Prevention of rotational speed drop during threading

Abstract

In a rolling mill (1), a metal workpiece (2) with a workpiece head (6) is to be rolled. A control unit (7) for the rolling mill (1) receives a starting point (t3), a rolling speed (vW), and a rolling torque (MW) before the workpiece head (6) enters the rolling mill (1). Before a starting point (t2) prior to the starting point (t3), the control unit (7) determines a torque profile (M) for a drive (4) that drives the rolling (3) of the rolling mill (1) via a drive train (5) for a period extending from the starting point (t2) to the starting point (t3). It controls the drive (4) during this period according to the determined torque profile (M).The control device (7) determines the torque profile (M) starting from an initial state (x2) of the drive (4), the drive train (5) and the driven rolls (3) of the rolling stand (1) at the initial time (t2) such that the driven rolls (3) of the rolling stand (1) are subjected to the rolling torque (MW) by the drive train (5) at the tapping time (t3) and rotate at a circumferential speed (vU) matched to the rolling speed (vW).

Description

field of technology

[0001] The present invention is based on an operating method for a rolling stand in which a metal rolling stock having a rolling stock head is to be rolled, wherein a control device for the rolling stand is informed of a tapping time, a rolling speed and a rolling torque before the rolling stock head enters the rolling stand, wherein the tapping time is the time at which the rolling stock head is to enter the rolling stand, and the rolling stock is to be rolled at the tapping time with the rolling speed and the rolling torque, wherein the control device determines a torque curve for a roll of the rolling stand via a drive train before a starting time lying before the tapping time for a period extending from the starting time to the tapping time and controls the drive during the period in accordance with the determined torque curve.

[0002] The present invention further relates to a control program comprising machine code which can be processed by a control device for a rolling stand for rolling a metal rolling stock having a rolling stock head, wherein the processing of the machine code by the control device causes the control device to carry out such an operating method.

[0003] The present invention further relates to a control device for a rolling stand for rolling a metal rolling stock having a rolling stock head, wherein the control device is programmed with such a control program so that it carries out such an operating method during operation.

[0004] The present invention further relates to a rolling device for rolling a metal rolling stock having a rolling stock head, wherein the rolling device comprises a rolling stand in which the rolling stock is rolled, wherein the rolling stand comprises a drive, a drive train and rollers, wherein the rollers can be driven by means of the drive via the drive train, wherein the rolling stand comprises such a control device for the rolling stand. State of the art

[0005] The above-mentioned topics are well known. For purely exemplary purposes, reference can be made to the technical paper "Impact Speed ​​Drop Compensation Procedure for a New Layout Wire Rod Mill" by Antonella Scaglia and Graziano Melandri, Proceedings of the 28th Annual Conference of the Industrial Electronics Society (IECON), 2002, pages 573 to 578, and to JP 2001 150 013 A. Summary of the invention

[0006] When rolling a metallic rolled stock, such as steel, aluminum, brass, or copper, the quality of the final product depends on many factors. The rolled stock can, in particular, be a flat rolled stock, such as strip or heavy plate. Factors influencing the quality of the final product include, among others, the material temperature, lubrication during rolling, the correct pressure conditions of the rolls, and the rotational speed or peripheral speed of the work rolls. Whether we speak of rotational speed or peripheral speed is equivalent, since the peripheral speed and the rotational speed can be converted into one another based on the diameter of the work rolls.

[0007] Especially with regard to the speed of the work rolls, it is important that the desired speed is maintained as precisely as possible. Any deviation leads to negative effects such as reduced product quality, material jams (e.g., the formation of a belt loop), inaccuracies in dependent process variables, difficulties in material handling, and more. The dimensions of the final product can also vary as a result.

[0008] During stable rolling, the required accuracy can be ensured through the use of simple and robust control systems. The transition from one stable operating range to another is considerably more difficult. This applies particularly to the pass-through time, i.e., the time at which the rolling stock head enters the rolling stand. At this time, the load on the rolling stand changes. In particular, the required rolling force and the required rolling torque increase significantly. The changes occur almost instantaneously at the pass-through time.

[0009] Due to the sudden increase in the required rolling torque, the current technology results in a drop in the speed of the driven rolls of the rolling stand. This drop is subsequently compensated for by the associated speed controller. However, this requires a settling time, the duration of which depends on the drive used, the drive train, and the controller characteristics. During this settling time, quality losses in the final product occur.

[0010] In the prior art, it is known to increase the target speed at which the work rolls or generally driven rolls rotate shortly before the tapping time. Such a procedure is explained, for example, in the aforementioned technical article by Antonella Scaglia and Graziano Melandri. It is also known to calculate an additional torque based on the drop in speed and the duration of this drop, which is applied to the torque controller or current controller underlying the speed controller at the tapping time. Such a procedure is described, for example, in JP 2001 150 013 A.

[0011] The current state of the art approaches do not adequately solve the problem of the speed of the driven rollers collapsing.

[0012] The object of the present invention is to create possibilities by means of which the collapse of the speed of the driven rollers can be completely or at least almost completely avoided.

[0013] The object is achieved by an operating method having the features of claim 1. Advantageous embodiments of the operating method are the subject of dependent claims 2 to 12.

[0014] According to the invention, an operating method of the type mentioned at the outset is designed in that the control device determines the torque curve starting from an initial state of the drive, the drive train and the driven rollers of the rolling stand given at the initial time in such a way that the driven rollers of the rolling stand are subjected to the rolling torque by the drive train at the time of penetration and rotate at a peripheral speed matched to the rolling speed.

