EP1721081B1 - Valve control for hydraulic actuators based on electrorheological liquids - Google Patents

Description

The invention relates to a method for valve control of hydraulic actuators based on electrorheological fluids according to the preamble of claims 1 and 3, and to an apparatus for carrying out the method according to the preamble of claims 7 and 8.

In pneumatics and hydraulics valves are subdivided into switching and continuous valves. Continuous valves are valves in which the output (e.g., valve spool travel, pressure, etc.) is proportional to the input signal (e.g., drive voltage). The operation can be done hand / mechanical / pressure / electric and electronic. Continuous valves are also commonly referred to as proportional, control and servo valves, due to differences in the accuracy and use of the valves.

Furthermore, there are valves based on electrorheological / magnetorheological fluids that can be assigned due to their properties to the class of continuous valves.

In the conventional hydraulic system, the control of a double-acting cylinder is generally carried out via a 4/3 servo valve, whereby the change in the volume flow in the 4/3 servo valve is effected by an electromechanical converter. The electromechanical transducer (as well as the subsequent hydraulic intermediate stages) generally represents the component limiting the dynamics of the valve.

The system, consisting of servo valve and cylinder, can be described here as a single-feed system with the input variable i _v (current through the servo valve) or with negligible dynamics of the servo valve x _v (position 4/3 servo valve) and the output variable s (position of the cylinder) become. The volume flows into the cylinder chambers q ₁ and q ₂ can not be used independently of each other as manipulated variables. Therefore The mathematical model of the cylinder loses properties that prevent the applicability of certain controller design methods.

Control theory discloses a variety of controller design methods for non-linear systems. For example, at this point the flatness-based controller design ( Joachim Rudolph: Contributions to the flatness-based follow-up control of linear and nonlinear systems of finite and infinite dimension, Shaker Verlag, 2003 ), the exact input state linearization ( Alberto Isidori: Nonlinear Control Systems, 3rd edition, Springer Verlag 2001 ) and the passivity-based controller design ( Romeo Ortega et al .: Passivity Based Control of Euler-Lagrange Systems, Springer Verlag London 1998 ) called.

When controlled via a 4/3 servo valve, the chamber pressures p ₁ and p ₂ in the cylinder are adjusted so that in the steady state the force on the piston caused by the chamber pressures is equal to the load force. By contrast, the absolute values of the pressures p ₁ and p ₂ themselves can not be specifically influenced.

Furthermore, valves based on electrorheological and / or magnetorheological fluids are known. Electrorheological valves are usually constructed of coaxial cylinder electrodes or of arrangements of parallel plates, between which the electrorheological fluid flows. By applied to the electrodes electrical voltage or by the electric field generated thereby, the effective viscosity of the electro-rheological fluid located between the electrodes and thus the flow resistance through the valve gap is controllable. In this case, with the pressure difference applied, the flow rate through the valve can be varied from complete opening (normal viscous flow) to complete blocking (solid state).

The operating principle is based on the fact that the particles of the electrorheological fluid form chains during application of an electric field, which impede the flow and thus change the flow resistance.

1. 1. Gavin, HP: Annular Poiseuille flow of ER and MR materials, Journal of Rheology, 45,4: 983-994, 2001 ;
2. 2. Parthasarathy M, Klingenberg DJ: Electrorheology: Machanisms and Models, Journal of Materials, Science and Engineering R, reports; 17,2: 57-103, 1995 ;
3. 3. Rajagopal KR, Wineman AS: Flow of electro-rheological materials, Acta Mechanica 91, 57-75, 1992 ;
4. 4. Whittle M, Atkin RJ, Bullough WA: Dynamics of an electrorheological valve, International Journal of modern Physics B, 10,23:2933-2950, 1996 ;
5. 5. Ru̇
   
   i
   
   ka M: Electrorheological Fluids: Modeling and Mathematical Theory. Lecture Notes in Mathematics, SpringerVerlag, Berlin, 2000.

