WO1991000189A1

WO1991000189A1 - Chassis control with servo-adjustment circuit

Info

Publication number: WO1991000189A1
Application number: PCT/DE1990/000401
Authority: WO
Inventors: Roland Kasper; Michael Panther; Karl Jaeker
Original assignee: Robert Bosch Gmbh
Priority date: 1989-06-29
Filing date: 1990-05-30
Publication date: 1991-01-10
Also published as: KR920700944A; JPH04500491A; DE3928993A1; EP0434787A1

Abstract

The invention concerns a device for regulating the chassis of vehicles, in particular passenger cars and utility vehicles. The device has sensors which determine the vehicle running status, the data from the sensors being fed to a control unit which controls a semi-active shock absorber working in conjunction with a spring unit. In order to be able to make use of existing control unit design and optimization procedures, a compensation circuit is proposed which is located between the control unit and the shock absorber (2) and the purpose of which is to adapt the non-linear control characteristics of the shock absorber (2) to the linear control voltage of the control unit.

Description

Chassis control with actuator adjustment circuit

State of the art

The invention relates to a device and a method for chassis control of vehicles, in particular passenger and commercial vehicles, with sensors that detect the driving state and whose data are fed to a controller that controls a semi-active damper that is connected in parallel to a spring arrangement.

The aim of a chassis control of vehicles is to carry out a chassis adjustment in such a way that both a high level of comfort and a high level of driving safety are provided. Comfort and driving safety are opposed to each other because comfort requires a soft damper characteristic of the damper provided on the chassis, which is connected in parallel with a spring arrangement, and on the other hand a tight chassis tuning, ie a hard damper characteristics, must be used for high driving safety.

It is known to detect the driving state of the vehicle using suitable sensors, the signals of which are fed to a controller. The controller controls the dampers of the undercarriage in such a way that a high level of driving comfort is available in non-critical situations (soft damper characteristics) and that in critical situations the suspension is appropriately taut by setting a hard damper characteristic.

There are three different types of dampers currently used. The so-called passive damper has a piston-cylinder arrangement, the piston dividing the cylinder space into two cylinder chambers which are filled with a flow medium. A displacement movement of the piston is braked by the medium which is displaced by it and passes through a constant flow cross section. With the so-called semi-active dampers, which are constructed similarly to the passive dampers, it is also possible to vary the flow cross-section of the communicating cylinder chambers. In this respect, the damper force can be adjusted. The setting can be from a soft to a hard damper characteristic. Finally, so-called active dampers are known in which the two cylinder chambers can be actively pressurized. This requires a significantly higher energy expenditure than the semi-active dampers, since a corresponding pressure build-up must be carried out. With semi-active dampers on the other hand, only the energy for adjusting the flow cross section has to be applied.

The well-known controller design and optimization methods are used to develop and design a chassis control. However, these known methods only apply to linear systems. If active actuators (dampers) are used in the chassis control, a high, constant system pressure is generally generated so that the pressure medium flow is linearly dependent on the control voltage supplied by the controller, since the metering valves for the pressure medium have a linear relationship between the control voltage and have their opening cross-section. If semi-active dampers are used, there is no system pressure. In addition to the controlled cross-section of the throttle, the pressure medium flow also depends on the differential pressure between the two cylinder chambers of the damper. Overall, this results in a nonlinear system component, so that the controller design and optimization methods that can only be used for linear systems cannot be used.

Advantages of the invention

The device according to the invention with the features mentioned in the main claim has the advantage that the known controller design and optimization methods can also be used to design the chassis control when using semi-active dampers. This is possible due to a compensation circuit between the controller and the damper is arranged and which adapts the non-linear control characteristic of the damper to the linear control voltage of the regulator. Thus, the design and layout can be carried out like an active system; in addition, only the compensation circuit according to the invention is necessary in the semi-active system.