[0015] Before the piercing point, there is no rolling stock in the roll gap of the rolling stand. Therefore, to simply maintain the speed of the driven rolls of the rolling stand, only a very low application torque is required, which is considerably less than the rolling torque required for rolling the rolling stock. The application torque is the torque with which the driven rolls of the rolling stand are driven. Furthermore, since the torque output by the drive cannot be increased to the rolling torque at will, the increase in torque to the rolling torque required for rolling the rolling stock before the piercing point causes the driven rolls of the rolling stand to accelerate. The extent of the acceleration is determined by the moving masses, the associated moments of inertia, and the rolling torque.For this reason, the peripheral speed of the driven rolls shortly before the pass-through time must also be lower than the peripheral speed corresponding to the rolling speed. However, due to acceleration, the peripheral speed reaches the desired peripheral speed corresponding to the rolling speed at the pass-through time. Deviations between the peripheral speed and the rolling speed before the pass-through time are not critical, since no material is being rolled in the roll gap of the rolling stand during this time.

[0016] The rolling speed and the peripheral speed of the driven rolls are not necessarily one and the same. In particular, during rolling, the rolled stock enters the rolling stand at an entry speed that is lower than the peripheral speed of the driven rolls. Likewise, during rolling, the rolled stock exits the rolling stand at an exit speed that is higher than the peripheral speed of the driven rolls. The corresponding speed relationships are known in technical circles as lag and lead. Depending on the individual case, the term "rolling speed" can refer to the entry speed, the exit speed, or a value between the entry speed and the exit speed. Regardless of the specific definition of the rolling speed, however, there is a fixed relationship between the rolling speed and the peripheral speed.

[0017] The starting time can be known to the control system, for example, based on conventional path tracking in conjunction with a (constant or time-varying) rolling stock speed. Path tracking is generally known to experts and therefore requires no further explanation. The starting time is usually determined indirectly by specifying the length of the period.

[0018] The driven rolls of the rolling stand are usually the work rolls of the rolling stand. In exceptional cases, backup rolls or intermediate rolls of the rolling stand may also be driven.

[0019] Preferably, the control device determines the torque curve such that, at the time of the tapping, the drive, the drive train, and the driven rolls of the rolling stand rotate at corresponding speeds, and changes in the speeds have a value of zero. This ensures that a transient condition, which must (still) exist before the tapping time, has just subsided at the time of the tapping.

[0020] It is currently particularly preferred for the control device to determine the torque curve using a model that models the drive, the drive train, and the rolls of the rolling stand driven by the drive based on mathematical and physical equations. This approach allows the control device to flexibly handle almost any boundary conditions.

[0021] In the simplest case, the drive, the drive train, and the driven rolls of the rolling stand are modeled as a rigid system. In this case, the speed of the drive is in a fixed ratio to the speed of the driven rolls, which remains the same at all times. In practice, however, torsions often occur, for example, between the drive and the drive train, or between the drive train and the driven rolls. It is therefore preferable for the model to model the drive, the drive train, and the driven rolls of the rolling stand as a multi-mass vibration system with a number of damped-elastically connected elements. This allows such torsions to be taken into account.

[0022] The number of damped-elastically interconnected elements can be determined as needed. Typically, it is at least three: the drive, the drive train (considered a single unit in this case), and the driven rollers. However, if necessary, the drive train can also be divided into several sections. In this case, the number of damped-elastically interconnected elements increases accordingly.

[0023] One way to determine the torque curve is for the control unit to set up an optimization problem in which the respective torque of the drive is used as input variables for a number of discrete points in time within the period. In this case, the control unit determines the torque curve by solving the optimization problem. By setting up and solving an optimization problem, additional conditions such as limit values ​​to be observed or other boundary conditions can be easily taken into account.

[0024] Alternatively, it is possible for the control device to determine the torque curve using tables stored in the control device, to which the penetration time, the rolling speed, the rolling torque, and the initial state of the drive, the drive train, and the driven rollers of the rolling stand are fed as input variables. It is also possible for the torque curve to be given to the control device by a parameterizable time function stored in the control device, and for the control device to determine the parameters of the function based on the penetration time, the rolling speed, the rolling torque, and the initial state of the drive, the drive train, and the driven rollers of the rolling stand.

[0025] These approaches are sometimes just as flexible as using a model and can often be executed in a shorter time, thus improving online capability. For example, the tables can be determined in advance and offline, using a model to determine the corresponding table entries for a large number of possible value combinations, and the results of the model-based calculations are stored in the control system. A similar approach is possible for determining the dependencies of the parameters on the possible value combinations.

[0026] Preferably, in the initial state, the drive, the drive train, and the driven rolls of the rolling stand rotate at corresponding speeds, and any changes in the speeds are zero. In this case, the drive, the drive train, and the driven rolls of the rolling stand are in a stable state at the initial time. This procedure facilitates the determination of the torque curve.

[0027] As already explained, the procedure according to the invention implies that the driven rolls of the rolling stand are accelerated at least shortly before the piercing time. At a certain point in time before the piercing time, the peripheral speed of the driven rolls must therefore be lower than the peripheral speed of the driven rolls coordinated with the rolling speed. It is possible that this situation does not yet exist at the initial time. In this case, the driven rolls of the rolling stand must first decelerate and then accelerate during this period. Preferably, however, this situation is already present in the initial state. This procedure can reduce both energy consumption and wear.For example, after rolling a rolling stock in the rolling stand, i.e. after the rolling stock foot of this rolling stock has left the rolling stand, the driven rolls can be allowed to run down slowly until they rotate at the desired (low) peripheral speed that is desired for the next rolling stock at the starting time.

[0028] Preferably, the control device determines the torque curve such that the torque generated by the drive and / or an applied torque is greater than a respective predetermined minimum value throughout the entire period. The applied torque is, as already mentioned, the torque with which the driven rolls of the rolling stand are driven.

[0029] Likewise, the control device preferably determines the torque curve in such a way that a rotational acceleration with which a rotational speed of the drive changes and / or a rotational acceleration with which a rotational speed of the driven rollers changes is greater than a respective predetermined minimum value during the entire period.

