JP5046512B2 - Apparatus and method for detecting abnormality of rotating body - Google Patents

Description

  The present invention relates to a rotating body abnormality detection apparatus and method for preventing an accident caused by a failure of a rotating body such as a tire.

  As such a device, there is known an internal pressure abnormality alarm device that issues an alarm and warns a driver when a tire internal pressure becomes a predetermined value or less. In this apparatus, an internal pressure is measured by an internal pressure sensor attached to a wheel of a tire, and a warning is issued when the pressure falls below a certain internal pressure.

  However, this internal pressure abnormality alarm device cannot detect a failure caused by a cause other than a decrease in internal pressure. For example, as an example of such a failure, separation between the tread and the belt, between the cords constituting the belt, between the side rubber and the carcass ply, breakage of the ply cord or the belt cord, and chunk-out of the tread rubber ( For example, there is a state in which the block land portion provided in the tread is torn off. If the vehicle continues to run with these tire failures, sudden tire bursts may occur, making it impossible to drive.

As a system for detecting failures other than such internal pressure, an abnormality detection system for a rotating body that knows tire abnormality by measuring tire vibration and sound data and comparing it with normal data obtained in advance is known. (For example, Patent Document 1).
JP 2003-80912 A

  In recent years, in such a rotating body abnormality detection system, real-time analysis is desired in practical use. Therefore, the time delay in the logic is shortened as much as possible, and the amount of memory is also reduced in terms of hardware implementation. There is a request to make as small as possible.

  An object of the present invention is to provide a rotating body abnormality detection apparatus and method capable of improving detection accuracy without increasing the time delay and the amount of memory during detection.

An abnormality detection device for a rotating body according to the present invention includes a measuring means for measuring various physical quantities of a rotating rotating body, an extracting means for extracting a signal synchronized with the rotation of the rotating body from data measured by the measuring means, A determination means for determining the state of the rotating body from the signal extracted by the extraction means; and an abnormality warning means for warning an abnormality when the abnormality is determined by the determination means. digital filter, said constructed from data measured by the measuring means in a plurality of adaptive digital filters to analyze one rotational order component generated by calculating the rotation period, from each of said adaptive digital filter, rotary the remaining extracts the synchronizing signals, from said and data measured by the measuring means signal synchronized with the rotation extracted by the adaptive digital filter, the extracted to Obtains a signal by using an adaptive algorithm, the signal obtained is characterized in that to optimize the coefficients of the adaptive digital filter.

  The rotating body abnormality detection method according to the present invention uses the rotating body abnormality detection device according to the present invention to extract a signal synchronized with the rotation of the rotating body from various physical quantities of the rotating rotating body and rotate the extracted signal from the rotation. It is characterized by detecting abnormalities in the body.

  In the case of a multistage serial adaptive digital filter, the convergence accuracy can be improved as compared with the case of a single stage adaptive digital filter. However, in order to analyze all the frequency bands from the first-order to higher-order (20th to 30th) components in each adaptive digital filter, the length of the coefficient (number of taps) of each adaptive digital filter is set to Since the digital filters are set to the same length, in the case of a multistage serial adaptive digital filter, a time delay that is times the number of stages of the adaptive digital filter occurs. The length of the coefficient of each adaptive digital filter is determined by the length (frequency) of the primary wavelength which is the lowest frequency order.

  For example, when considering the case where the frequency of the rotating body is 10 Hz (corresponding to about 75 km / hr in the case of a tire), assuming that the coefficient length of the adaptive digital filter secures one wavelength of the primary frequency (10 Hz), Since the period of 10 Hz is 0.1 second, the total length of the three-stage serial adaptive digital filter is 0.1 × 3 = 0.3 seconds.

  That is, in the case of the logic of a multistage serial adaptive digital filter, the accuracy of convergence can be improved as compared with a single-stage adaptive digital filter, but the number of stages of memory is required and there is also a time delay in the logic. There is an inconvenience of becoming larger.

  According to the present invention, a plurality of digital filters are configured so that one adaptive digital filter analyzes one order component when compared with a single-stage adaptive digital filter or a multi-stage serial adaptive digital filter, thereby converging. Further improvements in both speed and accuracy are possible.

  As another advantage, adaptive digital filters are arranged in the number of necessary order components, and each adaptive digital filter has only one order to be analyzed. Total length) can be shortened.

  Preferably, various physical quantities of the rotating body measured by the measuring means are vibration, sound, rotational speed, distortion or displacement, and the adaptive digital filter is realized with at least two taps.

