JPH0942432A - Control device for automatic transmission - Google Patents

Abstract

(57) [Summary] [Purpose] Accurately detect the actual shift start time in an automatic transmission to reduce shift shock during shift. [Arrangement] By the ignition timing retard amount detection means 11,
The ignition timing retard amount from the shift command to the start of the shift is detected, and the reference retard amount setting means 18 sets the reference retard amount based on the detected ignition timing retard amount.
This reference retard amount signal is input to the shift start detection detection means 13 via the engine rotation drop determination constant correction means 12, and when the shift start signal is calculated, the shift time detection means 1
4 and line pressure learning control means or learning control prohibition means 17
Is the engine rotation drop determination constant δ depending on the retard amount determined in relation to the engine load and the engine speed.
Was corrected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device for an automatic transmission for the purpose of reducing shift shock of a vehicle with an automatic transmission.

[0002]

2. Description of the Related Art Generally, in a vehicle with an automatic transmission, various frictional elements (such as clutches and brakes) are selectively hydraulically actuated by line pressure to select a predetermined number of gears, and the frictional element to be actuated is changed. Shifting to the gear position. However, at the time of this shift, a shift shock due to a change in the shift gear ratio is unavoidable.

As a conventional technique for reducing the shift shock, for example, the technique disclosed in Japanese Patent Laid-Open No. 2-38748 has been proposed.

According to this method, the output speed of the automatic transmission is multiplied by the gear ratio after shifting to obtain the pseudo input speed, and the torque indicated by the ratio between the pseudo input speed and the output speed of the engine. The pseudo speed ratio of the converter is calculated, the shift detecting means detects the start of the shift by the drop of the engine speed, and the time when the pseudo speed ratio of the torque converter reaches a predetermined set value is regarded as the end of the shift. Then, the shift shock is reduced by measuring the period as the shift time and learning-controlling the line pressure during the shift so that the measured shift time becomes the target time.

In the above-described conventional control device, whether or not the shift is started depends on whether or not the engine speed Ne has dropped from the maximum engine speed NeS of the transmission immediately before the shift by the engine drop determination constant δ or more. It is configured such that the engine drop determination constant δ is given as a fixed value regardless of the engine load and the engine speed due to the throttle opening.

[0006]

However, if the engine drop determination constant δ is set to a fixed value regardless of the engine load or the engine speed, the following problems occur.

I) When the value of the engine drop determination constant δ is increased, the engine load is small and the throttle opening (low T) is small.
At the time of VO), the shift start detection is delayed and the actual shift time cannot be accurately detected. As a result, the wrong shift time is learned and controlled.

Ii) When the value of the engine drop determination constant δ is reduced, the high throttle opening (high T
Even if the engine speed fluctuates due to ignition timing retardation such as knocking control during VO), the shift start is erroneously determined and an incorrect shift time is learned.

The reason will be described in detail below with reference to FIGS. 8 and 9.

First, FIG. 8 is a time chart showing the changing states of the engine speed and the output speed during gear shifting under low engine load and high engine load. Here, if the engine drop determination constant δ used for shift start detection is set to a fixed value and the actual shift times at the time of low engine load and high engine load are the same, the shift at low load is changed to that at high load. Since the difference in the engine speed between the front and the rear is small, the start of the shift is delayed with respect to the high load at low load.

Therefore, the shift time cannot be accurately detected, and the wrong shift time is learned and controlled.
The shift shock cannot be reduced.

FIG. 9 shows the relationship between the ignition timing retard change amount and the engine speed change amount based on the engine speed.

According to this, for example, when the retard change amount is set to 5 degrees, the engine fluctuates 15 revolutions when the engine is 2,000 revolutions, but at 5,000 revolutions, the same retard amount is doubled.
It fluctuates more than 0 revolutions.

As a result, when the engine drop determination constant δ is reduced in accordance with the low engine speed (low load), conversely, for high engine speed (high load), knocking control or the like is performed. Even if the engine speed changes due to ignition timing retard, it is erroneously determined that the shift is started,
The wrong shift time will be learned and controlled.

