Method for actuating a valve, and corresponding device

Abstract

A method for actuating a valve with an electromagnetic valve drive through which electric current is conducted to open or close the valve or to hold the valve in an open or closed position includes receiving an opening signal, determining an adapted opening signal which is adapted to physical limits of the valve or valve drive, and determining a feed-forward signal for feed-forward control of an electric current to drive the electromagnetic valve drive to open the valve in reaction to the adapted opening signal. The method further includes calculating an actuation signal for actuating the valve drive using the feed-forward signal, and outputting the actuation signal.

Claims

1. A method for actuating a valve using an electromagnetic valve drive, through which electric current is conducted in order to open the valve, to close the valve, or to hold the valve in an open or closed position, the method comprising: receiving an opening signal, the opening signal representing target values over time for a state of the valve; ascertaining, based on the opening signal, an adapted opening signal, the adapted opening signal representing values over time that the state of the valve that can physically be assumed according to physical limitations of the valve or the electromagnetic valve drive; ascertaining, based on the adapted opening signal, a pilot control signal for pilot control of the electric current to drive the electromagnetic valve drive to open the valve in reaction to the adapted opening signal; calculating an actuation signal to actuate the electromagnetic valve drive using the pilot control signal; and outputting the actuation signal.

2. The method as claimed in claim 1, further comprising: receiving an electric current signal representing the electric current through the electromagnetic valve drive; and wherein the actuation signal is calculated using the electric current signal and the pilot control signal.

3. The method as claimed in claim 1, further comprising: ascertaining the pilot control signal using a valve model.

4. The method as claimed in claim 3, wherein the valve model models a first-order delay behavior of the valve.

5. The method as claimed in claim 3, wherein a time constant of the valve model corresponds to a time constant of the electromagnetic valve drive.

6. The method as claimed in claim 3, wherein an electric resistance of the electromagnetic valve drive is estimated in the valve model.

7. The method as claimed in claim 3, wherein an inductance of the electromagnetic valve drive is estimated in the valve model.

8. The method as claimed in claim 1, wherein a pilot control and a regulator are used to calculate the actuation signal.

9. The method as claimed in claim 8, further comprising: applying an inverse PT1 element in the pilot control.

10. The method as claimed in claim 8, further comprising: applying a state variable filter in the pilot control.

11. The method as claimed in claim 1, wherein a control unit for actuating the valve is configured to carry out the method.

12. A non-transitory machine-readable storage medium that stores a computer program for actuating a valve using an electromagnetic valve drive, through which electric current is conducted in order to open the valve, to close the valve, or to hold the valve in an open or closed position, the computer program, when executed by a controller, causing the controller to: receive an opening signal, the opening signal representing target values over time for a state of the valve; ascertain, based on the opening signal, an adapted opening signal, the adapted opening signal representing values over time that the state of the valve that can physically be assumed according to physical limitations of the valve or the electromagnetic valve drive; ascertain, based on the adapted opening signal, a pilot control signal for pilot control of the electric current to drive the electromagnetic valve drive to open the valve in reaction to the adapted opening signal; calculate an actuation signal to actuate the electromagnetic valve drive using the pilot control signal; and output the actuation signal.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The disclosure and the technical environment are explained in greater detail hereinafter on the basis of the figures. It is to be noted that the figures and in particular the size relationships shown in the figures are only schematic. In the figures:

(2) FIG. 1: shows a schematic illustration of a described valve,

(3) FIG. 2: shows a regulation of a segment using a regulator and a pilot control,

(4) FIG. 3: shows a regulation of a PT1 segment using a regulator and a pilot control,

(5) FIG. 4: shows a control of a segment, having predetermined signal curve yt(t),

(6) FIG. 5: shows an example of a segment to be regulated,

(7) FIG. 6: shows an inverter with a segment,

(8) FIG. 7: shows a state variable filter without manipulated variable limiting,

(9) FIG. 8: shows a state variable filter with actuator limiting,

(10) FIG. 9: shows a jump response behavior solely with a pilot control (without regulator) but state variable filters having different data,

(11) FIG. 10: shows a jump response behavior corresponding to FIG. 9 in a different representation, and

(12) FIG. 11: shows a control concept having a segment observation for the pilot control and resistance estimation.

DETAILED DESCRIPTION

(13) FIG. 1 shows a valve 5 having a control unit 13 for actuating this valve 5. The valve 5 has a fluid block 14, which comprises, for example, a line, a bottleneck, and a valve body, using which the valve 5 can be opened and closed. The fluid block 14 or in particular the valve body of the fluid block 14 is controlled by a valve drive 6 of the valve 5. The valve drive 6 is in particular an electric coil or electromagnet, which can exert a magnetic force which acts on the fluid block 14 or in particular on a movable valve body in the fluid block 14 in order to open or close the valve 5 or to change the opening state of the valve 5 in general.