  Preferably, when extracting a signal synchronized with rotation by the extraction unit, an order component generated by calculating a rotation period from rotation information data among data measured by the measurement unit is used. In this case, an order component generation circuit for generating an order component is provided in a signal line between the rotation information data input unit from the measurement means and the adaptive digital filter.

  Preferably, when extracting the signal synchronized with the rotation by the extraction unit, the data measured by the measurement unit is variably sampled according to the data of the rotation speed information among the data measured by the measurement unit, and the apparent value is obtained. Make the upper period constant. In this case, a variable sampling circuit for executing variable sampling is provided in the data input section from the measuring means. As a result, the followability of the adaptive digital filter is improved even when the vehicle speed changes, and stable signal extraction becomes possible.

Embodiments of a rotating body abnormality detection apparatus and method according to the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram for explaining an example of a rotating body abnormality detection device of the present invention. In FIG. 1, 1 represents a tire as a rotating body, 2 represents a sensor for measuring vibration or sound attached to each tire 1, and 3 represents a tire 1 based on a signal from each sensor 2. This represents a central processing unit that detects abnormalities. The abnormality detection device for a rotating body of the present invention includes a measurement stage 11 that measures physical quantities such as vibration and sound by each sensor 2, and an extraction that extracts a signal synchronized with the rotation of the tire 1 from the data measured in the measurement stage 11. A stage 21, a determination stage 31 that determines the state of the tire 1 from the signal extracted in the extraction stage 21, and an abnormality warning stage 41 that warns the driver of an abnormality when the abnormality is determined by the determination stage 31. Yeah.

  The measurement stage 11 measures a sound or vibration around the tire 1 (measured by the sensor 2) or a rotation speed signal such as ABS (sensor 2 is not necessary), and inputs the measured data to the extraction stage 21 as a digital signal. When measuring sound, a microphone or the like is used as the sensor 2. When measuring vibration, an accelerometer, a speedometer, a displacement meter, or the like is used as the sensor 2. In addition, when the vehicle is equipped with an ABS (Anti-lock Brake System), an ABS rotation speed signal can be used. In this case, there is no need to provide the sensor 2, and a simple configuration can be achieved. Besides the ABS, the rotational speed can be measured by another method and the rotational speed signal can be used. Furthermore, when using an order component generation circuit and a variable sampling circuit, which will be described later, it is necessary to separately measure the wheel speed signal (ABS signal, etc.) of each wheel by a wheel speed signal measurement stage (not shown). Further, when measuring strain, a strain gauge or the like is used as the sensor 2, and when measuring displacement, a displacement meter or the like is used as the sensor 2.

  As the extraction stage 21, a periodic component synchronized with the rotation of the tire 1 is extracted using an adaptive digital filter (in this case, a notch type). As an example of the extraction stage 21, as shown in FIG. 2, a variable sampling circuit 32, an order component generation circuit 33, calculators 34-1, 34-2, 34-3,. . . , 34-n and adaptive digital filters 35-1, 35-2, 35-3,. . . , 35-n, and the order component generation circuit 33 may include primary component forming sections 33-1 to 3 to N (for example, 30) order component forming sections 33-n. In this case, the adaptive digital filters 35-1, 35-2, 35-3,. . . , 35-n are assigned for each order. As a method of changing the coefficients of the adaptive digital filters 25-1 to 25-3 using the error signal and the input signal, an LMS (Least Mean Square) method, a Newton method, or a maximum known conventionally as a filtering parameter update algorithm can be used. An adaptive algorithm such as a plunging method can be used. Other adaptive algorithms that can be suitably used include complex LMS algorithm (Complex Least means Square Algorithm), Normalized LMS algorithm (Normalized Least means Square Algorithm), projection algorithm, and SHARF algorithm (Simple Hyperstable Adaptive Recursive Filter Algorithm). , RLS algorithm (Recursive Least Square Algorithm), FLMS algorithm (Fast Least Mean Square Algorithm), adaptive filter using DCT (Adaptive Filter using Discrete Cosine Transform), SAN filter (Single Frequency Adaptive Notch Filter), neural network (Neural Network) ), Genetic Algorithm can be used.

