JP2928675B2 - Electronic suspension - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronically controlled suspension for improving ride comfort and durability of a suspension system.

[0002]

2. Description of the Related Art Some suspension systems for vehicles use not a metal spring having a fixed spring constant but an air spring whose damping force and spring constant are variable. The microcomputer controls spring characteristics according to running conditions. Such suspensions are commonly referred to as electronically controlled suspensions. This type of electronically controlled suspension can switch the spring constant by changing the air flow rate to the sub air chamber, and change the damping force by changing the flow rate of oil passing through the orifice inside the shock absorber. This makes it possible to obtain soft riding comfort and excellent shock absorbing performance with high-frequency vibration isolation characteristics. An actual spring constant switching mechanism has a configuration in which two large and small air passages are provided in an upper support portion in an air chamber, and the air passages are switched by an air valve driven by an actuator to switch the spring constant in three stages. The damping force switching mechanism is controlled by an actuator.
Three orifices are formed in a rotary valve integrated with the rod, and the orifice is opened and closed by rotating the rotary valve, and the damping force is switched between three stages by changing the flow rate of the passing oil. ing. For this reason, the actuator for driving the rotary valve of the damping force switching mechanism and the air valve of the spring constant switching mechanism is designed to be able to follow frequently changing traveling conditions.
A servo motor with a small starting current and a repetitive operation is used as a power source, and a pinion gear directly connected to the servo motor rotationally drives a rotary valve driving gear coaxial therewith via a sector gear. sometimes,
The air valve driving gear that meshes with the rotary valve driving gear rotates and the actuator operates.

A conventional electronic control suspension has a configuration in which an energization command is issued to a servomotor at the rise and fall of an output of a vibration sensor for detecting vibration from a road surface. Since the energizing command is issued twice, the response of the actuator may be too sensitive, the ride comfort may be impaired because it responds too faithfully to small vibrations, and the service life of the servo motor may be reduced. In view of the above, there is a problem that frequent operation is not preferable. Therefore, when the vibration sensor detects vibration, the output of the vibration sensor is silently killed until a certain time has elapsed thereafter, and by operating the actuator only several times out of the vibrations received several times, Electronically controlled suspensions have been introduced to improve ride comfort and the durability of servo motors.

In a conventional electronic control suspension 1 shown in FIG. 4, a vibration from a road surface detected by a vibration sensor 2 is shaped by a waveform shaping circuit 3, and a vibration peak portion exceeding a certain level is first shaped into a rectangular wave. I do. Next, the shaped vibration waveform is sent to the waveform widening circuit 4, where the trailing edge of the vibration waveform is extended by a certain vibration prediction period τo. Is regarded as one vibration. Therefore, as shown in FIGS. 5A to 5D, among the outputs of the waveform shaping circuit 3, two vibrations that are close to each other at an interval shorter than the vibration prediction period τo are regarded as one vibration. Since the servo motor (not shown) in the suspension device 5 operates by regarding the output of the waveform widening circuit 4 as the output of the vibration sensor 2, the number of times the servo motor in the suspension device 5 is energized is reduced. The ride comfort can be improved as much as the responsive response operation is suppressed.

[0005]

The conventional electronically controlled suspension 1 can reduce the number of operations of the suspension device 5 to extend the durable life and improve the riding comfort. For example, if the vibration prediction period τo is set relatively short, when the average number of vibrations per unit time is small and the vibration cycle is long, the suspension device 5 is set to a frequency close to the actual number of vibrations. If the vibration prediction period τo is set relatively long, the average vibration frequency per unit time is small and the vibration cycle is short. However, the suspension device 5 hardly responds to small vibrations, so that the cushioning effect is weakened and the ride comfort may be degraded. Since the vibration prediction period τo is fixed, Certain of ride comfort under loose running situation was having a problem that can not be maintained.

