阅读说明：本技术 用于运行测试台的方法 (Method for operating a test bench ) 是由 M·比尔 M·施密特 于 2018-12-21 设计创作，主要内容包括：本发明涉及一种用于控制驱动总成(2)的扭矩以在测试台(1)上实施测试运行的简单方法,该方法考虑了驱动总成(2)的有限的调节动态性,规定,由总成控制单元(6)以下述方式控制驱动总成(2)的内部有效扭矩(M<Sub>INT_EFF</Sub>)或有效扭矩(M<Sub>EFF</Sub>),即,预定内部有效扭矩(M<Sub>INT_EFF_SOLL</Sub>)或有效扭矩(M<Sub>EFF_SOLL</Sub>)的目标值并且在驱动总成于测试台(1)上运行期间确定内部有效扭矩(M<Sub>INT-EFF_IST</Sub>)或有效扭矩(M<Sub>EFF_IST</Sub>)的实际值以及通过传递函数(UF)以下述方式考虑在控制时驱动总成(2)的调节动态性,即,通过传递函数(UF)校正控制的目标值或使用驱动总成(2)的操纵变量的前馈控制来控制驱动总成(2)的内部有效扭矩(M<Sub>INT_EFF</Sub>)或有效扭矩(M<Sub>EFF</Sub>),其中,通过传递函数(UF)校正操纵变量的前馈控制值或控制的目标值。(The invention relates to a simple method for controlling the torque of a drive assembly (2) for carrying out a test run on a test bench (1), which method takes into account the limited control dynamics of the drive assembly (2) and provides that the internal effective torque (M) of the drive assembly (2) is controlled by an assembly control unit (6) in the following manner INT_EFF ) Or effective torque (M) EFF ) I.e. predetermined internal effective torque (M) INT_EFF_SOLL ) Or effective torque (M) EFF_SOLL ) And determining an internal effective torque (M) during operation of the drive assembly on the test bench (1) INT‑EFF_IST ) Or effective torque (M) EFF_IST ) And the control dynamics of the drive assembly (2) during the control are taken into account by means of the transfer function (UF) in such a way that the target value of the control is corrected by means of the transfer function (UF) or the drive assembly (2) is usedControlling an internal effective torque (M) of a drive assembly (2) by feedforward control of a manipulated variable INT_EFF ) Or effective torque (M) EFF ) Wherein the feedforward control value or the controlled target value of the manipulated variable is corrected by a transfer function (UF).)

1. A method for operating a test bench (1) for carrying out a test run, which test bench comprises a drive assembly (2) which is connected by a connecting shaft (3) to a dynamometer (4) for driving or loading the drive assembly (2), which dynamometer (4) is controlled on the test bench (1) by a control device (5) for carrying out the test run and the drive assembly (2) is controlled by an assembly control unit (6) for carrying out the test run, wherein for carrying out the test run a predetermined time profile of the rotational speed and the torque of the drive assembly (2) is simulated, characterized in that an internal effective torque (M) of the drive assembly (2) is controlled by the assembly control unit (6) in the following manner_{INT_EFF}) Or effective torque (M)_EFF) I.e. predetermined internal effective torque target value (M)_{INT_EFF_SOLL}) Or effective torque orderScalar value (M)_{EFF_SOLL}) And determining the actual value (M) of the internal effective torque during the operation of the drive assembly (2) on the test stand (1)_{INT-EFF_IST}) Or effective torque actual value (M)_{EFF_IST}) And the regulating dynamics of the drive aggregate (2) during the control are taken into account by means of the transfer function (UF) in such a way that the target value of the control is corrected by means of the transfer function (UF) or the internal effective torque (M) of the drive aggregate (2) is controlled by means of a feed-forward control using manipulated variables of the drive aggregate (2)_{INT_EFF}) Or effective torque (M)_EFF) Wherein the feedforward control value or the controlled target value of the manipulated variable is corrected by a transfer function (UF).

2. Method according to claim 1, characterized in that the actual rotational speed (N) is measured from the dynamometer (4) or from the drive assembly (2) or from the connecting shaft (3)_IST) And an effective actual torque (M) measured on the dynamometer (4), or on the drive assembly (2), or on the connecting shaft (3)_{EFF_IST}) And the known mass inertia (I) of the drive assembly (2)_A) Determining an internal effective actual torque (M)_{INT_EFF_IST}) Or by determining the effective actual torque (M) by measurement on the connecting shaft (3)_{EFF_IST})。

3. Method according to claim 2, characterized in that the internal effective actual torque (M) is determined in the following manner_{INT_EFF_IST}) I.e. for the measured actual rotational speed (N)_IST) Is derived with respect to time and is coupled to the known mass inertia (I) of the drive assembly (2)_A) Multiplying and multiplying the product with the measured effective actual torque (M)_{EFF_IST}) And (4) adding.

4. Method according to claim 1, characterized in that an internal combustion engine is used as the drive assembly (2) and the internal effective actual torque (M) is determined by means of a cylinder pressure indication on the internal combustion engine_{INT_EFF_IST})。

5. Method according to claim 4, characterized in that the actual torque (M) indicated is determined by_{INDI_IST}) And frictionTorque (M)_R) The difference between them to determine the internal effective actual torque (M)_{INT_EFF_IST}) Wherein the indicated actual torque (M) is determined by a cylinder pressure indication_{INDI_IST})。

6. The method according to any one of claims 1 to 5, characterized by determining the predetermined rotational speed profile of the drive assembly (2), the predetermined torque profile of the drive assembly (2) and the known mass inertia (I) of the drive assembly (2)_A) To determine an internal effective target torque (M)_{INT_EFF_SOLL})。

7. Method according to claim 6, characterized in that the internal effective target torque (M) is determined in the following way_{INT_EFF_SOLL}) That is, a predetermined speed curve is differentiated with respect to time and is compared with the known mass inertia (I) of the drive assembly (2)_A) Multiplying and adding the product to a predetermined torque curve of the drive assembly (2).

8. The method according to any one of claims 1 to 7, characterized in that accelerator pedal position (α) is used as manipulated variable.