Compared to conventionally controllable valves valves based on electrorheological / magnetorheological fluids are simpler, because they have no moving mechanical parts such as shut-off. Another advantage is that electrical signals can be implemented directly, so that very fast switching times can be realized with electro-rheological fluid valves and thus a much higher dynamics of the overall system, for example consisting of cylinders with electrorheological valve, is achieved. However, a further complicating factor for a subsequent controller design is that the electrorheological effect is an inherently nonlinear phenomenon. In addition, hysteresis occurs in certain operating ranges. The electrorheological effect is described in more detail, for example, in the literature listed below:

1. 1. Gavin, HP: Annular Poiseuille flow of ER and MR materials, Journal of Rheology, 45, 4: 983-994, 2001 ;
2. Second Parthasarathy M, Klingenberg DJ: Electrorheology: Machanisms and Models, Journal of Materials, Science and Engineering R, reports; 17,2: 57-103, 1995 ;
3. Third Rajagopal KR, Wineman AS: Flow of electro-rheological materials, Acta Mechanica 91, 57-75, 1992 ;
4. 4th Whittle M, Atkin RJ, Bullough WA: Dynamics of an electrorheological valve, International Journal of Modern Physics B, 10, 23: 2933-2950, 1996 ;
5. 5. Ru̇
   
   i
   
   ka M: Electrorheological Fluids: Modeling and Mathematical Theory. Lecture Notes in Mathematics, Springer Verlag, Berlin, 2000.

In one from the essay of G. Fees, "Static and Dynamic Properties of a Highly Dynamic ER Servo Drive" O + P Olhydraulik und Pneumatik, Vereinigte Fachverlage Krausskop, Mainz, DE, Vol. 45, No. 1, January 2001 (2001-01), pages 45-48 , XP000977828 ISSN: 0341-2660 , Known cylinder control of a cylinder each cylinder chamber is assigned in each case a half-bridge circuit consisting of two valves based on electrorheological and / or magnetorheological fluids. The four valves are interconnected to form a full bridge, in whose shunt branch the hydraulic actuator (cylinder) is located. The control of these valves takes place in such a way that the four valves with a common average electrical voltage u and a differential voltage .DELTA.u are driven, ie at the valve a and d is the electrical voltage u + Δ u , while at the valves b and c, the electrical voltage u -Δ u is applied. The mean voltage u is chosen so that the valve works in a middle operating point. The voltage Δ u now corresponds approximately to the position x _{v of} the 4/3 servovalve compared to conventional 4/3 servovalves. The electrorheological full bridge thus simulates the behavior of a 4/3-servo valve with an overlap, which with the help of u Both negative and positive can be set.

The disadvantage of this control and division, however, lies in the fact that not all degrees of freedom of the full bridge are used, with the consequence that this, for example, a compensation of inherent occurring due to the electrorheological effect Nonlinearities of the valves is impossible. In the above-described control of the valves based on electrorheological / magnetorheological fluids, the volume flows into the cylinder chambers q ₁ / q _{2 are} also not used independently of each other as a manipulated variable. For this reason, the mathematical model of the cylinder and the full bridge (consisting of electrorheological valves) loses properties that prevent the applicability of certain rule design methods. For example, it can be shown that the flatness-based controller design and the exact input state linearization are no longer applicable.

The disadvantages mentioned have the consequence that the achievable with electrorheological actuators control performance and dynamics have no visible advantage over classic hydraulic actuators with 4/3-servo valve.

The object of the present invention is to prevent the disadvantages mentioned above and to provide a valve control for actuators based on electrorheological fluids, which allows control and regulation operations with extremely high dynamics.

This object is achieved by the invention specified in the patent claims 1, 3, 7 and 8. Further developments and advantageous embodiments of the invention are specified in the dependent claims.

On the basis of the aforementioned control variables, which are much better suited for the controller design, the 4 electrical voltages for the control of the 4 valves are calculated. With the help of the new manipulated variables q ₁ and q _2, it is very easy to control the chamber pressures. q _{q, 1} and q _{q, 2} serve as manipulated variables for the supply pressure control and prevent a complete locking of the valves, as they ensure a minimum flow rate. q _Σ and q _Δ are used as manipulated variables for the decoupled control of the total pressure and the position or speed of the piston. Due to the independent specifications of the aforementioned manipulated variables, all linear and / or non-linear and / or adaptive single and multi-variable control methods with or without cascade structure can be used.

