JP6398826B2 - Model accuracy improvement method - Google Patents

Description

  The present invention relates to a method for improving the accuracy of a model, and more particularly to a method for improving the accuracy of a transmission model.

  In recent years, model-based development has become widespread, and the importance of vehicle and unit modeling has increased. Here, regarding modeling, for example, there is a technique described in Patent Document 1.

  Patent Document 1 discloses that each part of a driving force transmission device is modeled based on an equation of motion, these models are connected, and the transmission torque is estimated with high accuracy.

JP 2014-095444 A

  In the technique described in Patent Document 1, since modeling is performed using only the equation of motion, it is considered that there is a difference between the simulation result by the created model and the result by the actual machine. However, in the technique described in Patent Document 1, it is unclear how much the model is different from the actual machine and where the difference is generated. That is, it has been difficult to identify a problematic part in the created model.

  The present invention has been made against the background described above, and it is an object of the present invention to provide a model high-accuracy method capable of specifying a location where high-precision is required in a model.

  A model high-accuracy method according to one aspect of the present invention is a model high-accuracy method using a transmission bench, comprising the step of attaching an actual transmission to a transmission bench and randomly oscillating, and constituting the actual transmission A step of obtaining a transmission characteristic at the time of random vibration between parts to be performed, a step of exciting the simulation model having the same configuration as the transmission of the actual machine under the same condition as the random vibration by the transmission bench, and the simulation model The step of acquiring the transmission characteristic at the time of the random vibration between the parts which comprise this transmission, and the step which compares both transmission characteristics is included.

  ADVANTAGE OF THE INVENTION According to this invention, the model high precision improvement method which can pinpoint the location where high precision is required in a model can be provided.

It is a flowchart which shows the procedure of the model improvement method concerning embodiment. It is a schematic diagram which shows a mode that the transmission of the actual machine was attached to the transmission bench. It is a figure explaining the acquisition object of the transfer characteristic in a transmission bench. It is a figure which shows typically about a simulation model. It is a figure which shows an example of the transfer characteristic obtained by the bench characteristic, and the transfer characteristic obtained by simulation.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a flowchart illustrating a procedure of a model accuracy increasing method according to the embodiment. In this method, first, an actual transmission is attached to a transmission bench (S1: Step 1). FIG. 2 is a schematic diagram showing a state in which the transmission 10 as an actual machine is attached to the transmission bench. As shown in FIG. 2, in step 1, for example, the transmission 10 is connected to a dynamometer 20 that applies driving torque to the transmission 10 and dynamometers 21 and 22 that absorb output torque of the transmission 10. That is, the dynamometer 20 functions as a drive shaft. Further, the dynamometers 21 and 22 function as absorption axes and can be controlled in rotation.

  Next, random vibration is performed on the transmission bench (S2: Step 2). That is, random excitation is performed by the dynamometer 20. And the transmission characteristic at the time of the random vibration between the parts which comprise the transmission of a real machine is acquired (S3: step 3). For example, transfer characteristics are acquired for torque, angular velocity, and the like inside the transmission 10. It should be noted that the more information inside the transmission 10 such as torque and angular velocity, the easier it is to isolate the problem part of the model and increase the accuracy. Also, since the transmission characteristics change due to gear backlash, etc., if you want to eliminate the effect of backlash, while rotating the transmission with the rotation control by the dynamometer 21 or the dynamometer 22, the vibration is not crossed over 0 Nm. The transfer characteristic may be acquired. For example, the transfer characteristic may be acquired by setting the average torque to 100 Nm and the random excitation torque to 50 Nm.

  FIG. 3 is a diagram for explaining a transfer characteristic acquisition target in step 3. In FIG. 3, as an example of the internal configuration of the transmission 10, a torque converter 101, a primary turbine 102, a secondary turbine 103, a reduction mechanism 104, a differential device 105, and a drive shaft 106 are illustrated. Therefore, the driving force from the dynamometer 20 is transmitted to the dynamometer 21 through the torque converter 101, the primary turbine 102, the secondary turbine 103, the reduction mechanism 104, the differential device 105, and the drive shaft 106. In the example shown in FIG. 3, the dynamometer 22 is omitted for easy understanding.

