JP5658530B2 - Engine test equipment - Google Patents

Description

The present invention relates to an engine test apparatus , and more particularly to an engine test apparatus capable of performing control excellent in both responsiveness and robustness.

  An engine bench is known as a device for testing engine performance. The engine bench is a test device that evaluates whether a developed and manufactured test engine has a predetermined performance. The engine to be tested is attached to a bench test machine (engine bench). The output shaft of the engine is connected to a dynamometer via a torque meter or a tachometer, and the rotational force of the output shaft of the engine is absorbed by this dynamometer. Thereby, a bench test is performed, and the performance of the engine is measured and evaluated.

  By the way, in the engine bench, the engine speed and torque are normally set in the control unit as target values, and the control unit sets command values such as the throttle opening and the dynamometer current value according to the speed and torque. The engine and dynamometer are controlled according to the command value. Various sensors are attached to an actual machine part such as an engine or a dynamometer, and for example, an engine speed, torque, or the like is detected. This detection signal is fed back to the control unit, and the control unit performs PID control (or PI control) based on this detection signal.

  However, there is a problem that it is difficult to achieve both responsiveness and robustness with PID control alone. Therefore, in Patent Document 1, a torque-throttle opening map is created in advance for each engine speed, and the throttle opening is predicted using this map for control. Furthermore, in patent document 1, it correct | amends using fuzzy reasoning.

Patent No. 3028433

  However, Patent Document 1 requires that a map be created in advance for each engine, which is very time consuming, and can predict only the throttle opening created by the map, and simultaneously predicts a plurality of parameters. There was a problem that I could not. Furthermore, since Patent Document 1 uses fuzzy reasoning, there is a problem that accuracy is limited and testing cannot be performed with high accuracy.

  In addition, conventional engine test apparatuses including Patent Document 1 include communication means for an engine-side control system (hereinafter referred to as an En control system) and a dynamometer-side control system (hereinafter referred to as a Dy control system), respectively. However, since this dead time is different in both control systems, there is a problem that an error occurs and predictive control becomes difficult.

The present invention has been made in view of such circumstances, and an object of the present invention is to provide an engine test apparatus capable of performing control excellent in both responsiveness and robustness.

  In order to achieve the above object, the invention according to claim 1 includes an engine test apparatus comprising a dynamometer that applies a load to an engine that is a test target, and a controller that outputs a control command value to the engine and the dynamometer. A simulation unit that executes a simulation by inputting the control command value into a bench model obtained by modeling an actual machine unit including the engine and the dynamometer, and the controller inputs a result of the simulation. There is provided an engine test apparatus characterized by determining a control command value and outputting the determined control command value to the simulation unit and the actual machine unit.

  According to the present invention, simulation is performed using a bench model obtained by modeling a test apparatus itself including an engine and a dynamometer, and the simulation result is input to the controller. As a result, it is possible to control ideally and without dead time, and a control command value that can be stably controlled can be created. Therefore, predictive control can be performed by inputting the control command value to the actual machine unit, and feedforward control can be performed. . As a result, the control command value for the engine and the control command value for the dynamometer can be optimized simultaneously. In other words, when the simulation was performed only with the engine model as in the past, there was a problem that it was difficult to simultaneously optimize the control command value on the engine side and the control command value on the dynamometer side. In the invention, simulation is performed using a model of the entire test apparatus including the dynamometer as well as the engine bench to be tested, and control is performed based on the result, so that the control command value on the engine side and the dynamometer side These control command values can be optimized simultaneously.

  According to a second aspect of the present invention, in the first aspect of the present invention, a correction unit that corrects a dead time of the control command value output to the engine and / or the dynamometer is provided. As a correction function, for example, when the dead time generated by communication from the control device to the dynamometer is Δt1, and the dead time generated by communication from the control device to the engine is Δt2, the control command value on the engine side And a delay of Δt2 with respect to the control command value on the dynamometer side. Thereby, the timing controlled by the engine side and the dynamometer side becomes the same, and control without seemingly waste time can be performed. Therefore, when an ideal model with no dead time is created as a bench model and control is performed based on the ideal model, it is accurately reproduced in the actual machine unit. In the conventional case, there was a problem that even if an ideal model was created, it was different from the actual result. However, in the case of the present invention, there is a function for correcting the dead time, so the same as the ideal model. The situation can be created and the measurement accuracy can be improved.