[0006]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and comprises a vibration sensor for detecting a vibration received by a traveling vehicle from a road surface, and an average vibration per unit time based on an output of the vibration sensor. A vibration measurement circuit for measuring the number of times, a calculation circuit for dividing a fixed value by the average number of vibrations output by the vibration measurement circuit, and calculating a period in which vibration continuously occurs with a certain probability as a vibration prediction period; A waveform widening circuit that extends the output trailing edge of the vibration sensor until the vibration prediction period calculated by the circuit elapses, and a suspension device that performs a buffering operation by regarding the output of the waveform widening circuit as vibration from the road surface. The first feature is that, when the continuous occurrence probability of vibration is p as the fixed value, -ln (1
A second feature is to use a constant derived as -p).

[0007]

According to the present invention, the average number of vibrations per unit time is counted from the output of a vibration sensor, a fixed value is divided by the obtained average number of vibrations, and a period during which vibrations continuously occur with a certain probability is a vibration prediction period. As the calculated vibration prediction period elapses, the output trailing edge of the vibration sensor is extended, and the widened vibration waveform is regarded as vibration from the road surface and subjected to a buffering operation, so that the ride comfort and the suspension device are calculated. To improve the durability life. Further, assuming that the continuous occurrence probability of vibration is p as a fixed value, −l
By using a constant derived as n (1-p), a vibration prediction period supported by probability theory is adopted, and optimal buffer control is realized.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. FIG. 1 is a circuit block diagram showing one embodiment of the electronic control suspension of the present invention, FIG. 2 is a circuit diagram of the waveform widening circuit shown in FIG. 1, and FIG. 3 is a signal waveform of each part of the circuit shown in FIG. FIG.

The electronically controlled suspension 11 shown in FIG.
Expands the output waveform of the vibration sensor 2 in inverse proportion to the average number of vibrations per unit time, and regards the widened vibration waveform as the output of the vibration sensor 2 and a servo motor (not shown) in the suspension device 5. Is operated. Here, the output of the vibration sensor 2 is made to widen by a vibration prediction period τ where vibration is expected to occur continuously with a certain probability, and the vibration prediction period τ is a fixed value having an average number of vibrations λ per unit time. Is obtained by dividing. As the fixed value, a constant derived from traffic theory as -ln (1-p) is used, where p is the continuous occurrence probability of vibration. In the waveform shaping circuit 3, a vibration measuring circuit 12 for measuring the average number of vibrations λ and an arithmetic circuit 13 for calculating a vibration prediction period τ are connected in cascade. In order to measure the average number of vibrations λ per unit time, the vibration measurement circuit 12 counts clocks having a constant period between adjacent vibrations, and calculates the average number of vibrations λ as the reciprocal of the vibration period determined from the counting result. . Arithmetic circuit 13
Is obtained by dividing the above-mentioned fixed value −ln (1-p) by the average number of vibrations λ, and −ln (1-p) obtained by the division operation.
p) / λ is sent to the waveform widening circuit 14 as the vibration prediction period τ.

The waveform widening circuit 14 is triggered by the falling edge of the output of the waveform shaping circuit 3 as shown in FIG.
It comprises a monostable multivibrator 15 that outputs a pulse that lasts for a time τ according to the output of the arithmetic circuit 13, and an OR gate circuit 16 that performs an OR operation on the output of the monostable multivibrator 15 and the output of the waveform shaping circuit 3. Therefore, a pulse obtained by extending the trailing edge of the output of the waveform shaping circuit 3 by the vibration prediction period τ is obtained from the OR gate circuit 16. The monostable multivibrator 15 is a retriggerable type that can be retriggered. Further, the suspension device 5 receiving the output of the waveform widening circuit 14 issues an energization command to the servomotor at the rise and fall of the widened vibration waveform.

By the way, in the embodiment, a 90% probability (p = 0.
9) The time interval τ at which the vibration occurs one or more times is used again, and τ determined according to the arithmetic expression of ln10 / λ is the vibration prediction period.

Hereinafter, the probability Vk (τ) that the vibration will occur again k times or more within the time τ after the vibration ends will be derived based on the traffic theory. However, it is assumed here that the following three conditions are adopted as a premise of applying the traffic theory. 1) The probability of occurrence of vibration within an arbitrary short time Δτ is proportional to Δτ. 2) The probability of occurrence of vibration twice or more within an arbitrary minute time Δτ is a high-order minute amount equal to or more than the square of (Δτ). 3) Each vibration is independent of each other.