  Adaptive digital filters 35-1, 35-2, 35-3,. . . , 35-n can be realized with a minimum of 2 taps as long as each of them can be applied to a periodic signal (sine wave or cosine wave) of one period, and the total number of taps is 60 in the case of 1 to 30th order. Therefore, it is very advantageous from the viewpoint of hardware (reduction of memory amount). The adaptive digital filters 35-1, 35-2, 35-3,. . . , 35-n are responsible for only one order (sine wave), which is advantageous in terms of convergence accuracy and convergence speed.

  The determination stage 31 uses the value of the output signal of each adaptive digital filter in the extraction stage 21 to compare with the value of the normal state in each tire 1 or two wheels (front and rear, left and right) or four wheels of the tire 1. By performing the method of comparing the values, the tire 1 showing an abnormal value is determined to be abnormal. For that purpose, it is preferable to provide a computer having a database section in which the normal data of the tire 1 is recorded, and a stage for comparing the extracted data of the two wheels or four wheels and determining an abnormality from the difference therebetween.

  The abnormality warning stage 41 gives a warning to the driver when the determination stage 31 determines that there is an abnormality. As the abnormality warning stage 41, it is preferable to use a warning light or a warning alarm. Also, information indicating abnormality can be fed back to the vehicle control to switch to vehicle control (ABS, VSC, etc.) according to tire abnormality.

  In the rotating body abnormality detection device of the present invention having the configuration described so far, only a signal correlated with the rotation of the tire 1 is easily and effectively extracted as an output signal, even if there is no accurate tire rotation number signal. can do. That is, only a signal correlated with the rotation of the tire 1 such as a burst is included in the output signal, and a signal uncorrelated with the rotation of the tire 1 due to a single event such as climbing on the curb is not included in the output signal. . Therefore, accurate abnormality determination can be performed. In this embodiment, an example is described in which vibration and sound signals measured by the sensor 2 are used as input signals.

  Next, actual waveforms in the rotating body abnormality detection apparatus of the present invention will be described. As shown in FIGS. 3A and 3B, a vehicle having a front wheel (left) 1-1, a front wheel (right) 1-2, a rear wheel (left) 1-3, and a rear wheel (right) 1-4. In FIG. 1, a tire (a tire with a 1 mm protrusion) simulating an abnormality was prepared and attached to the left front wheel, and a normal tire was attached to the right front wheel and left and right rear wheels, and an actual vehicle running test was performed on an actual road.

FIG. 4 is a diagram illustrating an example of an input signal, an output signal, and an error signal of each tire on a general dry road surface. In this example, the vehicle steadily traveled on a straight road at a speed of 100 km / h, and the acceleration vibration of the protrusions of the two front wheels was measured.
In FIG. 4, vibrations (input signals) caused by the protrusions are indicated by input vibration signals (1) (front wheel left) and (2) (front wheel right), respectively, and the output signal of the adaptive digital filter is extracted signal (ADF). Signal) (3) (front wheel left) and (4) (front wheel right) respectively, and error signals are indicated by error signals (5) (front wheel left) and (6) (front wheel right), respectively.

  Comparing the extracted signal (3) and the extracted signal (4), it can be seen that the abnormal wheel (extracted signal (3)) has a larger amplitude, and the tire abnormality can be determined from the time waveform. Furthermore, when the output signal of the adaptive digital filter is fast Fourier transformed, peaks corresponding to the respective orders appear. With respect to the peak for each order, the fast Fourier transform result (Fourier transform of the extracted signal (3)) of the abnormal wheel (front wheel left) indicated by the solid line a in FIG. When the peak level difference of the fast Fourier transform result of the front wheel right) (Fourier transform of the extracted signal (4)) is taken and the average peak level difference from the 2nd to the 30th order is calculated, it becomes 5 to 6 dB. That is, it can be seen from this result that tire abnormality can be determined. FIG. 5 shows a graph obtained by fast Fourier transform analysis of the extracted signal (ADF output signal) in FIG.

  Since the order of the order in which the peak level difference differs depending on the type of abnormality of the tire, the type of abnormality can be roughly determined in addition to the abnormality alarm by looking at the change of each order. The primary component varies due to factors other than tire abnormality, for example, because the influence of tire wheel balance is very large, it is preferable to exclude the primary component from abnormality determination.

  As described above, in the case of the adaptive digital filter (notch type) according to the present invention, the convergence accuracy is improved as compared with the serial adaptive digital filter, and the overall (first to higher order components) convergence time is shortened. It leads to improvement of real-time property. Further, the total memory amount of the adaptive digital filter corresponding to the total length of the adaptive digital filter can be greatly reduced, leading to cost reduction and simplification of hardware, which is advantageous for practical use.

