Electrically excited synchronous machine with improved current determination

Abstract

The invention relates to an electrically excited synchronous machine (1) comprising a stator (2) having a stator winding (3) and a rotor (5) having an excitation winding (6). A control device (9) of the synchronous machine (1) operates the synchronous machine (1) with a torque (M) which assumes a high value during load phases (12) and, during load pauses (13) between the load phases (12), assumes a value which is at least one order of magnitude lower than the minimum value (M1) of the torque (M) during the load phases (12). The control device (9) rotates the rotor (5) during the load phases (12) with a constant direction of rotation for each load phase (12) and at a rotational speed (n) above a minimum rotational speed (nmin). The control device (9) determines, on the basis of an angular position (p) of the rotor (5), a d-coordinate and a q-coordinate of a coordinate system defined by the angular position (p) of the rotor (5). The d-coordinate coincides with the direction of a flux generated by an excitation current (Ie) flowing in the excitation winding (6). The q-coordinate is oriented orthogonally to the d-coordinate in the coordinate system. Prior to a load phase (12), still during the load pause (13), the control device (9) adjusts a q-axis current (Iq) to an initial value (Iqa). During the load phase (12), the control device (9) regulates the q-axis current (Iq) to at most twice the initial value (Iqa). No later than at the start of each load phase (12), the control device (9) activates a controller (17) which, during each load phase (12), tracks the d-axis current (Id) and the excitation current (Ie), but not the q-axis current (Iq). Tracking is performed such that an actual value (x) of the rotor (5) remains equal to a corresponding setpoint value (x*), and the d-axis current (Id) and the excitation current (Ie) satisfy the relationship: Id = - ü x Ie + F / Xh x cos(α). While the q-axis current (Iq) is being adjusted, the control device (9) adjusts the

Description

[0001] 202400311 1

[0002] Description

[0003] Title of the invention

[0004] Electrically excited synchronous machine with improved current measurement

[0005] field of technology

[0006] The present invention relates to an operating method for an electrically excited synchronous machine comprising a stator with a stator winding and a rotor with an excitation winding,

[0007] - wherein a control device of the synchronous machine operates the synchronous machine with a torque that has a high value during load phases and a value that is at least one order of magnitude smaller than the minimum value of the torque during the load phases during load breaks between load phases,

[0008] - wherein the control device rotates the rotor during the load phases with a uniform direction of rotation for the respective load phase and with a speed above a minimum speed,

[0009] - wherein the control device determines a d-coordinate and a q-coordinate of a coordinate system determined by the rotational position of the rotor, based on a rotational position of the rotor,

[0010] - where the d-coordinate corresponds to the direction of a flow caused by an excitation current flowing in the excitation winding,

[0011] - where the q-coordinate is determined by being oriented orthogonally to the d-coordinate in the coordinate system determined by the rotational position of the rotor.

[0012] The present invention further relates to a control program for a software-programmable control device of an electrically excited synchronous machine, wherein the control program comprises machine code that can be directly executed by the control device, wherein the execution of the machine code by the control device causes the control device to perform such an operating procedure.

[0013] The present invention further relates to a control device for an electrically excited synchronous machine, wherein the control device is programmed with such a control program, so that the control device performs such an operating procedure.

[0014] The present invention further assumes an electrically excited synchronous machine,

[0015] - wherein the synchronous machine has a stator with a stator winding and a rotor with an excitation winding, 202400311 2

[0016] - wherein the synchronous machine has a control device,

[0017] - wherein the synchronous machine has a position sensor connected to the control device for detecting the rotational position of the rotor,

[0018] - wherein the control device is designed as such a control device which controls the synchronous machine according to such an operating procedure.

[0019] State of the art

[0020] The aforementioned operating procedure corresponds to a field-oriented control system. Field-oriented control is generally known.

[0021] German patent application DE 10 2011 005 291 A1 discloses an operating method for an electrically excited synchronous machine comprising a stator with a stator winding and a rotor with an excitation winding. Among other things, it describes the tapping process associated with a load step in a rolling mill. According to DE 10 2011 005 291 A1, significant load fluctuations, including a reversal of the direction of the load torque, can occur during the load phases.

[0022] From DE 10 2017202 714 A1 an operating method for an electrically excited synchronous machine is known, which has a stator with a stator winding and a rotor with an excitation winding.

[0023] From the technical paper "Optimal dynamic current control for externally excited synchronous machines," by Johannes Reinhard et al., 2024 IEEE Conference on Control Technology and Applications (CCTA), August 21-23, 2024, Newcastle upon Tyne, UK, pages 146 to 152, an operating method for an electrically excited synchronous machine is known. The machine has a stator with a stator winding and a rotor with an excitation winding. The aforementioned paper poses an optimization problem to determine the optimal control of the synchronous machine for a given load requirement, including its change over time.

[0024] In each of the aforementioned writings, a conventional vector control is explained, so that the writings also reveal that a control device of the synchronous machine determines a d-coordinate and a q-coordinate of a coordinate system determined by the rotational position of the rotor, where the d-coordinate corresponds to the direction of a flux caused by an excitation current flowing in the excitation winding, and the q-coordinate is determined by being oriented orthogonally to the d-coordinate in the coordinate system determined by the rotational position of the rotor.

[0025] Summary of Invention 202400311 3

[0026] Synchronous machines are used in many applications. In some applications, phases of high load (load phases) occur repeatedly, interspersed with phases of low load (load breaks). The transition from a load phase to a load break, and especially vice versa, can be abrupt. One example of an application where such an abrupt change between load phases and load breaks occurs is the use of a synchronous machine to drive a rolling mill. At the start of the milling process, when the head of the material enters the mill, a load phase begins abruptly. Similarly, a load phase ends abruptly when the foot of the material exits the mill.