(14) The control unit 13 is configured to receive an opening signal 8. An item of information about how the opening state of the valve 5 is supposed to be changed is preferably contained in the opening signal 8. From the opening signal 8, the control unit 13 generates a suitable actuation signal 11, using which the valve drive is deliberately actuated in reaction to this opening signal 8. To produce the actuation signal 11 suitably, a valve model 12 and a state variable filter 23 are stored in the control unit 13. Further details on the linking of valve model 12 and state variable filter 23 are described hereinafter on the basis of the further figures. The control unit 13 can additionally be configured to detect valve parameters 15. Valve parameters 15 are measured actual parameters of the valve operation. The values of such parameters can be taken into consideration in the production of a suitable actuation signal 11. In addition, the control unit 13 can receive a current signal 10, which represents the current presently flowing through the valve drive 6. A current signal 10 of the valve drive 6 is also very advantageous to ascertain suitable actuation signals 11 for the valve drive 6.

(15) FIG. 2 shows a general illustration of a control loop having a pilot control 17 and a regulator 18 for controlling a segment 24. The segment 24 corresponds to the valve 5 or the valve drive. Particularly rapid, efficient, and accurate regulation of the segment 24 is achieved by the combination of pilot control 17 and regulator 18. The control unit 13 described in FIG. 1 is formed by the regulator 18 and the pilot control 17. An opening signal 8 is switched from the left on the regulator 18 and the pilot control 17. The pilot control acts directly on the segment 24. The segment 24 is monitored using a sensor 19. This can be, for example, a current sensor, which generates a current signal 10, which is based on the electric current 7, which is observed here as an output variable of the segment 24. A regulating error 20 is calculated from the current signal 10 and the opening signal 8, which is used as the input variable for the regulator 18. An actuation signal 11 for the segment 24 (the valve 5 or the valve drive of the valve 5) is determined from the pilot control signal ascertained by the pilot control 17 and the output of the regulator 18. It is assumed here that the regulator block also includes the actuator.

(16) The regulator 18 is also shown as part of the overall circuit in FIG. 3. However, embodiment variants without regulator 18 are also possible, in particular if the pilot control is sufficiently accurate or even possibly has an interference variable compensation, which enables it to compensate for interference variables in such a way that additional regulation is no longer necessary.

(17) FIG. 3 shows a variant of the control loop from FIG. 2, wherein identical elements are identified here by identical reference signs. The segment 24 is adopted here as the PT1 element 21.

(18) The pilot control here is a combination of state variable filter 23 and inverter 22. Such a pilot control is also referred to as predictive pilot control. The state variable filter 23 calculates a signal curve, which can also be called a “trajectory”. This signal curve or this trajectory corresponds to a desired curve in consideration of physical limits, which the state variables 7 are to fulfill. The trajectory is converted into the actual manipulated variable using the inverter 22.

(19) FIG. 4 illustrates a pilot control 17 alone for the control of a segment 24 without a regulator being provided. The identifications of the individual elements correspond to FIG. 2 and FIG. 3.

(20) The effect of the state variable filter 23 is to be explained on the basis of FIG. 4. The following variables are relevant for the state variable filter 23. y.sub.w(t)=time target value curve which the output variable is supposed to assume. This corresponds, for example, to the opening signal, y.sub.t (t)=time curve which the output variable can physically assume maximally, and this corresponds, for example, to the adapted opening signal, u(t)=manipulated variable u(t) so that y(t)=y.sub.t(t). This corresponds, for example, to the pilot control signal or the actuation signal.

(21) A PT1 behavior can be described as follows:

(22) $T_1.Math.\frac{d}{\mathrm{dt}}y\left(t\right)+y\left(t\right)=K.Math.u\left(t\right)$

(23) The segment described in FIGS. 2 to 4 is explained briefly for the case discussed here of a valve or a valve drive as the segment to be regulated in FIG. 5. FIG. 5 shows a diagram of the electrical properties of a valve drive having resistance R, inductance L, applied voltage U, and current i resulting from the voltage. The solenoid of the valve drive is thus electrically described. The solenoid valve coil can be described as shown in FIG. 5. The following differential equation thus results:

(24) $T_1.Math.\frac{d}{\mathrm{dt}}i\left(t\right)+i\left(t\right)=K.Math.u\left(t\right)$

(25) One now wishes to select u(t) so that the desired starting behavior results. One thus requires a block which outputs the desired signal u(t) as the output to achieve y(t). This is achieved by an inverter block, which has u(t) as the output but requires i(t) and

(26) $\frac{d}{d\left(t\right)}i\left(t\right)$
as input variables therefor.

