TEST BENCH AND METHOD FOR PERFORMING A TEST RUN ON A TEST BENCH

Description

[0001] The present invention relates to a method for carrying out a test run on a test bench, wherein a test specimen is connected to a load machine on the test bench, wherein a test bench automation unit specifies a target torque to a torque controller in accordance with the test run, whereby an actual torque of the load machine is set by the torque controller, and wherein a test specimen control variable is specified to the test specimen by a test specimen controller.Furthermore, the present invention relates to a test bench for carrying out a test run, wherein a test specimen is connected to a load machine on the test bench for carrying out the test run, wherein a test bench automation unit, a test specimen controller and a torque controller are provided, wherein the test specimen controller specifies a test specimen control variable to the test specimen and the test bench automation unit specifies a target torque to the torque controller in accordance with the test run, whereby the torque controller sets an actual torque of the load machine.

[0002] In the development of internal combustion engines, powertrains with internal combustion engines, and vehicles with internal combustion engines, the testing of emissions and fuel consumption plays a central role. In all phases of development, the testing of the internal combustion engine's emissions and fuel consumption is carried out on test benches, such as engine test benches, powertrain test benches, or chassis dynamometers. However, the legal framework for this testing is currently undergoing significant changes. While previously standardized driving cycles, such as the New European Driving Cycle (NEDC), were primarily used for this testing, testing under real-world driving conditions is now also required.For the emission behavior, so-called Real Driving Emission (RDE) test runs are used, in which no specific driving cycle is specified, but a more or less random driving route is covered, which only has to meet certain defined framework conditions.

[0003] Therefore, it is necessary, on the one hand, to collect measured values ​​from real test drives with a vehicle on a real road, such as GPS data, engine speed, accelerator pedal position, vehicle speed, etc. Second, the recorded measured values ​​must be used to create a test run representative of the test drive for a test bench, which can then be run in whole or in part on the test bench in order to record and evaluate, for example, the emissions or consumption behavior of the combustion engine.

[0004] On the test bench, a test object, such as an internal combustion engine, is operated according to the test run specifications, either alone (engine test bench) or in combination with other components (powertrain test bench, chassis dynamometer). To perform the test run, the internal combustion engine on the test bench is connected, directly or indirectly, to one or more load machines (dynamometers).

[0005] For the dynamometer, for example, the torque or speed can serve as the control variable. An actual torque or speed can be set according to a target torque or speed using a torque controller or speed controller. The dynamometer generates a load torque for the test specimen, for example, via a connecting test bench shaft. The target torque or speed is specified to the torque controller or speed controller by a test bench automation unit according to the specifications of the test run.

[0006] The setting of the actual torque or speed by the torque controller or speed controller can be performed in the form of a "control" in the automation sense, which means that there is no feedback of the actual torque or speed. Alternatively, the actual torque or speed can be fed back to the torque controller or speed controller, which corresponds to a "control" in the automation sense.

[0007] A test object controller is provided for the test object, which can essentially "regulate" or "control" the actual test object speed or the actual test object torque as a controlled variable or control variable. For this purpose, a manipulated variable can be specified for the test object. The manipulated variable can be a speed or torque, but also a manipulated variable that affects the speed or torque, such as an accelerator pedal position, an injection quantity, an effective cylinder mean pressure, an ECU engine torque, an injection quantity, an injection timing, an ignition timing, etc. The manipulated variable can also be specified for a test object, for example, a passively designed one, without setting a specific actual test object variable.

[0008] Thus, the test bench automation unit specifies a (discrete-time or continuous-time) time profile of the target torque, which was generated, for example, from a real test drive, to the torque controller based on the test run specifications. Likewise, the test object controller specifies a test object manipulated variable to the test object, which can be used to set an actual test object torque or an actual test object speed as the test object actual variable. The load machine and the test object form an open mechanical loop. This means that an actual test object variable (e.g., an actual test object torque) is not merely mechanically transmitted from one test object interface to another test object interface, but that the actual test object variable is fully supported by one or more load machines.To perform the test run, the test specimen is loaded directly by one or more loading machines via a test specimen interface. This direct loading ensures a rigid connection between the test specimen and the loading machine, thus ensuring dynamic and precise loading of the test specimen. Furthermore, the direct loading allows for easy operation of the test bench, while the test run is clearly traceable.