The present invention has been made in view of the above-mentioned problems, and an automatic transmission intended to accurately detect an actual shift time in the automatic transmission and reduce a shift shock at the time of shifting. The control device of

[0016]

In order to achieve such an object, the present invention provides an engine speed detecting means, a transmission output shaft speed detecting means, and a transmission output shaft speed after shifting. Pseudo speed ratio calculation means for calculating the pseudo speed ratio of the torque converter represented by the ratio of the pseudo input speed and the engine speed obtained by multiplying the gear ratio, means for detecting the engine load, and transmission output shaft rotation The speed change end determination means for determining the end of the speed change based on the torque converter pseudo speed ratio after the end of the speed change or the engine speed after the end of the speed change, which is set in advance by the number, and the engine speed has begun to decrease and has decreased more than the engine rotation drop determination constant. The shift detecting means for detecting the shift time from the time until the shift end determining means determines the shift end, and the shift detecting time detected by the shift detecting means. In a control device for an automatic transmission having a line pressure learning control means for learning and controlling a line pressure of a transmission so that a shift time is reached, at least an ignition timing retard amount of an engine from a shift command to the transmission to the start of the shift is set. Ignition timing retard amount detecting means for detecting, first reference retard amount setting means for setting a first reference retard amount, second reference retard amount setting means for setting a second reference retard amount, detected The ignition timing retard amount is larger than the first reference retard amount set by the first reference retard amount setting means and is less than or equal to the second reference retard amount set by the second reference retard amount setting means. At this time, a means for correcting the engine rotation drop determination constant according to the ignition timing retard amount is provided, and further, the detected ignition timing retard amount is When serial greater than the second reference retard amount, and a learning control prohibition means for prohibiting the learning control of the line pressure learning control means.

[0017]

According to the present invention, according to the present invention, when the ignition timing retard amount before the start of gear shift is larger than the first reference retard amount and is equal to or smaller than the second reference retard amount, the engine rotational drop determination constant is determined. As a result of δ being corrected according to the ignition timing retard amount, it is possible to accurately detect the shift start timing and reliably reduce the shift shock over the entire engine load.

Further, when the ignition timing retard which causes an erroneous detection when the ignition timing retard amount before the shift start is equal to or larger than the second reference retard amount is generated, the learning control of the shift time of the automatic transmission is prohibited, The quality of learning control was improved, and it became possible to reliably reduce shift shock.

[0019]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 (1) is a block diagram for explaining a control device for an automatic transmission according to the present invention, and FIG. 1 (2) is FIG. 1 (1).
1 is a system diagram for explaining the overall configuration of FIG. 1 (1), 1 is an engine, 2 is an automatic transmission, 3 is an engine control computer, and 4 is an automatic transmission control computer. The power of the engine 1 is input to the automatic transmission 2 via the torque converter 5, and the automatic transmission 2
Transmits the power to the output shaft 6 at a gear ratio corresponding to the selected number of gears in response to a shift control signal output from the automatic transmission control computer 4 to drive the vehicle.

The engine 1 is a multi-cylinder engine, and an ignition current is supplied to an ignition plug of each cylinder by an ignition signal from an engine control computer 3.
The engine control computer 3 receives a throttle opening signal from a throttle opening sensor 7 (engine load detecting means) for detecting the opening of a throttle valve and an idle SW.
Signal, engine speed signal from engine speed sensor 8 (engine speed detection means) for detecting engine speed, output shaft speed sensor 9 for detecting vehicle speed (transmission output shaft speed detection means). After the ignition signal of the engine 1 is calculated based on the output shaft rotation signal and the duty signal for performing comprehensive control such as the ignition timing retard signal between the engine control computer 3 and the automatic transmission control computer 4, Engine 1 by ignition signal
It is possible to drive.