As already shown, the damper has two cylinder chambers which are separated from one another by a damper piston and which communicate with one another via a throttle cross section which can be controlled by an actuating voltage of the compensation circuit and which are filled with a flow medium relationship

sets. Here V _q (u) denotes the total flow coefficient, p ₁ the pressure of the flow medium in one, first cylinder chamber, p ₂ the pressure of the flow medium in the other, second cylinder chamber, A ₂ the piston area with respect to the second cylinder chamber and Δ F _d the change in damper force . The damper force change Δ F _d results from the relationship

ΔF _d ≈ A ₂ · Δpp; Δp = p ₂ - p ₁ According to a development of the invention, it is provided that the total flow _{coefficient is} composed of the sum of a leakage current V _{q0 independent of the control} and a flow _coefficient V _q1 (u), so that for the volume flow q results:

To eliminate the non-linearity of the first summand of equation [3], an approach is preferably chosen such that the controllable flow of the pressure medium is proportional to the amount of a variable U ₁ to be defined.

The variable U ₁ is preferably the control voltage of the regulator.

According to a preferred embodiment of the invention, the approach is as follows:

in which

is the maximum volume flow.

The compensation to be carried out by the compensation circuit according to equations [4] results

The maximum volume flow can be determined from equation [5] in _that the maximum damper force for the damper force change Δ F _d

and the maximum values "1" are set for the control variables u and U ₁ (slide position u and control voltage U ₁ ).

If the implementation is carried out by a computer, scaling of the equations is provided for the nonlinear compensation, so that the computing range is not exceeded.

drawing

The invention is explained in more detail below with reference to the figures. Show it:

FIG. 1 shows a two-mass unicycle model of a vehicle provided with semi-active chassis control,

FIG. 2 shows a block diagram of the control scheme,

3 shows a model of a semi-active damper and

Figure 4 is a diagram showing the non-linearity of the

Represents control behavior of a semi-active damper.

According to FIG. 1, each wheel area of a vehicle provided with chassis control can be modeled as a two-mass unicycle model. The wheel mass m _r is coupled to the body mass m _{a of} the vehicle, which is proportional to the wheel area, via a spring arrangement 1 with the spring constant c _a . Parallel to the spring arrangement 1 is a semi-active damper 2, the Damping force F _d _is adjustable via an actuation voltage V _d . The elastic parts of the wheel are captured by a spring constant c _r . The wheel runs on route 3, the bumps of which are marked with S. These bumps S are related to an inertial system 4. The cycle path is designated x _r with respect to the inertial system 4; the relative cycle path x _rr lies between the cycle path 3 and the wheel mass m _r . Furthermore, the relative spring deflection x _{ar is} formed between the wheel dimensions m _r and the body mass m _{a of} the vehicle. The construction mass m _a is at a distance x _a (construction path) from the inertial system 4.

The dynamic driving conditions of the vehicle can be determined by sensors which are arranged both on the vehicle body and on the wheel support part. FIG. 2 shows - as a block diagram - the sensors 5 which are connected to a controller 6 which is connected to the semi-active damper 2 via a compensation circuit 7 according to the invention. For example, the sensor damping force F _d , relative deflection x _ar , wheel acceleration Ẍ _r and body acceleration ẍ _a can be used as input variables for the controller 6. The conversion of path variables into speed variables or acceleration variables and vice versa takes place in a known manner by means of differentiation or integration.

FIG. 3 shows the model of the semi-active damper 2. This has a cylinder 8, which is divided into two cylinder chambers 10 and 11 by the piston 9. A gas volume chamber 13 is divided above the cylinder chamber 10 by a separating piston 12. This contains the gas volume V, which stands for the volume compensation under a pressure P _gas . The damper 2 is connected to the body of the vehicle via a rubber bearing 14 provided with a spring constant c _g . The rubber bearing 14 can run the path x.

The upper, first cylinder chamber 10 has the pressure p _{1 of} a flow medium (e.g. oil). The first cylinder chamber 10 is also assigned the hydraulic capacity C ₁ and the piston cross section A _{1 of} the piston 9. The lower, second cylinder chamber 11 has the pressure p ₂ and the hydraulic capacity C ₂ . The piston surface with the piston cross section A _{2 is} assigned to it.

