JP4158175B2 - Abnormality detection device for automatic transmission - Google Patents

Description

  The present invention relates to an abnormality detection device for a stepped automatic transmission.

  A conventional abnormality detection device for an automatic transmission will be described with reference to the drawings. FIG. 8 is a diagram showing a configuration of a conventional abnormality detection device for an automatic transmission.

  In FIG. 8, 1 is an engine, 2 is a torque converter, 3 is an input shaft rotational speed sensor, 4 is an automatic transmission, 5 is an output shaft rotational speed sensor, and 6 is a reading device. Further, 10 is a control unit (abnormality detection device for automatic transmission), 11 is input shaft rotation speed detection means, 12 is output shaft rotation speed detection means, 13 is failure detection control means, 14 is failure information storage control means, 15 Is a shift control means for controlling the automatic transmission 4, and 16 is a learning control means for optimizing the rotational change during the shift control.

  Next, the operation of the conventional abnormality detection device for an automatic transmission will be described with reference to the drawings.

  In general, an automatic transmission performs a shift by a coupling element of an internal clutch. However, if a certain clutch is in an abnormal state and the coupling is incomplete, the clutch may be broken by frictional heat or the like, and the entire automatic transmission may be damaged due to a coupling state different from usual.

  Therefore, in general, the automatic transmission apparatus detects an abnormality in the shift state. As shown in FIG. 8, the abnormality detection is performed based on information on the input shaft rotational speed of the automatic transmission 4, so-called NT, and the output shaft rotational speed, so-called NO.

  That is, the failure detection control means 13 of the control unit 10 can calculate the theoretical NT by multiplying NO by the current gear ratio. By comparing this theoretical NT with the detected NT at the same time, it can be determined that the same or approximate value is normal, and if a difference greater than a certain number of revolutions occurs, it is determined to be abnormal.

  Then, the failure detection control means 13 recognizes this abnormal state as a failure in shift synchronization and records it in the failure information storage control means 14. After recognizing the failure, the shift control means 15 of the control unit 10 stops the shift control and performs a process of fixing to the gear position in the fail-safe state of the automatic transmission 4 (see, for example, Patent Document 1).

JP 2000-35115 A

  However, the above-described abnormality detection device for an automatic transmission as described above uses means that can only be determined after a failure has occurred. For this reason, when the control unit 10 recognizes an abnormality, the automatic transmission device 4 is already in a failure state and no gear shifting operation is performed.

  Further, when the abnormality of the automatic transmission 4 gradually proceeds, the abnormal state is detected even if the periodic inspection is surely performed by using the maintenance inspection method in the current general level. There was a problem that it was difficult.

  The present invention has been made to solve the above-described problems, and can detect and store an abnormal state before the automatic transmission completely fails, thereby facilitating the prediction of failure during maintenance. An object of the present invention is to obtain an abnormality detection device for an automatic transmission.

  An abnormality detection device for an automatic transmission according to the present invention includes a shift control means for controlling the automatic transmission, a learning control means for optimizing a rotation change during the shift control by the shift control means, and a learning control by the learning control means. Is executed, the learning control execution number counter is counted up, the presence or absence of a change in the learning value is determined, and if there is a change in the learning value, the learning value change number counter is counted up, and the learning control If the value of the execution number counter is equal to or greater than a predetermined first set value, a value obtained by subtracting the value of the learning value change number counter from the value of the learning control execution number counter is less than or equal to a predetermined second set value And an abnormality detection control means for recording as an abnormal state.

  The abnormality detection device for an automatic transmission according to the present invention has an effect that an abnormal state of a component such as a clutch in the automatic transmission can be detected before reaching a failure state.

Embodiment 1 FIG.
An automatic transmission abnormality detection device according to Embodiment 1 of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a configuration of an abnormality detection device for an automatic transmission according to Embodiment 1 of the present invention. In addition, in each figure, the same code | symbol shows the same or equivalent part.

  In FIG. 1, 1 is an engine, 2 is a torque converter, 3 is an input shaft rotational speed sensor, 4 is an automatic transmission, 5 is an output shaft rotational speed sensor, and 6 is a reading device.