[0027] In practice, it has been shown that abrupt changes, particularly from a load break to a load phase, can sometimes lead to an unstable state of the synchronous machine and, consequently, to malfunctions in the equipment driven by the synchronous machine. While these malfunctions can be avoided by using a larger drive, this solution is associated with increased costs. It is therefore only implemented when no other way to prevent the malfunctions exists.

[0028] The object of the present invention is to create possibilities by which an unstable state of the synchronous machine can be reliably avoided during the transition from a load break to a load phase.

[0029] The problem is solved by an operating method with the features of claim 1. Advantageous embodiments of the operating method are the subject of dependent claims 2 to 10.

[0030] According to the invention, an operating method of the type mentioned above is designed by:

[0031] - that the control device, before a load phase and during the load break, sets a q-current directed in the direction of the q-coordinate to an initial value,

[0032] - that the control device regulates the q-current to a maximum of twice the initial value during the load phase,

[0033] - that the control device activates a controller implemented by the control device at the latest at the beginning of the load phase following the load break, which adjusts the d-current and the excitation current, but not the q-current, during the load phase in such a way that the actual size of the rotor remains equal to the setpoint and the d-current and the excitation current follow the relationship 202400311 4 fulfill, and

[0034] - that the control device adjusts the d-current and the excitation current during the adjustment of the q-current to the initial value in such a way that a torque generated by the synchronous machine during the adjustment of the q-current to the initial value remains essentially constant,

[0035] - where Id is the d-current, ü is a turns ratio between the excitation winding and the stator winding, le is the excitation current, F is a resultant magnetic flux of the synchronous machine caused by the d-current, the q-current and the excitation current, Xh is a main inductance of the synchronous machine, a is a flux angle formed by the resultant magnetic flux with the d-coordinate, and the actual quantity and the set quantity are a speed or a torque.

[0036] The inventive method makes it particularly possible to build up a high q-current in the stator even before the load phase. This is advantageous because the q-current must be built up against the damping effect of the main inductance of the synchronous machine (equivalent to the term "main reactance of the synchronous machine") and therefore cannot be changed quickly. In contrast, the d-current and the excitation current can be built up and, if necessary, reduced very quickly if appropriately coordinated. It is therefore possible to make the corresponding adjustment, particularly of the excitation current, very quickly during the transition from the load pause to the load phase. This is important because the torque generated by the synchronous machine is proportional to the product of the excitation current and the q-current. The coordination of the d-current and the excitation current is achieved by adhering to the relationship

[0037] F

[0038] Id = —ü • le - cos(cz)

[0039] Xh is guaranteed.

[0040] Naturally, the synchronous machine must be able to generate the required torque during the respective load phase. The condition that the control unit regulates the q-current to a maximum of twice the initial value during the load phase therefore implies a corresponding determination of the initial value depending on the torque required during the load phase.

[0041] Preferably, the control device does not utilize the maximum possible increase in the q-current during the load phase. In particular, it is possible that the control device regulates the q-current during the load phase to a maximum of only 1.5 times or 1.2 times the initial value. In this case, the above statements do not apply with regard to 202400311 5.

[0042] Not double the initial value, but rather 1.5 times or 1.2 times the initial value. The initial value can therefore be correspondingly larger.

[0043] Preferably, the controller first determines the excitation current and then calculates the d-current using the aforementioned relationship. This has the particular advantage that the excitation current required to change the generated torque can be calculated first, and only then is the d-current determined based on this calculation. This allows for a faster adjustment of the generated torque to the required load torque.

[0044] Preferably, during the load break following each load phase, the control device keeps the controller activated at least until the d-current and the excitation current have reached their respective static values. This ensures, in particular, a stable transition from the respective load phase to the respective load break.

[0045] Unlike the transition from a load break to a load phase, the q-current no longer needs to be maintained at a high value from the start of the load break. Instead, the control unit can reduce the q-current during the load break following the respective load phase. This reduction can begin while the controller is still active. Alternatively, the control unit can reduce the q-current only during the load break or also during the period while the controller remains active. Reducing the q-current only during the period while the controller remains active is possible if a complete or generally sufficient reduction of the q-current to a desired value is achievable during this period. Otherwise, the q-current will continue to decrease even after this period has elapsed.

[0046] Preferably, the control device adjusts the rotor's direction of rotation during each load break according to the load phase following the load break. Adjusting the rotor's direction of rotation can be combined with either maintaining or inverting the rotor's direction of rotation, as required. Preferably, the control device adjusts the sign of the q-current to adjust the rotor's direction of rotation. In this case, inverting the rotor's direction of rotation is linked to inverting the sign of the q-current. Using the q-current to adjust the rotor's direction of rotation is advantageous because, in this case, the excitation current can always have the same sign, regardless of the rotor's direction of rotation.A current controller, by means of which the excitation current is set (for example, an inverter), therefore only needs to be able to cover a single quadrant (positive direction of rotation, positive current) and can thus be designed simply and cost-effectively. 202400311 6.

[0047] Preferably, the initial value of the q-current is determined by ensuring that the magnitude of the resulting magnetic flux lies within a predetermined range and / or that, at a predetermined rotational speed, the magnitude of the motor voltage applied to the stator winding lies within a predetermined range. This ensures, for example, that while the resulting magnetic flux is built up to a significant extent at the beginning of the load phase, the synchronous machine is not yet in saturation and / or that a sufficient voltage reserve is available during the load phase for building up the d-current and, if necessary, also for varying the q-current.The predetermined speed is preferably either the speed at which the rotor of the synchronous machine should rotate at the beginning of the load phase, or a maximum speed at which the rotor of the synchronous machine should rotate during the load phase. However, other speeds are also possible.

