JP3391486B2 - Tire pressure detector - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tire air pressure detecting device for detecting a tire air pressure condition of a vehicle.

[0002]

2. Description of the Related Art Conventionally, as a device for detecting the air pressure of a tire, there has been proposed a device in which a pressure responsive member which responds to the tire air pressure is provided inside the tire to directly detect the air pressure of the tire. However, in a device that directly detects the tire air pressure, there is a problem that the structure becomes complicated and the price becomes expensive because it is necessary to provide a pressure responsive member inside the tire.

Therefore, the fact that the tire radius changes (becomes shorter) when the tire air pressure decreases,
It has been proposed to indirectly detect the tire air pressure of a vehicle based on a detection signal of a wheel speed sensor that detects the wheel speed of each wheel.

[0004]

However, the tire radius to be detected is subject to individual differences due to wear and the like, and is easily influenced by running conditions such as turning, braking and starting. Further, radial tires, which have been widely used in recent years, have a small amount of deformation of a tire radius due to a change in tire air pressure (for example, when the tire air pressure is reduced by 1 kg / cm ² ,
The amount of deformation of the tire radius is about 1 mm. ). For this reason, the method of indirectly detecting the change in the tire air pressure from the deformation amount of the tire radius has a problem that the detection accuracy cannot be sufficiently secured.

The present invention has been made in view of the above points, and an object of the present invention is to provide a tire air pressure detecting device capable of indirectly detecting the tire air pressure and improving the detection accuracy. Is.

[0006]

In order to achieve the above-mentioned object, a tire pressure detecting device according to the present invention is used for running a vehicle.
Sometimes, a wheel speed sensor for detecting a rotational speed of the wheels, before
The output signal of the wheel speed sensor is frequency-analyzed and the tire
A gain distribution calculating means for calculating a gain distribution of frequencies including vibration frequency components, an extracting means for extracting a gain distribution in a predetermined frequency region from the gain distribution calculating means, and a vehicle based on the extracted gain distribution in the predetermined frequency region. Resonance frequency calculating means for calculating the unsprung resonance frequency and detecting means for detecting the air pressure state of the tire based on the resonance frequency.

[0007]

[Function] With the above configuration, the vibration frequency component of the tire is included
The signal is obtained using a wheel speed sensor and
Calculates the gain distribution for the frequency, to calculate the resonance frequency of the unsprung vehicle based from the gain distribution in the gain distribution of a predetermined frequency range, based on the calculated resonant frequency, the state of the tire air pressure is detected .

When the tire air pressure changes, the spring constant of the tire also changes accordingly. Since the resonance frequency in the vibration frequency component of the tire changes due to the change of the spring constant of the tire, it is possible to detect the tire air pressure state based on the extracted resonance frequency.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a configuration diagram showing the overall configuration of the first embodiment.

As shown in FIG. 1, each tire 1a of the vehicle is
A wheel speed sensor is provided corresponding to 1d. Each wheel speed sensor includes gears 2a to 2d and pickup coils 3a to 3d. Gears 2a-2d
Is coaxially attached to the rotating shafts (not shown) of the tires 1a to 1d and is made of a disk-shaped magnetic body. The pickup coils 3a to 3d have these gears 2a to 2d.
Mounted at a predetermined interval in the vicinity of the gears 2a to
2d, that is, an AC signal having a cycle corresponding to the rotation speeds of the tires 1a to 1d is output. Pickup coil 3
The AC signals output from a to 3d are waveform shaping circuits, R
Known electronic control unit (ECU) including OM, RAM, etc.
4 and is subjected to predetermined signal processing including waveform shaping. The result of this signal processing is input to the display unit 5, and the display unit 5 notifies the driver of the air pressure state of each tire 1a to 1d. The display unit 5 may independently display the air pressure state of each tire 1a to 1d, or may be provided with one warning lamp and light up when the air pressure of any one of the tires is lower than the reference air pressure. You may make it warn about it.

First, the principle of tire pressure detection in this embodiment will be described. When a vehicle travels on a paved asphalt road surface, for example, minute irregularities on the surface of the road surface receive vertical and longitudinal forces, which cause the tires to vibrate vertically and longitudinally. The frequency characteristic of the acceleration under the vehicle spring when the tire vibrates is as shown in FIG. As shown in FIG. 2, the frequency characteristic of acceleration shows a peak value at two points, point a is the resonance frequency in the vertical direction under the spring of the vehicle, and point b is the resonance frequency in the longitudinal direction under the spring of the vehicle. is there.