  The invention of claim 3 is characterized in that, in claim 1 or 2, the control command value to the engine is a throttle opening, and the control command value to the dynamometer is a current value.

  According to the present invention, since the simulation is performed using the bench model that models the test apparatus itself including the engine and the dynamometer, not only the control command value to the engine but also the control command value to the dynamometer Can be optimized.

The figure which shows schematic structure of the engine test apparatus to which this invention is applied. Control block diagram of engine test apparatus of the present embodiment Diagram showing an example of a bench model Explanatory diagram of wasted time Conventional control block diagram Explanatory drawing which shows the effect | action of this Embodiment

Hereinafter, preferred embodiments of an engine test apparatus according to the present invention will be described with reference to the accompanying drawings. FIG. 1 shows a schematic configuration mainly of an actual machine part of an engine test apparatus 10 to which the present invention is applied.

  The engine test apparatus 10 shown in FIG. 1 is an apparatus for measuring and evaluating the performance of the engine 12 to be tested. The engine test apparatus 10 mainly includes an engine 12, a dynamometer 14, a shaft member 32, and the like, a dynamo control unit 16, It is comprised with the control part which consists of the engine control part 18, the control apparatus 20, etc. FIG. In this embodiment, the dynamo control unit 16 and the engine control unit 18 are described as control parts, but they may be actual machine parts.

  The engine 12 is fixed to a gantry 22 and a combustion chamber (not shown) is provided in the engine 12. A suction pipe 24 for sucking air is connected to the combustion chamber, and a throttle 26 for adjusting the inflow amount is provided in the intake pipe 24. An exhaust pipe 28 is connected to the combustion chamber, and an exhaust gas purification catalyst mounting portion 30 is provided in the exhaust pipe 28.

  The output shaft of the engine 12 is connected to the dynamometer 14 via the shaft member 32. The shaft member 32 is configured by connecting a plurality of shaft members such as a main shaft, and a universal joint 34 is interposed in the connecting portion. A torque meter 36 is attached to the shaft member 32, and the torque is measured by the torque meter 36. In the present embodiment, the torque is measured by the torque meter 36. However, the present invention is not limited to this. For example, the torque may be detected from the output value of the dynamometer 14. In addition to the torque meter 36, a clutch, a transmission, various connecting means, and the like may be inserted depending on the purpose. Further, means for measuring the state of the engine 12 other than the torque (for example, the temperature of the exhaust gas) may be inserted.

  The dynamometer 14 is a device that applies a predetermined load torque to the engine 12, and can set the load torque by varying the current and voltage. As the dynamometer 14, a low inertia dynamometer is preferably used. By using the low inertia dynamometer, a stable output corresponding to a sudden change in the rotational speed from low speed rotation to high speed rotation can be obtained.

  A dynamometer control unit 16 is connected to the dynamometer 14. The dynamo control unit 16 is means for variably controlling the current / voltage applied to the dynamometer 14, and the load of the engine 12 connected to the dynamometer 14 is controlled by variably controlling the current / voltage with the dynamo control unit 16. Torque is controlled.

  On the other hand, the engine 12 is connected to the engine control unit 18. The engine control unit 18 is means for driving and controlling the engine 12 by giving a control signal such as throttle opening and ignition advance angle to the engine 12, and is usually an ECU or an engine control circuit in which a bypass circuit is added to the ECU. It is realized with. Instead of the ECU, a DSP (Digital Signal Processor) called a virtual ECU may be used. The engine control unit 18 gives a control signal (for example, a predetermined throttle opening) to the engine 12. As a result, the engine 12 rotates, and the rotation is transmitted to the dynamometer 14 via the shaft 32. The control signal given from the engine control unit 18 includes the engine speed, throttle opening, fuel injection amount, air injection amount, fuel / air mixture ratio, ignition time (in the case of a gasoline engine), fuel injection control. There are various control signals such as the method (in the case of a diesel engine).