[0048] Preferably, during the load break before a load phase, from the point at which the q-current, d-current, and excitation current are established—that is, when the synchronous machine has been "prepared" for the next load phase—the magnitude of the q-current is greater than the magnitude of the d-current and also greater than the excitation current scaled with the turns ratio between the excitation winding and the stator winding. The q-current thus represents the lion's share of the resulting motor current, the magnitude of which is the square root of the sum of the squares of the d-current and the q-current. In particular, during this period, the magnitude of the q-current can be at least three times, and preferably at least five times, greater than the magnitude of the d-current. The same applies to the excitation current scaled with the turns ratio between the excitation winding and the stator winding.In a preferred embodiment, the control device implements a flux or voltage regulator by means of which the resulting magnetic flux or the motor voltage is adjusted without affecting the torque generated by the synchronous machine and without affecting the flux angle. This allows for adjustments to the resulting magnetic flux or the motor voltage—albeit relatively slowly—without any adverse effects on the generated torque. Such adjustment of the resulting magnetic flux or the motor voltage can also be used, for example, in the context of so-called field weakening. Field weakening, i.e., a reduction of the q-current, is particularly necessary at higher speeds to ensure that the magnitude of the motor voltage remains within its permissible or achievable range.

[0049] Similarly, the control device preferably also implements a flux angle controller, by means of which the flux angle is set without affecting the torque generated by the synchronous machine and without affecting the resulting magnetic flux. This also makes it possible to adjust the flux angle – albeit relatively slowly – without having to fear any effects on the generated torque.

[0050] Preferably, during the adjustment of the q-current to its initial value, the control device adjusts the d-current and the excitation current in such a way that the torque generated by the synchronous machine remains essentially constant. The term "essentially constant" means that the generated torque fluctuates by 30% or less during the adjustment of the q-current to its initial value, but in any case is at most 50% of the minimum value of the torque generated by the synchronous machine during the subsequent load phase.

[0051] In a preferred application of the above principle, the synchronous machine drives the rolls of a rolling stand. In this case, the load phases correspond to times during which a metal stock is present in a roll gap of the rolling stand. Thus, the load breaks correspond to times during which no stock is present in the roll gap. Optionally, the load phases can also include short periods immediately before the start of the milling process, if the drive train of the rolling stand (synchronous machine, gearbox, driven rolls) is to be prepared for the start of the milling process by appropriately accelerating the synchronous machine.

[0052] The problem is further solved by a control program with the features of claim 11. According to the invention, the execution of the machine code by the control device causes the control device to execute an operating method according to the invention.

[0053] The problem is further solved by a control device with the features of claim 12. According to the invention, the control device is programmed with a control program according to the invention, such that the control device executes an operating method according to the invention.

[0054] The problem is further solved by an electrically excited synchronous machine with the features of claim 13. According to the invention, the control device is designed as a control device according to the invention, which controls the synchronous machine according to an operating method according to the invention.

[0055] Brief description of the drawings

[0056] The properties, features, and advantages of this invention described above, as well as the manner in which they are achieved, will become clearer and more readily understandable in connection with the following description of an exemplary embodiment, which is further explained in conjunction with the drawings. The drawings shown are: 202400311 8

[0057] FIG 1 an electrically excited synchronous machine including control,

[0058] FIGS. 2 and 3 Time diagrams,

[0059] FIG 4 a rolling mill, a rolled material and a synchronous machine,

[0060] FIGS. 5A and 5B show a flowchart.

[0061] FIG 6 a pointer diagram,

[0062] FIG 7 a time diagram,

[0063] FIG 8 a pointer diagram,

[0064] FIG 9 a block diagram and

[0065] FIG 10 a pointer diagram.

[0066] Description of the embodiments

[0067] According to FIG. 1, an electrically excited synchronous machine 1 has a stator 2 with a stator winding 3. A motor current Im can be fed into the stator winding 3 by a converter 4. The motor current Im is multiphase, usually three-phase. The synchronous machine 1 also has a rotor 5 with an excitation winding 6. An excitation current le can be fed into the excitation winding 6 by a current controller 7. The current controller 7 can be a converter. The excitation current le is usually a direct current.

[0068] The current rotational position p of the rotor 5 is known to a control unit 9 of the synchronous machine 1. It is possible that the control unit 9 determines the current rotational position p based on internal data. However, the synchronous machine 1 typically has a position sensor 8, which measures the current rotational position p of the rotor 5. In this case, the position sensor 8 is connected to the control unit 9 and continuously transmits the current rotational position p to the control unit 9. The control unit 9, in turn, continuously determines setpoints for the motor current Im and the excitation current le and controls the inverter 4 and the current controller 7 such that they provide the corresponding motor current Im and the corresponding excitation current le.

[0069] The control unit 9 is software-programmable. The control unit 9 is programmed with a control program 10. The control program 10 comprises machine code 11, which can be executed directly by the control unit 9. The execution of the machine code 11 by the control unit 9 causes the control unit 9 to execute an operating procedure, which is explained in more detail below. The execution of the operating procedure causes, in particular, the control unit 9 to control the synchronous machine 1 according to such an operating procedure. 202400311 9

[0070] The operating method according to the invention relates to a special operating mode of the synchronous machine 1. This operating mode is explained schematically below in conjunction with FIGS. 2 and 3.

