CN1950222A - Model free semi-active vehicle suspension system - Google Patents
Model free semi-active vehicle suspension system Download PDFInfo
- Publication number
- CN1950222A CN1950222A CN 200580013516 CN200580013516A CN1950222A CN 1950222 A CN1950222 A CN 1950222A CN 200580013516 CN200580013516 CN 200580013516 CN 200580013516 A CN200580013516 A CN 200580013516A CN 1950222 A CN1950222 A CN 1950222A
- Authority
- CN
- China
- Prior art keywords
- control
- damper
- shock absorber
- controller
- semi
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Vehicle Body Suspensions (AREA)
Abstract
A method for controlling a shock absorber system of a vehicle that includes a plurality of controlled shock absorbers (20) linearizes the system. The method includes transforming original control inputs of the shock absorbers into virtual damper force input signals based on a bilinear damper characteristic. The system dynamics are then decoupled into modal components using static decoupling matrices, and the system is controlled with a linear decentralized controller.
Description
Related application
The application requires the preceence of No. the 60/565th, 512, the U.S. Provisional Application submitted on April 27th, 2004, and its specification sheets is combined in herein as a reference.
Technical field
One embodiment of the present of invention are pointed to vehicle suspension system.More specifically, one embodiment of the present of invention are pointed to model free semi-active vehicle suspension system.
Background technology
The comfort level of passenger vehicle or any other vehicle and treatment of road surfaces performance mainly are by the decision of the damping force characteristics of the shock absorber on the vehicle.Passive shock absorber has the fixing damping force characteristics by their design decision.Yet, depend on road excitation, wish to adjust this characteristic to improve performance.Half initiatively and active suspension system provide along with surface evenness, by, for example, change the constraint of valve of one or two current control or the viscosity by the change magnetic flow liquid, change the possibility of damping force characteristics.
Active damper has the additional advantage that negative damping can be provided and can produce wider power when low speed, therefore may allow to improve system performance.Yet semi-active suspension compares simpler, more reliable with active suspension and commercial applicability is higher.They do not require external power supply (for example, Hydraulic Pump) and safer, because therefore their consumed energies only can not cause system's instability.
Existing several uses active or semi-active suspension come the linearity and the nonlinear method of Control of Automobile.As known linear method, they use the Linear Control strategy based on linear physics car model with as the shock absorber of ideal force sources modeling usually, and wherein linear physics car model is made up of lumped mass, Hookean spring and damper.Yet real car dynamics is more complex, and active damper is not ideal force sources, but has complicated non-linear dynamic characteristic.The employed unrealistic assumption of known method makes these linear control methods not too be suitable for practical application.
Will be applied to active suspension system such as the nonlinear control method that linear dimensions changes gain scheduling, backward induction method, adaptive control.These controllers are based on nonlinear physics automobile and the damper model with quantity of parameters.The tentative identification of these model parameters is complicated problems.In addition, it is difficult using the design and the adjustment of the gamma controller of these known methods, owing to do not have actv. standard technique or Software tool, so the use of nonlinear model and controller is design very consuming time.
People's such as Lauwerys " Design and experimental validation of linear robustcontroller for an active suspension of a quarter car ", Proceeding of theAmerican Control Conference (2004) discloses a kind of robust control technique of test method use linear Identification and to(for) the practicality of 1/4 auto test bed.Based on the linear model of the experimental identification of active suspension and 1/4 car dynamics, use that μ is synthetic to obtain linear robust implementation controller.The fairly simple structure of T/S and the linearity of active suspension make that may use linearity distinguishes and the controlling Design technology.Yet the behavioral characteristics of real car is more complex, and the running of semi-active suspension and active suspension has a great difference, because for example, when vibration (rattle) speed was zero, it became uncontrollable.
Above-mentioned based on producing the optimizing controller that is used for certain shock absorber and car model on the theoretical method of model.Yet, if not realizing that the application of their car load and high non-linearity semi-active damper is complicated and very difficult.