9. Method according to any one of claims 1 to 8, characterized in that the rotational speed, in particular the actual rotational speed (N), is determined from the rotational speed_Ist) Or a predetermined rotational speed, and a target torque, in particular an internal effective target torque (M)_{INT_EFF_soll}) Effective target Torque (M)_{EFF_soll}) Internal effective target torque (M) corrected by a transfer function (UF)_{INT_EFF_soll_UH}) Or an effective target torque (M) corrected by a transfer function (UF)_{EFF_soll_UH}) The feedforward control value of the manipulated variable is preferably determined from a characteristic map KF.

10. Method according to any of claims 1 to 9, characterized in that the feed-forward control value of the controlled target value or manipulated variable is shifted by the dead time (Δ t) on the time axis by means of a transfer function (UF).

11. Method according to claim 10, characterized in that the dead time (Δ t) is determined to be the same for all operating points of the drive assembly (2).

12. Method according to claim 10, characterized in that the dead time (Δ t) is determined depending on the operating point of the drive assembly (2).

13. Method according to claim 12, characterized in that the dead time (Δ t) for one working point of the drive assembly (2) is dependent on the internal effective target torque (M)_{INT_EFF_SOLL}) Or effective target torque (M)_{EFF_SOLL}) Is determined by the gradient of the curve of (a) in this operating point.

14. Method according to any of claims 11 to 13, characterized in that the dead time (Δ t) is determined by measuring the drive assembly (2) or a reference drive assembly on the test bench (1).

15. Method according to claim 14, characterized in that the dead time (Δ t) is measured in such a way that the manipulated variable of the drive assembly (2) is abruptly changed and the abrupt change in the manipulated variable and the resulting internal effective actual torque (M) are measured_{INT_EFF_IST}) Time between changes.

16. Test bench (1) for carrying out test runs, comprising a drive assembly (2) which is connected by a connecting shaft (3) to a dynamometer (4) for driving or loading the drive assembly (2), a control device (5) which controls the dynamometer (4) on the test bench (1) for carrying out test runs, and an assembly control unit (6) which controls the drive assembly (2) for carrying out test runs, the test bench (1) being provided for carrying out test runs in the form of predetermined time profiles of the rotational speed and the torque of the drive assembly (2), characterized in that the assembly control unit (6) controls an internal effective torque (M) of the drive assembly (2)_{INT_EFF}) Or effective torque (M)_EFF) Said assembly control unit (6) obtaining an internal effective torque target value (M) for control_{INT_EFF_SOLL}) Or effective torque target value (M)_{EFF_SOLL}) And the assembly control unit (6) determines the actual value of the internal effective torque (M) during the operation of the drive assembly (2) on the test stand (1)_{INT_EFF_IST}) Or effective torque actual value (M)_{EFF_IST}) And the assembly control unit (6) corrects the target value of the control for the control by means of a transfer function, in order to take account of the regulating dynamics of the drive assembly (2), or in order to control the internal effective torque (M) of the drive assembly (2)_{INT_EFF}) Or effective torque (M)_EFF) A feedforward control of a manipulated variable of a drive assembly (2) is provided, wherein a feedforward control value or a controlled target value of the manipulated variable is corrected by an assembly control unit (6) via a transfer function (UF).

17. Test bench (1) according to claim 16, characterized in that said test bench (1) comprises an observer (10) in hardware or software for determining an internal effective torque actual value (M)_{INT_EFF_IST})。

18. Test bench (1) according to claim 17, characterised in that the observer (10) is arranged for measuring the actual rotational speed (N) from the dynamometer (4) or from the drive assembly (2) or from the connecting shaft (3)_IST) And an effective actual torque (M) measured on the dynamometer (4), or on the drive assembly (2), or on the connecting shaft (3)_{EFF_IST}) And the known mass inertia (I) of the drive assembly (2)_A) Determining an internal effective actual torque (M)_{INT_EFF_IST})。

19. Test bench (1) according to claim 16, characterized in that an internal combustion engine is provided as the drive assembly (2) and that a cylinder pressure indication system for indicating the cylinder pressure of the internal combustion engine is provided on the test bench (1), wherein the internal effective actual torque (M) is determined from the cylinder pressure indication_{INT_EFF_IST})。

20. Test bench (1) according to claim 19, characterized in that the actual torque (M) indicated is_{INDI_IST}) And friction torque (M)_R) The difference between them determines the internal effective actual torque (M)_{INT_EFF_IST}) Wherein the indicated actual torque (M) is determined by a cylinder pressure indication_{INDI_IST})。

Technical Field

The invention relates to a method for carrying out a test run on a test bench having a drive assembly which is connected via a connecting shaft to a dynamometer for driving or loading the drive assembly, the dynamometer being controlled by a control device on the test bench for carrying out the test run and the drive assembly being controlled by an assembly control unit for carrying out the test run, wherein, for carrying out the test run, a predetermined time profile of the rotational speed and the torque of the drive assembly is simulated (i.e. reproduced). The invention also relates to a test bench for carrying out a test run.

Background

For many years test stands have been used for developing drive assemblies, such as combustion engines, electric motors or combinations of combustion engines and electric motors (so-called hybrid drives), the basic structure and mode of action of which are well known. An important requirement for such test stands is always to ensure that the predetermined speed/torque curve is simulated as accurately and reproducibly as possible on the output shaft of the drive assembly. For this purpose, the drive assembly is connected to a Dynamometer (dynameter, Dyno) via a connecting shaft.

Typically, the rotational speed is regulated on the test stand by a dynamometer and the torque is regulated by the drive assembly. Based on the limited availability of drive and measurement technology or control and regulation devices, initially it was essential to set and measure the static operating point (speed/torque combination). For many test runs it is only necessary to reach a static operating point. Due to the increasing demands on the drive assembly (e.g. high engine power, low consumption, low pollutant emissions of internal combustion engines) and the constant development in the above-mentioned technical field, and also due to the increasing test requirements and specifications of the drive assembly, it has become possible or desirable to set both the static operating point and the dynamic speed/torque curve on the test bench.