[0071] FIG. 2 shows the torque M generated and delivered by the synchronous machine 1 as a function of time t. According to FIG. 2, the torque M with which the control device 9 operates the synchronous machine 1 is high during load phases 12. Between the load phases 12, there are load breaks 13, during which the torque M with which the control device 9 operates the synchronous machine 1 is considerably lower.

[0072] The exact value of the torque M during each load phase 12 is of secondary importance. It can be a constant or a variable value. Likewise, the exact value of the torque M during the load breaks 13 is also of secondary importance. It can be a constant or a variable value. For this reason, only the corresponding ranges are shown in FIG. 2. The significant differences between them, however, are important. In particular, the minimum value M1 of the torque M during the load phases 12 is considerably larger than the maximum value M2 of the torque M during the load breaks 13. Specifically, there is at least one order of magnitude, i.e., a factor of 10, between the minimum value M1 of the torque M during the load phases 12 and the maximum value M2 of the torque M during the load breaks 13. Sometimes there are two or more orders of magnitude between them.However, an exception may apply during periods of the respective load break 13, during which a reversal of the direction of rotation of the synchronous machine 1 takes place.

[0073] Furthermore, the control unit 9 rotates the rotor 5 according to FIG. 3 during the load phases 12 with a uniform direction of rotation for each load phase 12. The rotational speed n can vary, but during the load phase 12 it always remains (in magnitude) above a minimum rotational speed nmin. The direction of rotation of the rotor 5 also remains constant during each load phase 12. During the load breaks 13, the rotational speed n can decrease below the minimum rotational speed nmin. Alternatively, the direction of rotation can be reversed during the load breaks 13. However, neither the decrease in the rotational speed n nor the reversal of the direction of rotation is mandatory.

[0074] FIG. 4 shows a typical application where such a change between load phases 12 and load breaks 13 occurs repeatedly. According to FIG. 4, the synchronous machine 1 drives rolls 14 of a rolling stand 15. In this case, the load phases 12 of FIGS. 2 and 3 correspond to times during which a metal stock 16 is located in a roll gap of the rolling stand 15. Likewise, the load breaks 13 correspond to times during which no metal stock 16 is located in the roll gap. 202400311 10

[0075] It is possible that the rolling stand 15 is part of a multi-stand rolling mill. In this case—see the curved arrow at the upper roll 14—the material 16 would always pass through the rolling stand 15 in the same direction. Thus, the direction of rotation of the rolls 14, and therefore also of the rotor 5 of the synchronous machine 1, would be the same for every rolling pass. However, it is also possible that the rolling stand 15 is a reversing stand. In this case—see the curved double arrow at the lower roll 14—the material 16 would pass through the rolling stand 15 from left to right in some cases and from right to left in others. Thus, the direction of rotation of the rolls 14, and also the direction of rotation of the synchronous machine 1, would change after each rolling pass or at least between some rolling passes.

[0076] The operating method according to the invention is explained below in conjunction with FIGS. 5A and 5B.

[0077] Figures 5A and 5B assume that the synchronous machine 1 is initially in a load pause 13. In this state, the control unit 9 is first informed of the rotational position p of the rotor 5 in step S1, as shown in Figures 5A and 5B. For example, in step S1, the control unit 9 can receive the rotational position p of the rotor 5 from the position sensor 8.

[0078] The rotational position p of the rotor 5 defines a two-dimensional coordinate system with a d-direction and a q-direction. The d-direction and the q-direction rotate with an (electrical) angular velocity w2 when the rotor 5 rotates with a (mechanical) angular velocity w1. The mechanical and the electrical angular velocities w1, w2 are, as is generally known to those skilled in the art, related to each other via the number of pole pairs of the synchronous machine 1. The d-coordinate corresponds to the direction of a flux induced by the excitation current le. The q-coordinate is determined by being oriented (electrically) orthogonal to the d-coordinate in the coordinate system. The control device 9 is therefore able to determine the d-coordinate and the q-coordinate in one step S2 based on the rotational position p of the rotor 5. This determination is generally known to those skilled in the art.

[0079] In a subsequent step S3, the control unit 9 determines a q-current Iq, a d-current Id, and the excitation current le. The d-current Id is directed in the direction of the d-coordinate, and the q-current in the direction of the q-coordinate. During the determination in step S3, as shown in FIG. 6, all three currents Iq, Id, and le have relatively small values. Consequently, the resulting magnetic flux F of the synchronous machine 1 caused by the d-current Id, the q-current Iq, and the excitation current le is relatively small. The same applies to the generated torque M of the synchronous machine 1. In FIG. 6, ü denotes the turns ratio between the excitation winding 6 and the stator winding 3, and Xh denotes a 202400311 11

[0080] Main inductance of synchronous machine 1. Furthermore, the resulting magnetic flux F forms a flux angle α with the d-coordinate. The dashed quarter circle shows the maximum value of the resulting magnetic flux F divided by the main inductance of synchronous machine 1. The latter also applies to the phasor diagrams of FIGS. 8 and 10. This allows, in particular, a qualitative comparison of the phasor diagrams of FIGS. 6, 8, and 10.

[0081] In step S4, the d-current Id and the q-current Iq are transformed into the motor current Im for the stator winding 3. This transformation is performed using the rotational position p. A conversion of the excitation current le is not required. In step S5, the inverter 4 and the current controller 7 are controlled accordingly to supply the motor winding 3 with the motor current Im and the excitation winding 6 with the excitation current le.

[0082] In step S6, the control unit 9 checks whether a transition to the next load phase 12 is imminent. As long as this is not the case, the control unit 9 returns to step S1. The repeated execution of steps S1 to S5 thus results in the synchronous machine 1 (more precisely: its rotor 5) rotating at a specific speed n (which can be high or low) with a low torque M in a (practically) load-free state.