  In step 3, for example, torque transmission characteristics between the dynamometer 20 and the dynamometer 21, angular speed transmission characteristics between the dynamometer 20 and the primary turbine 102, angular speed transmission characteristics between the primary turbine 102 and the secondary turbine 103, and The transfer characteristic of the angular velocity between the secondary turbine 103 and the dynamometer 21 is acquired.

  Next, the simulation model having the same configuration as that of the actual transmission is vibrated under the same conditions as the random vibration by the transmission bench in Step 2 (S4: Step 4). Specifically, as shown in FIG. 4, a transmission model 30 which is a transmission simulation model having the same configuration as that of the transmission 10 is combined with dynamometer models 40, 41, and 42 to perform random excitation. Here, the dynamometer model 40 is a model in which the characteristics of the dynamometer 20 are identified. The dynamometer model 41 is also a model in which the characteristics of the dynamometer 21 are completely identified. Furthermore, the dynamometer model 42 is also a model in which the characteristics of the dynamometer 22 are completely identified. Therefore, the dynamometer models 40, 41, and 42 have identified, for example, mechanical characteristics such as resonance frequency and control delay such as time constant and dead time.

  Then, as in step 3, transfer characteristics at the time of random vibration between the parts constituting the transmission model are acquired (S5: step 5).

  Next, the transfer characteristic by the bench test obtained in step 3 is compared with the transfer characteristic by the simulation obtained in step 5 (S6: step 6). FIG. 5 is a diagram illustrating an example of a comparison result between the transfer characteristic obtained by the bench test and the transfer characteristic obtained by the simulation. In FIG. 5, the two transfer characteristics are overlaid for comparison. In FIG. 5, the transmission characteristic by the bench test obtained in step 3 is represented by a broken line, and the transmission characteristic by the simulation obtained in step 5 is represented by a solid line. 5A is a Bode diagram showing torque transfer characteristics between the dynamometer 20 and the dynamometer 21 and torque transfer characteristics between the dynamometer model 40 and the dynamometer model 41. FIG. 5B is a board line showing the angular velocity transmission characteristics between the dynamometer 20 and the primary turbine 102 and the angular velocity transmission characteristics between the dynamometer model 40 and the primary turbine model (not shown) corresponding to the primary turbine 102. FIG. FIG. 5C shows the transmission characteristics of the angular velocity between the primary turbine 102 and the secondary turbine 103, and the angular velocity between the primary turbine model (not shown) corresponding to the primary turbine 102 and the secondary turbine model (not shown) corresponding to the secondary turbine 103. It is a Bode diagram which shows the transfer characteristic. Further, FIG. 5D is a board line showing the angular velocity transmission characteristics between the secondary turbine 103 and the dynamometer 21 and the angular velocity transmission characteristics between a secondary turbine model (not shown) corresponding to the secondary turbine 103 and the dynamometer model 41. FIG.

  As shown in FIG. 5, when the simulation model is different from the actual machine, it appears as a difference in transfer characteristics. For example, in the example shown in FIG. 5, the difference between the transmission characteristic by the bench test and the transmission characteristic by the simulation becomes significant in a specific frequency band of the gain diagram indicating the transmission characteristic between the dynamometer 20 and the primary turbine 102. Yes. This shows that the modeling between the dynamometer 20 and the primary turbine 102 is not sufficient. Therefore, it is understood that it is necessary to further increase the accuracy of this portion. As described above, in the present embodiment, it is possible to specify a location where high accuracy is required in the model.

  If an insufficient part is determined from the comparison result in Step 6, the model is adjusted so that the transfer characteristic by the simulation matches the transfer characteristic by the bench test (S7: Step 7).