"dynamic" here means in particular not only a static operating point but also in particular a rapid change in the rotational speed and/or torque. These curves can be, for example, legally prescribed measurement cycles for exhaust gas certification of internal combustion engines in order to provide evidence about compliance with pollutant emission limits. In order to optimize the power and consumption of the drive units, real, highly dynamic and non-standardized driving curves, for example, measured during driving tests with the vehicle when the drive units are used as vehicle drives, on roads or on test sections, are increasingly used. These dynamic curves impose extremely high requirements on the control of the test bench, which however cannot always be fully fulfilled.

Usually, a so-called control method N/M is used on the test bench_EFFWherein the dynamometer of the test bench controls the drive assembly speed N predetermined based on the target curve_MAnd the drive assembly controls a predetermined effective torque M on the connecting shaft between the dynamometer and the drive assembly_EFF. But these two parameters N_MAnd M_EFFIn the case of internal combustion engines, the manipulated variable of the drive assembly is, for example, the accelerator pedal position α, which directly influences the internal effective torque M_{INT_EFF}I.e. the torque that directly affects the mass inertia of the internal combustion engine. Effective torque M on the connecting shaft during acceleration and braking_EFFFrom an internal effective torque M_{INT_EFF}And the torque required to accelerate or brake the mass inertia of the engine to change the speed.

But internal effective torque M_{INT_EFF}Cannot be measured directly, so that up to now it has been the control of the measurable torque M on the connecting shaft_EFF. However, in particular in dynamic test operation, the effective torque M at the connecting shaft cannot be controlled independently of the rotational speed N_EFF. Usually consisting of a rotational speed N_MAnd effective torque M_EFFThe existing static characteristic map (measured static operating point) as input determines the manipulated variable of the drive aggregate (e.g. accelerator pedal position α in the case of an internal combustion engine.) this "feed-forward" control based on the characteristic map leads to incorrect values of the manipulated variable, because the effective torque M on the connecting shaft measured at one operating point in the dynamic test cycle_EFFIs different from the value at the corresponding operating point in the static operationThus, the method can be used for the treatment of the tumor. In addition, the turndynamik (Stelldynamik) of the drive assembly is typically significantly lower than that of a conventional test stand dynamometer.

Therefore, the torque of the internal combustion engine appears with a delay with respect to the rotation speed of the dynamometer. The control dynamics are understood here as how quickly a change in the manipulated variable influences the torque. In the case of internal combustion engines, the change in the position of the accelerator pedal does not affect the torque immediately, but usually after a certain time, mostly in the range of a few seconds. This is the main reason why the control of test runs on test benches in dynamic test runs at present sometimes yields poor results.

In the 2008 greuenbach, E et al publication "adaptive control of engine torque with input delay" (seventeenth international union world congress, seoul, korea, 6-11 d.7 of 2008), it was suggested to control the internal torque based on combustion in a test run on an engine test stand, but it was also stated that this is difficult in practice because the internal torque is a superposition of the individual expansion strokes during combustion in the cylinders of the internal combustion engine. In addition, the internal torque cannot be directly measured and must be estimated. Furthermore, test runs with dynamic speed profiles are not taken into account in this publication.

Document EP 3067681a1 describes a method for operating an engine or drive train test stand, in which an indicating device is used to detect the combustion chamber pressure. The combustion chamber pressure is converted by the crank angle precision (kurbelwinkelgenau) into the indicated torque and further into the effective torque of the crankshaft, which is used to control the dynamometer. However, this method has disadvantages in that: the combustion chamber of the internal combustion engine needs to be accessible for cylinder pressure measurement by machining, and the measurement method is very complicated and costly.

Disclosure of Invention

The object of the present invention is therefore to provide a simple method for controlling the torque of a drive assembly in order to carry out a test run on a test bench, which method takes into account the limited adjustment dynamics of the drive assembly.

According to the invention, the object is achieved as follows: the internal effective torque or effective torque of the drive assembly is controlled by the assembly control unit in such a way that a target value of the internal effective torque or effective torque is predefined and an actual value of the internal effective torque or effective torque is determined during operation of the drive assembly on the test stand, and the regulating dynamics of the drive assembly during the control are taken into account by means of the transfer function in such a way that the target value of the control is corrected by means of the transfer function or the internal effective torque or effective torque of the drive assembly is controlled using feed-forward control of a manipulated variable of the drive assembly, wherein the feed-forward control value of the manipulated variable or the target value of the control is corrected by means of the transfer function. By taking account of the adjustment dynamics, different delays in the torque build-up of different drive assemblies can be compensated, thereby improving the control accuracy.

Instead of using the effective torque, which is usually measured on the connecting shaft between the drive aggregate and the dynamometer, to control the torque of the drive aggregate, a so-called internal effective torque, which is a torque that is influenced by the acceleration with which the mass inertia of the drive aggregate is eliminated, relative to the effective torque, can also be controlled. This allows a substantial decoupling of the rotational speed and the torque at the test stand, so that the torque control can be improved. In contrast to the effective actual torque, the inner effective actual torque cannot be measured directly on the connecting shaft, but can be determined, for example, by means of an observer. As an observer, all known algorithms can be used which determine an internal effective torque value independent of the acceleration effect of the mass inertia. The advantage of the feed forward control is that the overall control unit only needs to correct much smaller deviations.

The internal effective actual torque is preferably determined from the actual rotational speed measured on the dynamometer or on the drive assembly or on the connecting shaft and the effective actual torque measured on the dynamometer or on the drive assembly or on the connecting shaft and the known mass inertia of the drive assembly, or the effective actual torque is determined by measurement on the connecting shaft. To determine the internal effective actual torque, the measured actual rotational speed may be differentiated with respect to time and multiplied by the known mass inertia of the drive assembly and added to the measured effective actual torque.

In the case of an internal combustion engine as drive assembly, the internal effective actual torque can be determined by means of a cylinder pressure indication on the internal combustion engine. For this purpose, the internal effective actual torque is preferably determined from the difference between the indicated actual torque and the friction torque, wherein the indicated actual torque is determined by means of the cylinder pressure indication.