On the other hand, when the air pressure of the tire changes, the spring constant of the rubber portion of the tire also changes, so that the resonance frequencies in the vertical direction and the front-back direction both change. For example, as shown in FIG. 3, when the tire air pressure decreases,
Since the spring constant of the tire rubber portion is also reduced, both the vertical and longitudinal resonance frequencies are reduced. Therefore, if at least one of the resonance frequency in the up-down direction and the front-rear direction in the unsprung part of the vehicle is extracted from the vibration frequency of the tire, it is possible to detect the tire air pressure state based on this resonance frequency.

Therefore, in this embodiment, the resonance frequencies in the up-down direction and the front-rear direction of the unsprung part of the vehicle are extracted from the detection signals of the wheel speed sensors. This is because, as a result of a detailed study by the inventors, it was found that the detection signal of the wheel speed sensor includes the vibration frequency component of the tire. That is, as a result of frequency analysis of the detection signal of the wheel speed sensor, it becomes clear that the peak value is shown at two points as shown in FIG. 4 and that when the tire air pressure is lowered, the two peak values are also lowered. It was

As a result, according to the present embodiment, the anti-skid control device (ABS), which has been increasing in the number of vehicles equipped in recent years,
In a vehicle or the like equipped with the above, since each wheel is already equipped with a wheel speed sensor, the tire air pressure can be detected without adding any new sensor. Further, in the practical range of the vehicle, since the amount of change in the resonance frequency is almost based on the change in the tire spring constant caused by the change in the tire pressure, it is not affected by other factors such as wear of the tire. Stable air pressure detection is possible.

FIG. 10 shows a flowchart showing the contents of processing executed by the ECU 4. Since the ECU 4 performs the same processing on each wheel 1a to 1d, the flowchart of FIG. 10 shows only the processing flow for one wheel. Further, in the following description, the subscripts of the respective reference numerals are omitted. Further, the flowchart shown in FIG. 10 shows an example in which it is detected that the tire air pressure has dropped below a reference value and a warning is given to the driver.

In FIG. 10, in step 100, after the waveform of the AC signal (FIG. 5) output from the pickup coil 3 is shaped into a pulse signal, the pulse interval is divided by the time interval between the wheel speeds v. Is calculated. As shown in FIG. 6, the wheel speed v usually includes many high frequency components including the vibration frequency component of the tire. In step 110, it is determined whether or not the calculated fluctuation width Δv of the wheel speed v exceeds the reference value v ₀ . At this time, if it is determined that the variation width Δv of the wheel speed v exceeds the reference value v ₀ , the process proceeds to step 120. Step 12
At 0, it is determined whether or not the time ΔT in which the fluctuation width Δv of the wheel speed v exceeds the reference value v ₀ exceeds the predetermined time t ₀ . The processing in steps 110 and 120 is performed to determine whether or not the road surface on which the vehicle is traveling is a road surface on which the tire air pressure can be detected by the detection method of this embodiment. That is, in the present embodiment, the tire air pressure is detected based on the change in the resonance frequency included in the tire vibration frequency component. Therefore, if the wheel speed v fluctuates to some extent and is not continued, sufficient data for calculating the resonance frequency cannot be obtained. In the determination in step 120, when the variation width Δv of the wheel speed v exceeds the reference value v ₀ , the predetermined time Δ
When t is set and the fluctuation width Δv of the wheel speed v again exceeds the reference value v ₀ within this predetermined time Δt, the measurement of the time ΔT is continued.

If both steps 110 and 120 are affirmatively determined, the operation proceeds to step 130, and if either one is negatively determined, the step 100 is performed.
Return to. In step 130, frequency analysis (for example, FFT calculation) is performed on the calculated wheel speed, and
The number of calculations N is counted. FIG. 7 shows an example of the result of performing this FFT operation.