  The dynamometer 14, dynamo control unit 16, engine control unit 18, and torque meter 36 described above are connected to the control device 20. The control device 20 has a function of feedforward controlling the dynamometer 14 and the engine 12 via the dynamo control unit 16 and the engine control unit 18, and as shown in FIG. 2, a controller 40, a simulation unit 42, and a correction unit. 44, an auxiliary controller 46 is provided.

  The controller 40 is a part that performs PID control, and includes a PID controller for the En control system and a PID controller for the Dy control system. When torque control is performed in the Dy control system, an FF controller may be further provided. The controller 40 is set (input) with a throttle opening target value of the engine 12 and a torque target value of the dynamometer 14, and a simulation result is input from a simulation unit 42 described later. The controller 40 has a throttle opening control command value (hereinafter referred to as a Th command value) such that a simulation result becomes a throttle opening target value and a torque target value, and a torque control command value (hereinafter referred to as a Dy command value) to the dynamometer 14. Are output to the simulation unit 42 and the actual machine unit (via the correction unit 44), respectively. Thereby, feedback control is performed in the simulation unit 42 by the same control command value, and feedforward control is performed in the actual unit. In this embodiment, the throttle opening of the engine 12 and the torque of the dynamometer 14 are used as the target value and the control command value. However, the present invention is not limited to this. The value of may be used.

  The simulation unit 42 includes a bench model obtained by modeling the above-described real machine unit, and simulation is performed using the bench model. The bench model is created as shown in FIG. 3 as an ideal two-mass model in which the engine 12 and the dynamometer 14 are connected, for example.

FIG. 3 is a block diagram showing an example of a bench model. In the figure, _JE is an inertia value of the engine 12, _JD is an inertia value of the dynamometer 14, K is a shaft torsion spring constant, and C is a shaft damper coefficient. Further, _{N E} is the rotational speed of the engine 12, _{N D} is the rotational speed of the dynamometer 14, _{T E} is the torque of the engine 12, _{T D} is the torque of the dynamometer 14, T is the shaft 50 Torque.

As shown in FIG. 3, engine model is made has an engine map as an input value of the engine map Th command value and the engine speed N _E, and outputs a driving torque T _E. Then, by integrating the difference between the load torque T from the output to the drive torque T _E and the shaft model, by dividing the inertia value J _E of the engine 12, it is obtained the rotational speed N _E.

Dynamo model has a dynamo map, this dynamo map as input dynamo rotational speed N _D and Dy command value (torque specified value), and outputs the drive torque T _D. Then, by integrating the difference between the load torque T from the output to the drive torque T _D and the shaft model, by dividing the inertia value J _D dynamometer 14, it is determined rotational speed N _D.

  The shaft model is a model of the shaft that connects the engine and the dynamo. The shaft model is twisted by the difference in the rotational speed between the engine and the dynamo and acts as a spring damper. The force generated by the spring / damper is output as a load torque T to the engine model and the dynamo model.

  The bench model configured as described above is a model that can be calculated by physical calculation, and is a model of an ideal state without considering dead time, rattling, noise, and the like. The bench model created in this way outputs the rotational speed and the torque value as simulation results by inputting the Dy command value and the Th command value, as in the actual machine part. The simulation result is output from the simulation unit 42 to the controller 40. The controller 40 receives this output, and the optimum Th command value and Dy command value such that the simulation result becomes the throttle opening target value and the torque target value are obtained. Is output to the correction unit 44.