The internal effective target torque may be determined from a predetermined profile of drive assembly rotational speed, a predetermined profile of drive assembly torque, and a known mass inertia of the drive assembly. Preferably, the predetermined speed profile is differentiated with respect to time and multiplied by the known mass inertia of the drive assembly and the product is added to the predetermined profile of the drive assembly torque. The predetermined profile may be determined, for example, from recorded measurement data of the drive assembly, legally prescribed measurement periods, or other sources. The mass inertia is selected according to the specific development objective in correspondence with the mass inertia of the drive assembly running in reference or in correspondence with the mass inertia of the drive assembly to be tested and is assumed to be known. The target value of the internal effective torque may also be determined from, for example, recorded data of an integrated control unit (e.g., an ECU of an internal combustion engine).

Advantageously, the accelerator pedal position is used as a manipulated variable for the feed-forward control.

The feedforward control value of the manipulated variable is preferably determined from the rotational speed, in particular the actual rotational speed or the predetermined rotational speed, and the target torque, in particular the internal effective target torque, the internal effective target torque corrected by the transfer function or the effective target torque corrected by the transfer function, preferably from a characteristic map.

In the simplest case, the correction can be carried out by means of a transfer function in such a way that the target value of the control or the feedforward control value of the manipulated variable is shifted by the dead time on the time axis. The dead time can be determined here to be the same for all operating points of the drive assembly or can be determined as a function of the operating points of the drive assembly. Therefore, different dynamic states of the internal effective torque or the effective torque during establishment can be compensated for different working points, and the control precision is further improved.

The different operating points can be taken into account in such a way that the dead time for an operating point of the drive assembly is determined as a function of the internal effective target torque or the gradient of the effective target torque curve in the operating point. No additional measurement effort is required by analyzing the internal effective target torque or the curve of the effective target torque. This allows, for example, different time delay characteristics of the drive assembly during torque increases and decreases.

However, the dead time can also be determined by measuring the drive assembly or a reference drive assembly at a test stand, preferably by abruptly changing the manipulated variable of the drive assembly and measuring the time between the abrupt change in the manipulated variable and the resulting change in the internal effective actual torque or the effective actual torque. For example, it is conceivable to create a family of dead time characteristic curves for drive units having similar desired control dynamics, for example as a function of displacement, supercharging pressure, number of cylinders, nominal rotational speed, etc.

Drawings

The present invention is explained in detail below with reference to fig. 1 to 7. The figures show advantageous embodiments of the invention by way of example, schematically and without limitation. The attached drawings are as follows:

FIG. 1 shows the general structure of a test station;

FIG. 2 illustrates the functionality of a viewer;

fig. 3a to 3c show a method according to the invention;

FIG. 4 shows a reference test run;

FIG. 5 illustrates a conventional control scheme N/M_EFFThe result of the time;

FIG. 6 shows a control scheme N/M_{INT_EFF}The result of the time;

FIG. 7 shows the target torque M effective at the inner part_{INT_EFF_SOLL}Control mode N/M under condition of moving dead time delta t_{INT_EFF}The result of the time;

FIG. 8 illustrates control strategy N/M with the feedforward control value of accelerator pedal position α shifted by the dead time Δ t_{INT_EFF}The results of (1).

Detailed Description

Fig. 1 shows a known common structure of a test bench 1 comprising a drive assembly 2 connected with a dynamometer 4 by a connecting shaft 3 for torque transmission, a control for controlling the dynamometer 4A device 5 and an assembly control unit 6 for controlling the drive assembly 2. The control device 5 and the overall control unit 6 can be implemented by means of suitable hardware and/or software (also on one common hardware). The drive assembly 2 has a device for measuring the rotational speed N of the assembly_MAnd the dynamometer 4 also has a device for measuring the dynamometer rotational speed N_BThe rotation speed measuring device 8. A connecting shaft 3 between the drive unit 2 and the dynamometer 4 is provided for measuring the effective torque M of the drive assembly 2_EFFThe torque measuring device 9.

The dynamometer 4 is understood not only as a conventional electric machine, such as a direct-current machine, an asynchronous machine or a three-phase synchronous machine, which is directly connected to the connecting shaft 3, but also, for example, as a combination of an electric machine and a transmission, such as in the form of a so-called drive test stand system (TRT). In this case, for example, two or more electric machines can be connected via a summation gear (summergeteblebe), which is itself connected to the connecting shaft 3 for driving or loading. In a step-up transmission, the power of two (or more) electric machines is added, if necessary, and a particular rotational speed level can also be shifted. Of course this is merely exemplary and all other suitable machines or combinations of machines and transmissions may be used as dynamometer machine 4. In order to determine the internal effective actual torque M of the drive assembly 2_{INT_EFF_IST}For example, a viewer 10 can be provided, which is in turn embodied as suitable hardware and/or software. As observer 10, all known algorithms for determining the elimination of the mass inertia I of drive assembly 2 can be used here_AIs used according to the invention as the internal effective actual torque M_{INT_EFF_IST}. The function of such a viewer 10 is known in principle, but for the sake of completeness the basic mode of operation will be briefly described below with reference to fig. 2.

If the drive assembly 2 is designed as an internal combustion engine, a cylinder pressure indicator system can also be used instead of the observer 10 to determine the internal effective actual torque M_{INT_EFF_IST}. Thus, the cylinder pressure in the combustion chamber of the internal combustion engine can be measured precisely at a crankshaft angle and the indicated actual torque M can be determined on the basis of the measured cylinder pressure by means of the thermodynamic law_{INT_IST}. If the indicated actual torque M is compared_{INT_IST}By eliminating the known internal friction of the internal combustion engine (which is present, for example, in the form of a characteristic map over the operating range of the internal combustion engine), the required internal effective actual torque M is achieved_{INT_EFF_IST}. The friction influence can be, for example, a friction torque M_RThe form of (d) is determined by a drag measurement of the internal combustion engine on the test bench 1 or by other suitable methods. Since cylinder pressure indication methods are well known, they will not be described in detail herein. A detailed description is known, for example, from EP 3067681a 1.