[0009] Either a target torque is specified for a torque controller of the dynamometer, or a target speed is specified for a speed controller of the dynamometer. If both a speed controller and a torque controller are present, switching between a target torque and a target speed is possible; however, the speed controller and torque controller cannot be active at the same time. However, if such switching occurs between the speed controller and torque controller, discontinuities can arise in the arrangement, which must be avoided. As a discontinuity, for example, the actuating torque of the dynamometer can change abruptly, which also causes the actual speed and the actual torque to change abruptly and in an unintended manner.

[0010] WO / 2018096085 A1 relates to a test bench with a test specimen and a dynamometer, in which two torque components are added to produce a dynamometer torque. However, speed limits are not taken into account.

[0011] DE 28 28 920 describes a test bench with a torque controller that can be deactivated depending on the speed. This ensures that the control is switched off if the speed reaches excessively high values.

[0012] It is an object of the present invention to provide a test bench for a test specimen which is directly loaded by one or more loading machines, which avoids the disadvantages described above and ensures the most robust control method possible.

[0013] This object is achieved according to the invention by determining the actual speed of the loading machine and determining at least one deviation of at least one attribute of the actual speed from at least one limit value, wherein based on the at least one deviation at least one additive torque correction value is determined and superimposed on the target torque.

[0014] Furthermore, the object is achieved in that a speed determination unit for determining a speed of the loading machine is provided on the test bench, wherein at least one comparison unit is provided which is designed to determine at least one deviation of at least one attribute of the actual speed from at least one limit value, wherein at least one correction unit is provided which is designed to determine an additive torque correction value based on the at least one deviation and to superimpose it on the target torque.

[0015] The test object can be, for example, a drive (torque generator), a drive train component, a drive train, a vehicle, etc. Inputs and / or outputs of the test object can be connected to one or more load machines, whereby the connection can also be made via adaptor gears.

[0016] Possible test benches include, for example, engine test benches, (differential) gearbox test benches, power packs (engine and gearbox), complete drive trains (up to the half shaft), roller test benches, drive train elements (clutch, converter, center differential, differential, ...).

[0017] According to the invention, it is assumed that a torque controller is provided for the load machine. Depending on the design, the torque controller can either "regulate" the actual torque in the automation sense or "control" the actual torque in the automation sense.

[0018] In the present application, the torque controller and the test object controller can each essentially perform open-loop and / or closed-loop control in the automation sense. This means that the torque controller and the test object controller do not necessarily have to be pure "controllers" in the automation sense, but can also be "controllers" in the automation sense.

[0019] Thus, the test bench automation unit specifies a target torque based on the test run to the torque controller in order to adjust, i.e., regulate or control, the actual torque. This can be done directly by specifying a control torque, or by specifying a control variable equivalent to the torque, or a control variable from which the torque can be derived. Other possible control variables include currents, voltages, valve positions, etc.

[0020] A test object controller is provided for the test object, which specifies a test object manipulated variable for the test object. The test object controller can perform a "closed-loop control" of the test object's actual variable (test object torque, test object speed) or a "closed-loop control" of the test object's actual variable, given the test object's target value. A test object target value can be provided; if the test object is passive (e.g., purely mechanical), it may be sufficient if only a test object manipulated variable and no test object target value is specified for the test object. For example, an accelerator pedal position can serve as the test object manipulated variable, which is referred to as T / α control or T / α open-loop control (T for the torque of the load machine and α for the accelerator pedal position). If a speed is specified as the test object manipulated variable for the test object instead of the accelerator pedal position, this is referred to as T / N control orT / N control, where T stands for the torque of the loading machine and N for the speed of the test specimen.

[0021] This list is only an example; any control variable can be specified by the test object controller, which is referred to as T / * control or T / * control, whereby the asterisk is used as a placeholder for any control variable of the test object.