  In the figure, 10A is a control unit (abnormality detecting device for automatic transmission), 11 is input shaft rotational speed detecting means, 12 is output shaft rotational speed detecting means, 13 is failure detection control means, and 14 is failure information storage. Control means 15 is a shift control means for controlling the automatic transmission 4, 16 is a learning control means for optimizing the rotation change during the shift control, 17 is a learning value change monitoring control means, 18 is a learning value recording control means, 19 Is an abnormality detection control means.

  Note that each means inside the control unit 10A is actually configured by software, and the relationship between each means is indicated by an arrow.

  Next, the operation of the abnormality detection device for the automatic transmission according to the first embodiment will be described with reference to the drawings.

The abnormality detection device for an automatic transmission according to the present invention performs abnormality detection based on the following criteria.
(1) Abnormality detection based on determination that the learning value is outside the upper and lower limit values (see FIG. 2).
(2) Abnormality detection based on determination that the learning value after learning convergence is outside the range of the upper and lower limit values (see FIGS. 3 and 4).
(3) Abnormality detection by learning direction judgment after learning convergence (FIGS. 3 and 5).
(4) Abnormality detection based on learning non-convergence determination (FIG. 6).
(5) Indirect abnormality detection by recording learning information (FIG. 7).

  FIG. 2 is a flowchart showing an abnormality detection operation based on determination of upper and lower limits of a learned value of the abnormality detection device for an automatic transmission according to Embodiment 1 of the present invention.

  The automatic transmission 4 shifts by controlling the hydraulic pressure with a solenoid controlled by the control unit 10A, operating the clutch with the hydraulic pressure, and holding the internal rotating components.

  However, the internal rotating components are subject to variations that are not handled when they are produced, and it is difficult to perform optimal control by uniform control.

  Therefore, learning control is generally known as control that has a target value (hydraulic pressure) in the content of the shift control and approaches the target value.

  The learning value (hydraulic pressure) obtained by this learning control is converged by absorbing the mass production variation when learning is performed from a state in which learning control is not performed. That is, the learning value represents the variation of the internal rotational component, but the state of the internal rotational component can be grasped by monitoring the change in this value.

  In general, a range that the value can take is considered for the learning value, that is, the learning control is converged in the learning range in a normal production variation state.

  Conversely, if the learning control tries to learn beyond that range, it is some kind of abnormality. Therefore, by detecting this, the abnormal state of the automatic transmission 4 can be detected before the failure state is reached.

  For example, when the shift behavior is slow, the hydraulic pressure should be increased and the clutch as the coupling element should be strongly grasped, so that the shift behavior should be accelerated, and learning is performed in the direction of increasing the hydraulic pressure by learning control. However, if the shift behavior is not improved at all, it is clear that the coupling element is abnormal. If the gear cannot be changed, it can be detected by the conventional failure detection control means 13, but it cannot be detected by the existing technology when the gear is still changed.

  Therefore, as shown in FIG. 2, a range (lower limit value to upper limit value) that can be taken by the learning value is set in the learning control, and when it exceeds the range, it is determined as an abnormal state.

  First, in step 200, the learning control means 16 performs learning control.

  Next, in step 201, the abnormality detection control means 19 determines whether or not the learning value is within the range between the upper limit value and the lower limit value.

  Then, in step 202, the abnormality detection control means 19 records an abnormal state when out of the range.

  Since the learning does not converge and exceeds an appropriate range, it can be estimated that the automatic transmission 4 is not productive. If these statistics are taken and the same phenomenon appears in the automatic transmission 4 at the same production time, a problem in the production process can be found.

  Subsequently, an abnormal state is detected by monitoring the number of learning and its change.

  FIG. 3 is a flowchart showing a learning convergence determination operation of the abnormality detection device for an automatic transmission according to Embodiment 1 of the present invention.

  When the number of learning is small, that is, at the initial stage of learning, as described above, the learning value changes to absorb the manufacturing variation, but then gradually converges.

  Therefore, as shown in FIG. 3, the number of times of no change in the learning value is counted and compared with a predetermined set value, and it is determined that convergence has occurred by not changing the number of times.

  First, in step 300, the learning control means 16 performs learning control.

  Next, in steps 301 to 303, the learning value change monitoring control means 17 determines whether or not there is a change in the learning value. When there is no change, the learning value non-change counter is counted up, and when there is a change, the counter is set to zero.