The method according to the invention is in principle not restricted to a specific drive assembly 2, but can be used for various drive assemblies 2, such as internal combustion engines, electric motors, combinations of electric motors and internal combustion engines (so-called hybrid drives), as long as the required parameters are available. The method can also be used, for example, in drive trains in which the drive assembly 2 can be connected to the connecting shaft 3 via a transmission, a clutch, a differential, a half shaft or the like.

Fig. 2 shows, by way of example, a block diagram of a known simplified operation of a viewing device 10, which is used, for example, for determining an internal effective torque M, using an internal combustion engine as drive train 2_{INT_EFF}. The engine speed N is measured by a speed measuring device 7 on the internal combustion engine_MAnd averaged over a working cycle (for example 720 crank angle in a four-stroke engine) and the number of cylinders of the internal combustion engine by means of the filter F. By averaging, an uneven torque input over a working cycle of the internal combustion engine, which is generated on the basis of the combustion in the cylinders of the internal combustion engine and the corresponding number of cylinders of the internal combustion engine and which leads to an engine speed N, is compensated_MA change in (c). For example, in a four-stroke engine, combustion occurs every 720 ° crankshaft angle in each cylinder, which produces a force on the piston and thus a torque input on the crankshaft. In a four-cylinder engine this means, for example, that torque is input once every 180 ° of crankshaft rotation, in a six-cylinder engine, for example every 120 ° of crankshaft rotation, etc. Based on the described speed N of the engine_MObtaining the filtered engine speed N_{M_FILT}. Similarly, such averaging orSaid that the filtering can also be applied to the effective torque M_EFFThereby obtaining a filtered effective torque M_{EFF_FILT}. The filtered engine speed N is then differentiated by a differentiator D_{M_FILT}Derived with respect to time, whereby angular acceleration is obtained

In the next step, the angular acceleration to be obtainedThe mass inertia I of the internal combustion engine assumed to be known is multiplied in a multiplier M_AAnd obtains a correction torque DeltaM_M. Now the correction torque Δ M to be obtained in the summer S_MAnd the effective torque M (e.g. filtered again over one working cycle and the number of cylinders of the internal combustion engine)_{EFF_FILT}Added to an internal effective torque M_{INT_EFF}. Thus, according to the engine speed N_MAnd the filtered engine speed N determined therefrom_{M_FILT}Time derivative or angular acceleration ofIncrease or decrease the effective torque M measured on the connecting shaft 3 and averaged over a working cycle and the number of cylinders of the internal combustion engine_{EFF_FILT}Whereby the mass inertia I of the internal combustion engine is taken into account_AThe dynamic influence of (2). This calculation can be used either "on-line" by means of the observer 10 to determine the internal effective actual torque M_{INT_EFF_IST}It is also possible to use "offline" or "online" for determining the internal effective target torque M for carrying out a test run on the test stand 1 from a predetermined reference speed/torque curve_{INT_EFF_SOLL}。

"on-line" in this case means that the internal effective actual torque M is determined during the test run carried out on the test bench 1_{INT_EFF_IST}And "offline" means that the internal effective target torque M is determined outside the test runs carried out on the test bench 1_{INT_EFF_SOLL}. However, the effective actual torque M can also be determined "on-line_{INT_EFF_IST}Time is savedThe filtering step, but basically the filtering is implicitly achieved by the characteristics of the controller used in the assembly control unit 6 and by the delay characteristics of the drive assembly 2. When the internal combustion engine is used in a vehicle, the internal effective target torque M can be determined, for example, from recorded measurement data (speed/torque curve) of actual driving tests or from other sources_{INT_EFF_SOLL}. The described observer method is of course not limited to use in internal combustion engines; it is also applicable to other drive assemblies 2, such as electric motors, hybrid drives, etc.

The mass inertia I of the drive assembly 2 can be assumed_AAre known. A plurality of different mass inertias I may also be used_ACan also be used to calculate the internal effective target torque M_{INT_EFF_SOLL}. For example, the known mass inertia I of the drive assembly 2 can be used on the test stand 1_A. However, the mass inertia I of the drive unit 2 from the reference operation that is to be reproduced on the test bench can also be used_A. This means that the mass inertia I of the drive assembly 2 on the test stand 1_ANot necessarily in line with the mass inertia of the drive assembly creating or measuring the reference run. For example, the internal power of the test object (the power in the combustion chamber of the internal combustion engine as drive assembly 2) corresponds well to the reference operation. If the reference run is reproduced on the test stand and the actual mass inertia I of the drive assembly 2 is used on the test stand_AThe power on the connection shaft 3 corresponds well to the reference operation.

Fig. 3 shows the basic procedure of the method according to the invention in a flow chart. In a first step, a block a represents the generation or provision of a speed and torque curve to be simulated for the test run to be carried out on the test stand 1 of the drive assembly 2. The assembly speed N of the drive assembly 2_{A_REF}Reference value of (d), effective torque M_{EFF_REF}Reference value and mass inertia I_A. These data can be provided, for example, by measurement data of the actual operation (reference operation), but they can also be predetermined by a legally prescribed measurement cycle or from other sources.

Represented in the next step by the block B for determining the internal effective actual torque M, as already described with reference to the observer 10 of fig. 2_{INT_EFF_IST}The same method as in (c) calculates the internal effective target torque M from the predetermined reference value_{INT_EFF_SOLL}. In the case of a drive assembly 2 designed as an internal combustion engine, the reference engine speed N is preferably determined over a working cycle and the number of cylinders of the internal combustion engine_{M_REF}Filtering and deriving it with respect to time, thereby obtaining a reference angular acceleration

According to engine speed N_{M_REF}The quality of the available reference data may also dispense with averaging over one working cycle and cylinder number, for example if such averaging calculations have taken place within the scope of determining the reference data or in the case of a drive assembly configured as a motor with a substantially uniform torque input per revolution. Then, the angular acceleration is referred toKnown mass inertia I of the drive assembly 2_A(e.g. I of internal combustion engine)_A) Multiplying to obtain a reference correction torque Δ M_{M_REF}. Finally, the correction torque Δ M will be referenced_{M_REF}With an effective reference torque M_{EFF_REF}(in the case of internal combustion engines with an effective reference torque M averaged over one working cycle and the number of cylinders of the internal combustion engine_{EFF_REF_FILT}) Adding up, thereby producing an internal effective target torque M_{INT_EFF_SOLL}Which is already available for controlling the drive assembly 2.