(27) The illustration of segment and inverter block is shown in FIG. 6, where the pilot control 17 is shown once again having the inverter 22 and the state variable filter 23. For example, adapted opening signals 27 are relayed by the state variable filter 23 to the inverter 22, which were ascertained by the state variable filter from opening signals 8 (step b11). The inverter 22 then carries out step b2).

(28) It is now necessary to calculate the input variables i(t) and

(29) $\frac{d}{d\left(t\right)}i\left(t\right)$
for a desired output behavior y(t) so that the desired behavior results or it is physically possible at all to set the desired behavior. The block which provides this is the described state variable filter 23, which can also be referred to as a signal generator. The state variable filter 23 has a signal y.sub.w(t) as an input and calculates the output variables i(t) and

(30) $\frac{d}{d\left(t\right)}i\left(t\right),$
so that they can be fed directly into the above-mentioned inverter block.

(31) This state variable filter 23 is shown once again in greater detail in FIG. 7. If the control segment has a PT1 behavior, a PT1 behavior can also be used for the state variable filter 23. The loop amplification of the state variable filter 23 is “1”, and the dynamic response of the behavior can be defined via a time constant of the state variable filter 23 “Tau_ZVF”=“time constant of the state variable filter”. Generally, the manipulated variable (in this case the voltage in the vehicle) is now limited. If one takes into consideration this manipulated variable limiting (i.e., the maximum possible voltage) directly in the state variable filter 23, this thus causes limiting of the maximum gradients of the current. The maximum possible voltage of the maximum possible gradients can form a vector, which is considered here as a whole as an adapted opening signal 27. In principle (in all embodiment variants), the adapted opening signal 27 can be a vector which comprises multiple individual variables relating to the opening state of the valve. In principle (in all embodiment variants) the opening signal can also be a vector, which comprises multiple individual variables relating to the opening state of the valve.

(32) FIG. 8 shows a further embodiment variant of a state variable filter 23, which generates the adapted opening signal 27 from the opening signal 8. In this variant of a state variable filter 23, manipulated variable limiting takes place, which is used for the purpose, for example, of taking into consideration physical limits. In combination of the state variable filter 23 with the inverter block, this has the effect that the manipulated variable limiting can be maximally utilized.

(33) The use of an actuation to actuate a valve drive using state variable filter 23 results in the jump response behavior shown in FIG. 9 and FIG. 10. According to FIG. 9 and FIG. 10, the response behavior is simulated in each case without using a regulator in addition to the pilot control. For comparison, the behavior with a regulator (3) is plotted in each of the diagrams.

(34) The images in FIG. 9 and FIG. 10 show the results of an overall system behavior using the actuation concept described here and the regulating method described here consisting of state variable filter, inverter block, and a segment having PT1 behavior.

(35) The upper portion of the diagrams in each of the figures shows the time behavior of the manipulated variable 4 with various time constants of the state variable filter. The lower portion of the diagrams in the figures shows the following signals: the setpoint jump (1), which is output as a specified value to the system consisting of pilot control and segment. The actual curve of the current (3), which results as a response to the setpoint jump (1). The current (3) was plotted here for comparison of the dynamic response and was regulated using a standard PI regulator. The calculated curve of the output current (2), which is achieved by simulation of the system made up of pilot control and segment.

(36) FIG. 9 and FIG. 10 show the illustration of a jump response with various time constants of the state variable filter without additional regulator. FIG. 10 shows the same behavior as FIG. 9. However, FIG. 10 shows the starting interval of the behavior shown in FIG. 9 in detail in an elongated illustration.

(37) The illustration of the jump responses is to show that the same dynamic response is achieved at the time of the actuation solely using a pilot control as using an actuation concept with PI regulator, which is very dynamically designed. The curve profiles 4 each show the simulated response behavior using a pilot control. The curve profiles 3 each show a real measured response behavior which can be achieved using a PI regulator.

(38) It can be seen from the simulation results in the lower plot that it is possible to achieve the same dynamic response by way of a corresponding pilot control (state variable filter and inverter of the segment) as using a regulator of a typical type without pilot control, which is very aggressively designed. To this end the time constant must be selected correspondingly at the state variable filter. The great advantage of the pilot control here, however, is that the pilot-controlled output variable without overshoots sets the setpoint value. This cannot be achieved using a conventional regulator at this speed.