[0030] In each of the above cases, the respective minimum value may in particular be zero.

[0031] By taking the minimum values ​​into account, the load on the drive, the drive train and the driven rolls of the rolling stand can be equalized.

[0032] This procedure is particularly advantageous in conjunction with the design in which, in the initial state, the peripheral speed of the driven rollers is lower than the peripheral speed of the driven rollers matched to the rolling speed.

[0033] Preferably, the control device determines the torque curve in such a way that energy consumption of the drive is minimized during the period.

[0034] Preferably, the period is between 50 ms and 200 ms, in particular between 75 ms and 150 ms. This period is sufficiently long to allow the loading torque required at the time of tapping to be built up in the manner according to the invention.

[0035] The object is further achieved by a control program having the features of claim 13. According to the invention, the execution of the control program causes the control device to execute an operating method according to the invention.

[0036] The object is further achieved by a control device having the features of claim 14. According to the invention, the control device is programmed with a control program according to the invention, so that the control device executes an operating method according to the invention during operation.

[0037] The object is further achieved by a rolling device having the features of claim 15. According to the invention, in a rolling device of the type mentioned at the outset, the control device is designed as a control device according to the invention. Short description of the drawings

[0038] The above-described properties, features, and advantages of this invention, as well as the manner in which they are achieved, will become clearer and more readily understood in connection with the following description of an embodiment, which is explained in more detail in conjunction with the drawings. FIG 1 shows a rolling device, FIG 2 shows a drive system, FIG 3 shows a flow chart, FIG 4 shows a model, FIG 5 shows a system of differential equations, FIG 6 shows the system of differential equations in matrix notation, FIG 7 shows an optimization problem and FIGS 8 to 23 show time diagrams. Description of the embodiments

[0039] According to FIG 1 A rolling facility comprises a rolling stand 1. In the rolling stand 1, a rolling stock 2 is to be rolled. The rolling stand 1 is often part of a multi-stand rolling mill. However, this is not absolutely necessary. Regardless of the number of additional rolling stands, only the procedure for the FIG 1 This is also possible and sufficient in the present case, since the operation of the rolling stand 1 of FIG 1 is independent of the operation of other rolling stands, for example one of the FIG 1 shown rolling stand 1 or a rolling stand arranged upstream of the rolling stand in FIG 1 The rolling stand 1 shown is arranged downstream of the rolling stand 1. If several rolling stands 1 are present, the inventive mode of operation of the rolling stand 1 can also be implemented in an analogous manner in the other rolling stands.

[0040] The rolling stand 1 of FIG 1has reels 3, which - see also FIG 2 - are driven by a drive 4 via a drive train 5. The driven rolls 3 are often the work rolls of the rolling stand 1. If the rolling stand 1 has other rolls in addition to the work rolls, in individual cases the driven rolls 3 may also be rolls other than the work rolls. However, even in this case, the driven rolls 3 are usually the work rolls of the rolling stand 1. The entirety of the drive 4, drive train 5, and driven rolls 3 is referred to below as the drive system.

[0041] It is possible that an upper driven roller 3 and a lower driven roller 3 each have their own drive 4 and their own drive train 5. However, often, as shown in FIG 2only one common drive 4 is present and the drive train 5 is split so that the drive 4 drives both driven rollers 3. Whether one or the other approach is taken is of secondary importance within the scope of the present invention.

[0042] The rolled stock 2 is made of metal. Most often, the rolled stock 2 is made of steel. However, it can also be another metal, such as aluminum, copper, or brass. Furthermore, the rolled stock 2 is usually a flat rolled stock, i.e. a rolled stock 2 that is rolled into a strip or heavy plate. In individual cases, however, it can also be a bar-shaped rolled stock, in particular bar steel. Regardless of the shape and material, the rolled stock 2 always has a rolling stock head 6. The rolling stock head 6 is the area of ​​the rolled stock 2 that reaches the rolling stand 1 first and runs into the rolling stand 1.

[0043] FIG 1shows a situation at a time t1, at which the rolling stock head 6 has not yet reached the rolling stand 1. The rolling stock head 6 is at this time at a distance in front of the rolling stand 1. In FIG 1 a distance in front of the rolling stand 1 is also shown in dashed lines. The rolling stock head 6 reaches the corresponding location at a time t2. Furthermore, the rolling stock head 6 reaches the roll gap of the rolling stand 1 at a time t3, i.e. it runs into the rolling stand 1 at this time. The time t1 is referred to below as the information time. The time t2 is referred to below as the starting time. The time t3 is referred to below, as is generally customary, as the penetration time. The starting time t2 is usually between 50 ms and 200 ms before the penetration time t3, in particular between 75 ms and 150 ms before the penetration time t3, for example approximately 100 ms before the penetration time t3.

[0044] The rolling stand 1 is controlled by a control device 7. The control device 7 is programmed with a control program 8. The control program 8 includes machine code 9 that can be executed by the control device 7. Programming the control device 7 with the control program 8 or executing the machine code 9 by the control device 7 causes the control device 7 to execute an operating method that is explained in more detail below.

[0045] The present invention relates to a procedure that is taken before the rolling stock head 6 enters the rolling stand 1. To implement the operating method according to the invention, the control device 7 must know the starting time t2 and the penetration time t3 in good time so that the calculations required for the operating method according to the invention are completed by the starting time t2 at the latest. For this purpose, the information time t1 lies before the starting time t2. The extent to which the information time t1 lies before the starting time t2 can be determined as required. It is crucial that calculations performed by the control device 7 and required for the correct control of the rolling stand 1 from the starting time t2 onwards are completed before the starting time t2.