Depending on the quality of the available reference data, the determination of the internal effective torque M over a working cycle and the number of cylinders of the internal combustion engine can also be dispensed with_{INT_EFF_SOLL}For example if such averaging has taken place within the range of the determined reference data or according to the design of the drive assembly 2 (e.g. as a motor). If the measurement data required for the process is not available, a similar method can be used in order to determine the internal effective target torque M from the available reference data_{INT_EFF_SOLL}。

For example, the measured vehicle acceleration can be calculated for the summation in a vehicle drive test range with the respective drive assembly 2The torque required by the vehicle mass is calculated and the internal effective torque M required for this purpose is calculated from the known mass inertia, transmission gear ratio, etc_{INT_EFF}And uses it as the internal effective target torque M_{INT_EFF_SOLL}. It is also conceivable to determine the internal effective target torque M of the drive aggregate 2 from stored data of an aggregate control unit, such as an Engine Control Unit (ECU) of an internal combustion engine_{INT_EFF_SOLL}. Alternatively, the internal effective target torque M may also be determined from the reference operational indicator data as described above_{INT_EFF_SOLL}The value of (c).

Resulting internal effective target torque M_{INT_EFF_SOLL}It is already directly available for controlling a drive assembly 2, such as an internal combustion engine, on the test bench 1, which is indicated by block D. For this purpose, as described above, the internal effective actual torque M can be determined, for example, in the observer 10 or in the case of an internal combustion engine, by means of a cylinder pressure indicator system during a test operation_{INT_EFF_IST}. The internal effective target torque M can then be corrected on the test stand 1 by means of a suitable controller, for example a simple PI controller_{INT_EFF_SOLL}And an internal effective actual torque M_{INT_EFF_IST}The deviation therebetween.

However, the control can also use a predetermined characteristic map KF of the manipulated variable in the feed-forward control, such as the accelerator pedal position α of the internal combustion engine as a function of the engine speed N_MAnd effective torque M_EFFOr internal effective torque M_{INT_EFF}A family of characteristic curves of (c). For this purpose, for example, by including an effective torque M_EFF(or effective torque M after filtration)_{EFF_FILT}) And engine speed N_MA characteristic map KF of (or summarized as) the assembly speed determines feedforward control values of manipulated variables, such as the accelerator pedal position α. the controller, preferably the assembly control unit 6, transmits the internal effective target torque M thereto_{INT_EFF_SOLL}And an internal effective actual torque M_{INT_EFF_IST}Deviation between them — the controller manipulated variable is then determined, by means of which only much smaller deviations caused by inaccuracies in the characteristic diagram KF need then be corrected.

Thus, the manipulated variables for the drive assembly 2 are manipulated variables and controller operations as feed forward control manipulated variables in a known mannerThis characteristic diagram KF can be determined, for example, by static test bench measurements in various operating points within the relevant operating range of the drive assembly 2, in the case of an internal combustion engine, for example, by means of the accelerator pedal position α and the engine speed N_MSetting static operating points and measuring the effective torque M on the connecting shaft 3 in the respective operating points_EFFAnd stored in the family KF of characteristics. Effective torque M due to lack of mass inertia dynamics_EFFCorresponding to an internal effective torque M in static operation_{INT_EFF}. The obtained family of characteristics is reversed, so as to obtain the internal effective torque M_{INT_EFF}And engine speed N_MCharacteristic map KF of accelerator pedal position α.

In principle, any suitable controller can be used as the controller, which can also be parameterized in a known manner, if necessary, specifically for the application, and is preferably implemented as hardware or software in the overall control unit 6.

The limited adjustment dynamics of the drive assembly 2 on the test bench 1 are taken into account during the test operation according to the invention. The test run is here with an internal effective torque M_{INT_EFF}Also as an effective torque M acting on the connecting shaft 3_EFFThe performance is not critical. If the internal effective torque M is used_{INT_EFF}It may be determined and used as described above. Effective torque M_EFFIt is possible to measure simply on the connecting shaft 3. Thus, the consideration of the regulation dynamics and the internal effective torque M_{INT_EFF}Is independent of the use and can therefore be achieved independently of the torque used. In an advantageous embodiment, the test runs with an internal effective torque M_{INT_EFF}The adjustment dynamics of the drive assembly 2 are taken into account during the test operation carried out on the test stand 1 and described below.

In order to take account of the regulation dynamics when controlling the drive assembly 2, a transfer function UF is used which corrects the temporal behavior of the drive assembly 2. The time characteristic of the drive assembly 2 basically describes the time delay of the controlled system (i.e. all the time between the manipulated variable setting and the torque build-up) and maps the delayed torque build-up of the drive assembly 2 onto the manipulated variable. Such as accelerator pedal positionα setting and internal effective torque M_{INT_EFF}(or effective Torque M_EFF) Is increased (or decreased).

Because of its physical mode of action, electric motors generally have a higher regulation dynamics than internal combustion engines, it is advantageous to take into account the regulation dynamics when carrying out test runs, in particular in the case of internal combustion engines. This is mainly because the internal combustion engine requires more time to achieve a torque request, i.e. the time between a predetermined manipulated variable, such as the accelerator pedal position alpha, and the actual torque build-up, based on the underlying physical process. Internal combustion engines with direct injection and exhaust turbocharging, for example, require sufficient time to establish boost, form a mixture, burn, etc. In contrast, fewer physical processes are required in the motor, for example, significantly less time is required to build up the magnetic field.

In a simple embodiment, the transfer function UF may be such that the internal effective target torque M is_{INT_EFF_SOLL}(or effective target Torque M_{EFF_SOLL}) Is shifted on the time axis by a so-called dead time deltat. Thereby obtaining an internal effective target torque M shifted by the dead time Deltat_{INT_EFF_SOLL_UH}(or effective target Torque M_{EFF_SOLL_UH}). This is indicated in fig. 3a and 3a by block C. M_{INT_EFF_SOLL_UH}(or M)_{EFF_SOLL_UH}) May be used as a target value for control (block D).