Since all the valves associated with the actuator are always flowed through with a minimum volume flow (q _{q, 1} or q _{q, 2} ), there are no undesirable hysteresis effects and the advantages of the technology based on electrorheological / magnetorheological fluids, ie the advantages of the fast reaction time and thus the higher dynamics are meaningfully usable.

Also advantageous is the possibility of regulating the supply pressure by means of the manipulated variables of the minimum volume flow q _{q, 1} / q _{q, 2} . This makes it possible to dispense with the arrangement of the otherwise necessary pressure control valve in the pressure supply.

In addition, it is now possible to adjust the supply pressure very simply in terms of time.

• Fig.1: eine Prinzipskizze eines Aktors auf Basis elektrorheologischer Flüssigkeiten, und
• Fig.2: eines schematische Darstellung eines Regelungskonzepts.

The invention will be explained in more detail with reference to an embodiment which is illustrated in the drawing. Show it:

• Fig.1 : A schematic diagram of an actuator based on electrorheological fluids, and
• Fig.2 : a schematic representation of a control concept.

Fig. 1 shows a schematic diagram of an actuator 1 based on electrorheological fluids consisting of a cylinder 2, in the transverse branch 3 one of four valves a, b, c, d based on electrorheological / magnetorheological Liquids existing full bridge is connected. The cylinder 2 is shown in the embodiment as a synchronous cylinder, but it could be used from any other cylinder design.

The cylinder 2 has a cylinder housing 4 with an axially displaceably mounted piston 5. The piston 5 divides the cylinder housing 4 into a first and a second variable-volume working chamber 6, 6 '. In the cylinder housing 4 each end an inlet / outlet opening for the pressure medium is introduced. The inlet / outlet opening of the first and second working chamber 6, 6 'is in each case coupled to a fluid line 7, 7' arranged in the transverse branch 3 of the valve full-bridge circuit. Each working chamber 6,6 'of the cylinder 2 is thus a half-bridge circuit consisting of two valves a, b / c, d assigned based on electrorheological / magnetorheological fluids. The first working chamber 6 associated valves a, b are, as can be seen from the illustrations, arranged in a first longitudinal branch 8 of the full bridge circuit. The second working chamber 6 'associated valves c, d are arranged in a second longitudinal branch 8' of the full bridge circuit. The full bridge circuit of the valves a, b, c, d is linked between the fluid connection of the valves a, c with a supply pressure line 9. The supply pressure is provided via a pump / storage arrangement not described in detail here. Since, in principle, only one volume flow in the direction of the pressure difference can flow in the case of valves based on electrorheological / magnetorheological fluids, the above-described interconnection to a full bridge is necessary. The flow direction of the volume flows is in the Fig. 1 represented by the arrows. The valves b, d are coupled to a tank 10.

Valves based on electrorheological / magnetorheological fluids are known in a variety of embodiments.

The valve based on electrorheological fluids consists in principle of a valve gap formed in a housing, which is bounded by electrically controllable electrode arrangements, so that an electrorheological fluid flowing through the valve gap can be changed by changing the electric field generated between the electrode arrangements with regard to the rheological properties. With the help of an impressed by a high voltage amplifier voltage to the electrode assemblies, an electric field can be generated and thus, with applied pressure difference, the volume flow through the valve can be varied. The valve thus represents an electrically adjustable throttle, which is shown schematically in the drawing of Fig. 1 is shown.

It is self-evident at this point that, when magnetorheological fluids are used instead of electrode arrangements, it is necessary to provide coil arrangements for generating a magnetic field.

An essential feature of the invention is that the four degrees of freedom of the valves a, b, c, d based on electrorheological / magnetorheological fluids are optimally utilized in the full bridge. By using the volume flows in the first and second working chamber q ₁ , q ₂ as independent control variables, the pressures in the two working chambers can be regulated. This can also be used in a cascade controller structure as a lower-level control loop, while in a higher-level control loop actual controlled variable (eg the position of the cylinder s or the pressure force) is regulated.