[0083] If the check in step S6 indicates that the next load phase 12 is imminent, the control unit 9 proceeds to step S7. In step S7, the control unit 9 sets the q-current Iq to an initial value Iqa. While the setting of the q-current Iq to the initial value Iqa still occurs during the respective load break 13, it usually takes place shortly before the subsequent load phase 12.

[0084] The initial value Iqa is determined by ensuring that the magnitude of the resulting magnetic flux F lies within a predetermined range and / or that, at a predetermined rotational speed n, the magnitude of the motor voltage Um applied to the stator winding 3 lies within a predetermined range. The motor voltage Um is a vector that has a component for each phase, or, in the dq coordinate system, for each coordinate. For example, the motor voltage Um can be at least 50% of the maximum possible motor voltage Ummax, as shown in FIG. 7. The initial value Iqa can also be determined such that the motor voltage Um is at an even higher value, for example, 60%, 66% (= 2 / 3), 80%, or 83% (= 5 / 6) of the maximum possible motor voltage Ummax. FIG. 8 shows, as a phasor diagram, the state in which the q-current Iq is set to the initial value Iqa. 202400311 12

[0085] From FIG. 8 – more precisely, from the comparison of FIGS. 6 and 8 – it can also be seen that the control device 9, while adjusting the q-current Iq to the initial value Iqa, adjusts in particular the excitation current le and consequently also the d-current Id. This adjustment is carried out in such a way that the torque M generated by the synchronous machine 1 while adjusting the q-current Iq to the initial value Iqa remains essentially constant. For this purpose, the excitation current le must be reduced. The d-current Id is adjusted accordingly.

[0086] Figure 7 also shows that – even during the load break 13 before a load phase 12 – once the q-current Iq is set to the initial value Iqa, the magnitude of the q-current Iq is greater than the magnitude of the d-current Id and also greater than the excitation current le scaled with the turns ratio ü. As a rule, the magnitude of the q-current Iq is at least three times the magnitude of the d-current Id and is also at least three times the magnitude of the excitation current le scaled with the turns ratio ü. Preferably, the factor is not only 3, but 5.

[0087] In step S7, the excitation current le and the d-current Id are coordinated in such a way that they establish the relationship This requirement can be met beforehand, but it is not absolutely necessary.

[0088] After the q-current Iq is set to the initial value Iqa (usually immediately afterwards), the control unit 9 activates a controller 17 in step S8 (see FIG. 9). Activation thus occurs at the latest at the beginning of the load phase 12.

[0089] After step S8 has been executed, and thus also during the load phase 12, the control unit 9 is informed of the rotational position p of the rotor 5 again in step S9. Therefore, in step S10, the control unit 9 is also able to determine the d-coordinate and the q-coordinate based on the rotational position p. Steps S9 and S10 correspond one-to-one with steps S1 and S2.

[0090] According to FIG. 9, the controller 17 receives an actual value x and a corresponding setpoint x* (or their difference). As shown in FIG. 9, the actual value x can alternatively be a rotational speed n or a torque M. Similarly, the setpoint x* can alternatively be a rotational speed n* or a torque M*. In any case, however, the actual value x is identical to the setpoint x*. Therefore, the actual value x and the setpoint x* are either both rotational speeds n, n* or both torques M, M*. 202400311 13

[0091] While active, controller 17 adjusts the d-current Id and the excitation current le so that the actual value x of rotor 5 remains equal to the setpoint x*. Controller 17 does not adjust the q-current Iq. When adjusting the d-current Id and the excitation current le, controller 17 observes the following relationship

[0092] The controller 17 adjusts the d-current Id and the excitation current le in such a way that the aforementioned relationship is maintained. In particular, this ensures that the torque M generated by the synchronous machine 1 is indeed changed by the correspondingly coordinated adjustment of the d-current Id and the excitation current le. This change is made for the purpose of maintaining the actual value x of the rotor at its setpoint x*. The controller 17, however, has no influence on the resulting magnetic flux F or on the flux angle α.

[0093] For example, as shown in FIG. 9, the controller 17 can first determine the excitation current le and then determine the d-current Id using the aforementioned relationship. In the flowchart of FIGS. 5A and 5B, the control unit 9 can, for example, receive the actual value x in step S11. Depending on whether the actual value x is greater or less than the setpoint x*, the control unit 9 increments or decrements the excitation current le in subsequent steps S12 to S15 using the controller 17. In a further step S16, the control unit 9 can then determine the d-current Id using the controller 17 according to the relationship explained above.

[0094] The tracking of the d-current Id and the excitation current le occurs, as already mentioned, from the activation of the controller 17 and thus also during the load phase 12. In contrast to the load break 13, however, the d-current Id and the excitation current le assume considerably larger values ​​in the load phase 12, as shown in FIG. 10. In particular, the d-current Id can be larger – possibly even considerably larger – than the q-current Iq. Due to the fact that the d-current Id and the excitation current le are linked by the relationship described above, this also applies analogously to the excitation current le, which is scaled with the transformation ratio ü.

[0095] In a subsequent step S17, the control unit 9 determines the q-current Iq. One possible method of determination will be discussed later.

[0096] In steps S18 and S19, the d-current Id and the q-current are transformed into phase currents for the stator winding 3, and the converter 4 and the current controller 7 are controlled accordingly. Steps S18 and S19 correspond 1:1 to steps S4 and S5. Analogous to steps S4 and S5, a conversion of the excitation current le is therefore not necessary.

[0097] In step S20, the control unit 9 checks whether the load phase 12 has ended. If the load phase 12 has not yet ended, the control unit 9 returns to step S9.