Alternatively, the internal effective target torque M can also be used_{INT_EFF_SOLL}(or effective target Torque M_{EFF_SOLL}) And a target rotational speed N_MThe associated, time-corrected manipulated variable, such as the accelerator pedal position α, is determined by means of the transfer function UF, for example, by including the internal effective target torque M_{INT_EFF_SOLL}(or effective target Torque M_{EFF_SOLL}) And a target rotational speed N_MDetermines the manipulated variable and shifts it by the dead time Δ t, as shown in fig. 3c, the thus determined, time-shifted manipulated variable α_UHAvailable for internal effective torque M_{INT_EFF}Or effective torque M_EFFFeed forward control of the control, which is indicated by block D.

In the simplest caseNext, the dead time Δ t may be a predetermined or parameterized constant time value. However, it is desirable to determine the dead time Δ t according to the operating point (torque/rotational speed) of the drive assembly 2. For this purpose, dead time Δ t can be determined, for example, by a characteristic map, in which dead time Δ t is determined, for example, as a function of a total rotational speed N of drive unit 2_MAnd an internal effective torque M_{INT_EFF}(Δt＝f(N_M，M_{INT_EFF}) Or effective torque M)_EFF(Δt＝f(N_M，M_EFF) The dead time Δ t is plotted. Such a characteristic map can be determined, for example, by measuring the drive assembly 2 in advance on the test stand 1 or can be determined approximately from empirical values or from measurements of a reference drive assembly of similar design. A similarly constructed internal combustion engine may, for example, be an internal combustion engine with comparable characteristic parameters, such as similar displacement, the same number of cylinders, the same supercharging design, the same mixture formation, etc.

In the determination of the dead time Δ t by measuring the drive assembly 2 in advance on the test stand 1, a characteristic map is preferably determined for the torque increase of the drive assembly 2 and the torque decrease of the drive assembly 2, respectively, in this case, a predetermined manipulated variable, such as a sudden change in the accelerator pedal position α of the internal combustion engine or a current change in the form of a short ramp (in the case of an internal combustion engine, in the form of a so-called α ramp), is preferably determined in the selected operating point of the drive assembly 2 and measured up to the internal effective torque M_{INT_EFF}Or effective torque M_EFFIs used, which time substantially represents a measure of the torque build-up delay of the drive assembly 2. Determination of the dead time Δ t by means of a ramp is intended not only for the internal effective torque M_{INT_EFF}Or effective torque M_EFFShould also be carried out for a sudden decrease thereof, two dead time characteristic maps result. The slope should be selected so steep that the maximum dynamics of the drive assembly 2 are required.

However, it is also possible for an operating point of the drive assembly 2 to be determined by evaluating the internal effective torque M_{INT_EFF}Or effective torque M_EFFThe dead time Δ t can be determined, for example, as a function of the internal effective torque M in the respective operating point_{INT_EFF}Determination of dead time by gradient of curveAt. This method is preferably chosen if no separate measurement can be made on the drive assembly 2 to determine the dead time Δ t or no such measurement is available.

However, the transfer function UF can also be designed in any other way, where the transfer function UF is normally the internal effective torque M_{INT_EFF}Or effective torque M_EFFI.e. UF ═ f (M)_{INT_EFF}Or M_EFF). The transfer function UF is preferably a function of the operating point of the drive assembly 2, i.e. UF ═ f (N, M)_{INT_EFF}Or M_EFF)。

Now the internal effective torque M is corrected by the transfer function UF_{INT_EFF_SOLL}Or effective torque M_{EFF_SOLL}In order to take into account the time characteristics (regulation dynamics) of the drive assembly 2, as will be explained below by way of example with the dead time Δ t as the transfer function UF.

To carry out a test run, the predetermined target value is shifted by the dead time Δ t, in particular shifted forward in time, and controlled on the test stand 1 as described above to carry out the test run.

A predetermined internal effective target torque M generated after moving the corresponding dead time Δ t_{INT_EFF_SOLL}Or effective target torque M_{EFF_SOLL}The curve of (c) may also be adjusted such that all data points having an absolute time value greater than their subsequent points are deleted. Thereby producing a continuously increasing time vector. In the next step, the target value curve generated is set to a value corresponding to the reference assembly speed N_{A_REF}The curves are based on a common time base so as to be suitable for the control of the drive assembly 2, indicated by block D, on the test bench 1.

Fig. 4 shows a measurement diagram of a reference test run, using a drive assembly 2 designed as an internal combustion engine as an example, in which a reference value curve N of the engine speed is plotted over time t in a dash-dotted line_{M_REF}Drawing a reference value curve M of the effective torque on the connecting shaft 3 with a solid line_{EFF_REF}And a reference value curve α of accelerator pedal position is plotted in dashed lines_{_REF}. In the present example, the reference test run represents a run with constant acceleration and includes three shifts and subsequent deceleration. The following tests with the aid of this referenceThe run illustrates the improvement achieved by the method according to the invention. A reference test run can be performed with the drive assembly 2 to be tested or another reference drive assembly. But the reference values can also come from other sources, such as legally prescribed measuring periods. For a more clear display, the results in the time period Z between time t1 of the reference test run and time t2 of the reference test run as shown in fig. 4 are shown.

FIG. 5 shows a method by means of a common N/M_EFFThe control scheme is the result of the first test run in time period Z between time t1 and time t 2. The engine speed N is controlled by means of a control device 5 of the dynamometer 4_MAnd the effective torque M on the connecting shaft 3 is controlled by the unit 6 by means of the manipulated variable of the accelerator pedal position α_EFF. The measured actual value curve of the first test run is compared with the reference value curve of the reference test run known from fig. 4. The engine speed N is again plotted in dash-dot lines_{M_REF}Is plotted against the reference value of (3) by a solid line_{EFF_REF}And the accelerator pedal position α is plotted in dashed lines_{_REF}Reference value curve of (2). Measured actual value N_{M_IST}、M_{EFF_IST}And α_{_IST}The corresponding curves of (a) are respectively marked with circles. It can be seen that the engine speed N can be controlled very precisely on the test stand 1_MThis is attributable to the powerful dynamometer 4 having corresponding control characteristics. Furthermore, the effective torque M can be seen_EFFAnd the reference and actual curves of the accelerator pedal position α, this may be due to the engine speed N, as mentioned at the outset_MAnd effective torque M_EFFBy mass inertia I of the internal combustion engine_AIs highly coupled.

