Pneumatic control device, process valve unit and method

Abstract

The invention relates to a pneumatic control device (1), in particular a position controller, for controlling a control variable (18) to a setpoint value (19) by providing at least one pneumatic control signal (21), the control device (1) being designed to perform a sliding-mode control using a sliding variable, and within the framework of the sliding-mode control, the control variable (18) being controlled by the pneumatic control signal (21) to the setpoint value (19). A control signal (u) for controlling the valve device (12) is calculated on the basis of a product of an amplification factor (k) and a sign function acting on the slip variable in order to provide the at least one pneumatic control signal (21) by means of the valve device (12).

Description

Technical Field

[0001] The invention relates to a pneumatic control device, in particular a position controller, for controlling a control variable to a given value by providing at least one pneumatic control signal. The control variable is, for example, the position of an actuator component of a pneumatic actuator, the pneumatic actuator being, for example, a process valve having a pneumatic drive and an accessory. The control variable is preferably the position of an actuator component of the pneumatic drive, by means of which a valve component of the accessory is actuated. The assembly consisting of the control device and the pneumatic actuator is also referred to as a controller. Background Art

[0002] When controlling pneumatic actuators, strong nonlinearities often occur due to the compressibility of the air and due to friction effects, for example in the pneumatic drive and / or in the accessories. There is also a wide variety of process valves. The large variety of designs and the pronounced nonlinearities make it difficult to design a suitable control algorithm for many different regulators, especially for any arbitrary regulator.

[0003] Traditionally, linear regulators, such as PID regulators, are used. Linear control concepts can only guarantee a high level of control performance to a limited extent while achieving high robustness. In addition, linear regulators require parameterization, for which (usually) complex identification processes must be carried out in order to identify suitable parameter values ​​for the respective regulator.

[0004] There are also nonlinear approaches for controlling pneumatic actuators. Nonlinear approaches generally allow shorter control times while achieving improved control quality and greater robustness against disturbances. However, these methods are much more complex than linear control approaches. As a rule, nonlinear approaches require a high level of model knowledge about the system to be controlled. This is often not the case in the process industry. On the one hand, the variety of models is too great, and on the other hand, the available sensor devices for automatic identification to achieve a sufficient model quality are limited.

[0005] A sliding mode control is known from research, which can also be called sliding mode control. In sliding mode control, a sliding variable σ(x) is defined, which depends on the state x of the system to be controlled. The control objective of the sliding mode controller is to control the sliding mode variable to zero within a finite time.

[0006] Huang, Kuo, and Chang: “Adaptive Sliding-Mode Control for Nonlinear SystemsWith Uncertain Parameters”, IEEE Transactions on Systems, Man, and Cybernetics—Part B: Cybernetics, Vol. 38, No. 2, April 2008. A method is proposed in which the amplification factor is continuously increased as long as the sliding level is not reached.

[0007] Fang, Tsai, Yan, Guo: “Adaptive Chattering-Free Sliding Mode Control of Chaotic Systems with Unknown Input Nonlinearity via Smooth Hyperbolic Tangent Function”, Problems in Engineering Mathematics, 2019, describes an adaptive algorithm that depends on the size of the sliding variable.

[0008] Plestan, Shtessel, Brégeault, Poznyak: “New methodologies for adaptive sliding mode control”, International Journal of Control, Vol. 83, July 2010, describes a method for approximating the average control variable near the sliding plane. In addition, a method is described in which the gain is changed dynamically in order to find the minimum permissible value of the amplification factor.

[0009] Edwards and Shtessel: “Adaptive continuous higher order Sliding mode control”, Automation, No. 65, 2016, describes an adaptive strategy that does not depend on the sliding variable.

[0010] Kuo, Huang, and Chang: “Sliding mode control I with self-tuning law for certainty linear systems”, ISA Journal 47, 2008. The discontinuous sign function is replaced by a continuous saturation function whose slope is adjusted around zero according to the process.

[0011] Ertugrul, Kerestiecioglu, Kaynak: “Gain adaptation in sliding mode control of robotic manipulators”, International Journal of System Science, Vol. 31, September 2000, describes the combination of approximating the average controller signal using a low-pass filter and an adaptive gain factor.