Based on above-mentioned, there is system and method demand for the Model free control of non-linear half active or active damper.
Summary of the invention
One embodiment of the present of invention are the methods that are used to control the vibration insulating system of the vehicle that comprises a plurality of controlled oscillating dampers.By the original control input of shock absorber being transformed into the virtual damper force incoming signal, with system linearization based on bilinear damper characteristic.Use the static decoupling matrix that system dynamics is decoupled into modal components, and adopt the linear dispersed controller to come control system.
Description of drawings
Fig. 1 is the passive shock absorber of expression prior art and the block diagram of semi-active damper according to an embodiment of the invention;
The diagrammatic sketch of Fig. 2 speed-mechanical resistance nit that to be expression record for the different received currents of CVSA valve;
Fig. 3 has the automobile of gamma controller and the block diagram of damper;
Fig. 4 is that linear controller is added in expression, and substitutes the block diagram of gamma controller with the LINEARIZED CONTROL device;
Fig. 5 illustrate the dumping force that calculates based on physics and bilinear model and the car body acceleration measured in a corner of automobile between many correlativitys;
Fig. 6 has four block diagrams of the passenger vehicle of semi-active damper according to an embodiment of the invention, and is shown as and needs four to take advantage of four of four MIMO controllers to take advantage of four multi-input multi-output systems.
Fig. 7 illustrates the frequency response function matrix of measurement of the system of Fig. 6;
Fig. 8 is the block diagram of mode controller and coupled system;
Fig. 9 illustrates the frequency response function matrix of measurement of the system of Fig. 7;
Figure 10 is the block diagram of adaptive controller.
The specific embodiment
One embodiment of the present of invention are Model free control structures, this Model free control structure is not target with the optimality directly, but combining many physically explainable parameters, these parameters can be according to the test criterion that provides of pilot signal and based on the easily online adjustment of test result.This method is usually based on the physical principle of semi-active damper and automobile, but do not require the model of its behavioral characteristics.Therefore, it is applicable to the automobile or the vehicle of any half active or active suspension system and any kind.
Fig. 1 is the passive shock absorber 10 of expression prior art and the block diagram of semi-active damper according to an embodiment of the invention 20.Shock absorber 10 comprises the cylinder body 12 of filling oil and the bar 14 that is connected to piston 16, and piston 16 comprises the piston valve 15 that calibration constraint is provided.Being moved in the cylinder body 12 or outer and volume change that cause by bar 14, is that seat valve 17 flows to or the oil that flows out reservoir 19 compensates by passing.Cause acting on dumping force on the piston 16 in seat valve 17 and piston valve 15 pressure drop on the two.
In operation, when bar 24 moved up (positive vibration velocity), piston check valve 28 closures and oil flow through CVSA valve 32.Because the volume of the bar 24 in cylinder body 22 reduces, so oil is pushed into the cylinder body 22 by base check-valve 30 from reservoir 31.Rattle displacement/velocity is the relative displacement/speed of bar 24 with respect to cylinder body 22.
When bar 24 moved down (negative vibration velocity), piston check valve 28 was opened.Because the increase in volume of the bar 24 in cylinder body 22, base check-valve 30 closures, and oil flows into the reservoir 31 by CVSA valve 32 from cylinder body 22.
Fig. 2 is the different received current is of expression for CVSA valve 32
vSpeed-mechanical resistance nit of measuring.Low/high electric current of giving CVSA valve 32 and the low/high damping ratio of generation little/limit corresponding greatly.The sinusoidal vibration displacement signal of the difference setting by will being used for control current is applied to CVSA valve 32 and obtains this characteristic.
In one embodiment, for obtaining hereinafter disclosed current parameters, the passenger vehicle that is equipped with four semi-active dampers is placed on four hydraulic oscillators, these four hydraulic oscillators can encourage four wheels of automobile independently with the surface evenness of expection.Use four accelerometers to come the vehicle body acceleration of measured automobiles.In addition, use Linear displacement transducer to measure the vibration displacement of whole four shock absorbers.In this proving installation, the interference of system input is the displacement of the oscillator under the automotive wheel.The control input is the electric current that flows to four semi-active damper CVSA valves.Although in the embodiment that describes, come control cock 32, in other embodiments, can come control cock 32 by different input such as voltage with electric current.