[0098] The repeated execution of steps S9 to S19 thus results in the synchronous machine 1 maintaining the actual value x as close as possible to the setpoint x*. Furthermore, during the load phase 12, the q-current Iq is either not changed or, if it is changed, not to a significant extent. In particular, during the load phase 12, the control unit 9 regulates the q-current Iq (more precisely: its magnitude) to a maximum of twice the initial value Iqa. Preferably, at least with regard to the magnitude, the change is even smaller, in particular to a maximum of 1.5 times or 1.2 times the initial value.

[0099] When the load phase 12 is complete, the control unit 9 proceeds to step S21. In step S21, with controller 17 still activated, the control unit 9 reduces the d-current Id and the excitation current le until they reach their respective static values. Because controller 17 is activated, it continues to use the relationship when reducing the excitation current le and the d-current Id.

[0100] In principle, the control unit 9 performs steps within step S20 that are analogous to steps S9 to S19. Preferably, step S21 is further configured such that the control unit 9 also reduces the q-current Iq. This is possible because the synchronous machine 1 only needs to generate a small amount of its torque M.

[0101] In step S22, the control unit 9 checks whether the d-current Id and the excitation current le have reached their respective static values. It is possible that step S22 is implemented such that the control unit 9 checks whether an additional time interval has elapsed since the end of the load phase 12. However, it is also possible that step S22 checks whether the condition to be achieved is met.

[0102] As long as the d-current Id and the excitation current le have not yet reached their respective static values, the control unit 9 returns to step S21. Otherwise, the control unit 9 proceeds to step S23. In step S23, the control unit 9 deactivates the controller 17. The control unit 9 can then return to step S1. 202400311 15

[0103] It is possible that control unit 9 will also stop lowering the q-current Iq when transitioning to step S23. Alternatively, it is possible that control unit 9 will lower the q-current Iq even further in step S23.

[0104] Typically, additional steps S24 and S25 are present, which are executed before the control unit 9 returns to step S1. In step S24, the control unit 9 is informed of the direction of rotation of the rotor 5 for the next load phase 12. In step S25, the control unit 9 adjusts the direction of rotation of the rotor 5 accordingly. In many cases, the direction of rotation of the rotor 5 is adjusted by the control unit 9 changing the sign of the q-current Iq.

[0105] As already mentioned, the q-current Iq is kept constant, or at least substantially constant, during the respective load phase 12. An exception to keeping the q-current Iq constant (completely or substantially) applies only if the synchronous machine 1 is operated, or must be operated, in field weakening mode.

[0106] If the q-current Iq is to be adjusted, the control device 9 preferably implements a flux controller 18 (see FIG. 9). The resulting magnetic flux F can be adjusted by means of the flux controller 18. With a suitable design of the flux controller 18, it is even possible to adjust the resulting magnetic flux F without affecting the torque M generated by the synchronous machine 1 and without affecting the flux angle α. The corresponding operation of the flux controller 18 is explained below.

[0107] Assume that at a given time, the three currents Iq, Id, and le have specific values. These values ​​are referred to below as Iq1, Id1, and Ie1. These values ​​result in a specific resultant magnetic flux F = F1 (more precisely, its magnitude) and a specific flux angle α. Now, the resultant magnetic flux F is to be changed from the value F1 to a new value F2 while maintaining the flux angle α. For this purpose, the flux controller 18 first determines a new value Iq2 for the q-current Iq. The new value Iq2 is calculated as follows:

[0108] F2 Iql = - sin(cz) .

[0109] Xh

[0110] This procedure is well-known and therefore does not need to be explained in detail.

[0111] Changing the q-current Iq to the aforementioned value Iq2 without simultaneously adjusting the excitation current le would result in a change in the torque M. However, the torque M should not be changed. Therefore, the 202400311 16 must first be additionally

[0112] The excitation current le is changed from its value Ie1 to a value Ie2, so that the relationship

[0113] Ie ■ Iql = Iq2 ■ lei remains.

[0114] Changing the q-current Iq and the excitation current le, coupled together, without adjusting the d-current Id, would maintain the torque M but change the flux angle a. However, the flux angle a should also remain unchanged. Therefore, starting from the new values ​​Iq2 for the q-current Iq and Ie2 for the excitation current le, the d-current must also be adjusted to the new value Id2.

[0115] F2

[0116] Id2 = -ü - Ie2 -i - cos(cz)

[0117] Xh can be changed. This change does not alter the torque M, since the d-current Id has no influence on the torque M. This approach makes it possible to adjust the resulting magnetic flux F without affecting the torque M generated by the synchronous machine 1 and also without affecting the flux angle α.

[0118] As an alternative to a flux regulator 18, the control unit 9 could also implement a voltage regulator. The procedure would be entirely analogous. In this case, the motor voltage Um could be adjusted without affecting the torque M and the flux angle α. By adjusting the resulting magnetic flux F or the motor voltage Um, the magnitude of the motor voltage Um can be adjusted as needed. In particular, the control unit 9 can slightly reduce the q-current Iq if otherwise the magnitude of the motor voltage Um would exceed its permissible limit. Conversely, the control unit 9 can slightly increase the q-current Iq. Furthermore, any necessary field weakening can be easily implemented.