  In steps 304 to 305, the learning value change monitoring control means 17 determines the learning value non-change counter value based on the set value, and if it is greater than or equal to the set value, records the learning value as converged.

  Next, threshold values (upper limit value and lower limit value) used for the abnormality determination of the behavior of re-change after learning convergence are set.

  FIG. 4 is a flowchart showing an abnormality detection operation based on determination of upper and lower limits of a learned value after learning convergence of the abnormality detection device for an automatic transmission according to Embodiment 1 of the present invention.

  As shown in FIG. 4, when learning is performed in the plus or minus direction when the learning value is 0 and converges, the learning change amount from the convergence point is set to a threshold value (upper limit value, Lower limit).

  First, in step 400, the learning control means 16 performs learning control.

  Next, in steps 401, 402 to 403, and 405 to 406, the abnormality detection control means 19 ends when the threshold value after convergence is not set and learning convergence does not occur, and when learning convergence converges (learning) For the learning convergence recording by the value change monitoring control means 17), a threshold value after learning convergence is set. That is, when learning converges in the positive direction, an upper limit value (plus side) and a lower limit value (minus side) are set around the learning value at the time of convergence. When learning converges in the minus direction, a lower limit value (minus side) and an upper limit value (plus side) are set around the learning value at the time of convergence.

  On the other hand, in steps 401, 404, and 407, when the threshold value after learning convergence has been set, the abnormality detection control means 19 determines the current learning value, and the threshold value after learning convergence (from the upper limit value to the lower limit value). When exceeding (range), it is recorded as an abnormal condition.

  FIG. 5 is a flowchart showing an abnormality detection operation by learning direction determination after learning convergence of the abnormality detection device for an automatic transmission according to Embodiment 1 of the present invention.

  Further, as shown in FIG. 5, the number of learning executions from when the learning value changes again after the learning convergence determination is set, and the setting value is set to the number of times that the learning control is reversed in the plus or minus direction within that number. When the number of inversions in the change direction of the learning value exceeding the set value is confirmed, it is determined as an abnormal state.

  First, in step 500, the learning control means 16 performs learning control.

  Next, in steps 501 to 502, the abnormality detection control means 19 determines whether or not there is a change in the learning value. If there is a change, the learning value change count counter is counted up.

  Next, in steps 504 and 506, the abnormality detection control means 19 determines whether the change direction of the learning value is the same or reverse between the previous time and the current time, and if it is the reverse, the learning value change direction inversion counter is counted up. .

  Next, in steps 508 to 510, the abnormality detection control means 19 determines whether or not the learning control has been converged (see learning convergence recording by the learning value change monitoring control means 17). The direction inversion counter value is determined by the set value 2, and if it is greater than or equal to the set value 2, it is recorded as an abnormal state.

  On the other hand, in steps 501, 503, 505, and 507, when the learning value is unchanged, the abnormality detection control means 19 counts up the number of times that the learning value has not changed, and the learning value unchanged number counter value is set to the set value. In the case of 1 or more, the learning value change direction inversion counter is set to 0.

  Thereby, it can be detected that the automatic transmission 4 that has been in a normal state is in an abnormal state for some reason, and the abnormal state can be found before failure by checking at the regular inspection.

  Subsequently, learning control is started from a state in which learning control is not performed, and when the learning value has changed but the value does not converge at a predetermined number of times of learning control execution, it is detected as abnormal.

  FIG. 6 is a flowchart illustrating an abnormality detection operation based on a learning non-convergence determination of the abnormality detection device for an automatic transmission according to the first embodiment of the present invention.

  As shown in FIG. 6, when the number of learning control executions from the start of learning control is counted and set to 1 or more, the setting value 2 is set to a value obtained by subtracting the number of learning control changes from the number of learning control executions. . When the set value is 2 or less, it is determined as an abnormal state.

  First, in step 600, the learning control means 16 performs learning control.

  Next, in steps 601 to 603, the abnormality detection control means 19 counts up the learning control execution number counter to determine whether or not the learning value has changed, and if yes, counts up the learning value change number counter. .

  Next, in steps 604 to 606, the abnormality detection control means 19 determines the number of times of learning control execution with the set value 1, and if it is greater than or equal to the set value 1, the value obtained by subtracting the number of learning value changes from the number of times of learning control execution is calculated. When learning is performed at a set value of 2 or less, that is, a certain number of times or more, and the learning value changes frequently, it is recorded as an abnormal state.

  Thereby, it is possible to estimate a production failure of the automatic transmission 4. If these statistics are taken and the same phenomenon appears in the automatic transmission 4 at the same production time, a problem in the production process can be found.

  Subsequently, the change of the learning value is stored.

  FIG. 7 is a flowchart showing the learning information recording operation of the abnormality detection device for an automatic transmission according to Embodiment 1 of the present invention.

  As shown in FIG. 7, the number of times of learning control execution from a state in which learning is not performed is counted, and when the learning value changes at that time, the count value and the learning direction, that is, the plus side or the minus side are recorded. Also, by recording the current count value, it is possible to analyze how and when learning has changed. For example, in the above-described determination of the abnormal state, if it is not determined to be abnormal but is not a normal learning change behavior, it becomes an indirect abnormality determination material.

  First, in step 700, the learning control means 16 performs learning control.

  Next, in steps 701 to 705, the learning value recording control means 18 counts up the learning execution number counter and determines whether or not there is a change in the learning value. Therefore, when there is a change in the learning value, the changed direction is determined. If the changed direction is positive, the counter value of the number of times of learning at that time and the learning direction are recorded as positive. Further, when it is determined that the changed direction is negative, the counter value of the number of learning executions at that time and the fact that the learning direction is negative are recorded.

  In step 706, the learning value recording control means 18 also records the total number of learning controls.

  By analyzing the stored contents at the time of abnormal state detection, failure, and complaint from the user, it is possible to analyze how the element is in an abnormal state, including the passage of time. Information. In particular, the content of complaints that are not easily reproduced among user complaints can be said to be particularly important information because of information stored in time series.

  It should be noted that the above-described abnormality determination contents and learning change transition records can be read by the reading device 6 connected to the outside. Thereby, the state of the automatic transmission 4 can be grasped | ascertained by confirming the value or result at the time of periodic inspection. In addition, by collecting statistics of the collected data, it is possible to confirm the manufacturing time, the state of production variation, and the state of aging of the automatic transmission itself, which can be used for future development and production technology. .

  As described above, according to the present invention, an abnormal state of a component such as a clutch in the automatic transmission can be detected before reaching a failure state. In addition, it is possible to confirm the record of the learning value over time.

It is a figure which shows the structure of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a flowchart which shows the abnormality detection operation | movement by the upper / lower limit judgment of the learning value of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a flowchart which shows the learning convergence determination operation | movement of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a flowchart which shows the abnormality detection operation | movement by the upper / lower limit judgment of the learning value after learning convergence of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a flowchart which shows the abnormality detection operation | movement by the learning direction judgment after learning convergence of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a flowchart which shows the abnormality detection operation | movement by the determination of the learning unconvergence of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a flowchart which shows the recording operation | movement of learning information of the abnormality detection apparatus of the automatic transmission which concerns on Embodiment 1 of this invention. It is a figure which shows the structure of the abnormality detection apparatus of the conventional automatic transmission.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Engine, 2 Torque converter, 3 Input shaft rotational speed sensor, 4 Automatic transmission device, 5 Output shaft rotational speed sensor, 6 Reading apparatus, 10A Control unit (automatic transmission device abnormality detection device), 11 Input shaft rotational speed detection means , 12 Output shaft rotation speed detection means, 13 Failure detection control means, 14 Failure information storage control means, 15 Shift control means, 16 Learning control means, 17 Learning value change monitoring control means, 18 Learning value recording control means, 19 Abnormality detection Control means.

Claims (1)

Shift control means for controlling the automatic transmission,
Learning control means for optimizing rotational changes during shift control by the shift control means;
When learning control is performed by the learning control means, the learning control execution number counter is counted up. When the learning value changes, the learning value change number counter is set. If the value of the learning control execution number counter is incremented and the value of the learning control execution number counter is equal to or greater than a predetermined first set value, a value obtained by subtracting the learning value change number counter value from the learning control execution number counter value is a predetermined first value. An abnormality detection device for an automatic transmission, comprising: an abnormality detection control means for recording an abnormality state when the value is equal to or less than a set value of 2.

Images