation defines:

• a line tension setpoint,

• a work roll bending force setpoint,

• a e_YPE = 1.9% based on e_TRM = 0.9% and e_TL = 0.9% after which the Rpo.2 = 352+/-8 MPa

• a rolling force

Comparative case C4

For this steel grade a e_YPE = 1.9% is set which results in a R.po.2 which is slightly higher than the Rpo.2 target but still in the Rpo.2 window. Preferably the Rpo.2 is not further increased by increasing the e_tot above the e_YPE = 1.9%. The initial setpoints are validated using the models (Figure 4 and Figure 5) which leads to a roughness Ra = 1.3 pm which is just within the Ra window but does not meet the Ra target of 1.5 pm.

Inventive case 14

With use of the invention case C4 can be adjusted. Following the setpoint optimization, adjusting the e_TRM to 1.6% and the e_TL to 0.2% will enable a target Ra = 1.5 pm while maintaining the Rpo.2 and YPE constant.

Brief description of the drawings

The invention will now be explained by means of the following, non-limiting figures.

The setpoint generation is shown in Figure 1 and is explained in a number of steps:

1. Coil: coil of certain chemical composition and dimensions (thickness, width, length) which has been cold rolled, annealed, and optionally coated.

2. Skin-pass rolling: rolling process in which a small thickness reduction is applied, a surface texture is created in the strip surface, and optionally strip shape is adjusted. The internal control system targets an elongation setpoint and adjusts the rolling force continuously.

3. Tension levelling: process of applying bulk strain by stretching and bending. The internal control systems target an elongation setpoint and adjust the tension and roll positions continuously.

4. Finishing: process where quality inspection is performed, surface roughness is measured and oil is applied.

5. Finished coil: finished product with properties according to the order.

Figure 1 also shows the different steps and the interaction between them as well as the feedback of process and product data into the control system

Figure 2 shows the roughness (Ra) values in dependence of the elongation in the temper rolling mill (e_TRM) for the IP grade.

Figure 3 shows the data used to construct Figure 2. Figure 4 shows the relation between the elongation in the temper rolling mill (e_TRM) and the elongation in the tension levelling (e_TL) and whether there is a YPE elongation (closed circles) or not (open circles).

Figure 5 shows the roughness (Ra) values in dependence of the elongation in the temper rolling mill (e_TRM) for the HSLA grade and whether there is a YPE elongation (closed circles) or not (open circles)

Figure 6 shows the data used to construct Figure 5.