[0119] Alternatively or additionally to the flux regulator 18 (or the voltage regulator), the control device 9 can implement a flux angle regulator 19, by means of which the flux angle α can be set and controlled. With a suitable design of the flux angle regulator 19, it is even possible to set the flux angle α without affecting the torque M generated by the synchronous machine 1 and without affecting the resulting magnetic flux F. The corresponding operation of the flux angle regulator 19 is explained below. 202400311 17

[0120] Assume that at a specific time – as before – the three currents Iq, Id, and le have specific values. These values ​​are again referred to below as Iq1, Id1, and le1. These values ​​result in a specific net magnetic flux F and a specific flux angle a = a1. Now, the flux angle a is to be changed from the value a1 to a new value a2 while maintaining the net magnetic flux F. For this purpose, the flux angle controller 19 first determines a new value Iq2 for the q-current Iq. The new value Iq2 is calculated as follows:

[0121] F

[0122] Iq2 = - sin(cz2) .

[0123] Xh

[0124] This procedure is well-known and therefore does not need to be explained in detail.

[0125] Changing the q-current Iq to the aforementioned value Iq2 without simultaneously adjusting the excitation current le would—just as before when adjusting the resulting magnetic flux F—result in a change in the torque M. However, the torque M should not be changed. Therefore, as before, the excitation current le must first be changed from its value le1 to a value Ie2, so that the relationship

[0126] Tel ■ Iql = Iq2 ■ Ie2 remains.

[0127] Changing the q-current Iq and the excitation current le, coupled without adjusting the d-current Id, would maintain the torque M but change the magnitude of the resulting magnetic flux F. However, the magnitude of the resulting magnetic flux F should also remain unchanged. Therefore, starting from the new values ​​Iq2 for the q-current Iq and Ie2 for the excitation current le, the d-current must also be adjusted to the new value Id2.

[0128] F

[0129] Id2 = -ü ■ Ie2 -i - cos(cz2)

[0130] Xh can be changed. This change does not alter the torque M, since the d-current Id has no influence on the torque M. This approach makes it possible to adjust the flux angle α without affecting the torque M generated by the synchronous machine 1 and also without affecting the magnitude of the resulting magnetic flux F. 202400311 18

[0131] The fact that the currents Iq, Id, and le are modified by controller 17, flux controller 18, and flux angle controller 19 only appears to be an inconsistency. In practice, the controls can be executed sequentially and therefore—where necessary—build upon the result of the preceding control.

[0132] This provides three controllers 17 to 19, by means of which - each decoupled from each other - the actual quantity x can be controlled with high dynamics and the resulting magnetic flux F and the flux angle a can be controlled with lower dynamics.

[0133] The present invention offers many advantages. The controller 17 is easy to program and commission. The same applies, if applicable, to the controllers 18 and 19. During the load phases 12, the synchronous machine 1 behaves very similarly to a compensated DC machine. The behavior of the synchronous machine 1 is therefore also easy to understand intuitively. Adjusting the current Id and the excitation current le, and thus maintaining the setpoint x*, is possible with high dynamics (generally in the range of low double-digit millisecond values). This is made possible by the fact that the resulting magnetic flux F is not changed by the coordinated change of the current Id and the excitation current le. Only stray inductances need to be overcome. An undesirable drop in speed n during the transition from a load break 13 to a load phase 12 can be avoided or at least minimized.Similarly, during the transition from a load phase 12 to a load break 13, an undesirable increase in rotational speed n can be avoided or at least minimized. Because the direction of rotation of the rotor 5 is set by the sign of the q-current Iq, the current controller 7 can be designed very simply. While adjusting the q-current Iq is not very dynamic, this is not critical as long as the synchronous machine 1 is in a load break 13. During a load phase 12, high dynamics are ensured by the coordinated adjustment of the d-current Id and the excitation current le. A tendency for the synchronous machine 1 to tilt can be avoided because the rotational position p of the rotor 5 can be measured using the position sensor 8. It is not necessary to know the motor parameters of the synchronous machine 1 exactly, nor is a flux model required.Losses can be minimized because only small currents flow in the so-called damper cage during the load phases 12 when adjusting the generated torque M. Losses are only generated in the damper cage for adjusting the resulting magnetic flux F and the flux angle α.

[0134] Although the invention has been further illustrated and described by the preferred embodiments, the invention is not limited by the disclosed examples and other variations can be derived from them by a person skilled in the art without departing from the scope of protection of the invention. 202400311 19

[0135] Reference symbol list

[0136] 1 synchronous machine

[0137] 2 Stator

[0138] 3 Stator winding

[0139] 4 inverters

[0140] 5 Rotor

[0141] 6 Excitation development

[0142] 7 current controllers

[0143] 8 Position sensors

[0144] 9 Control unit

[0145] 10 Tax program

[0146] 11 Machine code

[0147] 12 load phases

[0148] 13 breaks

[0149] 14 rollers

[0150] 15 Rolling mill

[0151] 16 Rolled goods

[0152] 17 to 19 regulators

[0153] F, F1, F2 resultant magnetic fluxes ld, Id1, Id2 d-current le, Ie1, Ie2 excitation current