Further, the automatic transmission control computer 4 is
The throttle opening signal and the idle SW signal from the throttle opening sensor 7, the engine rotation speed signal from the engine rotation speed sensor 8, the output shaft rotation speed signal from the output shaft rotation speed sensor 9, the engine control computer 3 and the automatic shift After calculating a shift control signal of the automatic transmission 2 based on a duty signal for performing comprehensive control such as an ignition timing retard signal with the machine control computer 4, the automatic transmission 2 is equipped with this shift control signal. The control valve 10 for gear shift control is controlled.

Next, the control device for the automatic transmission described above will be explained based on the block diagram shown in FIG. First, the retard signal of the engine control computer 3 is input to the retard amount detection means 11 for detecting the retard amount, and the reference retard amount setting means 18 sets the reference retard amount based on the detected retard amount detection signal. Then, the reference retard amount signal is input to the shift start detection means 13 via the engine rotation drop determination constant correction means 12. The shift start signal calculated by the shift start detecting means is input to the shift time detecting means 14 and the line pressure learning control means or the learning control prohibiting means 17 for correcting the line pressure of the torque converter.

On the other hand, the engine speed signal, the gear ratio after shifting and the output shaft speed signal are input to the torque converter pseudo speed ratio calculating means 15, and the torque converter pseudo speed ratio calculating means 15 causes the shift end judging means 16 to operate. The shift completion signal calculated through the above is input to the shift time detecting means 14. Then, the shift time detecting means 14 calculates a shift time based on the shift start signal input from the shift start determining means 13 and the shift end signal input from the shift end determining means 16, and the shift time is calculated by the torque converter. It is input to the learning control means or the learning control prohibiting means 17 which is the line pressure correcting means. The line pressure learning control means or the learning control prohibiting means 17 performs the learning control correction by increasing or decreasing the line pressure correction amount by a predetermined value according to the target shift time.

2 (a) and 2 (b) are a main flow chart executed by the control device for an automatic transmission according to the present invention and a flow chart for learning control of shift time, in which the engine speed decreases due to shift. It shows the shift time (inertia phase time) detection and learning control during the upshift shift, and these calculations are performed every 10 ms, for example.

First, at step 21, the engine speed sensor 8, the throttle opening sensor 7, and the output shaft speed sensor 9 are used to detect the engine speed Ne, the output shaft speed N0, the vehicle speed Vsp, the throttle opening Tvo at the present time. The basic physical quantity such as the idle SW Idle Sw is read and the process proceeds to step 22. The method of calculating each parameter is based on the method generally measured by a computer, and detailed description thereof will be omitted. In step 22, the shift start determination means 13 determines the gear position based on the throttle opening Tvo and the vehicle speed Vsp at the present time according to a preset shift pattern, and determines the shift. In step 23, it is determined whether or not the shift is an upshift. judge. If it is not an upshift, the engine speed NeS for shift start detection, the shift time detection timer Timer, the learning control determination (execution or prohibition determination) flag Flag associated with the shift time detection result, the maximum retard amount storage variable rmax, the shift in step 24 The start (engine rotational speed drop determination) detection end determination flag Ip SFlag is cleared (set to 0) to prepare for a future shift-up shift.

If it is determined in step 23 that the shift is an upshift, then in step 25, the engine speed drop determination constant δ is corrected or learning is prohibited, and step 2 is performed.
6, the method according to the first embodiment shown in FIG. 4 and the method according to the second embodiment shown in FIG. 5 is used for the correction of the engine speed drop determination constant δ or the learning prohibition determination at step 25. Since it exists, it will be described in detail later.

In step 26, the shift start detection means 13 determines whether or not the shift has started (the inertia phase has started). Basically, when an upshift shift is started after a shift command is issued to the automatic transmission 2, what has been increasing the engine rotation until then drops toward the engine rotation after the shift, so this is detected. . That is, if the engine speed Ne> NeS, it is determined that the engine speed is still increasing, and NeS is updated (NeS ← Ne) in step 27, and the process ends. As a result, NeS serves to detect the maximum engine speed during the gear shift. On the other hand, if the engine speed Ne> NeS is not satisfied at step 26, the routine proceeds to step 28, at which it is determined whether or not the gear shift has started. That is,
Here, in order to avoid erroneous determination, (NeS-Ne) ≧ δ
(Engine speed drop determination constant) is determined. Then, it is determined that the shift is started when the maximum engine speed NeS is decreased by δ rpm or more.