[0046] A suitable determination and adherence to the information time t1 is readily possible because the location of the rolling stock head 6 can be detected in good time before the time t1 and this can be made known to the control device 7. A usual tracking of the path of the rolling stock head 6 can then take place based on the detected location and the constant or time-dependent rolling stock speed v of the rolling stock 2. The current and future rolling stock speed v can be readily known to the control device 7. All of this is generally known to those skilled in the art and will therefore not be explained in more detail below. The rolling stock speed v is not referred to here as the rolling speed. The reason for this is that the rolling stock speed v in this case relates to the period of time at which the rolling stock head 6 has not yet reached the rolling stand 1, i.e., at which rolling is not yet taking place in the rolling stand 1.

[0047] According to FIG 3In a step S1, the control device 7 is informed of the penetration time t3, a rolling speed vW, and a rolling torque MW. The rolling speed vW is the speed at which the rolled stock 2 is to be rolled in the roll stand 1, at least at the penetration time t3. The rolling torque MW is the associated torque with which the driven rolls 3 are to be driven by the drive 4 via the drive train 5. It is possible for the values ​​t3, vW, MW to be specified externally to the control device 7. Alternatively, it is possible for them to be determined by the control device 7 itself. The determination of the penetration time t3, the rolling speed vW, and the rolling torque MW will not be explained in more detail here. They are generally known to those skilled in the art. Step S1 is carried out by the control device 7 at the information time t1.

[0048] In a step S2, the control device 7 determines a torque curve M for the drive 4 for a period extending from the initial time t2 to the piercing time t3. The torque curve M is a function of time t. Possible embodiments of step S2 will be explained in more detail later. In any case, however, the control device 7 starts from an initial state x2 of the drive system to determine the torque curve M. The initial state x2 is the state x of the drive system that exists at the initial time t2. In any case, the determination of step S2 is also carried out in such a way that the driven rollers 3 are subjected to the rolling torque MW by the drive train 5 at the piercing time t3 and rotate at a peripheral speed vU that is coordinated with the rolling speed vW.The peripheral speed vU is linked to the rolling speed vW via the lead or lag, depending on whether the rolling speed is the entry-side or exit-side rolling speed. The torque curve M is preferably determined such that, at the starting time t3, the drive 4, the drive train 5, and the driven rolls 3 of the rolling stand 1 rotate at corresponding speeds, and changes in the speeds have the value zero. Step S2 is executed by the control device 7 before the starting time t3, or more precisely, even before the starting time t2.

[0049] It is possible for the initial state x2 to be specified to the control device 7. Alternatively, it is possible for the control device 7 to determine the initial state x2 and control the drive 4 such that the initial state x2 is present at the initial time t2. It is also possible for the initial time t2 to be specified to the control device 7. However, the control device 7 generally knows how long the period between the initial time t2 and the tapping time t3 should be, so that the control device 7 can determine the initial time t2 itself.

[0050] In step S3, the control device 7 waits for the initial time t2. Step S3 is repeatedly executed until the initial time t2 is reached. Starting at the initial time t2, the control device 7 repeatedly executes steps S4 and S5.

[0051] In step S4, the control device 7 controls the drive 4 according to the determined torque curve M for the respective time t. For example, the control device 7 can FIG 1 control the drive 4 by outputting corresponding control signals to a converter 4', which supplies an electric motor 4" with electrical energy. The representation of the converter 4' in FIG 1 Thyristor control is purely exemplary. It could also be transistor control.

[0052] In step S5, the control device 7 checks whether the tapping time t3 has been reached. If the tapping time t3 has not yet been reached, the control device 7 returns to step S4. Steps S4 and S5 are therefore repeatedly executed by the control device 7 until the tapping time t3 is reached. When executing each step S4, the control device 7 naturally takes the progress over time t into account.

[0053] Once the piercing time t3 is reached, the control device 7 proceeds to step S6. In step S6, the "normal" rolling of the rolling stock 2 takes place. Step S6 can be implemented in the same way as in the prior art.

[0054] Within the scope of the implementation of step S2, the drive 4, the drive train 5 and the driven rollers 3 can, in the simplest case, be considered as being completely rigidly connected to one another. In this case, the speed of the drive 4 at any time t (possibly taking into account a gear ratio) corresponds to the speed of the driven rollers 3. In this case, an analytical determination of the torque curve M may even be possible under certain circumstances. In any case, however, it is possible for the control device 7 to determine the torque curve M using a model 10 (see FIG 4), which models the drive system based on mathematical-physical equations.

[0055] Within the framework of model 10, it is generally assumed that the drive 4 is connected to the drive train 5 via a coupling 11 and that the drive train 5 is also connected to the driven rollers 3 via a coupling 12. The two couplings 11, 12 are, as in FIG 4 indicated by a respective spring component 13 and a respective damper component 14, each damped-elastic coupling.

[0056] FIG 4thus shows a design in which the drive system is modeled as a three-mass oscillator. In individual cases, one of the two couplings 11, 12 can be disregarded. In this case, the drive system can be modeled as a two-mass oscillator. If both couplings 11, 12 are disregarded, the drive system is modeled as a completely rigid system. In other cases, a subdivision into more than three damped-elastically coupled elements occurs. In this case, in addition to the two couplings 11, 12, further damped-elastic couplings are present.

[0057] In the case of modeling as a multi-mass oscillator, the drive system can be modeled as described below in connection with FIG 5 is explained.

[0058] For the foremost element of the drive system (in this case drive 4) a differential equation of the form J 1 ω ˙ 1 = M − c 1 Δφ 1 − d ω 1 − ω 2 − D 1 ω 1 For all elements arranged between the front and rear elements (for example the drive train 5), a differential equation of the form J i ω ˙ i = c i − 1 Δφ i − 1 + d i − 1 ω i − 1 − ω i − c i Δφ i − d i ω i − ω i − 1 − D i ω i For the last element of the drive system (i.e. the driven rollers 3) a differential equation of the form J n ω ˙ n = c n − 1 Δφ n − 1 + d n − 1 ω n − 1 − ω n − D n ω n − ML be set.