[0022] During operation of a T / * closed-loop or T / * open-loop control, usually only the actual torque of the load machine can be adjusted, which provides little flexibility and makes it impossible to react to changed, particularly increased, actual speeds. At increased actual speeds, there is a risk that components of the test bench, in particular the test object, could be damaged. Therefore, according to the invention, the actual speed is determined using a torque determination unit. Furthermore, at least one attribute of the actual speed is compared with a limit value, and a deviation of the attribute from the limit value is determined. Based on the current deviation, a torque correction value is determined and superimposed on the target torque. This allows the actual speed of the test object to be influenced via the target torque.

[0023] Advantageously, the value of the actual speed is used as a first attribute of the actual speed, and an upper and / or lower speed limit is provided as the limit value. A first correction unit can comprise a speed correction controller, preferably a PI controller, and a first superposition unit. The speed correction controller receives a speed deviation of the actual speed from a speed limit from a first comparison unit and, based on the speed deviation, determines a torque correction value and superimposes it on the actuating torque via the first superposition unit.

[0024] This allows the actual speed value to be monitored as an attribute. This ensures that an upper / lower speed limit is not exceeded / undershot by superimposing the additive torque correction value on the target torque.

[0025] The speed correction controller can be activated, preferably when a first speed limit is exceeded, and / or deactivated, preferably when a second speed limit is undershot. This activation and / or deactivation can be performed by a user and / or the test bench automation unit.

[0026] Preferably, an actual speed gradient of the actual speed is used as a second attribute of the actual speed, with an upper and / or lower speed gradient limit being provided as the limit value. A second correction unit can comprise a speed gradient correction controller, preferably a PI controller, and a second superposition unit. The speed gradient correction controller receives a first deviation of the actual speed gradient from a speed gradient limit from a second comparison unit, determines a torque correction value from the first deviation, and superimposes the torque correction value on the actuating torque via the first superposition unit.

[0027] If an actual speed gradient is considered an attribute, a rapid change in speed can be responded to by adjusting the target torque. This ensures that the rate of change of speed does not exceed a limit. The actual speed gradient can be determined, for example, by comparing two consecutive absolute speed values.

[0028] The speed gradient correction controller can be activated, preferably when a first speed gradient limit is exceeded, and / or deactivated, preferably when a second speed gradient limit is undershot. This activation and / or deactivation can be performed by a user and / or the test bench automation unit.

[0029] Advantageously, an actual torque is determined and fed back to the torque controller for controlling the actual torque. Accordingly, a torque determination unit can be provided, which is designed to determine the actual torque and is connected to the torque controller in order to feed back the actual torque for controlling the actual torque.

[0030] In automation terms, this results in closed-loop control of the actual torque. If, as is typical in state-of-the-art control, the actual torque were fed back to the torque controller to control the actual torque, and a non-controllable target torque were specified for the torque controller, it would only be possible to react to an attribute of the actual torque to be controlled. However, it may happen that high actual speeds occur on the load machine, which do not necessarily have to be accompanied by an increased actual torque. In particular, an increased actual speed can occur when a clutch between the combustion engine and the load machine is opened. Current state-of-the-art control of the actual torque would therefore not be able to react to such a change in speed.However, since according to the invention the actual speed of the load machine is determined, at least one deviation of the at least one attribute of the actual speed from the at least one limit value is determined, and an additive torque correction value is determined based on the at least one deviation and superimposed on the target torque, a response to a changed actual speed is possible. Thus, an attribute of the actual speed can be kept within the specified limit values ​​via the actual torque.

[0031] Advantageously, the test bench automation unit provides the test object controller with a setpoint value in order to adjust the actual value of the test object by specifying the test object's manipulated variable. Thus, the test bench automation unit can be configured to specify a setpoint value to the test object controller, and the test object controller can be configured to adjust the actual value of the test object by specifying the test object's manipulated variable.