Feedback linearization
A target of feedback linearization is (being input to the electric current of semi-active damper CVSA valve) to be imported in the original control of system be transformed into the virtual controlling input, so that the dynamic relationship linearization between the output (vehicle body acceleration of measurement) of the system that the control that these are new is imported and will be controlled.If the relation between system's input and output is linear (or enough linear), but then because successful Application is known and support computer aided control system to design the Linear Control designing technique of (" CACSD "), so can simplify controlling Design and adjustment.
In the prior art, conversion comprises physical damper model and imports corresponding to the new control of dumping force.Yet one embodiment of the present of invention are to use the alternative conversion of bilinear model, and it causes not being another new control input of dumping force, but it makes system linearityization better.
Only need to consider an angle of automobile, how to disclose embodiments of the invention in conjunction with feedback linearization and Linear Control.Shock absorber is according to vibration velocity v
rWith the dash pot valve current i
vGenerate power f
dDisturbing input is pavement displacement x
aThe output of measuring is vehicle body acceleration a
bFig. 3 has the automobile 52 of gamma controller 50 and the block diagram of damper 51.Be the vehicle body (that is, reducing vehicle body acceleration) of Control of Automobile 52, gamma controller 50 is with vehicle body acceleration a
bFeed back to damping control current i
v
The design of such gamma controller 50 and adjustment are not direct.Because semi-active damper is the most nonlinear element of system, so introduced the LINEARIZED CONTROL device, this LINEARIZED CONTROL device calculates suitable damping current i
v, make for given vibration velocity v
r, realize the dumping force f that expects
cFig. 4 is that linear controller 60 is added in expression, and LINEARIZED CONTROL device 62 substitutes the block diagram of gamma controller.Use linear controller 60 based on the vehicle body acceleration a that measures
bProduce the dumping force f of expection
c LINEARIZED CONTROL device 62 is based on reverse damping force characteristics.Consider two specific characters: the bilinear characteristics that the speed-force characteristic of the measurement of the physical damper model that the representative shown in Fig. 2 is simplified and analysis obtain.
The art methods of feedback linearization is based on the LINEARIZED CONTROL device 62 of physical model shock absorber or damper.Use this method, physical model can be simplified to all 2D lookup table as shown in Figure 2, this 2D lookup table is with dumping force f
dWith control current i
vWith vibration velocity v
rConnect.By using the 2D interpositioning to obtain reverse model, making can be from vibration velocity v then
rDumping force f with expection
cCalculate damping current i
v
Unlike the prior art, embodiments of the invention are based on the LINEARIZED CONTROL device 62 of bilinear damper characteristic.Particularly, semi-active damper is to transmit and vibration velocity v
rWith control signal i
vRelevant power f
cEquipment (equation 1).(v when bar is not mobile
r=0), because semi-active damper can not transmit any power, so can pass through with coefficient F
0And F
10Being arranged to 0 bilinearity of simplifying (equation 3) this relation (equation 2) is similar to.Equation 4 and 5 expression forward directions and reverse damping similarity relation.Approximate based on this bilinearity, the product linearly dependent of dumping force and vibration velocity and the control signal that is biased.
f
c＝F(v
r，i
v) (1)
f
c＝F
0+v
rF
01+i
vF
10+v
ri
vF
11 (2)
f
c～v
r(i
0+i
v) (4)
Because convergent-divergent is the linear operation by the linear controller compensation, so can create and vibration velocity v
rWith the control current i that is biased
v+ i
0The new input f that equates of product
cThis input no longer has the physics dimension of dumping force.Therefore be referred to as virtual damper force.It should be noted that this LINEARIZED CONTROL device only comprises a parameter i
0, the control current biasing, controller will be operated around it.