[0059] In the equations, the symbols used have the following meaning: J i is the moment of inertia of the i-th element. ω i is the angular velocity of the i-th element. M is the torque applied by the drive 4, hereinafter also referred to as motor torque M. ci is the spring constant with which the i-th element is coupled to the respective subsequent element. Δϕ i is the twist of the i-th element relative to the respective subsequent element. di is the material damping with which the i-th element is coupled to the respective subsequent element. D i is the friction coefficient of the i-th element. ML is the torque acting on the driven rollers 3, which ultimately drives them, hereinafter also referred to as the applied torque ML.

[0060] Any gears (including their gear ratios) were neglected above. They can be considered at any time. Their inclusion merely results in scaling. The angular velocity ω i of an element is related to its rotational speed by a factor of 2π. Therefore, no distinction is made between rotational speed and angular velocity below.

[0061] Below, as shown in FIG 6 with x ω T = ω 1 , … , ω n and x φ T = Δφ 1 , … , Δφ n − 1

[0062] partial states and with x T = x ω T x φ T is the state x of the modeled drive system. In this case, the system of differential equations, as determined by equations (1) to (3), can be written in matrix notation in the form x ˙ ω x ˙ φ = A ω A ωφ A φω 0 ⋅ x ω x φ + b ω 0 M + e 0 ML The matrices A ω , A ωϕ and A ϕω are known and constant. The same applies to the vectors b ω and e, which are given by b ω T = 1 J 1 , 0 , … , 0 and e T = 0 , … , 0 , − 1 J n are given. These to FIG 5equivalent representation is known and familiar to experts.

[0063] Various conditions must be considered to determine the torque curve M. One of the most important conditions is that at the cutting time t3, the speed ω n of the work rolls 3 is equal to a target speed w*: ω n t 3 = ω ∗ .

[0064] The target speed w* corresponds to the peripheral speed vU, which the driven rollers 3 must have for the desired rolling speed vW.

[0065] A further condition is that at the penetration time t3, the applied torque ML is equal to the desired rolling torque MW. This condition can be formulated using equation (3) as follows: c n − 1 Δφ n − 1 t 3 + d n − 1 ω n − 1 t 3 − ω n t 3 − D n ω n t 3 = MW .

[0066] Strictly speaking, equation (11) only applies if the speed of the driven rollers 3 is to be kept constant after the piercing time t3. If the driven rollers 3 are to be accelerated or decelerated after the piercing time t3, an additional torque can be considered on the right-hand side of equation (11) in addition to the rolling torque MW.

[0067] As already mentioned, the motor torque M applied by drive 4 cannot be changed arbitrarily quickly. Furthermore, certain elasticities, damping, and losses generally exist, as described by equations (1) to (3). The torque curve M of drive 4 must therefore be determined appropriately to ensure that the conditions according to equations (10) and (11) are met.

[0068] Furthermore, transient conditions should already be adjusted when accelerating the driven rollers 3 to the target speed w*. Therefore, one can require as further conditions that the angular velocities ω i of all upstream elements of the drive system have also reached the target speed w* at the tapping time t3, i.e., the equations ω i t 3 = ω ∗ (for i = 1, 2, ..., n-1) and the angular accelerations, i.e. the time derivatives of the angular velocities ω i , are all 0. This allows the conditions to be determined which the rotations Δϕ i and the motor torque M must fulfill at the starting time t3, namely Δφ n − 1 t 3 = MW + D n ω ∗ / c n − 1 , Δφ i t 3 = c i + 1 Δφ i + 1 t 3 + D i + 1 ω ∗ / c i (for i = 1, 2, ..., n-2) and M t 3 = c 1 Δφ 1 t 3 + D 1 ω ∗ .

[0069] Thus, the required state x3 of the drive system at the penetration time t3—hereinafter referred to as the target state x3—is uniquely determined. Since the rolling stock head 6 is located immediately before the penetration time t3 but still in front of the roll stand 1, but not in the roll stand 1, an acceleration still acts in the drive system immediately before the penetration time t3. Thus, starting from the initial state x2 of the drive system, it is necessary to determine the corresponding torque curve M, by means of which the drive system is transferred to the target state x3.

[0070] The initial state x2 is known to the control device 7. As a rule, the initial state x2 is a stable state, i.e. a state in which the drive 4, the drive train 5 and the driven rolls 3 of the rolling stand 1 rotate at mutually corresponding speeds ω i and changes in the speeds ω i have the value zero. The peripheral speed vU of the driven rolls 3 can, in principle, have any desired value in the initial state x2. The peripheral speed vU of the driven rolls 3 can - purely theoretically - be greater than the value matched to the rolling speed vW, have the value 0 or even be negative. In practice, however, it is generally advantageous if the peripheral speed vU of the driven rolls 3 is positive, but smaller than the peripheral speed vU of the driven rolls 3 matched to the rolling speed vW.The reasons for this will become clear later. The states of the drive system before the initial time t2 are not relevant here.

[0071] To determine the moment curve M (= implementation of step S2 of FIG 3 ) various approaches are possible. In some simple cases, it may be possible to determine the torque curve M analytically. This can be particularly possible if the drive system can be considered as a rigid system. It may also be possible to determine the torque curve M using a so-called flatness-based trajectory calculation. Such calculations are known to experts as such. In any case, however, it is possible, according to the representation in FIG 7 To determine the moment curve M, an optimization problem must be set: min M ∫ t 2 t 3 x t − x ′ T Q x t − x ′ + M t − M ′ R M t − M ′ dt . x' is a state to be aimed for if possible. The desired state x' can be constant or time-dependent. The torque curve M can be influenced by the state x'. Q is a weighting matrix with which the deviation of the respective state x(t) from the desired state x' is weighted. The weighting matrix Q is positive (semi-) definite. M' is a motor torque to be aimed for if possible. The desired motor torque M' can be constant or time-dependent. The torque curve M can also be influenced by the desired motor torque M'. In the case of a general optimization problem, R is a positive definite weighting matrix, in this case a positive weighting factor, with which the deviation of the motor torque M(t) from the desired motor torque M' is weighted.