(39) Moreover, in a concept with a pilot control comprising a state variable filter, the dynamic response can be deliberately selected/adapted situationally, so that either very high dynamic response or very low dynamic response can be predetermined. The deliberate setting of a low dynamic response can be advantageous to achieve noise-optimized switching.

(40) Various manipulated variable curves 4 can be seen in the upper plot of FIG. 9 and FIG. 10, which show the curve of the manipulated variable at various time constants of the state variable filter. If the state variable filter has the same time constant as the segment itself, the jump at the input is also output as the jump at the output, however with the corresponding amplification which is necessary to reach the final value (here u=R.Math.y w). If the time constant of the state variable filter (with consideration of the actuator limiting) is less than the time constant of the segment, it can thus be recognized on the upper plot that the manipulated variable is nearly automatically kept at the maximum of the actuator as long as necessary.

(41) The concept described here of the pilot control for valve actuations can be used in particular for braking control systems. In braking control systems, it is possible to use the concept in particular for the following purposes: valve actuation solely using pilot control valve actuation using pilot control and regulator (regulator optimized for regulating out the regulation deviation or interference variables) as a valve actuation having special requirements for the current curve (trajectory planning of the current).

(42) In addition, this concept is usable in nearly all applications with solenoid valves. It is appropriate in particular if the requirements for the time constant of the segment are high.

(43) Moreover, additionally carrying out further above-described additional method steps f) and g) is advantageous, comprising returning the estimated or calculated system parameter (in particular the system parameter resistance). It can be ensured by these additional steps that depending on the design of the observer, the resistance adapts within a few milliseconds. This means that a deviation of the resistance can already be regulated out within one ABS actuation cycle. Such an actuation cycle has a length, for example, of approximately 200 ms.

(44) A linearized implementation of the underlying concept of the method described here in hardware can be described by the following equations.

(45) Step 1 calculates the setpoint trajectory of the current.

(46) $I_{\mathrm{setpointk}}=I_0+\frac{\mathrm{dt}}{{\tau }_{\mathrm{ZVF}}}.Math.\left(I_{\mathrm{setpoint}}-I_0\right)$

(47) Step 2 calculates the required (actuation) voltage for the setpoint trajectory of the current.

(48) $U_k=R.Math.\left(\frac{{\tau }_{\mathrm{LR}}}{\mathrm{dt}}.Math.\left(I_{\mathrm{setpointk}}-I_0\right)+I_0\right)$

(49) The third step limits the voltage to the maximum available voltage
U.sub.lim=lim(min,max,U.sub.k)

(50) The fourth step now calculates the maximum possible current which is possible due to the limiting of the voltage.

(51) $I_{\mathrm{actualk}}=I_0+\frac{\mathrm{dt}}{{\tau }_{\mathrm{LR}}}.Math.\left(\frac{U_{\lim }}{R}-I_0\right)$

(52) Since deviations arise in real use, due to tolerances and component variations and naturally also due to changes of the environmental conditions, which cannot all be detected, a combination of pilot control and regulator is advisable, as is also shown in particular in FIG. 3.

(53) The regulation approach provided here is to regulate out the changes (resistance, voltage, etc.), which cannot be compensated for by the pilot control, in the case of a valve actuation as occurs in many applications. In the first approach, a Luenberger observer structure is used, which is supplemented, however, by a resistance estimation.

(54) FIG. 11 shows the comprehensible structure of real segment 24 comprising the valve 5 and the valve 6 and the valve model 12 adjacent to them, which is implemented by this implementation in hardware in a control unit 13. The valve model 12 can also be supplemented by a so-called observer, which permanently monitors the real segment 24 and brings the valve model 12 and the real segment 24 into correspondence. The observer is used to observe the behavior of the segment 24 and/or of the valve 5 and/or of the valve drive 6. The observer is also implemented in the valve model 12.

(55) In the first step, the current deviation is interpreted as a resistance deviation, the correction variable and unit is contained in the amplifier gain K_Observ. Subsequently, the ΔR is integrated and the term R+ΔR is formed. A fracdiff is now furthermore calculated therefrom by skilled reconfiguration, which is then also incorporated in the observer.

(56) Step 1:

(57) $\frac{R+\Delta R}{L}$

(58) Step 2:

(59) 0$\frac{R+\Delta R}{L}.Math.I=\frac{R+\Delta R}{L}.Math.\frac{U}{R+\Delta R}$

(60) Step 3:

(61) $\frac{R+\Delta R}{L}.Math.\frac{U}{R+\Delta R}=\frac{U}{L}=\frac{\mathrm{di}}{\mathrm{dl}}$