As shown in FIG. 7, when the FFT calculation is performed on the wheel speed obtained by actually traveling the vehicle on a general road, it is usual that the frequency characteristic becomes extremely random. This is because the shape (size and height) of the minute irregularities present on the road surface is completely irregular, and therefore the frequency characteristic varies for each wheel speed data. Therefore, in the present embodiment, in order to reduce the variation of the frequency characteristic as much as possible, the average value of the FFT calculation results of a plurality of times is calculated. Therefore, in step 140, step 130
It is determined whether or not the number of FFT operations N in (4) has reached a predetermined number n ₀ . Then, when the number of calculations N has not reached the predetermined number n ₀ , the processing from step 100 to step 130 is repeated. On the other hand, when the number of calculations N has reached the predetermined number n ₀ , step 1
Proceeding to 50, averaging processing is performed. In this averaging process, as shown in FIG. 8, the average value of each FFT calculation result is obtained, and the average value of the gain of each frequency component is calculated. By such an averaging process, the FFT depending on the road surface
It is possible to reduce the fluctuation of the calculation result.

However, with the above-mentioned averaging process alone, the gain of the resonance frequency in the up-down direction and the front-rear direction of the unsprung part of the vehicle does not always reach the maximum peak value as compared with the gain of the frequencies in the vicinity thereof due to noise or the like. There is a problem that it is not limited. Therefore, in the present embodiment, following the averaging process described above, the following moving average process is performed in step 160.

This moving average processing is carried out by obtaining the gain Y _n of the _nth frequency by the following arithmetic expression.

[0021]

## EQU1 ## Y _n = (y _{n + 1} + Y _n-1 ) / 2 That is, in the moving average processing, the gain Y of the nth frequency is obtained.
_n is the (n + 1) th gain y in the previous calculation result
_{n + 1} and the gain Y of the n-1th frequency already calculated
_It is an average value with _n-1 . As a result, the FFT calculation result shows a waveform that changes smoothly. FIG. 9 shows the calculation result obtained by this moving average processing.

The waveform processing here is not limited to the moving average processing described above, but low-pass filter processing may be performed on the FFT calculation result after the averaging processing, or the FFT calculation of step 130 is performed. Before this, the wheel speed v may be differentially calculated, and the FFT calculation may be performed on the differential calculation result.

Next, at step 170, the resonance frequency f in the unsprung front-rear direction of the vehicle is calculated based on the FFT calculation result smoothed by the moving average processing. However, the maximum peak gain does not completely match the resonance frequency. This tends to occur when the number of times N of calculation of the averaging process is not sufficiently large, but when the number of times N of calculation is increased, it takes time to determine the air pressure, which causes a problem of impracticality. Therefore, here, the following estimation of the resonance frequency (hereinafter referred to as shape estimation processing) is performed to solve the above problem. A flowchart of this shape estimation processing is shown in FIG.

At step 171, the presence of the resonant frequency
Maximum peak gain G in the expected frequency band_P
To calculate. Then, in step 172, as shown in FIG.
Its peak gain G_PA few percent (eg 70%) of
Gain is threshold G_SAs a predetermined lap
Wave number band F_bFrequency f at the point where the threshold of₁, F₂Calculate
Put out. In step 173, the point p at which the threshold is cut
Check if there are two, and only if there are only two
The operation proceeds to step 174 and the calculated frequency f ₁, F₂
Intermediate frequency f_mIs the resonance frequency f.

In step 175, if there are two or more points at which the threshold value is cut, the data is a data that contains a lot of noise components. Therefore, the resonance frequency obtained immediately before as the value of the resonance frequency at this time. Is used as it is, or is estimated (for example, its average value) from a plurality of resonance frequency values obtained in the past.

Then, in step 180, a decrease deviation (f ₀ −f) from the initial frequency f ₀ which is set in advance corresponding to the normal tire air pressure is obtained, and this decrease deviation (f ₀
-F) is compared with the predetermined deviation Δf. This predetermined deviation Δf
Is an allowable lower limit value of the tire air pressure (for example, 1.4 k, based on the initial frequency f ₀ corresponding to the normal tire air pressure).
g / m ² ). Therefore, in step 180, the decrease deviation (f ₀ −f) is equal to the predetermined deviation Δf.
If it is determined that the tire air pressure has fallen below the allowable lower limit value, the routine proceeds to step 190,
The display unit 5 displays a warning to the driver.