[0032] Advantageously, the actual size of the test object can be determined and fed back to the test object controller for regulating the actual size. Thus, an actual size determination unit can be provided to determine the actual size of the test object, with the actual size determination unit being connected to the test object controller in order to feed back the actual size for regulating the actual size. Thus, the test object controller is designed as a "controller" in the automation sense for the actual size of the test object.

[0033] Advantageously, at least one limit value is predefined. By specifying at least one fixed limit value, a fixed limit for an attribute of the actual speed can be implemented. If a lower limit value is provided, any undershoot of this lower limit value can be monitored. If an upper limit value is provided, any overshoot of this upper limit value can be monitored.

[0034] Advantageously, at least one limit value can be changed, preferably depending on a known system variable. The timing of the test run, for example, can be used as a known system variable. For example, a lower speed limit can be kept at zero before activating the test object. After activating the test object, the lower speed limit can be raised to prevent the test object from stopping during the test run.

[0035] If an engine is calibrated as a test specimen on an engine test bench, a target torque is specified for the load machine and an accelerator pedal position (which in turn influences the engine torque) is specified for the engine. As part of the calibration, the test bench automation unit varies, for example, the injection timing. Since the injection timing changes not only the combustion quality but also the engine torque, the operating point is not completely stable, which can also cause the engine speed to change. At extreme values, the engine can run away or stall, which can be prevented by monitoring the actual speed and limiting it via an intervention.Therefore, it is not necessary for the motor to be thermally stabilized again after a shift in the operating point, since the speed limits can be dynamically adjusted to keep the motor always close to the intended operating point.

[0036] For example, if dynamic operation of a powertrain test bench with a speed-controlled test object (engine, prime mover, etc.) and a torque-controlled load machine is planned, it may happen that the test object's torque is unexpectedly temporarily not maintained, for example, because a clutch has been opened. The load machine would then react to an increased actual speed depending on the specified speed limit. However, if it is planned that no reaction will occur in this specific case, the speed limit can be temporarily shifted.

[0037] The present invention is described below with reference to the Figures 1 to 3 which show exemplary, schematic and non-limiting advantageous embodiments of the invention. Fig. 1 a test bench with a torque controller for the loading machine, Fig. 2 a comparison unit and correction unit, Fig. 3 a special embodiment of a first and second correction unit.

[0038] In Fig.1 A typical test bench 10 for a test object 1 is shown. Examples of test benches 10 that can be used are engine test benches, (differential) transmission test benches, power packs (engine and transmission), complete drive trains (up to the half shaft), roller test benches, drive train elements (clutch, converter, center differential, differential, etc.).

[0039] On test bench 10, test object 1 can, for example, be physically constructed as part of a vehicle, an internal combustion engine. However, test object 1 can also be a drivetrain, a drivetrain component, an entire vehicle, or another component to be tested. An input and / or output of test object 1 is connected to one or more load machines 2, for example, via a test bench shaft. The connection between test object 1 and the load machine can also be established via matching gears.

[0040] A test object controller R1 and a torque controller R2 are provided on the test bench 10. The test object controller R1 and / or the torque controller R2 can be implemented as an integral part of the test bench automation unit 5, as a standalone unit, as part of a converter of a test object controller or load machine controller, etc. The function of the test object controller R1 and / or the torque controller R2 can be based on field-oriented control, particularly when a synchronous machine serves as test object 1 and / or load machine 2. Field-oriented control can run on a real-time system and provide switching pulses to transistors. The test object controller R1 and / or the torque controller R2 can also be integrated into such a real-time system. The correction unit B, B1, B2 according to the invention can also be integrated into such a real-time system.

[0041] A torque controller R2 can control a single load machine 4 or a combination of several load machines 4. The test specimen 1 can thus be connected to several load machines 4, for example, via a test bench summing gear.

[0042] To perform the test run, a test bench automation unit 5 determines the corresponding temporal progression of a target torque T_soll for the load machine 2 and makes the target torque T_soll available to the torque controller R2. To generate the progression of the target torque T_soll, simulation hardware and / or simulation software can be provided on the test bench automation unit 5, with which the test drive of a vehicle is simulated. For this purpose, a simulation model, which includes, for example, a driver model, a vehicle model, and an environmental model, can be implemented in the test bench automation unit 5. Other models, such as a tire model, a road model, etc., can also be implemented.