The present invention is not limited to the above-described embodiment, and many changes and modifications can be made.
For example, in the example shown in FIG. 1, the present invention will be described by taking a tire mounted on a vehicle as a rotating body as an example. However, for the purpose of detecting an abnormality of a rotating rotating body, the rotating body other than a tire may be used. Similarly, the present invention can be applied.

  In the above-described embodiment, the example using the vibration and sound signals measured by the sensor as the input signal has been described. However, the present invention can be similarly applied to the case where the ABS rotation speed signal is used as the input signal. it can. For example, the rotation speed signal composed of a sine wave of a predetermined period includes a signal when the signal is lifted up on a burst and a curb as described above. Even in such a case, the rotation speed signal is input. By passing the extraction stage as a signal, an output signal including only a signal correlated with the rotation of the tire can be obtained.

  The rotating body abnormality detection device of the present invention is suitable for all applications that detect abnormalities of a rotating body, particularly abnormalities such as tire bursts and tread peeling, at an initial stage, which can be captured as periodic signals during rotation. Can adapt.

It is a figure for demonstrating an example of the abnormality detection apparatus of the rotary body of this invention. It is a block diagram of an example of the extraction stage in this invention. It is a figure which shows an example of the input vibration signal, extraction signal, and error signal of each tire in a general dry road surface. It is the graph which carried out the fast Fourier transform analysis of the extraction signal (ADF output signal) of FIG. It is a figure for demonstrating abnormality determination of a tire.

Explanation of symbols

1 Tire 1-1 Front wheel (left)
1-2 Front wheel (right)
1-3 Rear wheel (left)
1-4 Rear wheel (right)
2 Sensor 3 Central processing unit 11 Measurement stage 21 Extraction stage 32 Variable sampling circuit 33, 62, 72 Order component generation circuit 33-1, 33-2, 33-3,. . . , 33-n synthetic waveform shaping sections 34-1, 34-2, 34-3,. . . , 34-n computing units 35-1, 35-2, 35-3,. . . , 35-n Adaptive digital filter 31 Judgment stage 41 Abnormal warning stage 51 Burst part

Claims (8)

Measuring means for measuring various physical quantities of the rotating rotating body, extracting means for extracting a signal synchronized with the rotation of the rotating body from the data measured by the measuring means, and the rotating body from the signal extracted by the extracting means A determination unit that determines a state; and an abnormality warning unit that warns of an abnormality when the determination unit determines that there is an abnormality. The extraction unit is a data that is measured by the measurement unit by one adaptive digital filter. composed of a plurality of adaptive digital filters to analyze one rotational order component of the rotation period is generated by calculating from the data of the rotational information of the, from each of said adaptive digital filter, a signal synchronized with the rotation is extracted from the synchronization signal and the rotation extracted with data and said adaptive digital filter measured by said measuring means, obtains the extracted remaining signal, Using response algorithm, the abnormality detecting device of a rotating body, characterized in that to optimize the coefficients of the adaptive digital filter according to a signal obtained.   2. The rotating body abnormality detection device according to claim 1, wherein the various physical quantities of the rotating body measured by the measuring means are vibration, sound, rotational speed, distortion, or displacement.   2. The rotating body abnormality detection device according to claim 1, wherein the adaptive digital filter is realized by at least two taps. 2. The rotating body abnormality detection device according to claim 1, wherein the adaptive digital filter is realized by a SAN filter (Single Frequency Adaptive Notch Filter). The order component generation circuit for generating an order component, any of the preceding claims, characterized in that provided on the signal line between the adaptive digital filter and the input portion of the data of the rotation information from said measuring means The abnormality detection apparatus of the rotary body of Claim 1 .   In extracting the signal synchronized with the rotation by the extraction means, the data measured by the measurement means is variably sampled according to the data of the rotation speed information among the data measured by the measurement means, and the apparent cycle is determined. The rotating body abnormality detection device according to claim 1, wherein the abnormality detection device is constant. 6. The rotating body abnormality detection device according to claim 5 , wherein a variable sampling circuit for performing variable sampling is provided in a data input section from the measuring means.   Using the abnormality detection device for a rotating body according to any one of claims 1 to 7, a signal synchronized with the rotation of the rotating body is extracted from various physical quantities of the rotating rotating body, and the rotating body is extracted from the extracted signal. An abnormality detection method for a rotating body, characterized by detecting an abnormality in the rotating body.

Images