Measuring and testing – Simulating operating condition – Marine
Reexamination Certificate
2003-05-27
2004-12-28
Lefkowitz, Edward (Department: 2855)
Measuring and testing
Simulating operating condition
Marine
Reexamination Certificate
active
06834541
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to an engine testing apparatus and further relates to a map preparing method for the engine testing apparatus or a vehicle testing apparatus. More particularly, the invention relates to a novel preparing method in which learned data (actual machine data) or a learning map for determining a throttle (accelerator) opening degree, which is a target valve for controlling an engine under test or a vehicle under test, is defined as an exponential function or a multiple-degree equation function. Thereby, a peculiar point of the learned data is determined, and the particular point is automatically corrected when preparing the learning map.
The present invention further relates to a novel preparing method that uses data from a torque curve obtained by varying the throttle valve from a fully closed position to a fully open position while maintaining a constant engine rotation number and that uses data from a torque curve obtained by varying the throttle valve from a fully open position while maintaining a constant engine rotation number of a constant. The data is used to prepare a learning map for determining a throttle (accelerator) opening degree which is a target value for controlling an engine under test of a vehicle under test.
DESCRIPTION OF THE PRIOR ART
A conventional vehicle simulation system carried out on a stage includes a function for learning an engine under test (simply “engine”, hereinafter), and a learning map is prepared from the learned data. The engine is controlled based on the learning map.
The learned data are prepared by varying the throttle valve with an engine at an arbitrary rotation number and by storing an output torque (see FIG.
6
).
FIG. 6
shows a torque curve (actual measured value) obtained by varying the throttle valve from a fully closed position to a fully open position while maintaining the engine rotation number at a constant. From the torque curve, a torque value is determined at one point with respect to the throttle valve opening degree at a certain engine rotation number (e.g., 2000 rpm).
However, since torque curves A, B, C, D and E of various engine rotation numbers (1000 rpm, 1500 rpm, 2000 rpm, 2500 rpm, 3000 rpm) intersect at a low throttle opening degree in some cases, a peculiar portion is generated in a learning map based on the learned data, and accuracy of the control is deteriorated. For example, a peculiar driving state in which the throttle is closed for acceleration is generated. A first invention has been accomplished in view of the above circumstances, and an object of the first invention is to provide a map preparing method for an engine testing apparatus or a vehicle testing apparatus capable of preventing a peculiar driving state from being generated.
A throttle valve is varied while maintaining an engine at an arbitrary rotation number (e.g., 1500 rpm), and an output torque curve at the arbitrary rotation number is stored. The obtained output curves are determined as learned data
40
, and the learning map is prepared based on the learned data
40
(see FIG.
8
). Table 1 shows the learning map prepared by the conventional method.
TABLE 1
N
m
Rpm 1500
Rpm 2000
Rpm 2500
Rpm 3000
Rpm 3500
−40
0
0
0
0
0
−35
0
0
0
0
0
−30
0
0
0
0
0
−25
0
0
0
163
276
−20
0
43
110
260
343
−15
28
132
200
325
430
−10
80
190
279
392
460
−5
133
242
339
407
475
0
175
295
362
430
512
20
324
430
512
587
678
40
459
558
654
737
813
60
572
678
775
888
978
80
721
813
910
1045
1143
100
925
1060
1157
1261
1345
120
1359
1525
1524
1592
1675
140
4096
4096
4096
3054
2552
160
4096
4096
4096
4096
4096
180
4096
4096
4096
4096
4096
200
4096
4096
4096
4096
4096
From the learning map of Table 1, an output torque (simply “torque”, hereinafter) is determined at one point with respect to a particular engine rotation number and a particular throttle opening degree. For example, when the target engine rotation number is 1700 rpm and the desired target torque N
m
is 30, the throttle opening degree for generating the target torque N
m
can be determined from the values 329, 464, 435 and 563 by referring to the learning map of Table 1.
Conventionally, the throttle opening degree is controlled by varying the throttle valve from the fully closed position to the fully opened position, and the throttle opening degree is increased stepwise, for example, at 5% increments. Whenever the throttle opening degree is increased by 5%, it is necessary to wait until the torque is stabilized. The torque value is stored when it is stabilized.
However, when the engine is held at a constant rotation number and the throttle openings are the same, a torque output resulting from a throttle valve operated in the opening direction differs from a torque output resulting from the throttle valve operated in the closing direction. On the other hand, the conventional learned data can be obtained only when the throttle valve is fixed and the torque is stabilized as described above. Therefore, it is not possible to obtain a learning map corresponding to a variation in speed during a running speed pattern of a driving mode.
For example, it can be found from
FIG. 11
that actual vehicle data
38
, which exhibits variations in throttle opening degree of an actual vehicle running on a chassis dynamo based on a running speed pattern I of a driving mode, intersects data
39
, which is data simulated according to the conventional method. In
FIG. 11
, the pattern I is constituted by constant speed straight lines f, h, k, o, r, w, x, acceleration straight lines g, j, l, q and deceleration straight lines i, p, s and u.
That is, from
FIG. 11
, the following points can be found:
(1) For example, with respect to acceleration straight line j, data
38
does not coincide with data
39
. That is, since an output torque with respect to a throttle opening degree operated while referring to the learning map and a torque necessary for acceleration do not coincide with each other, data
39
is deviated higher than data
38
in the first half. In order to correct the deviation of vehicle speed, data
39
is deviated lower than the data
38
in the latter half.
(2) A reversal exists in the vertical relation between data
38
and data
39
in the case of the acceleration straight line j and the vertical relation between data
38
and data
39
in the case of the deceleration straight line p.
(3) The same phenomenon exists in the acceleration straight line g and the deceleration straight line s.
In this manner, since the accuracy of the simulation is poor, it is difficult to accurately drive an engine with respect to the running speed pattern I of the driving mode.
A second invention has been accomplished in view of the above circumstances, and an object of the second invention is to provide a map preparing method for an engine testing apparatus or a vehicle testing apparatus capable of enhancing the simulation accuracy.
To verify the performance of an automobile engine, there exists an engine testing apparatus comprising a dynamometer connected to an output section of an engine which is to be tested, a dynamo controller for controlling the dynamometer, and an actuator for controlling a throttle opening degree of the engine under test. The engine testing apparatus controls the dynamo controller and the actuator to adjust the output of the engine under test.
In the conventional engine testing apparatus, the rotation of the dynamometer is controlled by the dynamo controller, the throttle valve of the engine under test is controlled and operated, and the output torque of the engine under test is controlled, thereby simulating the actual vehicle running.
However, the conventional engine testing apparatus does not have a function for controlling the temperature of the peripheral portions of the engine under test such as engine cooling water temperature, fuel temperature, air intake temperature, exhaust gas temperature and lubricant temperature. Therefore, the temperature environment of an actual vehicle can not be reproduced, and e
Nakanishi Hideki
Noguchi Shinji
Ogawa Yasuhiro
Horiba Ltd.
Lefkowitz Edward
Stevens Maurice
Swienton Brian F.
LandOfFree
Engine testing apparatus and map preparing method for an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Engine testing apparatus and map preparing method for an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Engine testing apparatus and map preparing method for an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3299332