[0043] During the test run, an actual torque T_actual is established on the load machine 2. In the illustrated embodiment, the actual torque T_actual is determined by a torque determination unit 22 and fed back to the torque controller R2. The torque determination unit 22 can be, for example, a torque sensor on the test bench shaft or an observer that estimates the actual torque T_actual from other measured variables. The actual torque T_actual can, of course, also be derived from another (measured) variable. The torque controller R2 further determines a control torque T and transmits this to the load machine 2, at which, as mentioned, the actual torque T_actual is established.

[0044] The test object 1 is also operated on the test bench 10 according to the specifications of a test run, with the test object controller R1 specifying a test object manipulated variable SW for the test object 1. This can be done, for example, to obtain information regarding certain measured variables. The measured variables can be pollutant emissions, (fuel) consumption, acoustic behavior of the test object 1, etc. Depending on the measured variable, a corresponding measuring unit, such as an emissions measuring unit which receives exhaust gas from the test object, can be provided on the test bench 1. An emissions measuring unit measures the emission of at least one pollutant, such as CO 2 , CO, NO x , and / or the total mass of hydrocarbons (THC) and / or a particle number (such as soot particles). A consumption measuring unit which measures the fuel consumption of the test object can also be provided as a measuring unit.

[0045] In the illustrated embodiment, the test bench automation unit 5 is designed to specify a test object target value W to the test object controller R1 according to the test run. Furthermore, in the illustrated embodiment, the test object controller R1 is designed to set a test object actual value M of the test object 1 by specifying the test object manipulated variable SW, and an actual value determination unit 11 is provided to determine the test object actual value M of the test object 1 and feed it back to the test object controller R1 for controlling the test object actual value M. Thus, the test object controller R1 is designed to regulate the test object actual value M of the test object 1 to the test object target value W by specifying a test object manipulated variable SW to the test object 1. An actual engine torque, an actual engine speed, etc. can serve as the test object actual value M, whereas a target engine torque or a target engine speed can serve as the test object target value W.For example, an accelerator pedal position α can be determined as the test object control variable SW, which is passed to the engine control unit ECU for controlling the test object 1.

[0046] The variables actual torque T_actual, target torque T_target, actuating torque T, test object actual variable M, test object target variable W, test object manipulated variable SW, actual speed n, etc. can also be understood as time-continuous or time-discrete curves during the test run.

[0047] In Fig. 2a speed determination unit 21 is provided for detecting the actual speed n. The actual speed n is fed to a comparison unit V, which determines at least one deviation a of at least one attribute of the actual speed n from at least one limit value G and forwards it to a correction unit B. The correction unit B is designed to determine an additive torque correction value Tk based on the at least one deviation a and to superimpose it on the target torque T_soll, thereby resulting in a modified target torque T_soll'. The torque controller R2 now regulates the actual torque T_ist to the modified target torque T_soll', thereby resulting in a modified actual speed n'. The modified target torque T_soll' thus affects the actual speed n via a modified actual torque T_ist', thereby resulting in a modified actual speed n'.This ensures that the attribute of the (now modified) actual speed n does not exceed the specified limit value G.

[0048] Correction unit B is connected upstream of torque controller R2, preventing torque controller R2 from correcting the torque correction value Tk again, i.e., from adjusting the now modified actual torque T_actual' back to the unmodified actual torque T_actual. Since the target torque T_set is corrected upstream of torque controller R2 by adding it to the corrected target torque T_set', torque controller R2 does not even notice this intervention in the target torque T_set.

[0049] The additive torque correction value Tk can of course assume positive and negative values ​​depending on the deviation a, whereby a superposition on the target torque T_soll can lead to an increased or decreased modified target torque T_soll'.

[0050] If, as shown, the actual torque T_actual of the test object is controlled according to a target torque T_target using the torque controller R2, not only can a change in the actual torque T_actual be responded to, but an attribute of the actual speed n can also be monitored. By superimposing the torque correction value Tk, it can also be ensured that the attribute of the actual speed n does not exceed or fall below its limit value G.