$${\mathrm{q}}_{\mathrm{2}}\mathrm{=}{\mathrm{q}}_{\mathrm{c}}\mathrm{-}{\mathrm{q}}_{\mathrm{d}}$$

q_a = Volumenstrom durch das Ventil a
q_b = Volumenstrom durch das Ventil b
q_c = Volumenstrom durch das Ventil c
q_d = Volumenstrom durch das Ventil dFrom the schematic representation of Fig. 1 It can be seen that the volume flows q ₁ , q ₂ to and from the working chambers respectively form the difference between the valve volume flows of the first and second longitudinal branch 8, 8 'of the full bridge: $${\mathrm{q}}_{\mathrm{1}}\mathrm{=}{\mathrm{q}}_{\mathrm{a}}\mathrm{-}{\mathrm{<figure-callout\; id="2"\; label="q\; b\; q"\; filenames="imgf0001.png"\; state="\{\{state\}\}">q\; </figure-callout>}}_{\mathrm{b}}$$

$${\mathrm{q}}_{}$$$${\mathrm{}}_{\mathrm{2}}\mathrm{=}{\mathrm{q}}_{\mathrm{c}}\mathrm{-}{\mathrm{q}}_{\mathrm{d}}$$

q _a = volume flow through the valve a
q _b = volume flow through the valve b
q _c = volume flow through the valve c
q _d = volume flow through the valve d

As already explained in more detail above, hysteresis effects occur in the transition region to completely block the valve based on electrorheological fluids. To prevent this, the valves a, b, c, d are always operated with a minimum volume flow. It is essential that the minimum volume flow through the valves a and b (q _{q, 1} ) and c and d (q _{q, 2} ) are the same size, because then the volume flows q ₁ and q ₂ are not affected. The minimum volume flow through the first longitudinal branch 8 is referred to below as q _{q, 1} , the minimum volume flow through the second longitudinal branch 8 'as q _{q, 2} .

If now the volumetric flows q ₁ / q _{2 are used} as manipulated variables, then the distribution of the volumetric flows to the four valves a, b, c, d of the full bridge takes place as follows: $$\begin{array}{l}\begin{array}{ll}{\mathrm{q}}_{\mathrm{a}}& \mathrm{=}\mathrm{sg}\left({\mathrm{q}}_{\mathrm{1}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{.}\mathrm{1}}\\ q_{\mathrm{b}}& \mathrm{=}\mathrm{sg}\left(\mathrm{-}{\mathrm{q}}_{\mathrm{1}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{.}\mathrm{1}}\end{array}\\ \begin{array}{ll}{\mathrm{q}}_{\mathrm{c}}& \mathrm{=}\mathrm{<figure-callout\; id="2"\; label="sg\; q"\; filenames="imgf0001.png"\; state="\{\{state\}\}">sg\; </figure-callout>}\left({\mathrm{q}}_{}\right)\left({\mathrm{}}_2\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{.}2}\mathrm{with\; sg}\left(\mathrm{q}\right)\mathrm{=}\mathrm{\{}\begin{array}{c}\mathrm{q}\mathit{f}\ddot{u}\mathit{r}q\mathrm{\ge }\mathrm{0}\\ \mathrm{0}\mathit{otherwise}\end{array}\\ q_{\mathrm{d}}& \mathrm{=}\mathrm{sg}\left(\mathrm{-}{\mathrm{<figure-callout\; id="2"\; label="q"\; filenames="imgf0001.png"\; state="\{\{state\}\}">q</figure-callout>}}_{\mathrm{2}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{.}2}\end{array}\end{array}$$

The value of the function q (q) for positive q corresponds to the argument q, while the value for negative q is identical to 0.

With a symmetrical bridge, it makes sense to select q _{q, 1} and q _{q, 2 the} same size. It follows immediately from the above drive strategy that q _a ≥ q _{q, 1} , q _b ≥ q _{q, 1} , q _c ≥ q _{q, 2} and q _d ≥ q _{q, 2} , and thus the valves are never completely disabled.