  The correction unit 44 corrects the output timing with respect to the Th command value output from the controller 40 to the engine control unit 18 and the Dy command value output from the controller 40 to the dynamo control unit 16. Correction processing is performed so that there is no difference in dead time between the control system and the Dy control system. Specifically, the dead time Δt1 in the Dy control system and the dead time Δt2 in the En control system are obtained in advance by measurement or the like and input to the correction unit 44. Here, the dead time Δt1 is a delay time that occurs mainly due to the communication means in the Dy control system. For example, as shown in FIG. It is obtained by measuring the time until the torque measurement value actually starts to rise. Similarly, the dead time Δt2 is a delay time that occurs mainly due to communication means in the En control system. For example, as shown in FIG. 4B, the Th command value is changed stepwise, It is obtained by measuring the time until the torque measurement value starts to rise. Since these dead times Δt1 and Δt2 are individually generated due to the communication means, etc., each becomes an independent value and becomes a non-linear element.

  Therefore, the correction unit 44 performs a correction process such that the difference between the dead times Δt1 and Δt2 is apparently eliminated. For example, the Th command value that is the En control system is delayed by the dead time Δt1 of the Dy control system, and the Dy command value that is the Dy control system is delayed by the dead time Δt2 of the En control system. As a result, a time delay of Δt1 + Δt2 occurs in both control systems, so that control is performed simultaneously in both control systems. That is, an ideal situation with no dead time can be reproduced in the actual machine unit. Since this is the same state as the above-mentioned bench model, the behavior of the actual machine part and the simulation result of the bench model coincide with each other with high accuracy.

  Note that the correction method in the correction unit 44 is not limited to the above, and any correction process may be used as long as there is no difference between the dead times Δt1 and Δt2 between the En control system and the Dy control system. For example, the larger one of Δt1 and Δt2 may be used as a reference, and the other control system may be subjected to a correction process for delaying by the difference between Δt1 and Δt2. Specifically, when Δt1> Δt2, the Dy command value that is the Dy control system is not corrected, and the correction process that delays the Th command value that is the En control system by (Δt1−Δt2) is performed. As a result, a delay of Δt1 occurs in both control systems, so that it is possible to apparently eliminate the time lag.

  The control command value from the auxiliary controller 46 is added to the Th command value and the Dy command value output from the correction unit 44, respectively. The auxiliary controller 46 receives the above-described Th target value and Dy target value, and also receives a feedback signal (a torque measurement value and a rotation speed measurement value) from the actual machine unit. The auxiliary controller 46 includes an I (integration) controller for the En control system and an I (integration) controller for the Dy control system, and performs I control. Thereby, it can prevent that the measured value of a real machine part shift | deviates largely from a target value. That is, in the present embodiment, the controller 40 performs feedforward control based on the simulation result of the simulation unit 42, and control is performed without using the measurement value of the actual machine unit, which may cause a large error. This can be prevented by providing the auxiliary controller 46 for correction. In addition, since it will become unstable if I control is too good, it is good to restrict | limit a control amount with a fixed value. Further, if the simulation accuracy is improved, an aspect without the auxiliary controller 46 is possible.

  Next, the operation of the engine test apparatus 10 configured as described above will be described. FIG. 5 is a block diagram showing the configuration of a conventional engine test apparatus. Similar to the engine test apparatus 10 of the present embodiment, the conventional engine test apparatus shown in the figure includes a controller 40 that performs feedback control. The controller 40 includes a PID controller for the En control system and a PID controller for the Dy control system, and an FF controller as necessary. However, in the case of a conventional engine test apparatus, a torque measurement value and a rotation measurement value measured by the actual machine unit are input as feedback signals. That is, in the conventional engine test apparatus, feedback control is performed based on the measured value in the actual machine unit.

  In the conventional engine test apparatus configured as described above, if there is a difference between the dead time Δt2 of the En control system and the dead time Δt1 of the Dy control system, the control timing depends on the difference in the dead time of the two control systems. The problem of shifting occurs. Since this is a non-linear element, it is difficult to perform predictive control with a conventional engine test apparatus, which causes a problem that ideal control cannot be performed and responsiveness cannot be improved. Further, since there is a trade-off relationship between responsiveness and robustness, in the case of the conventional device, if the responsiveness is forcibly improved, the problem that the robustness is lowered occurs.