The piston 9 has a connecting channel 16 provided with a throttle 15. The throttle opening cross section can be controlled by applying an actuating voltage V _d , by means of which the volume flow q of the flow medium passing through the throttle cross section can be changed. A position-controlled magnetic actuation system, which can be described by a second-order system, is preferably used for the adjustment of the throttle 15 (cf. the function framed in FIG. 3). The slide position u of the throttle 15 can be adjusted via the magnetic actuating system by means of the actuating voltage V _d . The total flow coefficient of the throttle 15 is not linearly dependent on the slide position u.

This non-linearity is evident from FIG. 4. It leads to the fact that the known controller design and optimization methods for the control loop of the driving factory regulations cannot be applied, as they only apply to linear systems. According to the invention, it is therefore provided that the nonlinearity shown is eliminated by the compensation circuit 7, so that the known design and optimization methods can also be used for the semi-active damper 2 used here.

The relationship for the volume flow q results from the model of the damper 2 described

where Δ F _{d is} the damper force change. The relationship applies to these

Δ F _d ≈ A ₂ · Δp;

Δp = P ₂ - P ₁

It can be seen from FIG. 4 that a leakage flow of the flow medium can be assumed since the characteristic curve does not go through the zero point of the coordinate system. This leakage current V _q0 is independent of the control (slide position u). For the total flow coefficient V _q (u), an approach is now taken in which this fact is taken into account. It applies

V _q (u) = V _q0 ÷ V _q1 (u).

Now putting this expression in equation [2] gives the relationship

V _{q0 represents} the leakage current _coefficient and V _q1 the flow _coefficient .

The known controller design and optimization methods already mentioned require a linear system model to determine the controller coefficients of the multivariable controller (controller 6). Overall, it can be seen that the throttle equation [3] has a substantial non-linearity. In order to eliminate this, a nonlinear compensation is carried out according to the invention, so that a linear-like relationship is obtained for the first summand in equation [3]

The controllable flow should accordingly be proportional to the amount of a variable U ₁ which has not yet been defined. This variable is the control voltage U ₁ (output voltage of the regulator 6). The direction of flow of the flow medium is determined by the direction of the damper force change Δ F _d . In the above approach, it is taken into account that when using a semi-active damper, the direction of flow is always in the direction of the momentary damper force change Δ F _d . The type of nonlinear compensation can now be determined by solving equation [4] for V _q1 (u). Then it applies

The elimination of the non-linearity becomes clear when comparing equation [3] with equation [5], since the root expression is once in the numerator and once in the denominator, so that it is omitted.

The maximum volume flow

can be derived from equation [5] by taking the maximum damper force for the damper force change ΔF _d

and uses its maximum values "1" for the control variables u or U ₁ . This leads to the equation

If a computer is used to implement the nonlinear compensation, the equation [5] must be scaled so that the calculation range is not exceeded. The procedure is as follows: Equation [6] is replaced by equation [5] and the damper force change Δ F _{d is} replaced by the ska gated relationship Δ F _dm · F _d . The whole equation is then related to the maximum flow V _q1 (1). This leads to the relationship

If we now solve the scaled equation [7] according to the size u, we get

The quantities for the leakage current _coefficient V _q0 , the flow _coefficient V _q1 , and the maximum medium flow

are known as parameters of the damper 2 used. The following values result for a damper used in the test:

V _g0 = 5.93 e ^-8 [m ⁴ / s /

V _q1 = 6.46 e ^-7 [m ⁴ / s /

= 1.50 e ^-3 [m ³ / s].