For with embodiments of the invention with test based on the performance of the prior art LINEARIZED CONTROL device of physical damper model and compare, wherein the LINEARIZED CONTROL device is based on bilinear damper characteristic in an embodiment of the present invention, testing automobile is placed on the dynamic oscillation device, and encourages with following signal:
Be applied to the uncorrelated pink noise surface evenness displacement signal of four oscillators under the automotive wheel.
Be applied to the uncorrelated white noise electric current of four CVSA valves of semi-active damper.
Measure following signal:
Four accelerometer's signals on four angles of automobile.
Four vibration displacements of four shock absorbers.
Come four (virtual) dumping forces of computed offline based on physics and bilinear damper model.
Should note, do not verify the performance of damper model by (virtual) dumping force of relatively real (measurement) and calculating, because embodiments of the invention do not need to reproduce this power, and because bilinear model generates the virtual damper force that no longer has this physical meaning because of scaled and skew.But embodiments of the invention calculate the signal higher than original control signal with the linearly dependent of vehicle body acceleration.
Fig. 5 illustrates dumping force and the many correlativitys between the car body acceleration of measuring on the angle of automobile based on physics and bilinear model calculating.The correlativity of bilinear model acquisition is higher according to an embodiment of the invention, therefore, illustrates higher than the behavioral characteristics of the linearity that obtains with the prior art physical damper model.
A problem of physical damper model is the characteristic of attempting to compensate the nonlinear electric current-speed-Li of shock absorber, and it is more complicated more than the lookup table shown in Fig. 2.The lookup table of Fig. 2 is that the particular harmonic pumping signal of the scope that is provided with fixed current generates.Yet Nonlinear system response can depend on amplitude.Therefore, model is only effective to having with the pumping signal of the similar amplitude level that is used for its evaluation.On the contrary, bilinear model does not comprise these nonlinear characteristics according to an embodiment of the invention, but it is approximate to carry out bilinearity, the approximate feasible wideband excitation signal linearization more such as the random road surface excitation of bilinearity.Bilinear damper model only comprises a parameter, " i
0", and lookup table damper model is made up of the mass data point of the nonlinear characteristic that is used to describe damping.
The model decoupling zero
Fig. 6 is the block diagram with passenger vehicle of four aforesaid semi-active dampers according to an embodiment of the invention, and is shown as and needs four to take advantage of four of four MIMO controllers 72 to take advantage of four multiinput-multioutput (" MIMO ") system 70.The control input (that is the output of controller 72) of system is the fictitious force f of semi-active damper
c dThe system outlet of measuring (that is the input of controller 72) is at automobile a
c bThe vehicle body acceleration at place, four angles.In other embodiments of the invention, the system outlet of measurement is the vehicle body acceleration at the place, three angles of automobile, rather than whole four angles, and calculates the 4th angle based on the output of three measurements.
Fig. 7 illustrates frequency response function (" the FRF ") matrix of measurement of the system of Fig. 6.Fig. 7 locates from fictitious force (f at four angles of automobile
d) to vehicle body acceleration (a
b) four take advantage of four coupled systems (fl: preceding-left side, fr: preceding-right side, rl: a back-left side, rr: the back-right side).Off diagonal element is with respect to the bigger indication close coupling of diagonal entry.
Static decoupling matrix D (equation 6) and pseudo inverse matrix D thereof
+Be used for system decoupling is become that its mode motion by 80 controls of the mode shown in Fig. 8 (diagonal angle) controller travels, wigwag motion, jolts.The conversion of system control input (output of controller 80) is to affact three virtual modal forces f on the automobile by four semi-active dampers
d dThe conversion control output (input of controller 80) of system is three mode motion a of automobile
d b
Will be by transformation matrix D and D
+Come the decoupling zero automobile of symmetry fully with δ=0.Each row of this matrix are corresponding to the position of each sensor on the automobile: preceding-left side, preceding-right, a back-left side and the back-right side.Each row is represented mode motion: travel (all homophases), wigwag motion (left and right sides is anti-phase), jolt (front and back are anti-phase).