On the basis of the four valve volume flows (and the associated pressure drops), the electrical voltages of the valves can now be clearly defined. The four electrical voltages at the valves are the actual manipulated variables of the actuator.

$${\mathrm{q}}_{\mathrm{\Delta }}=\frac{1}{V_{01}+A_{1s}}{\mathrm{q}}_1-\frac{1}{V_{02}-A_{2s}}{\mathrm{q}}_2$$

mit dem Summenvolumenstrom q_Σ, dem Differenzvolumenstrom q_Δ sowie den Arbeitskammervolumina (für s=0) v₀₁ und V₀₂ und den effektiven Kolbenflächen A₁ und A₂ des Kolbens zu parametrieren. Die Terme V₀₁ + A₁s und V₀₂ - A₂s dienen dazu, die durch die Stellung des Kolbens s bedingte Nichtlinearität durch die unterschiedlichen Volumina in den beiden Kammern zu kompensieren. Mit dieser Gleichung werden sämtliche Bauformen, wie Gleichgangzylinder (A₁ = A₂), Zweistangenzylinder mit unterschiedlicher Kolbenfläche oder Differenzialzylinder erfasst. Falls die durch diese Terme bedingten Nichtlinearitäten keinen besonders großen Einfluss auf das dynamische Verhalten haben, beispielsweise wenn der Weg des Kolbens s sehr klein ist, können diese in der Gleichung auch entfallen.If, for example, one wishes to regulate the position of the piston 5 of the cylinder 2 and the sum pressure or the force on the piston 5 and the total pressure or other equivalent variables, it makes sense to design the volume flows q ₁ and q ₂ as follows: $${\mathrm{q}}_{\mathrm{\Sigma }}=\frac{1}{V_{01}+A_{1s}}{\mathrm{q}}_1+\frac{1}{V_{02}-{\mathrm{<figure-callout\; id="2"\; label="A"\; filenames="imgf0001.png"\; state="\{\{state\}\}">A</figure-callout>}}_{2s}}{\mathrm{q}}_2$$

$${\mathrm{q}}_{\mathrm{\Delta }}=\frac{1}{V_{01}+A_{1s}}{\mathrm{q}}_1-\frac{1}{V_{02}-{\mathrm{<figure-callout\; id="2"\; label="A"\; filenames="imgf0001.png"\; state="\{\{state\}\}">A</figure-callout>}}_{2s}}{\mathrm{<figure-callout\; id="2"\; label="q"\; filenames="imgf0001.png"\; state="\{\{state\}\}">q</figure-callout>}}_2$$

with the sum volume flow q _Σ , the differential volume flow q _Δ and the working chamber volumes (for s = 0) v ₀₁ and V ₀₂ and the effective piston areas A ₁ and A _{2 of} the piston to parameterize. The terms V ₀₁ + A ₁ s and V ₀₂ - A ₂ s are used to describe the compensate for non-linearity due to the position of the piston due to the different volumes in the two chambers. With this equation, all types, such as synchronous cylinders (A ₁ = A ₂ ), two-rod cylinder with different piston area or differential cylinder are detected. If the non-linearities caused by these terms do not have a particularly great influence on the dynamic behavior, for example if the travel of the piston s is very small, these can also be omitted in the equation.

The above-mentioned manipulated variable transformation now has the advantage that with the sum volume flow q _Σ directly the total pressure and with the differential volume flow q _Δ directly the position or speed of the piston or the force on the piston can be decoupled influenced.

For the new virtually formed manipulated variables q _Σ , q _Δ , the distribution of the volume flows to the four valves a, b, c, d of the full bridge takes place as follows: $$\begin{array}{l}{\mathrm{q}}_{\mathrm{a}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(\frac{\mathit{q}\mathrm{<figure-callout\; id="2"\; label="\Delta "\; filenames="imgf0001.png"\; state="\{\{state\}\}">\Delta </figure-callout>}}{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{02}}+A_{\mathrm{1}S}\right)\mathrm{+}{q}_{q\mathrm{.}1}\\ {\mathrm{q}}_{\mathrm{b}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(-\frac{\mathit{<figure-callout\; id="2"\; label="q\; \; \Delta "\; filenames="imgf0001.png"\; state="\{\{state\}\}">q\; </figure-callout>}\Delta }{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(-\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{01}}+A_{\mathrm{1}S}\right)\mathrm{+}{q}_{q\mathrm{.}1}\\ {\mathrm{q}}_{\mathrm{c}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(-\frac{\mathit{<figure-callout\; id="2"\; label="q\; \; \Delta "\; filenames="imgf0001.png"\; state="\{\{state\}\}">q\; </figure-callout>}\Delta }{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{02}}\mathrm{-}{\mathrm{<figure-callout\; id="2"\; label="A"\; filenames="imgf0001.png"\; state="\{\{state\}\}">A</figure-callout>}}_{\mathrm{2}S}\right)\mathrm{+}{q}_{q\mathrm{.}2}\\ {\mathrm{q}}_{\mathrm{d}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(\frac{\mathit{q}\mathrm{<figure-callout\; id="2"\; label="\Delta "\; filenames="imgf0001.png"\; state="\{\{state\}\}">\Delta </figure-callout>}}{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(-\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{02}}\mathrm{-}{\mathrm{<figure-callout\; id="2"\; label="A"\; filenames="imgf0001.png"\; state="\{\{state\}\}">A</figure-callout>}}_{\mathrm{2}S}\right)\mathrm{+}{q}_{q\mathrm{.}2}\end{array}\mathrm{with\; sg}\left(\mathrm{q}\right)=\{\begin{array}{c}\mathit{q}f\ddot{u}\mathit{rq}\ge 0\\ 0\mathrm{otherwise}\end{array}$$