[0072] The initial state x2 and the target state x3 are given as mandatory boundary conditions for the optimization problem in each case: x t 2 = x 2 and x t 3 = x 3 .

[0073] Furthermore, the optimization problem includes as a boundary condition the motor torque M, which is to be delivered by the drive 4 at the tapping time t3: M t 3 = M 3 .

[0074] The engine torque M3 is uniquely determined by the system of equations according to equations (10) to (15).

[0075] Furthermore, the optimization problem includes as a boundary condition the motor torque M, which is to be delivered by the drive 4 at the initial time t2: M t 2 = M 2 .

[0076] The numerical value can be freely chosen within certain limits, but it must be specified.

[0077] As a further condition, it is generally taken into account that the engine torque M is within the permissible limits at any time t between the initial time t2 and the starting time t3, i.e. it does not fall below a minimum value Mmin and does not exceed a maximum value Mmax: M min ≤ M t ≤ M max .

[0078] In the general case, additional equality and inequality constraints g, h can be specified, which should be observed by the state x between the initial time t2 and the tapping time t3: g x t , M t = 0 and h x t , M t ≤ 0 .

[0079] Examples of possible inequality constraints h can be, for example, requirements that certain changes in angular velocities ω i or certain rotations Δϕ i are not exceeded. It can also be specified that the respective torque acting between successive elements of the modeled drive system does not exceed certain limits.

[0080] When updating the state x, the modeling of the drive system is taken into account, for example according to equation (24), which is a modification of equation (7): x ˙ ω x ˙ φ = A ω A ωφ A φω 0 ⋅ x ω x φ + b ω 0 M .

[0081] The optimization problem according to equation (16) can be solved analytically, if possible. It is always possible to solve the optimization problem numerically. The numerical solution implies that equation (16) is not solved in a time continuum, but that the period extending from the initial time t2 to the tapping time t3 is divided into small time steps (discretized). As a result, an optimization problem is set up in which the respective torque M of the drive 4 is used as input variables for a plurality of discrete times t within the specified period. The solution found corresponds to the desired torque curve M. The torque curve M is thus determined by minimizing or solving the optimization problem.

[0082] The most elegant way is when the model 10 is shown in FIG 1is implemented within the control device 7 and thus the optimization problem is also solved by the control device 7. Alternatively, it is possible, as also in FIG 1shown that the control device 7 determines the torque curve M using tables 15 which are stored in the control device 7. In this case, the piercing time t3, the rolling speed vW, the rolling torque MW and the initial state x2 of the drive system are fed to the tables 15 as input variables. The distance between the initial time t2 and the piercing time t3 can be known to the control device 7 in advance. Alternatively, it is possible, as also shown in FIG, for a function fa of the time t to be stored in the control device. In this case, the function fa can be parameterized with parameters a. In this case, the torque curve M is given to the control device 7 by the function fa. In this case, the control device 7 determines the parameters a of the function fa based on the piercing time t3, the rolling speed vW, the rolling torque MW and the initial state x2 of the drive system.

[0083] Various modifications are possible within the scope of the inventive procedure. For example, it is possible for the control device 7 to determine the torque curve M such that the torque M generated by the drive 4 and / or the applied torque ML applied to the driven rollers 3 is greater than a respective predetermined minimum value throughout the entire period extending from the initial time t2 to the piercing time t3. This does not mean that one considers the result "torque curve M" or "curve of the applied torque ML," then determines the respective minimum value, and then concludes that the entire corresponding curve lies above the respective minimum value. This would correspond to a procedure according to the motto "you shoot an arrow; where it hits, that's the target."Conversely, the respective minimum value is specified before the determination of the torque curve M, i.e. before the execution of step S2 of . FIG 3 The determination of the torque curve M is then carried out in such a way that the respective minimum value is maintained. The predetermined minimum value is therefore a condition to be observed for the torque curve M. If the determination of the torque curve M is carried out by solving the optimization problem of FIG 7 , the specified minimum value is entered, for example, as the minimum value Mmin in equation (21). The minimum value can, in particular, be 0 or greater than zero.

[0084] In an analogous manner, it is also possible for the control device 7 to determine the torque curve M such that an angular acceleration with which the rotational speed ω 1 of the drive 4 changes, and / or an angular acceleration with which the rotational speed ω n of the driven rollers 3 changes, is greater than a respective predetermined minimum value during the entire period extending from the initial time t2 to the piercing time t3. Here, too, this is a condition to be observed for the torque curve M, not merely an ensuing result. Here, too, the minimum value can, in particular, be 0 or greater than zero.

[0085] Furthermore, it is possible for the control device 7 to determine the torque curve M in such a way that the energy consumption of the drive 4 is minimized during the period extending from the initial time t2 to the tapping time t3. This can be achieved, in particular, by appropriately specifying the motor torque M'.

[0086] The following presents test results obtained by simulating typical state-of-the-art procedures and procedures according to the invention. The drive system was modeled as a three-mass oscillator in all cases.

[0087] In the FIGS 8 to 23 The time in seconds is given on the abscissa, with the piercing time t3 occurring at time 1.0 s. From time t3 onwards, the rolling torque MW must be applied. Before the piercing time t3, only small moments are required for uniform rotation of the driven rollers 3. On the ordinate, in the FIGS 8 to 23 -each standardized - a speed or a torque is specified.