  In the transmission model 30, when the resonance frequency is reproduced, it is preferable to take the following procedure. First, as parameters such as inertia and spring rigidity in the transmission model 30, values that are adjusted to suit the phenomenon to be reproduced are not set, but design values are set. Then, the resonance frequency is grasped based on the gain peak. When investigating whether or not the recognized resonance frequency is reproducible, it is preferable to attempt to reproduce it by temporarily removing nonlinear elements included in the transmission model 30. This is because when a non-linear element is included, resonance may be buried and may not be manifested. If the resonance frequency of the actual machine and the resonance frequency of the simulation model are different, it means that there is an excess or deficiency of inertia / spring elements in the transmission model 30 or the design value is incorrect.

  In order to improve the reproduction accuracy in the transmission model 30, it is preferable to take the following procedure. After identifying the problem part of the transmission model 30, the vibration mode is checked to confirm which spring stiffness is oscillating before and after. Then, when matching the absolute value of the transmission characteristics in the transmission model 30 with that of the actual machine, adjust the damping of the spring stiffness specified when examining the vibration mode, and add nonlinear elements such as hysteresis and drag Adjust the gain and phase to the actual machine. When adjusting the attenuation, it is preferable to set a realistic value and to add a non-linear element with a ground.

  Next, this embodiment will be described using a specific example. Here, a CVT (Continuously Variable Transmission) transmission model is taken as an example. Specifically, it is assumed that the change in crankshaft rotation fluctuation when the engine is misfired by one cylinder is reproduced by a transmission model. In order to reproduce this, it is necessary to improve the accuracy of the transmission model connected to the engine. Here, it is assumed that the frequency range for which high accuracy is desired is 8 to 25 Hz. In this way, when making a model highly accurate, it is difficult to reproduce the behavior of the entire frequency band with the model, so the model is refined only to the frequency band where the phenomenon to be reproduced occurs. It is preferable.

  In order to increase the accuracy of this transmission model, first, as described above, the transmission characteristics of the actual machine are acquired by random excitation using a transmission bench. Next, even in a transmission model to be highly accurate that has the same configuration as that of the actual machine, a simulation is performed under the same conditions as the transmission bench, and transfer characteristics are acquired. Then, both transfer characteristics are compared. Here, it is assumed that resonance exists in the range of 8 to 25 Hz in the actual machine. In this case, for example, nonlinear elements such as damper hysteresis are temporarily removed from the transmission model, and simulation is performed again. Thereby, resonance can also occur in the simulation result. If the occurrence of resonance is also confirmed in the simulation, vibration mode analysis is performed. Here, it is assumed that the vibration is analyzed before and after the drive shaft and the lock amplifier damper by the vibration mode analysis. Finally, the damping of the drive shaft and lockup damper is adjusted, and nonlinear elements such as hysteresis are added to the transmission model to adjust the transmission characteristics and gain and phase of the actual machine obtained by the transmission bench. This increases the accuracy of the transmission model.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and can be modified as appropriate without departing from the spirit of the present invention.

DESCRIPTION OF SYMBOLS 10 Transmission 20, 21, 22 Dynamometer 30 Transmission model 40, 41, 42 Dynamometer model 101 Torque converter 102 Primary turbine 103 Secondary turbine 104 Reduction mechanism 105 Differential apparatus 106 Drive shaft

Claims (1)

A model high accuracy method using a transmission bench,
Attaching the actual transmission to the transmission bench and randomly oscillating;
Obtaining transfer characteristics at the time of random vibration between the parts constituting the transmission of the actual machine;
Exciting the simulation model having the same configuration as the actual transmission under the same conditions as the random excitation by the transmission bench,
Obtaining transfer characteristics at the time of random excitation between the parts constituting the transmission of the simulation model;
Comparing both transfer characteristics ;
A step of temporarily removing nonlinear elements from the simulation model, confirming the occurrence of resonance by performing the random excitation again, and performing vibration mode analysis;
Identifying a vibration occurrence location by the vibration mode analysis;
In the simulation model, the method includes adjusting the attenuation of the site where the occurrence occurs, adding the nonlinear element, and adjusting the transfer characteristics of the real machine and the transfer characteristics of the simulation model to match each other. Method.

Images