[0012] Hall, Shtessel: “Sliding mode disturbance observer-based control for areusable launch vehicle,” AIAA Journal, Guidance, Control, and Dynamics, vol. 29, 2006, describes the use of an observer for system uncertainty to adjust amplification factors.

[0013] In addition, a super-twist algorithm is known - a second-order sliding mode controller that depends only on the sliding variable. The following publications describe adaptive variants of the super-twist algorithm: Davila, Moreno, Fridman: "Variable Gains Super-Twisting Algorithm: A Lyapunov Based Design", American Control Conference, 2010; Shtessel, Moreno, Plestan, Fridman, Poznyak: "Super-twisting Adaptive Sliding Mode Control: a Lyapunov Design", 49th IEEE Decision and Control Conference, December 2010; Utkin, Poznyak: "Adaptive Sliding mode control with application to super-twist algorithm: Equivalent control method", Automatica 49, 2013. Summary of the invention

[0014] The object of the present invention is to provide a control device that can be used conveniently and flexibly.

[0015] This object is achieved by a control device according to claim 1. The control device is designed to carry out a sliding mode control with a sliding variable and to calculate a control signal for controlling a valve device within the framework of the sliding mode control based on a product of an amplification factor and a sign function applied to the sliding variable in order to provide at least one pneumatic control signal with the valve device.

[0016] It has been found that sliding mode control is particularly suitable for use in pneumatic control devices, such as position controllers. Sliding mode control can be used for a variety of different controller variants without the need to adapt the sliding mode control to the respective variant. As a result, the control device can be used easily and flexibly. Sliding mode control can be conveniently performed without an identification process for identifying controller parameters, in particular taking place outside or before operation. Furthermore, sliding mode control preferably does not require any expert knowledge of the customer for specifying possible limitations.

[0017] Advantageous further developments are the subject matter of the dependent claims.

[0018] The invention further relates to a process valve unit comprising a position controller designed as a pneumatic control device and a process valve having a drive and fittings, wherein the position controller is designed to control the process valve.

[0019] The invention also relates to a method for operating a pneumatic control device, comprising the step of calculating a control signal based on the product of an amplification factor and a sign function. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Further exemplary details and exemplary embodiments are explained below with reference to the drawings.

[0021] Figure 1 is a schematic diagram of an assembly having a pneumatic control device and a pneumatic actuator;

[0022] Figure 2 It is a block diagram of regulation;

[0023] Figure 3 A method for calculating a control signal is shown. DETAILED DESCRIPTION

[0024] Figure 1 An arrangement 10 is shown, which comprises a pneumatic control device 1 and a pneumatic actuator 2. The arrangement 10 represents an exemplary application environment for the pneumatic control device 1. The control device 1 can also be provided alone, ie in particular without the pneumatic actuator 2.

[0025] By way of example, the component 10 is designed as a regulator, in particular a process valve unit. The control device 1 is designed by way of example as a position controller and can also be referred to as a positioner. The pneumatic actuator is designed by way of example as a process valve. The process valve unit is used to control or regulate the fluid flow in a process technology facility.

[0026] First, let me introduce the control device 1:

[0027] The control device 1 is used in industrial automation, in particular process automation. The control device 1 is designed to control a pneumatic actuator 2 .

[0028] The control device 1 preferably has a valve device 12, through which the control device 1 can ventilate and / or exhaust the first pressure chamber 7 of the pneumatic actuator 2. The valve device 12 is designed, for example, as an electropneumatic converter. The valve device 12 advantageously has one or more pilot valves.

[0029] Optionally, the control device 1 has an input unit 17 , and a user can input to the control device 1 through the input unit 17 .

[0030] The control device 1 preferably has a control unit 14 which is designed, for example, as a microcontroller. The control unit 14 preferably controls the valve device 12 .

[0031] Optionally, the control device 1 has a communication interface 36, which is used in particular for communication with a higher-level controller, for example to receive a setpoint value 19, in particular for controlling the position of the actuator component 3 of the pneumatic actuator 2. The communication interface 36 advantageously makes the setpoint value 19 available to the control unit 14.