If it is determined in step 28 that the shift is started,
In step 29, the shift start detection end determination flag Ip SF
Set lag (Ip SFlag ← 1). This flag I
p SFlag is provided to detect the ignition timing retard amount (= constant δ correction or learning prohibition determination) only from the shift command to the shift start detection. Therefore IpS
After setting Flag, the shift time timer Timer is incremented (Timer ← Timer + 1) in step 30 (where shift time = inertia phase time = timer value × computation cycle).

Next, the torque converter pseudo speed ratio calculation means 15 determines the torque converter pseudo speed ratio for determining the end of the shift in order to determine the shift end. First, in step 31, Then, the torque converter pseudo speed ratio set value e0 (slice level) at the end of the shift is searched with reference to a two-dimensional table showing the relationship between the output shaft rotation speed and the throttle opening. This two-dimensional table shows, for example, a shift from 1st speed to 2nd speed for each type of shift,
It shall be set for each shift from 2nd speed to 3rd speed. And
Proceeding to step 32, where the output shaft speed No is multiplied by the post-shift gear ratio and divided by the engine speed Ne to obtain the torque converter pseudo speed ratio e, then in step 33,
The shift end determination means 16 determines whether or not e ≧ e0. Here, if e ≧ e0, it is determined that the shift is completed, and in step 34, the learning control based on the shift time detected by the shift time detecting means 14 is performed according to the learning control flowchart of FIG. This completes the detection of the shift time in the upshift gearshift, and in the next calculation, it is not considered in step 23 that the upshift gearshift is in progress, and this calculation is not repeated.

Next, the learning control routine will be described with reference to FIG. First, in step 41, the learning control or learning control prohibiting means 17 causes the type of shift (for example, 1
A target shift time (inertia phase target time) preset according to the throttle opening is calculated. Then, in step 42, this target shift time is compared with the shift time measured this time. If the shift time is longer than the target shift time, the line pressure correction amount is increased by a predetermined value in step 44, and if the shift time is shorter than the target shift time, the line pressure correction amount is decreased by a predetermined value in step 43. The learning control correction that has been conventionally performed, such as, is performed.

Although the line pressure control at the time of shifting is not described in detail here, it is basically assumed that the optimum line pressure (the optimum hydraulic pressure for various friction elements) is given according to the shifting. It is assumed that the hydraulic pressure is corrected by the learning control as described above.

On the other hand, FIG. 4 is a learning control routine showing a second embodiment for performing the engine speed drop determination constant δ correction or the learning inhibition determination. In this embodiment, first, at step 61, the learning control determination (execution or inhibition determination) flag Flag is checked. Where F
When lag = 1, it is determined that the ignition timing retard larger than the specified value has occurred, learning control is prohibited, and the process ends. If Flag = 1 is not satisfied, the routine proceeds to step 62, where the shift start detection end determination flag Ip SFlag is checked. Here, when Ip SFlag = 1 is determined that the shift start detection is completed, the retard amount is not detected and the routine proceeds to step 67.

If Ip SFlag = 1 is not satisfied, the retard amount r1 is read in step 63, and in the next step 64, from the one-dimensional table shown in FIG.
Based on the engine speed or the engine load (for example, the throttle opening), the detection determination reference retard r2 is searched.

Next, at step 65, the presence or absence of erroneous detection of engine rotation drop is determined using the retard value r1 read at step 63 and the first reference retard value r2 retrieved at step 64. If r1> r2 is not satisfied, it is determined that no erroneous detection has occurred, the process proceeds to step 73, and shift time detection and learning control are performed as in step 26 and subsequent steps in the main flow of FIG.