On the basis of the four valve flow rates q _a, q _b, q _c, q _d can now electric voltages for controlling the four valves a, b, c, d clearly defined.

In Fig. 2 is shown in a schematic representation of the entire control concept. In the controller 11 (designed by means of known controller throwing method), the manipulated variables q _{q, 1} , q _{q, 2} as well as the manipulated variables q ₁ , q ₂ or q _Σ and q _{Δ are formed} , as further defined above in the text. In block valve control 12, the valve flow rates q _a, q _b, q _c, q _d through the above-mentioned equations in response to the manipulated variables used q ₁ and q ₂ or q _Σ, and calculates q _Δ are calculated from the predetermined control values.

In a subsequent voltage calculation 13, the corresponding voltages of the valves a, b, c, d are calculated from the valve volume flows. These values (the real manipulated variables) are supplied to the high-voltage amplifier 14. The valves (shown here by block 15) are now controlled according to the calculated voltages, so that the previously calculated valve volume flows q _a , q _b , q _c , q _d set and the first working chamber 6 with the volume flow q ₁ and the second working chamber 6 'are charged with the flow q ₂ .

For the implementation of an actuating law, a measurement or observation (in the sense of control engineering) of certain system sizes is generally necessary. Which system variables are necessary for the control generally depends on the selected controller concept. For the proposed control of electrorheological valves in any case, the knowledge of the pressure drop along the individual valves is necessary.

The pressures p ₁ and p ₂ in the working chambers 6, 6 ', which are referred to below as state variables as a function of the volume flows q ₁ and q ₂ , are recorded in or on the cylinder 2, ie measured or monitored by means of corresponding sensors , The captured State variables 16 such as pressures, travel of the piston and / or speed or forces are supplied to the controller 11 as actual values and compared with the predetermined desired values. A corresponding control deviation is corrected accordingly.

In the schematic representation, the pressure supply is shown as block 17. The valve full-bridge circuit 15 is supplied with the predetermined supply volume flow 18 via the supply pressure line.

It goes without saying that the valve control according to the invention of hydraulic actuators which has been described above with reference to a full-bridge circuit can also be used for a half-bridge circuit. In this case, only q ₁ for a working chamber / pressure medium chamber and q _{q, 1} can be used as manipulated variables as the minimum volume flow of the longitudinal branch of the half bridge. From these manipulated variables, the distribution of the volume flows for the valves a and b in q _a and q _{b takes place} analogously.

In principle, as an actuator instead of the cylinder-piston arrangement described above, arrangements may be provided which have means of a membrane delimiting pressure medium spaces.