[0088] The FIGS 8 to 11 show the results of a state-of-the-art approach. These results occurred when the drive system was maintained at the target speed w* corresponding to the rolling speed vW before the pass time t3, and the drop in speed occurring at the pass time t3 was compensated for after the pass time t3 by the torque and speed control of roll stand 1. FIG 8 shows the time course of the speed or angular velocity ω 1 of the drive 4, FIG 9 the temporal progression of the rotational speed or angular velocity ω n of the driven rollers 3. In an analogous manner, the FIGS 10 and 11The temporal progression of the motor torque M and the applied torque ML. The speed ω n of the driven rollers 3 clearly drops after the piercing time t3. The same applies, with a delay, to drive 4. More than 0.2 s elapses until the speed drop is compensated.

[0089] The FIGS 12 to 15 also show results from a state-of-the-art approach. These results occurred when the drive system was maintained at the target speed w* before the piercing time t3, which is above the roll peripheral speed vU corresponding to the rolling speed vW, and the target speed w* was reduced approximately 50 ms after the piercing time t3 to the value corresponding to the rolling speed vW. FIG 12 shows the time course of the speed or angular velocity ω 1 of the drive 4, FIG 13the temporal progression of the rotational speed or angular velocity ω n of the driven rollers 3. In an analogous manner, the FIGS 14 and 15 the time course of the motor torque M and the loading torque ML. It is evident that the speed ω n of the driven rollers 3 also drops here, although not quite as much as in the procedure of the FIGS 8 to 11 . With a delay, the angular velocity ω 1 of drive 4 also drops. Here, too, a time period of more than 0.2 s is required to compensate for the drop in speed. The transient of the motor torque M that occurs when the target speed w* is reduced is particularly disadvantageous.

[0090] The FIGS 16 to 19show the results of an inventive procedure. These results occurred when the drive system was maintained at the target speed w*, which corresponds to the rolling speed vW, before the initial time t2. The initial time t2 is approximately 0.1 s before the piercing time t3. FIG 16 shows the time course of the speed or angular velocity ω 1 of the drive 4, FIG 17 the temporal progression of the rotational speed or angular velocity ω n of the driven rollers 3. In an analogous manner, the FIGS 18 and 19the temporal progression of the motor torque M and the loading torque ML. It is evident that the speed ω n of the driven rollers 3 does not collapse at the piercing time t3. The same applies to the angular velocity ω 1 of the drive 4. Furthermore, it is evident that the speed ω n of the driven rollers 3 increases shortly before the piercing time t3 and that the - in terms of magnitude - maximum motor torque M is below the maximum of the procedures according to the FIGS 8 to 11 or according to the FIGS 12 to 15 lies.

[0091] The approach of the FIGS 16 to 19 The drive 4 and the driven rollers 3 must be temporarily braked shortly after the initial time t2. This results in both the motor torque M and the applied torque ML becoming negative shortly after the initial time t2. With the procedure of FIGS 20 to 23 , which also show results of an inventive procedure, this can be avoided.

[0092] The approach of the FIGS 20 to 23 The starting time t2 is also approximately 0.1 s before the penetration time t3. However, before the starting time t2, the drive system was maintained at a target speed w* that is significantly below the speed corresponding to the rolling speed vW, specifically at approximately 70% of this speed. The corresponding reduction in speed can be easily achieved, for example, by gradually allowing the drive 4 to coast down after a rolling stock rolled before the rolling stock 2 under consideration here has coasted down, until the drive system reaches the desired (low) target speed w*.

[0093] The FIG 20 shows the time course of the speed or angular velocity ω 1 of the drive 4, FIG 21 the temporal progression of the rotational speed or angular velocity ω n of the driven rollers 3. In an analogous manner, the FIGS 22 and 23the temporal progression of the motor torque M and the loading torque ML. It is evident that not only is the drop in the speed ω n of the driven rollers 3 at the piercing time t3 avoided. In addition, a continuous increase in the speeds ω 1 , ω n of the drive 4 and the driven rollers 3 can be achieved, whereby the loading torque ML also continues to rise continuously and, in addition, the motor torque M remains above 0 at any time t. The maximum of the motor torque M is still below the maximum of the procedures according to the FIGS 8 to 11 or according to the FIGS 12 to 15 . The procedure of the FIGS 20 to 23 is both more robust and more energy efficient than the state-of-the-art approaches and the approach according to the FIGS 16 to 19The present invention has many advantages. In particular, within the scope of the present invention, the drop in speed during the first pass is not subsequently corrected, but rather, by means of suitable pre-control, it is ensured from the outset that the peripheral speed vU of the driven rolls 3 and the load torque ML acting on the driven rolls 3 are already set at the first pass time t3 such that they correspond exactly to the desired values ​​vW, MW. This is ensured by the inventive predictive determination of the torque curve M for the period from the initial time t2 to the first pass time t3. By means of the inventive procedure, the drop in speed during the first pass can be completely or at least almost completely avoided. The inventive procedure can also be easily retrofitted to existing rolling stands.The procedure is independent of the control system underlying control device 7. Commissioning and testing, for example, for stability, can be performed independently of other controllers.