If r1> r2, it is determined that an erroneous detection has occurred, and the routine proceeds to step 66, where the one-dimensional table shown in FIG. 7B is used to determine the engine speed or engine load (for example, throttle opening). Based on this, the second reference retard value r3 for the constant δ correction or learning control prohibition determination is retrieved.
The second reference retard value r3 is set with reference to the correction limit of the constant δ at each engine rotation (or engine load), and if the correction is made too large, the shift start detection will cause an actual shift start. On the other hand, it is delayed and the shift time cannot be detected accurately. Therefore, if the ignition timing retard exceeds the correction limit,
The learning control is prohibited.

Next, after searching the reference retard value r3,
In step 67, it is determined whether the constant δ is corrected or learning is prohibited. If r1> r3, it is determined that a retard exceeding the correction limit has occurred, and in step 68, the learning control prohibition flag F
Set lag (Flag ← 1) and end. Also, r1
If not> r3, in step 69, the magnitude of the retard amount detected this time is determined. That is, a comparison is made with the maximum value of the retard amount detected from the shift determination to the previous time, and if rmax <r1 is not satisfied, the process proceeds to step 70 if rmax <r1 and rmax is updated (rmax ← r
1) Yes.

Next, in step 71, a correction amount Δδ for correcting the constant δ from the two-dimensional map shown in FIG. 7 (c) consisting of the engine rotation or engine load (for example, throttle opening) and the ignition timing retard value r1. Will be searched. Then, when the correction amount Δδ is retrieved, the routine proceeds to step 72, where the correction of the constant δ (δ = δ + Δδ) is performed, and the routine proceeds to step 73, where shift time detection is performed in the same manner as in step 26 and after in the main flow of FIG. 2A. And implement learning control.

FIG. 5 is a learning control routine showing a second embodiment for making a decision to prohibit learning control when the retard amount of the engine is larger than a specified value. In the figure, first, at step 51, a flag for line pressure learning control determination (execution or inhibition determination) is checked. As will be described later, this flag Flag is provided to prohibit learning control when the ignition timing retard amount is larger than a specified value.

Here, when Flag = 1, it is judged that the ignition timing retard larger than the specified value has occurred, and the learning control or learning control prohibiting means 17 prohibits the learning control and ends. If Flag = 1 is not satisfied, the routine proceeds to step 52, where the shift start detection end determination flag Ip SFlag is checked. This shift start detection end flag I
As described in detail in the description of the main flow of FIG. 2A, p SFlag is used to detect the ignition timing retard amount (= constant δ correction or learning prohibition determination) only from the shift command to the shift start detection. It was installed in. Where Ip
If SFlag = 1, it is determined that the shift start detection is completed, and the retard amount is not detected, and the routine proceeds to step 57.

If Ip SFlag = 1 is not satisfied, the retard amount r1 is read in step 53, and in the next step 54, the engine speed or engine load (for example, throttle opening) is read from the one-dimensional table shown in FIG. The first reference retard r2 for detection determination is searched based on the frequency. The first reference retard value r2 searched here is set based on the constant δ used for the engine drop determination, and an erroneous determination occurs at the constant δ at each engine rotation (or engine load). The ignition timing retard value (limit value) is set.

Next, at step 55, the presence or absence of erroneous detection of engine rotation drop is determined using the retard value r1 read at step 53 and the reference retard value r2 retrieved at step 54. If r1> r2, it is determined that an erroneous detection has occurred, the learning prohibition flag Flag is set (Flag ← 1) in step 56, and the process ends. If r1> r2 is not satisfied, it is determined that no erroneous detection has occurred, and step 5
7, the gear shift time detection and learning control are carried out in the same manner as in step 26 and thereafter in the main flow of FIG.

In the control device according to the present invention, the calculation of the engine speed is explained by the cycle measuring method using the ignition signal, but the same effect can be obtained by the frequency measuring method. Further, although the end of the shift is determined using the torque converter pseudo speed ratio, the present invention is not limited to this, and it may be determined using the engine speed after the shift.

[0044]

As described above in detail, in the control device for an automatic transmission according to the present invention, the shift start detection constant δ which is a constant for determining the engine rotation drop is determined in relation to the engine load and the engine speed. As a result of being corrected according to the ignition timing retard amount, it is possible to reliably reduce the shift shock over the entire engine load, and when ignition timing retardation that may lead to false detection occurs, the automatic transmission Since learning of the shift time is prohibited, the shift time can be detected more accurately.