  On the other hand, the engine test apparatus 10 according to the present embodiment performs a simulation using a bench model obtained by modeling the ideal state of the actual machine unit, and inputs the result to the controller 40, and the controller 40 performs optimal control. The command value is determined and output to the simulation unit 42 and the actual unit (via the correction unit 44). Accordingly, the simulation unit 42 performs feedback control using the control command value, and the actual unit performs feedforward control using the same control command value. Thereby, since control is performed by physical calculation without a non-linear element, predictive control can be performed and responsiveness and robustness can be improved.

  In addition, according to the present embodiment, since the deviation of the dead times Δt1 and Δt2 is corrected by the correction unit 44, the deviation of the control timing in the two control systems can be eliminated, and the same ideal state as the bench model, That is, an ideal state that is not affected by the difference between the dead times Δt1 and Δt2 can be reproduced by the actual machine unit. Therefore, the simulation result by the bench model that models the ideal state is equal to the behavior of the actual machine part after the correction process, and the ideal state can be controlled. Thereby, the responsiveness of both the En control system and the Dy control system can be increased at the same time, and the performance excellent in robustness can be exhibited.

  FIG. 6 shows the results of a comparison test between this example and the conventional example. In this test, the torque responsiveness of the two is made equal, the throttle opening target value is constant, and the torque target value is changed stepwise. The change in torque over time is shown in FIG. The change with time is shown in FIG.

  As shown in these figures, in the conventional example, when the torque increases, the rotational speed decreases and the robustness deteriorates. This is considered to be because the controllability has deteriorated due to the dead time. On the other hand, in this embodiment, even if the torque increases, the rotational speed does not decrease and the robustness is excellent.

  In addition, although embodiment mentioned above was demonstrated in the example which applied this invention to the engine test apparatus which has two control systems (En control system and Dy control system), it is applied also in the case of three or more control systems. can do. For example, in the case of three control systems, if the respective dead times are Δt1, Δt2, and Δt3, the correction unit 44 delays the control command value by adding the dead times of the other two control systems. Output. That is, the Δt1 side control system gives a time delay of Δt2 + Δt3, the Δt2 side control system gives a time delay of Δt1 + Δt3, and the Δt3 side control system gives a time delay of Δt1 + Δt2. As a result, the three control systems have a time delay of Δt1 + Δt2 + Δt3, and the three control systems can reproduce the ideal state with no time lag. In this case, the control may be performed in accordance with the largest dead time.

  DESCRIPTION OF SYMBOLS 10 ... Engine test apparatus, 12 ... Engine, 14 ... Dynamometer, 16 ... Dynamo control part, 18 ... Engine control part, 20 ... Control apparatus, 22 ... Mount, 24 ... Suction pipe, 26 ... Slot, 28 ... Exhaust pipe, DESCRIPTION OF SYMBOLS 30 ... Catalyst mounting part, 32 ... Shaft member, 34 ... Universal joint, 36 ... Torque meter, 40 ... Controller, 42 ... Simulation part, 44 ... Correction part, 46 ... Auxiliary controller

Claims (3)

In an engine test apparatus comprising: a dynamometer that applies a load to an engine to be tested; and a controller that outputs a control command value to the engine and the dynamometer.
A simulation unit that executes a simulation by inputting the control command value into a bench model that models an actual machine unit including the engine and the dynamometer;
The engine test apparatus is characterized in that the controller inputs a result of the simulation to determine a control command value, and outputs the determined control command value to the simulation unit and the actual machine unit.   The engine test apparatus according to claim 1, further comprising a correction unit that corrects a dead time of the control command value output to the engine and / or the dynamometer. The engine test apparatus according to claim 1 or 2, wherein the control command value to the engine is a throttle opening, and the control command value to the dynamometer is a current value .

Images