When the resonance frequency is estimated in step 170 described above, if the waveform shape near the resonance frequency is not smooth, there are two or more points p at which the threshold value is cut in step 173, resulting in an erroneous determination. Therefore, step 1 above
The moving average process at 60 may be replaced with a more powerful smoothing average process as follows. That is, when obtaining the n-th frequency gain Y _n , an average of gains before and after a plurality of n (for example, average of 5 points before and after shown in Formula 2) is at least 1.
Repeat (the process becomes smoother).

[0028]

## EQU2 ## Y _n = (y _n-2 + y _n-1 + y _n + y _{n + 1} + y _{n + 2} ) / 5 (where y _n-2 and y _n-1 ── are before processing) It is a gain, and if the process is repeated and the sth time is performed, it is the s-1th gain.) Or, in this modification, a moving average that combines the gains before and after the process may be used. Repeating the process makes it smoother than the above method. An example using 5 points in Equation 3 is shown.

[0029]

## EQU3 ## Y _n = (Y _n-2 + Y _n-1 + y _n + y _{n + 1} + y _{n + 2} ) / 5 (where, in the case of the s-th processing, y _n , y _{n + 1} , y _{n +2}
Is the gain after the s-1st processing, and Y _n-2 and Y _n-1 are s
It is the gain after the second processing. In addition, in both Equations 2 and 3, the denominator 5 may be omitted for convenience of processing. In addition, a smooth waveform can be obtained by obtaining an envelope (for example, a spline function) instead of these averaging processes, and thus this may be used in combination with step 160 or as an alternative.

In the above-described first embodiment, an example in which a decrease in tire air pressure is detected based only on the resonance frequency of the unsprung part of the vehicle in the front-rear direction has been shown. The decrease in tire air pressure may be detected based only on the frequency, or may be detected based on both the longitudinal and vertical resonance frequencies.

Further, in the above-described first embodiment, the fact that the tire air pressure is lower than the allowable lower limit value is particularly detected, but the relationship between the tire air pressure and the resonance frequency is shown in FIG. Store as a map for each tire,
The resonance frequency f may be calculated in the same manner as in the first embodiment, and the tire air pressure itself may be directly estimated from the calculated resonance frequency f.

Next, a second embodiment will be described. Above
In the resonance frequency estimation (step 170) of the first embodiment,
Was performed by setting a single threshold for the gain used for estimation.
However, if there is a margin in the memory capacity of the ECU 4, this second implementation
The estimation accuracy can be further improved by an example. Sanawa
Then, in step 172, the threshold value is set to several steps (for example, 7
0%, 60%, 50%)
The frequency at which the threshold is cut is calculated for each value. That
Then, in step 174, the above-mentioned procedure is performed for each threshold value.
Intermediate frequency (f_m1, F _m2, F_m3) Is calculated and
The average value of is the resonance frequency f. Also, calculate this average value
When there is a large variation in each intermediate frequency,
You can do it on your own, or weight it with a threshold
You may go.

In the second embodiment, the processing contents in the ECU 4 differ from the first embodiment only in the above-mentioned parts, and the other processing contents and the configuration thereof are the same as those in the first embodiment.

Next, a third embodiment will be described. Although the shape estimation process shown in FIG. 12 was performed in the resonance frequency estimation (step 170) of the first embodiment, the center of gravity of the waveform near the resonance frequency is calculated in the third embodiment, and the center of gravity is calculated. Is used as the resonance frequency f (hereinafter referred to as center-of-gravity processing). That is, in the third embodiment, the processing of FIG. 12 of the first embodiment is changed to the processing shown in FIG.

In FIG. 14, in step 176, the total gain sum S of the predetermined frequency band is calculated. Then, in step 177, the sum of gains is sequentially obtained from the end of the frequency band, and the frequency at the point where the sum of gains S becomes S / 2 is calculated. This frequency is the frequency having the center of gravity, and this frequency is defined as the resonance frequency f (see FIG. 15).

In the frequency band used in step 176, if the memory capacity of the ECU 4 has a margin, the peak of the resonance frequency characteristic will be between valleys on both sides (the bottom of the mountain when the peak is regarded as a mountain). It is better to use.

In the third embodiment, the processing contents in the ECU 4 differ from the first embodiment only in the above-mentioned parts, and the other processing contents and the configuration thereof are the same as those in the first embodiment.