[0032] The control device 1 advantageously comprises a housing 37 in which the valve device 12, the control unit 14 and / or the pressure sensor device 11 are preferably arranged. By way of example, the input unit 17 and / or the communication interface 36 are arranged in and / or on the housing 37.

[0033] The control device 1 (in particular designed as a position controller) is fastened, in particular by means of its housing 37 , to the pneumatic actuator 2 , by way of example.

[0034] Now let's introduce pneumatic actuator 2:

[0035] The pneumatic actuator 2 advantageously comprises a pneumatic drive 38 and a fitting 39 which can be actuated by the pneumatic drive 38. The fitting 39 is in particular a process fitting.

[0036] The pneumatic actuator 2, in particular the driver 38, has an actuator member 3. The pneumatic actuator 2, in particular the driver 38, has a piston device 4. By way of example, the piston device 4 forms the actuator member 3. The pneumatic actuator 2, in particular the driver 38, has a first pressure chamber 7. The first pressure chamber 7 is bounded by the piston device 4. By applying compressed air to the first pressure chamber 7, the position of the actuator member 3 can be changed. Merely as an example, the pneumatic actuator 2 is designed as a single-acting actuator. The pneumatic actuator 2, in particular the driver 38, comprises a spring element 8, which provides a spring force acting on the piston device 4.

[0037] Alternatively, the pneumatic actuator 2 , in particular the drive 38 , can be designed as a double-acting actuator.

[0038] The drive 38 is designed as a rotary drive, for example. Exemplarily, the drive 38 is designed as a piston drive. Alternatively, the drive 38 can be designed as a film drive. The fitting 39 includes a valve member 41, which is actuated by the drive 38. The fitting 39 can be designed as a flap valve, a slide valve or a ball valve, for example. The drive 38 has a drive element 42 that can be driven by the piston device 4. The drive element 42 is connected to the valve member 41 so that the position of the valve member 41 can be changed by the drive element 42. The piston device 4, the drive element 42 or the valve member 41 advantageously form the actuator member 3.

[0039] By way of example, the assembly 10 has a sensor device, which advantageously comprises a pressure sensor device 11 and / or a position sensor device 9. The sensor device, in particular the pressure sensor device 11 and / or the position sensor device 9, is designed to detect a manipulated variable 18 to be regulated by the control device and to provide it as a feedback variable 22, in particular to the control unit 14.

[0040] By way of example, the pressure sensor device 11 is part of the control device 1 . The pressure sensor device 11 is preferably used to detect the current pressure of the first pressure chamber 7 .

[0041] By way of example, the position sensor device 9 is arranged on the pneumatic actuator 2. The position sensor device 9 is used to detect the current position of the actuator member 3. Advantageously, the current position of the actuator member 3 can be detected by the position sensor device 9 over the entire movement path of the actuator member 3, i.e. from the first end position to the second end position.

[0042] The control performed by the pneumatic control device 1 will be discussed in more detail below. Figure 2 A block diagram of an exemplary implementation of this regulation is shown.

[0043] The pneumatic control device 1 is used to control the control variable 18 to a set value 19 by providing at least one pneumatic control signal 21. Exemplarily, the control variable 18 is the position of the actuator member 3. In addition, the control variable 18 can be a pressure, such as the pressure of the first pressure chamber 7, a flow, such as the flow through the fitting 39, a liquid level or a force. The set value 19 is provided, for example, by a superior controller and transmitted to the control device 1. In addition, the set value 19 can be provided by the control device 1, for example, by calculation. The pneumatic control signal 21 is output by the valve device 12, for example, into the first pressure chamber 7.

[0044] By way of example, the control is performed using a control unit 14. The control unit 14 advantageously comprises a subtraction component 51 and / or a control unit 52, which is designed, for example, as a control algorithm. The subtraction component 51 calculates the control deviation 26 as the difference between the setpoint value 19 and the feedback variable 22. The control unit 52 calculates a control signal u based on the control deviation 26, with which the valve device 12 is controlled. The valve device 12 outputs at least one pneumatic control signal 21 according to the control signal u, in particular to the pneumatic actuator 2, in order to change the control variable 18 towards the setpoint value 19. The current value of the control variable 18 is returned as the feedback variable 22.