Fig. 9 illustrates the FRF matrix of this decoupled system of measuring with δ=0.Fig. 9 is from virtual modal forces (f
d) to mode vehicle body acceleration (a
b) the three FRF matrixes of taking advantage of three decoupled systems.The obvious less indication of the relative size of off diagonal element allows a little less than the coupling of decentralised control.Based on the FRF matrix in Fig. 7 and 9, can not deduction whether be worth δ ≠ 0 and can obtain the more excellent decoupling zero that is used to control.Therefore, δ is introduced as controlled variable, this parameter can onlinely be adjusted, so that realize the symmetric car dynamics that the vertical eccentricity owing to center of gravity causes.
Control policy
As mentioned above, the linearized and decoupling zero of system now can come CONTROLLER DESIGN based on the system after the conversion now.In one embodiment, use known skyhook (skyhook) principle to come CONTROLLER DESIGN.Yet, can use any other design of Controller principle in conjunction with the system of linearization in the embodiments of the invention and decoupling zero.
Linear controller among embodiment is made up of some feedbacks and feed-forward module, and each module solves specific comfort level or operational issue.All the output summation of module is the Anticipation Mode virtual damper force.Target is to suppress the automobile mode motion to improve passenger's comfort level.Feedback linearization controller and mode decoupling zero conversion allow directly to specify the Anticipation Mode power of the next self-metering mode motion that is transmitted by shock absorber.Based on the skyhook principle, the diagonal angle module controller is made up of three low-pass first order filters, bandwidth f that can online this filter of adjustment
bAnd gain, to satisfy trading off between optimum comfortable specification of expection and the input degree of saturation specific humidity.
Wheel bounce is that the wheel of vehicle moves the static relatively mode of resonance of vehicle body maintenance simultaneously significantly with respect to road.Because the tire contact force changes greatly, so the operating characteristic of this phenomenon damage car.Can be by increasing control current biasing i
0(equation 5 that sees above) weakens wheel bounce, and wherein control current is around control current biasing i
0Change.
When driving in turning, because the centnifugal force automobile can swing, centnifugal force is directly proportional with the curvature that the sum of squares of the speed of driving is turned.By using feedforward controller to compensate this motion that swings, feedforward controller is added to the mode power of swinging the dumping force of expection, this mode swing power and the reverse movement that swings that causes by turning, and be directly proportional with driving velocity squared and the steering angle measured.
When quickening or brake, automobile can jolt respectively backward and forward, is directly proportional with the transverse acceleration that applies.By using feedforward controller to compensate balleting, feedforward controller is added to the mode pitch force dumping force of expection.The expection pitch force of this interpolation is directly proportional with the destructive force of measurement and the wheel torque of expection (both is effective in the controller local area network (" CAN ") of automobile).
Controlled variable is adjusted
In one embodiment, there is not model can be used for adjusting control device (for example, in simulation).It is very difficult and uncertain will being transformed into classical control specification (for example, bandwidth and stabilization time) as the target problem of comfort level and road surface operability.Therefore, an embodiment of the controller of exploitation is equipped with many parameters, and these parameters can be based on the suggestion that is provided by the experienced test ride personnel of sailing that drive a car on the demarcation test tracks, online adjustment respectively.Whole adjustable parameters have physical interpretation, make them very clear to the influence of the overall performance of suspension.Different controlled variable hereinafter is discussed, their physical interpretation and they influences to automotive performance.
Mode decoupling zero matrix D comprises a parameter δ who represents the center of gravity vertical misalignment.Adjust this parameter to obtain the similar balance vehicle response of front and rear behavioral characteristics.
The integral feedback controller, it is made up of three low-pass first order filters, comprises six parameters: three gains and three bandwidth.Increase these gains and bandwidth and improve the low cut of the mode motion of automobile, also begin to make certain point of high frequency noise deterioration until them.