[0094] Although the invention has been illustrated and described in detail by the preferred embodiments, the invention is not limited by the disclosed examples and other variations may be derived therefrom by those skilled in the art without departing from the scope of the invention. List of reference symbols

[0095] 1Roll stand 2Rolling stock 3Driven rolls 4Drive 4'Converter 4"Electric motor 5Drive train 6Rolling stock head 7Control device 8Control program 9Machine code 10Model 11, 12Couplings 13Spring elements 14Damper elements 15Tables aParameter A, A ω , A ωϕ A ϕω Matrices b ω , eVectors c 1 , ci ,,c n-1 Spring constants d 1 , di , d n-1 Material damping D 1 , D i , D n Friction coefficients fa Function J 1 , J i , J n Moments of inertia MTorque curve MminMinimal value MmaxMaximum value M, M2, M3, M'Motor torque MLActuating torque MWRolling torque QWeighting matrix RGeighting factor S1 to S6Steps tTime / Time points general t1, t2, t3Time points TPeriod vRolling material speed vWRolling speed x, x2, x3, x', x ω , x ϕ States Δϕ 1 , Δϕ i , Δϕ n-1 twists ω*, ω 1 , ω i , ω n angular velocities

Claims

1. Operating method for a rolling stand (1) in which a metal rolling stock (2) having a rolling stock head (6) is to be rolled, - wherein a control device (7) for the rolling stand (1) is provided with a starting time (t3), a rolling speed (vW) and a rolling torque (MW) before the rolling stock head (6) enters the rolling stand (1), - wherein the starting time (t3) is the time at which the rolling stock head (6) is to enter the rolling stand (1), and the rolling stock (2) is to be rolled at the starting time (t3) with the rolling speed (vW) and the rolling torque (MW), - wherein the control device (7) is to be rolled before a starting time (t2) prior to the starting time (t3) for a period extending from the starting time (t2) to the starting time (t3),a torque curve (M) is determined for a drive (4) driving rolls (3) of the rolling stand (1) via a drive train (5) and the drive (4) is controlled during the period according to the determined torque curve (M), characterized by that the control device (7) determines the torque curve (M) starting from an initial state (x2) of the drive (4), the drive train (5) and the driven rollers (3) of the rolling stand (1) given at the initial time (t2) in such a way that the driven rollers (3) of the rolling stand (1) are subjected to the rolling torque (MW) by the drive train (5) at the piercing time (t3) and rotate at a peripheral speed (vU) matched to the rolling speed (vW).

2. Operating method according to claim 1, characterized by thatthe control device (7) determines the torque curve (M) in such a way that at the piercing time (t3) the drive (4), the drive train (5) and the driven rollers (3) of the rolling stand (1) rotate at mutually corresponding speeds and changes in the speeds have the value zero.

3. Operating method according to claim 1 or 2, characterized by that the control device (7) determines the torque curve (M) using a model (10) which models the drive (4), the drive train (5) and the rolls (3) of the rolling stand (1) driven by the drive (4) based on mathematical-physical equations.

4. Operating method according to claim 3, characterized by that the model (10) models the drive (4), the drive train (5) and the driven rolls (3) of the rolling stand (1) as a multi-mass vibration system with a number of damped-elastically connected elements.

5. Operating method according to claim 3 or 4, characterized by that the control device (7) uses an optimization problem to determine the torque curve (M), into which the respective torque (M) of the drive (4) is entered as input variables for a plurality of discrete times (t) within the period, and determines the torque curve (M) by solving the optimization problem.

6. Operating method according to claim 1 or 2, characterized by that the control device (7) determines the torque curve (M) using tables (15) stored in the control device (7), to which the piercing time (t3), the rolling speed (vW), the rolling torque (MW) and the initial state (x2) of the drive (4), the drive train (5) and the driven rolls (3) of the rolling stand (1) are supplied as input variables, or that the torque curve (M) is transmitted to the control device (7) by a parameterizable function (fa ) of time (t) and the control device (7) determines the parameters (a) of the function (f a ) is determined based on the piercing time (t3), the rolling speed (vW), the rolling torque (MW) and the initial state (x2) of the drive (4), the drive train (5) and the driven rolls (3) of the rolling stand (1).

7. Operating method according to one of the above claims, characterized by that in the initial state (x2) the drive (4), the drive train (5) and the driven rollers (3) of the rolling stand (1) rotate at corresponding speeds and changes in the speeds have the value zero.

8. Operating method according to one of the above claims, characterized by that in the initial state (x2) the peripheral speed (vU) of the driven rollers (3) is smaller than the peripheral speed (v) of the driven rollers (3) matched to the rolling speed (vW).

9. Operating method according to one of the above claims, characterized by that the control device (7) determines the torque curve (M) in such a way that the torque (M) generated by the drive (4) and / or an application torque (ML) with which the driven rolls (3) of the rolling stand (1) are driven is greater than a respective predetermined minimum value during the entire period.

10. Operating method according to one of the above claims, characterized by that the control device (7) determines the torque curve (M) in such a way that a rotational acceleration with which a rotational speed of the drive (4) changes and / or a rotational acceleration with which a rotational speed of the driven rollers (3) changes is greater than a respective predetermined minimum value during the entire period.

11. Operating method according to one of the above claims, characterized by thatthe control device (7) determines the torque curve (M) in such a way that energy consumption of the drive (4) is minimal during the period.

12. Operating method according to one of the above claims, characterized by that the period is between 50 ms and 200 ms, in particular between 75 ms and 150 ms.

13. Control program comprising machine code (9) which can be processed by a control device (7) for a rolling stand (1) for rolling a rolling stock (2) made of metal having a rolling stock head (6), wherein the processing of the machine code (9) by the control device (7) causes the control device (7) to carry out an operating method according to one of the above claims.

14. Control device for a rolling stand (1) for rolling a rolling stock (2) made of metal having a rolling stock head (6), wherein the control device is programmed with a control program (8) according to claim 13, so that it carries out an operating method according to one of claims 1 to 12 during operation.

15. Rolling device for rolling a rolled stock (2) made of metal and having a rolling stock head (6), - wherein the rolling device has a rolling stand (1) in which the rolled stock (2) is rolled, - wherein the rolling stand (1) has a drive (4), a drive train (5) and rollers (3), - wherein the rollers (3) can be driven by means of the drive (4) via the drive train (5), - wherein the rolling device has a control device (7) for the rolling stand (1), - wherein the control device (7) is designed as a control device (7) according to claim 14.

Images