Claims (8)

1. A method for controlling the valves of hydraulic actuators (2), wherein the actuator (2) comprises at least two working chambers (6, 6') that are separated by a resilient or moveable element (5) and wherein the working chambers (6, 6') are linked to four valves a, b, c, d (15) that are based on electro-rheological / magneto-rheological fluids and interconnected in the form of a full bridge, characterized in that the valves (15) are controllable independently, wherein the valves a and b are operated with a minimum flow rate q_q,1 of the same size and the valves c and d are operated with a minimum flow rate q_q,2 of the same size, wherein the flow rate q₁ into a first working chamber and the flow rate q₂ into a second working chamber (6') and also the minimum flow rate q_q,1 and the minimum flow rate q_q,2 are used as controlling variables for the corresponding regulation process, and the apportionment of the valve flow rates q_a, q_b, q_c, q_d through the valves (15) and thus the computation of the voltage signals for the control of the valves a, b, c, d are computed from these controlling variables q₁, q₂, q_q,1, q_q,2.
2. A method for controlling the valves of hydraulic actuators (2) in accordance with Claim 1, characterized in that the apportionment of the valve flow rates q_a, q_b, q_c, q_d through the valves (15) and hence the voltage signals for the control of the valves a, b, c, d are computed using the formulae: $$\begin{array}{l}\begin{array}{ll}{\mathrm{q}}_{\mathrm{a}}& \mathrm{=}\mathrm{sg}\left({\mathrm{q}}_{\mathrm{1}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{,}\mathrm{1}}\\ q_{\mathrm{b}}& \mathrm{=}\mathrm{sg}\left(\mathrm{-}{\mathrm{q}}_{\mathrm{1}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{,}\mathrm{1}}\end{array}\\ \begin{array}{ll}{\mathrm{q}}_{\mathrm{c}}& \mathrm{=}\mathrm{sg}\left({\mathrm{q}}_2\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{,}2}\mathrm{where\; sg}\left(\mathrm{q}\right)\mathrm{=}\mathrm{\{}\begin{array}{c}\mathrm{q}\mathit{for}q\mathrm{\ge }\mathrm{0}\\ \mathrm{0}\mathit{otherwise}\end{array}\\ q_{\mathrm{d}}& \mathrm{=}\mathrm{sg}\left(\mathrm{-}{\mathrm{q}}_{\mathrm{2}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{,}2}\end{array}\end{array}$$

   
3. A method for controlling the valves of hydraulic actuators (2), wherein the actuator (2) comprises at least two working chambers (6, 6') that are separated by a resilient or moveable element (5) and wherein the working chambers (6, 6') are linked to four valves a, b, c, d (15) that are based on electro-rheological / magneto-rheological fluids and interconnected in the form of a full bridge, characterized in that the valves (15) are controllable independently, wherein the valves a and b are operated with a minimum flow rate q_q,1 of the same size and the valves c and d are operated with a minimum flow rate q_q,2 of the same size, wherein the differential flow rate q_Δ into the two working chambers (6, 6') and the sum flow rate q_Σ into the two working chambers (6, 6') and the minimum flow rate q_q,1 and the minimum flow rate q_q,2 are used as controlling variables, and the apportionment of the valve flow rates q_a, q_b, q_c, q_d through the valves (15) and thus the voltage signals for the control of the valves a, b, c, d (15) are computed from these controlling variables q_Δ, q_Σ, q_q,1, q_q,2.
4. A method for controlling the valves of hydraulic actuators in accordance with Claim 3, characterized in that the apportionment of the valve flow rates q_a, q_b, q_c, q_d through the valves (15) corresponds to the following formulae whence the voltage signals for the control of the valves a, b, c, d are computed: $$\begin{array}{l}{\mathrm{q}}_{\mathrm{a}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(\frac{\mathit{q}\Delta }{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{02}}\mathrm{-}{A}_{\mathrm{1}S}\right)\mathrm{+}{q}_{q\mathrm{,}1}\\ {\mathrm{q}}_{\mathrm{b}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(-\frac{\mathit{q}\Delta }{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(-\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{01}}+A_{\mathrm{1}S}\right)\mathrm{+}{q}_{q\mathrm{,}1}\\ {\mathrm{q}}_{\mathrm{c}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(-\frac{\mathit{q}\Delta }{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{02}}\mathrm{-}{A}_{\mathrm{2}S}\right)\mathrm{+}{q}_{q\mathrm{,}2}\\ {\mathrm{q}}_{\mathrm{d}}\mathrm{=}\mathrm{\left(}\begin{array}{c}\mathit{sg}\left(\frac{\mathit{q}\Delta }{\mathrm{2}}\right)\mathrm{+}\mathit{sg}\left(-\frac{\mathit{q}\Sigma }{\mathrm{2}}\right)\end{array}\mathrm{\right)}\left(V_{\mathrm{02}}\mathrm{-}{A}_{\mathrm{2}S}\right)\mathrm{+}{q}_{q\mathrm{,}2}\end{array}\mathrm{mit\; sg}\left(\mathrm{q}\right)=\{\begin{array}{c}\mathrm{q\; for\; q}\ge 0\\ 0\mathrm{otherwise}\end{array}$$