[0038]

As described above, according to the present invention,
Calculate the gain distribution for the frequency in the signal containing the vibration frequency component of the tire, extract the gain distribution in the predetermined frequency region from this gain distribution, and determine the unsprung resonance frequency of the vehicle based on the extracted gain distribution in the predetermined frequency region. The tire pressure is calculated based on the calculated resonance frequency. Here, the resonance frequency changes according to the spring constant of the tire, and the spring constant of the tire changes substantially only depending on the air pressure of the tire. Therefore, according to the present invention, it is possible to improve the detection accuracy while indirectly detecting the tire air pressure.

Further, since the unsprung resonance frequency of the vehicle is calculated on the basis of the gain distribution in the predetermined frequency region, if the predetermined frequency region is set to a frequency band near the resonance frequency, unnecessary calculation is performed in other frequency regions. There is an effect that the unsprung resonance frequency can be calculated promptly without performing any processing.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is a characteristic diagram showing frequency characteristics of unsprung acceleration of a vehicle.

FIG. 3 is a characteristic diagram showing how the resonance frequencies of the unsprung part of the vehicle change in the up-down direction and the front-rear direction due to changes in tire air pressure.

FIG. 4 is an explanatory diagram showing the principle of tire pressure detection according to the first embodiment.

FIG. 5 is a waveform diagram showing an output voltage waveform of a wheel speed sensor.

FIG. 6 is a waveform diagram showing a variation state of wheel speed v calculated based on a detection signal of a wheel speed sensor.

7 is a characteristic diagram showing a result of performing a frequency analysis calculation on the wheel speed v having the waveform shown in FIG.

FIG. 8 is an explanatory diagram illustrating an averaging process according to the first embodiment.

FIG. 9 is a characteristic diagram showing a frequency analysis result after performing a moving average process in the first example.

FIG. 10 is a characteristic diagram showing processing contents of the electronic control unit of the first embodiment.

FIG. 11 is a characteristic diagram showing a relationship between tire pressure and resonance frequency.

FIG. 12 is a flowchart showing the processing contents of the shape estimation processing of the first embodiment.

FIG. 13 is an explanatory diagram illustrating a shape estimation process according to the first embodiment.

FIG. 14 is a flowchart showing the processing contents of the center of gravity processing of the third embodiment.

FIG. 15 is an explanatory diagram for explaining the processing of the center of gravity processing of the third embodiment.

[Explanation of symbols]

1 tire 2 gears 3 pickup coils 4 Electronic control unit (ECU) 5 Display

Continued front page    (72) Inventor Masahiko Kamiya               1-1, Showa-cho, Kariya city, Aichi Japan               Denso Co., Ltd. (72) Inventor Kenji Fujiwara               1-1, Showa-cho, Kariya city, Aichi Japan               Denso Co., Ltd.                (56) References JP-A-59-26029 (JP, A)                 JP 62-149503 (JP, A)                 "Automotive Technology Handbook" Volume 1                 Basics / Theory, 1st edition, incorporated association               Automotive Engineering Society December 1, 1990 p264-               288

Claims (3)

(57) [Claims] 1. Rotational speed of wheels is detected when the vehicle is running.
A tire speed sensor, and a tire by frequency-analyzing the output signal of the wheel speed sensor.
Based on the gain distribution of the extracted predetermined frequency region, the gain distribution calculation unit that calculates the gain distribution of the frequency including the vibration frequency component, the extraction unit that extracts the gain distribution of the predetermined frequency region from the gain distribution calculation unit. A tire air pressure detection device comprising: a resonance frequency calculation means for calculating an unsprung resonance frequency of a vehicle; and a detection means for detecting an air pressure state of the tire based on the resonance frequency. 2. The extracting means is arranged in the gain distribution.
Peak gain calculating means for calculating the peak gain
The specified frequency range based on the peak gain
And a setting means for setting as a wave number region.
The tire pressure detection device according to claim 1. 3. The resonance frequency calculation means is configured to provide the predetermined frequency.
When there are multiple gain peak values in the wavenumber domain
Is a contract characterized by prohibiting the calculation of the resonance frequency.
The tire air pressure detection device according to claim 2.