[0051] If the actual speed n itself is used as the first attribute of the actual speed n, the actual speed n can be compared with an upper speed limit no or lower speed limit nu as a limit value G. If the first deviation a1 between the actual speed n and the upper or lower speed limit no or nu exceeds (or falls below), a negative or positive additive torque correction value Tk is determined, which is superimposed on the target torque T_soll and leads to a reduced or increased modified target torque T_soll.

[0052] The upper speed limit no or the lower speed limit nu can be predefined or adjusted during the test run. It is also possible to adjust the upper speed limit no or the lower speed limit nu depending on other system variables, e.g., the system variables of test object 1. If an upper speed limit no and a lower speed limit nu are specified as limit values ​​G for the actual speed n, the actual speed n of the load machine 2 can, for example, be controlled such that it remains within a specified speed range between the upper speed limit no and the lower speed limit nu.

[0053] If an actual speed gradient of the speed dn is used as the second attribute of the actual speed n, the actual speed gradient dn can be compared with an upper or lower speed gradient limit dn o or dn u as a limit value G in order to determine a second deviation a2. If the second deviation a2 exceeds (or falls below) the upper or lower speed gradient limit dn o , a negative or positive additive torque correction value Tk is determined, which is superimposed on the target torque T_soll and leads to a reduced or increased modified target torque T_soll.

[0054] The selection of the lower speed gradient limit dn u , but especially the upper speed gradient limit dn o can serve to protect the test object 1. If exceeding a certain actual speed gradient dn is considered harmful to the test object, the upper speed gradient limit dn o can be selected such that reaching this certain actual speed gradient dn is prevented.

[0055] If an upper speed gradient limit dn o and a lower speed gradient limit dn u are specified as limit values ​​G for the actual speed gradient dn, the actual speed n of the loading machine 2 can, for example, be controlled such that the speed gradient dn remains within a specified speed gradient range between the upper speed gradient limit dn o and the lower speed gradient limit dn u.

[0056] Basically, it can also be provided that the at least one additive torque correction value Tk is superimposed on the target torque T_soll only if the at least one deviation a exceeds a maximum deviation and / or falls below a minimum deviation.

[0057] Fig. 3represents a special embodiment of the present invention, wherein a first correction unit B1 comprises a speed correction controller Rn, preferably an upper and a lower PI controller (shown here only as a PI controller), and a first superposition unit B11. A first comparison unit V1 determines a first deviation a1 of the actual speed n from a speed limit dn o , dn u as at least one limit value G and transfers this first deviation a1 to the speed correction controller Rdn. The speed correction controller Rdn determines a torque correction value Tk from the first deviation a1 and transmits this to the first superposition unit B11, which superimposes the torque correction value Tk on the target torque T.

[0058] The first superposition unit B11 intervenes additively on the target torque T_soll. This means that the torque correction value Tk is superimposed on the target torque T_soll by the superposition unit B11, i.e., added or subtracted depending on the sign, resulting in a modified target torque T_soll'.

[0059] The control range, i.e. working range, of the upper and / or lower PI controller is preferably dependent on the first deviation a1 and the target torque T_soll, with which a suitable torque correction value Tk is determined. The control range of the upper PI controller can be selected such that it only intervenes when an upper limit value G (upper speed limit dn o ) is exceeded, while the control range of the lower PI controller can be designed such that it only intervenes when a lower limit value G (lower speed limit dn u ) is undershot. The lower speed limit dn u is preferably zero. The controller parameters of the upper and lower PI controllers can preferably be defined by a common parameter set.