   

   
   wherein

   A₁ = the effective piston surface area of the first working chamber 6;

   A₂ = the effective piston surface area of the second working chamber 6';

   V₀₁ = the volume of the first working chamber 6;

   V₀₂ = the volume of the second working chamber 6' and

   S = the path of the piston 5.
5. A method for controlling the valves of hydraulic actuators (2), wherein the actuator (2) comprises at least one working chamber (6) that is divided by a resilient or moveable element (5) and wherein the working chamber (6) is linked to two valves a, b that are based on electrorheological / magneto-rheological fluids and interconnected in the form of a half bridge, characterized in that the valves a, b are controllable independently, wherein the valves a and b are operated with a minimum flow rate q_q,1 of the same size, wherein the flow rate q₁ into a first working chamber (6) and the minimum flow rate q_q,1 are used as controlling variables for the corresponding regulation process, and the apportionment of the valve flow rates q_a, q_b through the valves (15) and hence the voltage signals for the control of the valves a, b are computed from these controlling variables q₁, q_q,1.
6. A method for controlling the valves of hydraulic actuators in accordance with Claim 5, characterized in that the apportionment of the valve flow rates q_a, q_b through the valves a, b and hence the voltage signals for the control of the valves a, b are computed using the formulae: $$\begin{array}{|c|}\hline \begin{array}{ll}{\mathrm{q}}_{\mathrm{a}}& \mathrm{=}\mathrm{sg}\left({\mathrm{q}}_{\mathrm{1}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{,}\mathrm{1}}\\ q_{\mathrm{b}}& \mathrm{=}\mathrm{sg}\left(\mathrm{-}{\mathrm{q}}_{\mathrm{1}}\right)\mathrm{+}{\mathrm{q}}_{\mathrm{q}\mathrm{,}\mathrm{1}}\end{array}\mathrm{where\; sg}\left(\mathrm{q}\right)\mathrm{=}\left|\mathrm{\right\{}\begin{array}{c}\mathrm{q}\mathit{for}\mathrm{q}\mathrm{\ge }\mathrm{0}\\ 0\mathit{otherwise}\end{array}\\ \hline\end{array}$$

   
7. A device for controlling valves using hydraulic actuators (2), wherein the actuator (2) comprises at least two pressure medium spaces (6, 6') that are separated by a resilient or moveable element (5) and wherein the pressure medium spaces (6, 6') are linked to four valves a, b, c, d (15) that are based on electro-rheological / magneto-rheological fluids and interconnected in the form of a full bridge, and wherein the valves (15) are controllable independently in accordance with any of the methods according to Claims 1 to 4, characterized in that there is provided a regulator 11 for the control process, wherein q₁, q₂ or q_Δ, q_Σ, and q_q,₁ and q_q,2 are used as controlling variables for the corresponding regulation process, and the apportionment of the valve flow rates q_a, q_b, q_c, q_d through the valves (15) is determined from these controlling variables by a valve control means (12) and the voltage signals for the control of the valves a, b, c, d (15) are computed by a subsequent voltage computation process (13).
8. A device for controlling valves using hydraulic actuators (2), wherein the actuator (2) comprises at least one pressure medium space (6) that is divided by a resilient or moveable element (5) and wherein the pressure medium space (6) is linked to two valves a, b (15) that are based on electro-rheological / magneto-rheological fluids and interconnected in the form of a half bridge, and wherein the valves a, b (15) are controllable independently in accordance with either of the methods according to Claim 5 or 6, characterized in that there is provided a regulator 11 for the control process, wherein q₁ and q_q,1 are used as controlling variables for the corresponding regulation process, and the apportionment of the valve flow rates q_a, q_b, through the valves (15) is determined from these controlling variables q₁, q_q,1 by a valve control means (12) and the voltage signals for the control of the valves a, b (15) are computed by a subsequent voltage computation means (13).

Images