FIG. 6 shows the control mode N/M by means of the invention_{INT_EFF}The results of the second test run in time period Z between time t1 and time t 2. The engine speed N is controlled by means of a control device 5 of the dynamometer 4_MAnd the internal effective torque M is controlled by the assembly control unit 6 via the manipulated variable of the accelerator pedal position α_{INT_EFF}. In thatThis compares the measured actual value curve of the second test run with the reference value curve of the reference test run known from fig. 4. The engine speed N is again plotted in dash-dot lines_{M_REF}Is plotted against the reference value of (3) by a solid line_{EFF_REF}And the accelerator pedal position α is plotted in dashed lines_{_REF}Reference value curve of (2). Measured actual value N_{M_IST}、M_{EFF_IST}And α_{_IST}The corresponding curves of (a) are respectively marked with circles. It can be seen that the effective torque M on the connecting shaft 3_EFFThere is a qualitatively better match between the reference and actual curves of the accelerator pedal position α, but the time offset t of the reference curve and the actual curve can also be seen_v. The offset t_vThe time characteristic of the described internal combustion engine transfer function UF, i.e. the delay in the torque build-up between the signal, which is essentially the manipulated variable, and the actually measurable torque build-up, is mainly attributed.

As mentioned above, it is advantageous to do so by applying an internal effective target torque M_{INT_EFF_SOLL}The forward movement dead time Δ t takes into account the time characteristics of the engine transfer function UF, as will be shown below with reference to fig. 7.

FIG. 7 shows the control mode N/M by means of the invention_{INT_EFF}Results of a third test run in time period Z between time t1 and time t2, wherein the internal effective target torque M is determined_{INT_EFF_SOLL}Moving forward by a constant dead time Δ t of 100 ms. The engine speed N is controlled by means of a control device 5 of the dynamometer 4_MAnd the internal effective torque M is controlled by the assembly control unit 6 via the manipulated variable of the accelerator pedal position α_{INT_EFF}. The measured actual value curve of the third test run is compared with the reference value curve of the reference test run known from fig. 4. The engine speed N is again plotted in dash-dot lines_{M_REF}Is plotted against the reference value of (3) by a solid line_{EFF_REF}And the accelerator pedal position α is plotted in dashed lines_{_REF}Reference value curve of (2). Measured actual value N_{M_IST}、M_{EFF_IST}And α_{_IST}The corresponding curves of (a) are respectively marked with circles. The effective torque M can be seen_EFFAnd a significantly better match between the reference and actual value curves for accelerator pedal position α_{EFF_IST}And accelerator pedal position α_{_IST}Is excessively high in the actual value curve in the present case, for example, attributable to the controller used in the assembly control unit 6 correcting the target value (target torque M) of the dead time Δ t by correcting the dead time Δ t_{INT_EFF_SOLL}) The manipulated variable (accelerator pedal position α) is therefore increased too much by the integral (I) component in the controller used.

However, this effect can also be avoided by: controlling internal effective torque M using feed forward control_{INT_EFF}And correction of the time characteristic is not applied to the target torque M_{INT_EFF_SOLL}(or M)_{EFF_SOLL}) But rather a feedforward control value of a manipulated variable applied to a feedforward control. For this purpose, the internal effective target torque M is determined, for example, by means of a characteristic diagram KF_{INT_EFF_SOLL}And a target rotational speed N_{M_SOLL}The feedforward control value (accelerator pedal position α) is then shifted by the dead-time Δ t_{INT_EFF_SOLL}By the assembly control unit 6 (see fig. 3c) controls the internal effective torque M_{INT_EFF}And the feedforward control value (accelerator pedal position α) shifted by the dead time Δ t is added to the controller output of the integrated control unit 6 controller the result is shown in fig. 8 and can be seen at the effective torque M_{EFF_IST}And accelerator pedal position α_{_IST}Substantially without excessive height in the curve of the actual values of (a). However, as an alternative, the internal effective target torque M_{INT_EFF_SOLL}It is also possible to correct the target torque M by means of the transfer function UF and to use the characteristic diagram KF for correction_{INT_EFF_SOLL}And a target rotational speed N_{M_SOLL}To determine a feedforward control value.

According to a particularly advantageous embodiment of the invention, the internal effective torque M is adjusted_{INT_EFF_SOLL}Move forwardThe dead time deltat is selected in dependence on the operating point of the drive assembly 2. The effective torque M can thereby be further improved_EFFFor this purpose, as already described, a characteristic map relating to the operating point can be created for the dead time Δ t, which characteristic map can be determined, for example, by measuring the drive assembly 2 beforehand on the test stand 1, as already explained with reference to fig. 3.

If this cannot be measured beforehand, it is also possible in one operating point of the drive assembly 2, for example, depending on the internal effective target torque M in the respective operating point_{INT_EFF_SOLL}The gradient of the curve determines the dead time at. However, an approximately constant dead time Δ t may also be selected, as described with reference to the third test result in fig. 7. Of course, a family of characteristic curves for the dead time Δ t may also be created based on empirical values or based on measurements on a reference drive assembly. The reference internal combustion engine may in this case be, for example, an internal combustion engine of similar construction, such as an internal combustion engine with comparable characteristic parameters, such as similar displacement, the same number of cylinders, the same supercharging design, the same mixture formation, etc. Although the method according to the invention is described by way of example with reference to measurements of an internal combustion engine, it should again be pointed out here that the method is also suitable for other drive assemblies 2, such as electric motors, hybrid drives, drive trains, etc.