The bilinear damper model involving vibrations speed that is used for the linearized system behavioral characteristics is used for coming the calculation control electric current based on the virtual damper force of expection.Therefore with low-pass filter the vibration displacement of measuring is carried out difference and filtering, amplify to prevent high frequency noise.The bandwidth of this filter is an important parameter, adjusts this parameter to optimize trading off between controller bandwidth and noise susceptibility.
Control current biasing i
0(equation 5 that sees above) determines the average magnitude of damping in the system, and this control current biasing i
0Mainly be adjusted to optimize the operating characteristic of automobile.Increase this value and make automobile have better tire contact force, but damaged comfort of passenger.Experimental adjustment display optimization value depends on road category: flat road surface allows soft setting, and rough riding surface needs more intense setting.
The control current biasing i that causes the optimal compromise between comfort level and road surface operating characteristic
0Depend on road-surface roughness.Target is the change of restriction tire contact force, so that keep a certain amount of operating characteristic, is not subjected to the domination of road input.Since can't these tire contact forces of on-line measurement, so the average magnitude of the kinetic energy of hypothesis wheel is relevant with the average magnitude of tire contact force variation.Because wheel bounce resonance, the absolute velocitye of wheel can be similar to vibration velocity, and the vehicle body of automobile keeps static with respect to wheel.This causes following adaptive control rule to keep constant operating characteristic: make control current biasing i
0Self adaptation makes the kinetic energy average magnitude of wheel remain unchanged.
By filter vibration velocity v with low-pass first order filter
rSquare and measurement E that (whole four wheels) come online calculating to be directly proportional with the average magnitude (mobile) of the kinetic energy of wheel.Shown in equation 7, this filter time constant τ determines the time of calculating mean value.
Road and control current biasing i for particular type
0Fixed value, the life period constant tau makes the aviation value E of kinetic wheel converge a constant (in given range).Increase i
0Cause stronger shock absorber, and E is increased.Can have control input i with being considered as
0Automobile with the system G of the output E that measures is modeled as the first-order system with timeconstant and negative DC-gain A, shown in following equation 8.Proportional feedback controller P causes having time constant
Gain with DC-
New first-order system.Figure 10 is that adaptive controller passes through with proportional regulator control current biasing i
0, the mean kinetic energy E of the wheel relevant with average tire contact force variation is remained on constant level E
0Adjustable parameters P and benchmark E
0Obtain the irrelevant operating characteristic of road of desired amount.
As mentioned above, understand, obtained controller architecture according to an embodiment of the invention based on physics under situation about obviously not using a model to automobile and semi-active suspension behavioral characteristics.In one embodiment, controller architecture is made up of three foundations.At first by initial current control input is transformed into the virtual damper force incoming signal with system linearization.Illustrate,, can improve this linearization if use the nonlinear characteristic of bilinear damper characteristic replacement based on physical model.Use the static decoupling matrix system dynamics to be decoupled into their modal components then.Can not static decoupling although actual car dynamics was both asymmetric also, the off-diagonal element of FRF matrix is compared enough little with diagonal element after static decoupling.
Use single parameter to finely tune the vertical misalignment of center of gravity, so that realize the car dynamics of symmetry.At last, control this linearization and the system of decoupling zero by linear decoupling controller, this linearity decoupling controller is made up of the module that several all solve specific comfort or operational issue.The whole parameters that obtained by this Model free control structure all have physical significance, and therefore can come online adjustment intuitively based on test ride personnel's suggestion.Because the roughness of road excitation is depended in the optimum average damping of system, so use adaptive controller to come the regulating control current offset.
For the semi-active suspension system that obtains according to embodiments of the invention, can pass through, for example, change the constraint of damper in-to-in, or, adjust the dumping force of each damper by changing the characteristic of damper in-to-in oil.Use a computer and in time calculate the expection damping level of each damper on each moment automobile.The damping level of these expections is calculated in motion, condition of road surface and the chaufeur input of embodiments of the invention by considering auto body.Different with prior art systems, the algorithm of embodiments of the invention neither use about the physical message of damper also with the physical message of using automobile itself.