Here, an event in which the vibration ends at time to is denoted by A, its probability is denoted by P (A), an event in which the vibration occurs k times in time τ is denoted by B, and its probability is denoted by P (B). Then, the conditional probability P (B | A) that the vibration will occur again within the time τ after the vibration ends is P (B | A) = P (B, A) / P (A). Here, P (B, A) is the probability that events A and B occur at the same time, that is, the simultaneous probability P (B, A). From the assumption that vibrations and shocks are independent of each other, P (B, A) = P (B) · P (A) holds. Therefore, P (B | A) = P (B) · P (A) / P (A) = P (B). That is, the calculated probability P (B | A) is the probability that vibrations will occur k times within τ seconds from an arbitrary point in time, and is expressed as Vk (τ).

P (B) = Vk (τ) Here, the probability Vk (τ + Δτ) that the vibration occurs k times in τ + Δτ seconds from an arbitrary time point is r (r = 0, r = 0,
1,. . k), and then add the probability of the occurrence of kr vibrations at time Δτ for r. That is, Vk (τ + Δτ) = V0 (τ) · Vk (Δτ) + V1 (τ) · Vk -1 (Δτ) + V2 (τ) · Vk−2 (Δτ)... + Vk-1 (τ) · V1 (Δτ) + Vk (τ) · V0 (Δτ)

[0015]

(Equation 1)

Here, according to Assumption 2, the probability Vj (Δτ) in which j exceeds 2 is a high-order minute amount equal to or larger than the square of Δτ, and the sum of these minute amounts is calculated as

[0017]

(Equation 2)

[0018] By expressing

[0019]

(Equation 3)

Further, from Assumption 1, V1 (Δτ) = λΔτ. However, as described above, λ is the average number of vibrations generated per unit time, and is data obtained when the vehicle equipped with the electronic control suspension 11 actually travels. Fluctuate.

By the way, the probability V0 (Δτ) that no vibration shock occurs once after Δτ time can be obtained by subtracting from 1 the probability that vibration shock occurs one or more times after Δτ time.

[0022]

(Equation 4)

And eventually

[0024]

(Equation 5)

## EQU1 ## Therefore,

[0026]

(Equation 6)

When this is transformed and arranged,

[0028]

(Equation 7)

Then, when Δτ approaches 0 as much as possible,
That is, if Δτ → 0, dVk (τ) / dτ = λ {Vk-1 (τ) -Vk (τ)} where V-1 (τ) = 0, V0 (0) = 1, Vk
(0) = 0.

By solving the obtained recurrence formula, Vk
(Τ) can be obtained. Therefore, the following probability generating function U (τ, z) is introduced.

[0031]

(Equation 8)

First, both sides of the above equation are differentiated by τ, and the converging series is transformed using the fact that the order of operations can be exchanged.

[0033]

(Equation 9)

Here,

[0035]

(Equation 10)

Therefore, dU (τ, z) / dτ = λ (z−1) U (τ, z) This differential equation is expressed as an initial value U (0, z), that is,

[0037]

[Equation 11]

Solving with

[0039]

(Equation 12)

Here, the definition formula of the probability generating function

[0041]

(Equation 13)

Comparing with the coefficient of z,

[0043]

[Equation 14]

Is obtained.

That is, there is a probability that the vibration occurs again k times within τ time after the vibration.

On the other hand, after the end of the oscillation,
The probability that it will never occur is

[0047]

(Equation 15)

Therefore, the probability that the vibration occurs once or more again during the time τ after the vibration ends is

[0049]

(Equation 16)

Is represented by

Then, the time interval τ at which the vibration occurs once or more with a probability of 90% is

[0052]

[Equation 17]

From this, it can be said that at least one vibration occurs with a probability of 90% within τ = ln10 / λ. Therefore, when the average number of vibrations λ per unit time is given, by operating the suspension device 5 in units of ln10 / λ, the durable life of the servomotor can be more than doubled. is there.