[0154] In the motor current lq, Iq1 , Iq2 q-current

[0155] IQA initial value

[0156] M, M', M1, M2 Torques n, n* Speeds nmin Minimum speed

[0157] P rotation position

[0158] S1 to S25 steps t time

[0159] Um, Ummax motor voltages ü gear ratio x actual value x* target value

[0160] Xh main inductance a, a1 , a2 flux angle

Claims

202400311 20 Claims 1. Operating method for an electrically excited synchronous machine (1) comprising a stator (2) with a stator winding (3) and a rotor (5) with an excitation winding (6), - wherein a control device (9) of the synchronous machine (1) operates the synchronous machine (1) with a torque (M) which has a high value during load phases (12) and has a value during load breaks (13) between the load phases (12) which is at least one order of magnitude smaller than the minimum value (M1) of the torque (M) during the load phases (12), - wherein the control device (9) rotates the rotor (5) during the load phases (12) with a uniform direction of rotation for the respective load phase (12) and with a speed (n) above a minimum speed (nmin), - wherein the control device (9) determines a d-coordinate and a q-coordinate of a coordinate system determined by the rotational position (p) of the rotor (5) based on a rotational position (p) of the rotor (5), - where the d-coordinate corresponds to the direction of a flow caused by an excitation current (le) flowing in the excitation winding (6), - wherein the q-coordinate is determined by being oriented orthogonally to the d-coordinate in the coordinate system determined by the rotational position (p) of the rotor (5), characterized by , - that the control device (9) before a load phase (12) and during the load break (13) sets a q-current (Iq) directed in the direction of the q-coordinate to an initial value (Iqa), - that the control device (9) regulates the q-current (Iq) during the load phase (12) to a maximum of twice the initial value (Iqa), - that the control device (9) activates a controller (17) implemented by the control device (9) at the latest at the beginning of the load phase (12) following the load break (13), which adjusts the d-current (Id) and the excitation current (le), but not the q-current (Iq), during the load phase (12) such that the actual value (x) of the rotor (5) remains equal to the setpoint (x*) and the d-current (Id) and the excitation current (le) follow the relationship fulfill, and - that the control device (9) adjusts the d-current (Id) and the excitation current (le) during the adjustment of the q-current (Iq) to the initial value (Iqa) in such a way that a torque (M) generated by the synchronous machine (1) during the adjustment of the q-current (Iq) to the initial value (Iqa) remains essentially constant, 202400311 21 - where Id is the d-current (Id), ü is a turns ratio (ü) between the excitation winding (6) and the stator winding (3), le is the excitation current (le), F is a resultant magnetic flux (F) of the synchronous machine (1) caused by the d-current (Id), the q-current (Iq) and the excitation current (le), Xh is a main inductance (Xh) of the synchronous machine (1), a is a flux angle (a) formed by the resultant magnetic flux (F) with the d-coordinate, and the actual quantity (x) and the set quantity (x*) are a rotational speed (n, n*) or a torque (M, M').

2. Operating method according to claim 1, characterized in that the controller (17) first determines the excitation current (le) and then determines the d-current (Id) based on the relationship determined.

3. Operating method according to claim 1 or 2, characterized in that the control device (9) keeps the controller (17) activated during the load pause (13) following the respective load phase (12) at least until the d-current (Id) and the excitation current (le) have reached a respective static value.

4. Operating method according to claim 3, characterized in that the control device (9) reduces the q-current (Iq) only during or also during the period during which the control device (9) keeps the controller (17) activated during the load pause (13) following the respective load phase (12).

5. Operating method according to one of the above claims, characterized in that the control device (9) during a respective load break (13) sets the direction of rotation of the rotor (5) according to a load phase (12) following the load break (13) and that the control device (9) sets the sign of the q-current (Iq) to set the direction of rotation of the rotor (5).

6. Operating method according to one of the above claims, characterized in that the magnitude of the initial value (Iqa) of the q-current (Iq) is determined by the fact that the Be- 202400311 22 the resulting magnetic flux (F) is within a predetermined range and / or at a predetermined rotational speed (n) the magnitude of a motor voltage (Um) applied to the stator winding (3) is within a predetermined range.

7. Operating method according to one of the above claims, characterized in that during the load break (13) before a load phase (12) from the point at which the q-current (Iq) is set to the initial value (Iqa) the magnitude of the q-current (Iq) is greater than the magnitude of the d-current (Id) and greater than the excitation current (le) scaled with the turns ratio (ü) between the excitation winding (6) and the stator winding (3), in particular at least three times and preferably at least five times as large as the magnitude of the d-current (Id) and in particular at least three times, preferably at least five times as large as the excitation current (le) scaled with the turns ratio (ü) between the excitation winding (6) and the stator winding (3).

8. Operating method according to one of the above claims, characterized in that the control device (9) implements a flux or voltage regulator (18) by means of which the resulting magnetic flux (F) or the motor voltage (Um) is adjusted without affecting the torque (M) generated by the synchronous machine (1) and without affecting the flux angle (a).

9. Operating method according to one of the above claims, characterized in that the control device (9) implements a flux angle controller (19) by means of which the flux angle (a) is set without influencing the torque (M) generated by the synchronous machine (1) and without influencing the resulting magnetic flux (F).

10. Operating method according to one of the above claims, characterized in that the synchronous machine (1) drives rolls (14) of a rolling stand (15), that the load phases (12) correspond to times during which a rolled material (16) made of metal is located in a rolling gap of the rolling stand (15), and that the load breaks (13) correspond to times during which no rolled material (16) is located in the rolling gap.

11. Control program for a software-programmable control device (9) of an electrically excited synchronous machine (1), wherein the control program comprises machine code (11) that can be directly executed by the control device (9), wherein the execution of the machine- 202400311 23 nencodes (11) by the control device (9) causes the control device (9) to execute an operating procedure according to one of the above claims.

12. Control device for an electrically excited synchronous machine (1), wherein the control device is programmed with a control program (10) according to claim 12, such that the control device performs an operating procedure according to one of claims 1 to 10.

13. Electrically excited synchronous machine, - wherein the synchronous machine has a stator (2) with a stator winding (3) and a rotor (5) with an excitation winding (6), - wherein the synchronous machine has a control device (9), - wherein the synchronous machine has a position sensor (8) connected to the control device (9) for detecting the rotational position (p) of the rotor (5), - wherein the control device (9) is configured as a control device (9) according to claim 12, which controls the synchronous machine according to an operating method according to one of claims 1 to 10.

Images