This paper specifies also/or some embodiment of the present invention have been described.However, it should be understood that modification of the present invention or change to be covered by technique scheme, and within the scope of the appended claims, and do not depart from spirit of the present invention and predetermined category.
Claims (18)
1. the control method of the shock absorber system of a vehicle that comprises a plurality of controllable dampers, described method comprises:
Based on bilinear damper characteristic, be transformed into the virtual damper force incoming signal by original control input, with described system linearization with described shock absorber;
Use the static decoupling matrix that system dynamics is decoupled into modal components; With
Control described system with the linear dispersed controller.
2. the method for claim 1 wherein uses model-free to import described system linearization.
3. the method for claim 1, described modal components comprise the mode decoupling zero matrix of parameter of the vertical misalignment of the center of gravity with the described vehicle of representative.
4. the method for claim 1 is wherein by generating the control signal of each described controllable damper, i
v, control described system, wherein
5. the method for claim 1, wherein said shock absorber is a semi-active damper.
6. the method for claim 1, wherein said shock absorber is the current control shock absorber.
7. the semi-active damper of a Vehicle damper system comprises:
Cylinder body;
Be inserted into the bar in the described cylinder body;
Be coupled to the piston of described bar;
Be coupled to the controlled valve of described cylinder body; With
Be coupled to the controller that is used to control described valve of described controlled valve, described controller comprises:
Linear controller;
Be coupled to the LINEARIZED CONTROL device of described linear controller, described LINEARIZED CONTROL device is based on bilinear damper characteristic.
8. semi-active damper as claimed in claim 7, wherein said controller uses virtual damper force incoming signal, i
v, control described valve, and wherein
And v wherein
rBe the vibration velocity of described shock absorber, f
cBe the relevant power of described shock absorber, i
0It is the control current biasing.
9. semi-active damper as claimed in claim 7, described controller also comprise mode decoupling zero matrix.
10. semi-active damper as claimed in claim 9, described mode decoupling zero matrix comprises the parameter of the vertical misalignment of the center of gravity of representing described vehicle.
11. semi-active damper as claimed in claim 7, wherein said controlled valve are the current control valves.
12. semi-active damper as claimed in claim 7, wherein said controller is based on the skyhook principle.
13. a vibration insulating system that comprises the vehicle of a plurality of controllable dampers, described system comprises:
Be used for being transformed into the device of virtual damper force incoming signal by control input with described system linearization with described shock absorber based on bilinear damper characteristic;
Be used to use the static decoupling matrix system dynamics to be decoupled into the device of modal components; With
Be used to use the linear dispersed controller to control the device of described system;
14. vibration insulating system as claimed in claim 13 wherein uses model-free to import described system linearization.
15. vibration insulating system as claimed in claim 13, described modal components comprise mode decoupling zero matrix, described mode decoupling zero matrix has the parameter of the vertical misalignment of the center of gravity of representing described vehicle.
16. vibration insulating system as claimed in claim 13 is wherein by generating the control signal of each described controllable damper, i
v, control described system, wherein
And v wherein
rBe the vibration velocity of described shock absorber, f
cBe the relevant power of described shock absorber, i
0It is the control current biasing.
17. vibration insulating system as claimed in claim 13, wherein said shock absorber is a semi-active damper.