[0060] Furthermore, a second correction unit B2 is provided, which comprises a speed gradient correction controller Rdn, preferably a PI controller, and a second superposition unit B12. An actual speed gradient dn is determined from the actual speed n, which can be done by the second comparison unit V2 or an upstream determination unit. Fig. 2the actual speed gradient dn is specified to the second comparison unit V2 and is thus determined from the actual speed n by a determination unit (not shown). The second comparison unit V2 determines the second deviation a2 between the actual speed gradient dn and an upper and / or lower speed gradient limit dn o , dn u , which is present as a limit value G. This second deviation a2 is fed to the speed gradient correction controller Rdn, which processes it and, based on this, transmits a torque correction value Tk to a second superimposition unit B21. The second superimposition unit B21 superimposes the torque correction value Tk on the target torque T_soll, which leads to a modified target torque T_soll'. The speed gradient correction controller Rdn can be designed as described above with regard to the speed correction controller Rn.Thus, an upper and lower PI controller can also be provided, whose control ranges can also be selected such that the respective PI controller only intervenes when the speed gradient dn exceeds an upper speed gradient limit dn o and / or falls below a lower speed gradient limit dn u . The actual speed gradient dn can be provided with a sign, with the lower speed gradient limit dn u having a negative sign.

[0061] Thus, a first superposition unit B11 can modify the target torque T_soll by superimposing a torque correction value Tk on the modified target torque T_soll' and the second superposition unit B21 can further modify the already modified target torque T_soll' by superimposing a (further) torque correction value Tk or - if the second superposition unit B21 is arranged upstream of the first superposition unit B11 - vice versa.

[0062] The comparison units V1, V2 and / or speed correction controller Rn and / or speed gradient correction controller Rdn, etc. can be designed as an integral part of the test bench automation unit 5, as an independent unit, as a function of the test object controller R1 and / or torque controller R2, etc.

[0063] If the first superposition unit B11 does not modify the target torque T_soll (e.g., because the speed correction controller Rn does not initiate this due to the first deviation a1 being too small), the second superposition unit can still determine an additive torque correction value Tk and superimpose it on the target torque T_soll. Similarly, if the second superposition unit B11 does not modify the actuating torque T (e.g., because the speed gradient correction controller Rn does not initiate this due to the second deviation a2), the first superposition unit B11 can still determine an additive torque correction value Tk and superimpose it on the target torque T_soll.

[0064] Although a speed correction controller Rn and a speed gradient correction controller Rdn are provided in the illustrated embodiment, a modification of the target torque T_toll using only one speed correction controller Rn or only one speed gradient correction controller Rdn is of course also conceivable.

[0065] The speed correction controller Rn can be configured to be activated or deactivated. For example, the speed correction controller Rn can be activated when the actual speed n exceeds a first speed limit and deactivated when a second speed limit is undershot. The first and second speed limits can, of course, be identical. If they do not match, a hysteresis is generated.

[0066] The speed gradient correction controller Rdn can be configured to be activated or deactivated. For example, the speed gradient correction controller Rn can be activated depending on whether the actual speed gradient dn exceeds a first speed gradient limit and deactivated when it falls below a second speed gradient limit. The first and second speed gradient limits can, of course, be identical. If they do not, a hysteresis is generated.

Claims

1. A method for carrying out a test run on a test bench (10), wherein a test object (1) is connected to a load machine (2) on the test bench (10), wherein a target torque (T_soll) is specified for a torque controller (R2) by a test bench automation unit (5) according to the test run, whereby an actual torque (T_ist) of the load machine (2) is adjusted by the torque controller (R2), and wherein a test object control variable (SW) is specified for the test object (1) by a test object controller (R1), characterized in that an actual rotational speed (n) of the load machine (2) is determined and at least one deviation (a, a1, a2) of at least one attribute of the actual rotational speed (n) from at least one threshold value (G) is determined, and in that, based on the at least one deviation (a, a1, a2), at least one additive torque correction value (Tk) is determined and superimposed on the target torque (T_soll).

2. The method according to claim 1, characterized in that the value of the actual rotational speed (n) is used as a first attribute of the actual rotational speed (n), and in that an upper and / or lower rotational speed threshold (no, nu) is provided as a threshold value (G).

3. The method according to claim 1 or 2, characterized in that an actual rotational speed gradient (dn) is used as a second attribute of the actual rotational speed (n), and in that an upper and / or lower rotational speed gradient threshold (dno, dnu) is provided as a threshold value (G).