The vibration prediction period τ widened by the waveform widening circuit 14 is the average number of vibrations λ per unit time.
3A to 3E, when the average number of vibrations λ per unit time is large (the first half of the time axis in FIG. 3), as shown in FIGS. Is shortened, and it is possible to quickly respond to changes in the road surface as many times as possible. Conversely, when the average number of vibrations λ per unit time is small (the latter half of the time axis in FIG. 3), The vibration prediction period τ also becomes longer, and by operating the suspension device 5 relatively slowly as long as the ride comfort is not impaired, an effective life extension of the durability life can be achieved. Can be reliably improved.

In the above embodiment, when determining the vibration prediction period τ, the time interval τ at which the vibration / impact occurs once or more with a probability of 90% is obtained. Can be obtained as a time interval τ at which one or more occurrences occur, in which case,

[0056]

(Equation 18)

Therefore, the vibration prediction period may be set as τ = −ln (1-p) / λ.

[0058]

As described above, according to the present invention, the average number of vibrations per unit time is counted from the output of the vibration sensor, the fixed value is divided by the obtained average number of vibrations, and the vibration is generated with a certain probability. The continuous occurrence period is calculated as the vibration prediction period, and the output trailing edge of the vibration sensor is extended until the calculated vibration prediction period elapses, and the widened vibration waveform is regarded as vibration from the road surface to perform the buffer operation. With this configuration, when the vibration sensor detects one vibration, a plurality of vibrations generated during a vibration prediction period following the end of the vibration are included in the first vibration to be one vibration. Unlike the method in which the suspension is operated for each vibration, it is possible to improve ride comfort against small vibrations, and at the same time, the durability of the suspension is reduced by the number of times of operation of the suspension. The vibration prediction period can be changed in inverse proportion to the average number of vibrations per unit time, so if the average number of vibrations per unit time is large, the vibration prediction period can be shortened. However, it is possible to quickly respond to changes in the road surface as many times as possible, and conversely, if the average number of vibrations per unit time is small, the vibration prediction period is set longer, and the ride comfort is not impaired. By operating the suspension relatively slowly, the effective life of the durable life can be effectively extended, and thereby excellent effects such as both the riding comfort and the durable life of the shock absorber can be reliably improved. To play.

Further, according to the present invention, assuming that the continuous occurrence probability of vibration is p as a fixed value,
Since a constant derived as ln (1-p) is used, for example, vibrations that continuously occur with a probability of 90% (p = 0.9) are collectively generated within one period within ln10 / λ. The vibration can be captured as a vibration, and thereby, the vibration prediction period supported by the probability theory is adopted, and effects such as an optimum buffer control can be realized.

[Brief description of the drawings]

FIG. 1 is a circuit block diagram showing one embodiment of an electronic control suspension of the present invention.

FIG. 2 is a circuit diagram of the waveform widening circuit shown in FIG.

FIG. 3 is a signal waveform diagram of each part of the circuit shown in FIG.

FIG. 4 is a circuit block diagram showing an example of a conventional electronic control suspension.

5 is a signal waveform diagram of each section of the circuit shown in FIG.

[Explanation of symbols]

 2 Vibration sensor 5 Suspension device 11 Electronic control suspension 12 Vibration measurement circuit 13 Operation circuit 14 Waveform widening circuit

Claims (2)

(57) [Claims] 1. A vibration sensor for detecting a vibration received by a traveling vehicle from a road surface, a vibration measurement circuit for measuring an average number of vibrations per unit time from an output of the vibration sensor, and the vibration measurement circuit outputting the vibration measurement circuit. An arithmetic circuit that divides a fixed value by the average number of vibrations and calculates a period during which vibrations continuously occur with a certain probability as a vibration prediction period, and an output of the vibration sensor until the vibration prediction period calculated by the calculation circuit elapses. An electronic control suspension comprising: a waveform widening circuit that extends a trailing edge; and a suspension device that performs a buffering operation by regarding an output of the waveform widening circuit as vibration from a road surface. 2. The method according to claim 1, wherein the fixed value is a continuous occurrence probability of the vibration.
2. The electronically controlled suspension according to claim 1, wherein the constant is derived as -ln (1-p) from traffic theory.