The following table results for the scaled inverse flow line:

V ' _q1 (U) u (pressure) u (tension)

[P ₂ -P ₁ <0] [p2 - p1> 0]

[1] [1] [1]

0. 00 0.0000 0.0000

0. 05 0.1541 0.0882

0. 10 0.2643 0.1830

0. 15 0.3680 0.3003

0. 20 0.4772 0.4047 0.25 0.5492 0.5010

0.30 0.6307 0.6197

0.35 0.7407 0.7081

0.40 0.7911 0.7833

0.45 0.8365 0.8276

0. 50 0.8672 0.8582

0.55 0.8854 0.8789

0.60 0.8971 0.8936

0.65 0.9080 0.9062

0.70 0 .9200 0.9188

0.75 0.99325 0.9316

0.80 0.9450 0.9447

0.85 0 .9569 0.9581

0.90 0 .9682 0.9719

0.95 0 .9789 0.9859

1.00 0 .9894 1.0000

Equations [3] and [4] apply to the volume flow q of the semi-active system. If an active damper is used instead of the semi-active damper, the following linear relationship should apply to the volume flow q

It can be seen that the direction of the volume flow q is completely independent of the damping force. Energy can thus be withdrawn or supplied to the active system. If the signs of U _{1 and} ΔFd agree with each other, there is an identity between equations [3] and [9]. This means that the active system for which the known controller design and optimization methods are valid - in this operating range - can be fully replaced by the semi-active system. In the other operating range, the two equations [3] and [9] cannot be matched. In this case, the throttle can either be fully closed or fully opened, so that as little power as possible is implemented in the damper and thus the smallest possible deviation from the active system is ensured.

Claims

Expectations

1.Device for chassis control of vehicles, in particular of passenger and commercial vehicles, with sensors which detect the driving state and whose data are fed to a controller which controls a semi-active damper interacting with a spring arrangement, characterized by a between the controller (6) and the damper (2) switched, the adaptation of the non-linear control characteristic of the damper (2) to the linear control voltage of the controller (6) serving compensation circuit (7).

2. Device according to claim 1, characterized in that the damper (2) two, separated from each other by a damper piston (9), communicating with one another via a actuating voltage (V _d ) of the compensation circuit (7) controllable throttle opening cross-section, filled with a flow medium, cylinder chambers (10, 11) and that the volume flow q of the flow medium depending on the throttle cross-section dependent on the slide position (u) depends on the relationship

sets, where V _q (u) is the total flow coefficient, p ₁ the pressure of the flow medium in one, first cylinder chamber (10), p ₂ the pressure of the flow medium in the other, second cylinder chamber (11), A ₂ the piston surface with respect to the second cylinder chamber (11) and Δ F _{d is} the change in damper force (Δ F _d ≈ A ₂ · Δ p; Δ p = p ₂ - p ₁ ).

3. Device according to one of the preceding claims, characterized in that the total flow coefficient (V _q (u)) is composed of the sum of a leakage current _independent of control (V _q0 ) and a flow _coefficient (V _q1 ), so that for the volume flow ( q) results in:

4. Device according to one of the preceding claims, characterized in that in order to eliminate the non-linearity of the first summand of the equation [3], an approach is chosen such that the controllable flow of the pressure medium is proportional to the amount of a variable to be defined (U ₁ ).

5. Device according to one of the preceding claims, characterized in that the size (U ₁ ) is the control voltage of the regulator (6).

6. Device according to one of the preceding claims, d a d u r c h g e k e n n z e i c h n e t that the approach is:

in which

is the maximum volume flow.

7. Device according to one of the preceding claims, d a d u r c h g e k e n n z e i c h n e t that the compensation to be performed by the compensation circuit (7) according to equation [4] results

8. Device according to one of the preceding claims, characterized in that the maximum volume flow

thereby determining from equations [5] _that the maximum damper force for the damper force change (Δ F _d )

and the maximum values "1" are used for the control variables (u and U ₁ ) (slide position u and control voltage U ₁ ).

9. Device according to one of the preceding claims, d a d u r c h g e k e n n z e i c h n e t that a scaling of the equations is carried out for the implementation of the non-linear compensation in a computer.