18. vibration insulating system as claimed in claim 13, wherein said shock absorber are the current control shock absorbers.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US56551204P | 2004-04-27 | 2004-04-27 | |
| US60/565,512 | 2004-04-27 | ||
| US11/111,832 | 2005-04-22 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1950222A true CN1950222A (en) | 2007-04-18 |
| CN100478201C CN100478201C (en) | 2009-04-15 |
Family
ID=38019320
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005800135169A Expired - Fee Related CN100478201C (en) | 2004-04-27 | 2005-04-26 | Model free semi-active vehicle suspension system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100478201C (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063122A (en) * | 2010-11-10 | 2011-05-18 | 哈尔滨工业大学 | Spatial six-degree-of-freedom motion platform modal control method |
| CN107065553A (en) * | 2017-04-20 | 2017-08-18 | 南开大学 | Many rotation excitations translate oscilator system Non-linear coupling self-adaptation control method |
| CN110712491A (en) * | 2019-10-16 | 2020-01-21 | 湖南大学 | Layered control method, system and medium for vehicle modal decoupling |
-
2005
- 2005-04-26 CN CNB2005800135169A patent/CN100478201C/en not_active Expired - Fee Related
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063122A (en) * | 2010-11-10 | 2011-05-18 | 哈尔滨工业大学 | Spatial six-degree-of-freedom motion platform modal control method |
| CN102063122B (en) * | 2010-11-10 | 2012-09-05 | 哈尔滨工业大学 | Spatial six-degree-of-freedom motion platform modal control method |
| CN107065553A (en) * | 2017-04-20 | 2017-08-18 | 南开大学 | Many rotation excitations translate oscilator system Non-linear coupling self-adaptation control method |
| CN107065553B (en) * | 2017-04-20 | 2019-12-03 | 南开大学 | More rotation excitations translate oscilator system Non-linear coupling self-adaptation control method |
| CN110712491A (en) * | 2019-10-16 | 2020-01-21 | 湖南大学 | Layered control method, system and medium for vehicle modal decoupling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100478201C (en) | 2009-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4926945B2 (en) | Method for controlling vehicle shock absorber system, semi-active shock absorber and shock absorber system | |
| CN110154666B (en) | Vehicle suspension system self-adaptive reverse-thrust control method capable of realizing road condition prediction | |
| Sam et al. | LQR controller for active car suspension | |
| Kruczek et al. | A full-car model for active suspension-some practical aspects | |
| CN110597063B (en) | Active suspension output feedback control method based on nonlinear extended state observer | |
| Darus et al. | Modeling and control active suspension system for a full car model | |
| US10005457B2 (en) | Driving force control device for a vehicle | |
| US8086377B2 (en) | Suspension control apparatus | |
| CN110597064B (en) | Active suspension output feedback control method based on nonlinear and uncertain models | |
| CN102582389B (en) | Vehicle suspension control method based on virtual damper | |
| Zhang et al. | Semi‐active control of air suspension with auxiliary chamber subject to parameter uncertainties and time‐delay | |
| Gordon | Non-linear optimal control of a semi-active vehicle suspension system | |
| CN103754081A (en) | Optimal fuzzy complex control method of vehicle nonlinear suspension system | |
| CN102975587A (en) | Vehicle semiactive suspension based on double controllable dampers and control method thereof | |
| CN100478201C (en) | Model free semi-active vehicle suspension system | |
| CN114590090A (en) | Direct-drive semi-active suspension control system construction method based on self-adaptive LQR (Low-speed response) wheel hub | |
| CN113467233B (en) | Time-lag finite frequency domain output feedback control method based on fuzzy model | |
| Dorling | Integrated control of road vehicle dynamics | |
| Chen et al. | Stiffness-damping matching method of an ECAS system based on LQG control | |
| Craft et al. | Fuzzy logic control algorithms for MagneShock semiactive vehicle shock absorbers: design and experimental evaluations | |
| Lauwerys et al. | Model free control design for a semi-active suspension of a passenger car | |
| Lauwerys et al. | A model free control design approach for a semi-active suspension of a passenger car | |
| Zhu | An investigation into the roll control of vehicles with hydraulically interconnected suspensions | |
| Gaikwad et al. | Overview of Active Suspension System | |
| CN116834493A (en) | Semi-active damping coordination control method for whole heavy vehicle provided with hydro-pneumatic suspension |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090415 Termination date: 20150426 |
|
| EXPY | Termination of patent right or utility model |