4. The method according to any of claims 1 to 3, characterized in that an actual torque (T_ist) is determined and fed back to the torque controller (R2) for controlling the actual torque (T_ist).

5. The method according to any of claims 1 to 4, characterized in that a target value (W) is specified for the test object controller (R1) by the test bench automation unit (5) in order to set an actual test object variable (M) of the test object (1) by specifying the test object control variable (SW).

6. The method according to claim 5, characterized in that the actual test object variable (M) of the test object (1) is determined and fed back to the test object controller (R1) for controlling the actual test object variable (M).

7. The method according to any of claims 1 to 6, characterized in that at least one threshold value (G) is fixed and / or in that at least one threshold value (G), preferably depending on a known system variable, can be changed.

8. A test bench(10) for carrying out a test run, wherein a test object (1) is connected to a load machine (2) on the test bench (10) to carry out the test run, wherein a test bench automation unit (5), a test object controller (R1) and a torque controller (R2) are provided, wherein the test object controller (R1) specifies a test object control variable (SW) for the test object (1) and the test bench automation unit (5) specifies a target torque (T_soll) for the torque controller (R2) according to the test run, whereby the torque controller (R2) adjusts an actual torque (T_ist) of the load machine (2), characterized in that a rotational speed determination unit (21) is provided for determining an actual rotational speed (n) of the load machine (2), in that at least one comparison unit (V, V1, V2) is provided, which is configured to determine at least one deviation (a, a1, a2) of at least one attribute of the actual rotational speed (n) from at least one threshold value (G), and in that at least one correction unit (B, B1, B2) is provided, which is configured to determine, based on the at least one deviation (a, a1, a2), at least one additive torque correction value (Tk) and superimpose it on the target torque (T_soll).

9. The test bench according to claim 8, characterized in that a first correction unit (B1) comprises a rotational speed correction controller (Rn), preferably a PI controller, and a first superposition unit (B11), wherein the rotational speed correction controller (Rd) receives a first deviation (a1) of the actual rotational speed (n) from a rotational speed threshold (no, nu) from a first comparison unit (V1) and, based on the first deviation (a1), determines a torque correction value (Tk) and superimposes it on the control torque (T) via the first superposition unit (B11).

10. The test bench according to claim 9, characterized in that the rotational speed correction controller (Rn) can be activated, preferably when a first rotational speed threshold is exceeded, and / or deactivated, preferably when the speed falls below a second rotational speed threshold.

11. The test bench according to claims 8 to 10, characterized in that a second correction unit (B2) comprises a rotational speed gradient correction controller (Rdn), preferably a PI controller, and a second superposition unit (B12), wherein the rotational speed gradient correction controller (Rdn) receives a second deviation (a2) of an actual rotational speed gradient (dn) from a rotational speed gradient threshold dnu) from a second comparison unit (V2), determines a torque correction value (Tk) from the second deviation (a2) and superimposes the torque correction value (Tk) on the control torque (T) via the second superposition unit (B21).

12. The test bench according to claim 11, characterized in that the rotational speed gradient correction controller (Rdn) can be activated, preferably when a first rotational speed gradient threshold is exceeded, and / or deactivated, preferably when the speed falls below a second rotational speed gradient threshold.

13. The test bench according to any of claims 8 to 12, characterized in that a torque determination unit (22) is provided, which is configured to determine the actual torque (T_ist) and connected to the torque controller (R2) in order to feed the actual torque (T_ist) back to control the actual torque (T_ist).

14. The test bench according to claims 8 to 13, characterized in that the test bench automation unit (5) is designed to specify a target value (W) for the test object controller (R1), and in that the test object controller (R1) is configured to specify an actual test object variable (M) of the test object (1) by specifying the test object control variable (SW).

15. The test bench according to claim 14, characterized in that an actual variable determination unit (11) is provided in order to determine the actual test object variable (M) of the test object (1), wherein the actual variable determination unit (11) is connected to the test object controller (R1) to feed the actual test object variable (M) back in order to control the actual test object variable (M).