Therefore, the quality of the learning control is improved, and the shift shock can be reliably reduced.

[Brief description of drawings]

FIG. 1 is a block diagram for explaining a control device for an automatic transmission according to the present invention, and FIG. 1 is a system diagram for explaining the overall configuration of (1).

FIG. 2 (a) is a main flowchart executed by a control device for an automatic transmission according to the present invention.
(B) is a flowchart for learning control of shift time.

FIG. 3 is a two-dimensional table showing the relationship between the output shaft speed and the throttle opening according to the present invention.

FIG. 4 is a learning control routine showing a first embodiment for making a judgment to prohibit engine speed drop judgment constant correction or learning control when the retard amount of the engine is larger than a prescribed value.

FIG. 5 is a learning control routine showing a second embodiment for making a judgment to prohibit engine speed drop judgment constant correction or learning control when the retard amount of the engine is larger than a prescribed value.

FIG. 6 is a one-dimensional table for searching a reference retard for detection determination based on the engine speed or engine load.

7A is a one-dimensional table for searching a reference retard for detection determination based on engine speed or engine load, and FIG. 7B is a reference retard value for constant δ correction or learning control inhibition determination. One-dimensional table to search, Fig. 7
(C) is a two-dimensional table for searching a correction amount for correcting the engine rotation drop determination constant.

FIG. 8 is a time chart showing a change state of an engine speed, a pseudo speed ratio change state, a transmission output rotation speed, and the like during a shift of the conventional automatic transmission during low engine load and high engine load.

FIG. 9 is an explanatory diagram showing a relationship between an ignition timing retard change amount and a change amount of the engine speed based on the conventional engine speed.

[Explanation of symbols]

1 Engine 2 Automatic Transmission 3 Engine Control Computer 4 Automatic Transmission Control Computer 5 Torque Converter 6 Output Shaft 7 Throttle Opening Sensor (Means to Detect Engine Load) 8 Engine Rotation Sensor (Engine Rotation Speed Detection Means) 9 Output Shaft Rotation Sensor (transmission output shaft rotation speed detection means) 12 Engine rotation drop determination constant correction means 17 Line pressure learning control means or learning control prohibition means

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display area F16H 59:40 59:42 59:74

Claims (2)

[Claims] Claim: What is claimed is: 1. An engine speed detecting means, a transmission output shaft speed detecting means, and a ratio of a pseudo input speed and an engine speed obtained by multiplying the transmission output shaft speed by a gear ratio after shifting. The pseudo speed ratio calculation means for obtaining the pseudo speed ratio of the torque converter, the means for detecting the engine load, and the torque converter pseudo speed ratio or the end of the speed change after the speed change preset by the transmission output shaft speed. The shift end determining means for determining the end of the shift based on the subsequent engine speed, and the period from the time when the engine speed starts to decrease and decreases by the engine rotation drop determining constant or more until the end of the shift is determined by the shift end determining means. A gear shift detection unit that detects the gear shift time, and a line pressure that learns and controls the line pressure of the transmission so that the gear shift time detected by the gear shift detection unit becomes a target gear shift time. In an automatic transmission control device having learning control means, ignition timing retard amount detection means for detecting at least an ignition timing retard amount of an engine from a gear shift command to the transmission to start of the gear shift, and a first reference retard amount. The first reference retard amount setting means, the second reference retard amount setting means for setting the second reference retard amount, and the detected ignition timing retard amount from the first reference retard amount setting means. Greater than the set first standard retard amount,
Further, when the second reference retard amount is less than or equal to the second reference retard amount set by the second reference retard amount setting means, a means for correcting the engine rotation drop determination constant according to the ignition timing retard amount is provided. Transmission control device. 2. The learning control inhibiting means according to claim 1, wherein the learning control inhibiting means inhibits the learning control of the line pressure learning controlling means when the detected ignition timing retard amount is larger than the second reference retard amount. A control device for an automatic transmission characterized by: