JP2002366023A - Method and device for predicting abnormality in controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To precisely predict the abnormality occurrence time of the constitution portion of a controller and to securely set an abnormality-checking time. SOLUTION: In a processing part 64, a secular deterioration prediction part 90 obtains the predictive date of secular deterioration by comparing the obtained time-series data of the constitution site of a riding simulation device 10 with past fault history data. A maintenance time prediction part 92 obtains the predictive date of a maintenance time by comparing the accumulated using time of the device 10 with maintenance adaptation reference. Then, an earlier time of the predictive date of these secular deterioration and predictive date of the maintenance time to be the abnormality-checking time.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for estimating an abnormality in a control apparatus for estimating the time of occurrence of an abnormal state in each component of the control apparatus.

[0002]

2. Description of the Related Art For example, in a riding simulation device, which is one of the control devices, the operation state of its components is monitored in real time, and when it is detected that the operation state exceeds a predetermined allowable range, the device is activated. Emergency stop, repair of defective parts and replacement of broken parts. In this case, the riding simulation device cannot be used from the time the trouble occurs until the maintenance staff arrives at the site and the repair is completed. I do.

[0003] In addition, the maintenance and inspection work of the control device is usually performed.
Maintenance personnel go to the site for a certain period of time, check the condition of parts, and perform necessary parts replacement etc. based on experience, but when maintenance personnel confirm, they can always find faulty parts Not necessarily. Therefore, it is pointed out that it is extremely difficult to perform maintenance work at an appropriate time.

[0004]

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above-mentioned problems. The present invention predicts the occurrence of an abnormality in a component of a control device with high accuracy, or accurately determines the timing of an abnormality inspection. It is an object of the present invention to provide a method and an apparatus for detecting an abnormality in a control device that can be set to the following.

[0005]

In order to solve the above-mentioned problems, the present invention provides a first step of acquiring state information of each component when a control device is in a stopped state, and the acquired state information. A second step of storing the time series data as time-series data, and a third step of comparing the time-series data with a plurality of pieces of failure history data indicating a past failure state to predict an abnormality occurrence time of the component. It is characterized by becoming.

Further, the present invention provides a state information detecting means for detecting state information of each component of the control device, a time series data storing means for storing the detected state information as time series data, Failure history data storage means for storing a plurality of failure history data indicating a state, and an abnormality occurrence time prediction means for comparing the time-series data and the failure history data, and predicting an abnormality occurrence time of the component part. It is characterized by having.

In this case, it is possible to select failure history data exhibiting characteristics closest to the time-series data, determine the time at which a failure has occurred from the data, and predict that time as an abnormality occurrence time.

Further, after obtaining the cumulative operation time data of the control device, predicting the maintenance time of the control device from the temporal change of the cumulative operation time data, the abnormality occurrence time and the earlier maintenance time are determined as the abnormal inspection time. The maintenance work of the control device can be performed at an appropriate time before the occurrence of the abnormality.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An abnormality prediction method and apparatus in a control device according to the present invention will be described in detail below with reference to the accompanying drawings, showing preferred embodiments.

FIG. 1 shows a riding simulation device 10 which is an example of a control device. The device 10 includes a simulated vehicle body 14 disposed on a base 12. The simulated vehicle body 14 includes a sensor unit 40 (state information detecting unit) that detects each operation state of the occupant 11 and an operation unit 16 that operates according to the operation state detected by the sensor unit 40, and is controlled by the control unit 60. Is done.

As shown in the control block diagram of FIG. 2, the sensor unit 40 includes a throttle sensor 42 for detecting a throttle operation amount by the occupant 11, a clutch sensor 44 for detecting a clutch switching state, and a front brake. A front brake sensor 46 for detecting an operation amount, a rear brake sensor 48 for detecting an operation amount of a rear brake,
Steering sensor 50 for detecting steering angle of steering
A lean sensor 52 for detecting the inclination angle of the simulated vehicle body 14 due to a shift in the weight of the occupant 11, a temperature sensor 56 for detecting the temperature of each part of the simulated vehicle body 14, and a reference voltage supplied to each sensor. And a reference voltage detection unit 54 that performs the operation.

The operating unit 16 includes a speedometer 18 and a tachometer 20 for displaying the speed of the simulated vehicle body 14 and the engine speed based on the value detected by the throttle sensor 42;
Steering motor 24 for applying a predetermined torque to the steering based on the detection value of steering sensor 50
And a rolling motor 28 and a pitching motor 32 for setting a roll angle and a pitch angle of the simulated vehicle body 14 based on each detection value of the sensor unit 40.

The sensor section 40 supplies each detected value to a processing section 64 via an input section 72. In addition, the processing unit 64 supplies a control signal generated based on a value detected by the sensor unit 40 to the operating unit 16 via the output unit 70. It should be noted that the output unit 70 and the steering motor 2
4. Rolling motor 28 and pitching motor 32
Driver boards 22, 26, and 30 for converting a control signal into a drive signal are provided between.

The processing unit 64 constituting the control unit 60 stores the respective detection values detected by the sensor unit 40 as time-series data, and also stores the past failure state of each component constituting the riding simulation apparatus 10. A memory 62 (time-series data storage means, failure history data storage means) for storing a plurality of failure history data shown, and a time-lapse deterioration prediction unit for processing the time-series data to predict the time-dependent deterioration of the components of the simulated vehicle body 14 90 (abnormality occurrence prediction means), and a maintenance time prediction unit 92 (maintenance time prediction means, abnormality check time setting means) for predicting a maintenance time from the abnormality occurrence time and the predicted maintenance time of the component parts.
And the monitor 84, the operation unit 86, and the printer 88 are connected.

The riding simulation apparatus 10 of the present embodiment is basically configured as described above. Next, the operation thereof will be described with reference to FIG.

First, the power of the riding simulation apparatus 10 is turned on by an operator (step S1). Next, the operator sets a simulation mode suitable for the occupant 11 to be trained (step S2). In the simulation mode, for example, the attendance number of the occupant 11, setting of a driving mode for determining whether the vehicle is a two-wheeled vehicle or four-wheeled vehicle, setting of a simulated traveling course, selection of an automatic vehicle or a manual vehicle, and the like are performed. After the setting of the simulation mode is completed, the operator instructs the start of the simulation using the simulated vehicle body 14 (step S3).

At this time, prior to the operation by the occupant 11, the processing section 64 acquires a detection value from the sensor section 40 via the input section 72 (step S4). Each of the acquired detection values is stored in the memory 62 as time-series data (Step S5).

FIG. 4 is an explanatory diagram of the detection values stored in the memory 62 as time-series data. Each detection value is stored in the memory 62 in the order of the acquired time. As time series data, detected date data A _{1 to An} , detected time data B _{1 to} B _n , cycle mode data C _{1 to} C _n , cycle ON total time data D _{1 to} D _n , servo ON total time data E ₁ to E _n, traveling ON cumulative time data F ₁ to F _n, temperature data G ₁ ~G _n detected by the temperature sensor 56, the sensor reference detected by the reference voltage detection unit 54 voltage data H ₁ to H _n, throttle data I ₁ ~I _n detected by the throttle sensor 42, the clutch data J ₁ through J _n detected by the clutch sensor 44, the front brake data K ₁ ~ detected by the front brake sensor 46
K _n , rear brake data L _{1 to} L _n detected by the rear brake sensor 48, steering data M _{1 to} M _n detected by the steering sensor 50, and lean data N _{1 to} N _n detected by the lean sensor 52. Obtained and stored.

The cycle mode data C _{1 to} C _n are data indicating whether the state of the device for acquiring the detected value is before the start of the travel or after the end of the travel. The cycle ON cumulative time data D _{1 to} D _n are data representing the cumulative time when the riding simulation apparatus 10 is activated (power is on). The servo ON cumulative time data E ₁ to E _n, and data representing the cumulative time of the state in which the simulated vehicle body 14 can be actuated, the travel ON cumulative time data F ₁ to F _n, by occupants 11 This is data representing the total time during which the simulated running is being performed.

After the time series data is stored in the memory 62, the simulated traveling by the occupant 11 is started (step S6).

During the simulated traveling, the processing section 64 reads a detection value from the sensor section 40 based on the operation of the occupant 11, generates a predetermined control signal according to the detection value, and outputs an operation section via the output section 70. 16 to operate the speedometer 18, the tachometer 20, the steering motor 24, the rolling motor 28, and the pitching motor 32. Thereby, the simulated traveling is executed.

During the simulated traveling, the processing section 64 compares the detection value from the sensor section 40 with a predetermined set value in real time to determine whether or not the value is within a desired allowable range (step S7). If it is within the allowable range, the simulated traveling is continued until the operator gives an instruction to terminate traveling (step S8).

When the simulated running is completed, the processing section 64 obtains detection values from the sensor section 40 via the input section 72 as in the case of step S5 (step S9), and outputs the obtained detection values. The data is stored in the memory 62 as series data (step S10). In the time-series data in this case, the cycle mode data C _{1 to} C _n are set as data indicating that the state of the riding simulation apparatus 10 is after the end of traveling.

On the other hand, in step S7, the sensor unit 4
When it is determined that there is an abnormality in the detection value from 0, the processing unit 64 abnormally stops the simulation operation (step S11), and stores a message code indicating the abnormal state in the memory 62 together with the time-series data (step S11). Step S
12) Display an abnormal state on the monitor 84 (step S1)
3) Notify the operator.

As described above, the simulated traveling is performed, and the time series data of the state of each component of the riding simulation apparatus 10 before and after the simulated traveling is recorded.

Next, an abnormality prediction method of the riding simulation apparatus 10 based on the obtained time-series data will be described with reference to a flowchart shown in FIG.

In step S 30, the processing section 64 reads the time series data and the failure history data from the memory 62 and transfers them to the temporal deterioration prediction section 90.

The temporal deterioration prediction section 90 predicts the temporal deterioration of each component of the riding simulation apparatus 10 and calculates a temporal deterioration prediction date D1 (step S3).
1).

Here, the details of the temporal deterioration prediction processing will be described with reference to FIG. It should be noted that the failure history data is data indicating a temporal deterioration variation pattern acquired in the past of each component, and a plurality of patterns are set. Waveforms A, B, and C in FIG. 6 show, for example, past aging deterioration patterns of the plurality of throttle sensors 42. A waveform X is a plot of the throttle data I ₁ ~I _n of the throttle sensor 42 in the riding simulation system 10 obtained.

The degradation over time prediction unit 90, a waveform X and deterioration over time fluctuation pattern of the throttle data _{I 1 ~I n (A, B} ,
C), and the temporal deterioration fluctuation pattern (A, B, C) which is closest to the fluctuation pattern up to the collection date D0 of the waveform X
Search for. In this case, it is determined that the temporal deterioration variation pattern C shows a change closest to the waveform X.

Assuming that the permissible limit value of the throttle sensor 42 from the reference value is a time-dependent deterioration reference (a change value determined to have led to time-dependent deterioration) L1, the retrieved time-dependent deterioration pattern C leads to time-dependent deterioration. (Determination of time-dependent deterioration) D1 required for this can be obtained from FIG.
Therefore, this day D1 is set as a time-dependent deterioration prediction date on which the throttle sensor 42 having the waveform X is predicted to deteriorate with time.

That is, the aging of a part often progresses with a certain regularity, and if the aging deterioration pattern of the part at a certain point in time approximates a certain pattern in the past, it starts from that point. Can be predicted to approximate a certain pattern in the past, so that the predicted time-dependent deterioration D1 can be obtained from the past fluctuation pattern.

The past aging deterioration variation patterns A,
B and C may be generated from the product specifications of the parts, or data obtained in the design evaluation stage may be used. Of course, the time series data may be diverted. Further, by providing a greater number of the aging deterioration variation patterns, the accuracy of approximation can be improved, and the reliability of the aging deterioration prediction date D1 can be further improved.

The processing section 64 receives the temporal deterioration prediction date D1 from the temporal deterioration prediction section 90 and stores it in the memory 62 (step S32).

Further, the processing section 64 reads out the time series data from the memory 62 and transfers it to the maintenance time prediction section 92 (step S33). Maintenance time prediction unit 9
2 is a maintenance time (maintenance time) set for each component of the riding simulation apparatus 10
Is calculated, and a maintenance time prediction date D2 is calculated (step S34).

Here, details of the maintenance time prediction processing will be described with reference to FIG. The waveform Y in FIG. 7 is, for example, the cycle-on cumulative time data D _{1 to} D _n
Is plotted.

Normally, in a control device such as the riding simulation device 10, an allowable use time is set, and a maintenance application reference time L2 is set in advance as the use allowable time.

On the other hand, at the delivery destination of the riding simulation apparatus 10, for example, at a driving school or the like, the training time of the day is determined to some extent. Time is almost constant. Therefore, the cycle ON cumulative time can be considered to increase substantially in a linear curve.

Therefore, in FIG.
The variation pattern of the cycle ON cumulative time data D _{1 to} D _n up to 0 is approximated by a linear expression, and the day at the point where the extension line intersects the maintenance application reference time L 2 is determined as the maintenance time prediction date D 2.

The processing section 64 includes a maintenance time prediction section 9
Then, the maintenance date prediction date D2 is received from No. 2 and stored in the memory 62 (step S35).

Next, the processing unit 64 compares the predicted aging deterioration date D1 with the predicted maintenance time D2 (step S36), and determines the smaller value, that is, the smaller number of days, as the next maintenance date (step S36). Step S37
Or step S38). The processing unit 64 outputs the determined next maintenance date to the monitor 84 and the printer 88.

The maintenance time prediction process described above can be basically executed at any timing, and how often and when to perform the process or how to process the detection result is determined by the processing speed. Product performance such as
It is determined by the product specifications including the MTBF (mean failure interval) of the parts, the maintenance system, and the like. For example, the determination may be performed when the riding simulation apparatus 10 is started or terminated, or may be performed only when a maintenance person visits. Is also good.

The riding simulation apparatus 10 including the abnormality prediction apparatus according to the present invention may be configured as shown in FIG. 8 or FIG.

That is, in the riding simulation apparatus 10a shown in FIG. 8, the aging deterioration prediction section 90 and the maintenance time prediction section 92 of the riding simulation apparatus 10 shown in FIG. Connected data extraction unit 14
The time series data stored in the memory 62 is output to a storage medium 142 such as a floppy (registered trademark) disk via the control unit 0. The storage medium 142 storing the time-series data is loaded into an abnormality prediction device at a remote location or in another work room at an arbitrary timing, and the abnormality prediction device performs abnormality prediction processing. Can be.

Further, in the riding simulation apparatus 10b shown in FIG. 9, the time series data is transmitted from the communication section 150 via the communication line 154 to the external abnormality prediction apparatus 1
52 to perform an abnormality prediction process. In this case, the abnormality prediction process can be performed more quickly than in the riding simulation apparatus 10a illustrated in FIG.

Riding simulation apparatus 10
By configuring a and 10b as shown in FIGS. 8 and 9,
Although the speed of the abnormality prediction processing is slightly lower than that of the riding simulation apparatus 10, it is not necessary to provide the aging deterioration prediction section 90 and the maintenance time prediction section 92 for each riding simulation apparatus 10, thereby simplifying the apparatus configuration. Cost reduction can also be achieved.

As described above, the abnormality prediction apparatus according to the present embodiment may be incorporated in the riding simulation apparatus 10 as shown in FIG. 2 and executed in real time, or as shown in FIG. 8 or FIG. You may comprise so that it can be implemented by an external device. Further, in the configuration of FIG.
Needless to say, the configurations in FIG. 8 or FIG. 9 can be combined.

[0048]

As described above, according to the present invention, it is possible to accurately predict the time-dependent deterioration of the components of the control device, thereby performing maintenance at an appropriate time. Therefore, the occurrence of an abnormal state during the operation of the control device can be avoided, and as a result, the failure rate of the control device can be reduced and the reliability can be significantly improved. Further, since the maintenance and inspection work by the maintenance staff can be performed effectively, the effect of reducing the load required for the maintenance work can be obtained.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a riding simulation apparatus according to the present invention.

FIG. 2 is a control circuit block diagram including an abnormality prediction device of the riding simulation device according to the present invention.

FIG. 3 is an operation flowchart of the riding simulation apparatus according to the present invention.

FIG. 4 is an explanatory diagram of time-series data stored in a storage unit.

FIG. 5 is a flowchart illustrating a procedure of a temporal deterioration prediction process.

FIG. 6 is an explanatory diagram showing a comparison between time-dependent deterioration data and failure history data.

FIG. 7 is an explanatory diagram for comparing accumulated time data and maintenance application reference time data.

FIG. 8 is another control circuit block diagram of the riding simulation apparatus according to the present invention.

FIG. 9 is a block diagram of another control circuit of the riding simulation apparatus according to the present invention.

[Explanation of symbols]

10, 10a, 10b: Riding simulation device 14: Simulated vehicle body 16: Operating unit 40: Sensor unit 60: Control unit 62: Memory 64: Processing unit 70: Output unit 72: Input unit 90: Temporal deterioration prediction unit 92: Maintenance time Prediction unit 140 Data extraction unit 142 Storage medium 150 Communication unit 152 Abnormality prediction device 154 Communication line

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Kimio Wada 2-1-1 Minamiaoyama, Minato-ku, Tokyo Inside Honda Motor Co., Ltd. (72) Inventor Ken Masaki 1-10-1 Shinsayama, Sayama City, Saitama Honda E Engineering Co., Ltd.

Claims (6)

[Claims] A first step of acquiring state information of each component when the control device is in a stopped state; a second step of storing the acquired state information as time-series data; A third step of comparing a plurality of pieces of failure history data indicating past failure states and predicting an abnormality occurrence time of the component part. 2. The method according to claim 1, wherein a fourth step of obtaining accumulated operation time data of the control device as the time-series data; and a maintenance time of the control device based on a temporal change of the accumulated operation time data. A fifth step of predicting the abnormality, and a sixth step of comparing the abnormality occurrence time with the maintenance time and setting an earlier time as an abnormality inspection time. Method. 3. The method according to claim 1, wherein the abnormality occurrence time is selected from the failure history data indicating the time-dependent deterioration characteristic closest to the time-series data, and the abnormality occurrence time related to the failure history data is selected. An abnormality prediction method in the control device, wherein the method is set as: 4. A state information detecting means for detecting state information of each component of the control device; a time series data storage means for storing the detected state information as time series data; Failure history data storage means for storing the failure history data, and abnormality occurrence time prediction means for comparing the time-series data and the failure history data and predicting the abnormality occurrence time of the component. Abnormality detection device in the control device. 5. The apparatus according to claim 4, wherein maintenance time prediction means for predicting a maintenance time of the control device from a temporal change of accumulated operation time data of the control device, which is one of the time series data, An abnormality inspection time setting means for comparing the abnormality occurrence time with the maintenance time, and setting an earlier time as an abnormality inspection time. 6. An apparatus according to claim 4, wherein said status information detecting means acquires said status information when said control device is started and / or terminated. .