[0045] The control device 1 is designed to perform a sliding mode control with a sliding variable σ(x). Sliding mode control can also be referred to as sliding mode control. The sliding variable can also be referred to as a sliding variable. The sliding variable σ(x) depends on one or more states x of the system to be controlled. For example, the sliding variable σ(x) is defined as the difference between a setpoint 19 and a control variable 18. The sliding variable is advantageously equal to the control deviation 26. The sliding variable will also be referred to as σ below.

[0046] The subspace bounded by the condition σ(x) = 0 is called the sliding plane. If the system moves on this sliding plane, its dynamics will not be affected by disturbances or parameter inaccuracies. Since the sliding variable can be set, the dynamics on the sliding plane can also be set. This means that the system is asymptotically stable and converges once σ(x) = 0 is satisfied. The control goal of sliding mode control is to control the sliding mode variable to zero in a finite time.

[0047] The control device 1 is designed to calculate a control signal u for controlling the valve device 12 within the framework of sliding mode control based on the product of the amplification factor k and the sign function applied to the sliding variable, so as to provide at least one pneumatic control signal 21 by means of the valve device 12. The control signal u is, for example, an electrical control signal and is outputted by the control unit 14 to the valve device 12, exemplarily to control the output of the pneumatic control signal 21.

[0048] Exemplarily, the control device 1 is designed to calculate the control signal u according to the following formula (1):

[0049] u=-k sig(σ) (1)

[0050] In formula (1), "u" is the control signal, "k" is the amplification factor, and sign(σ) is the sign function applied to the sliding variable σ. The amplification factor k is preferably positive.

[0051] The control device 1 is advantageously designed to control the sliding variable σ to zero, in particular based on the formula (1), which advantageously represents a discontinuous control law.

[0052] By way of example, the controller unit 52 comprises a control signal calculation component 61 which is designed to calculate the control signal u based on the control deviation 26 and the amplification factor k, in particular according to formula (1).

[0053] The pneumatic control device 1 is preferably designed to adjust the amplification factor k during the sliding mode control. Such a sliding mode control in which the amplification factor k is adjusted should also be referred to as an adaptive sliding mode control. The pneumatic control device 1 is preferably designed to continuously adjust the amplification factor k. In particular, the pneumatic control device 1 is designed to calculate the value of the amplification factor k for each calculation of the value of the control signal u, so as to use this value of the amplification factor k to calculate the value of the control signal u.

[0054] By way of example, the controller unit 52 comprises an amplification factor calculation component 62 which is designed, in particular to continuously adjust the amplification factor k. The amplification factor calculation component 62 is advantageously designed to calculate, in particular to adjust, the amplification factor k based on the replacement variable ξ.

[0055] By adjusting the amplification factor k, the control can be adapted quickly to changes without endangering the stability of the control. For example, by adjusting the amplification factor k, it can be avoided in particular that disturbances in the control variable 18 are not adequately compensated (due to too small amplification factor k) and / or that (due to too large amplification factor k) "chatter" - an uncontrolled overshoot of the control variable 18 - occurs. In particular, the control device 1 is designed to change the amplification factor k during operation so that, although disturbances and / or inaccuracies are compensated, the amplification factor k is not overestimated.

[0056] The pneumatic control device 1 is preferably designed to calculate the amplification factor k based on the square of the substitution variable ξ. For example, the pneumatic control device 1 is designed to calculate the amplification factor k according to the following formula (2):

[0057]

[0058] In formula (2), ξ is a substitution variable.

[0059] By using the substitution variable ξ in this way, it can be ensured that the amplification factor k always remains positive. For this purpose, it is not necessary to introduce a lower limit for the amplification factor k. Advantageously, no lower limit is defined for the amplification factor k.

[0060] Furthermore, the pneumatic control device 1 can be designed to calculate the amplification factor k as the product of the square of the substitution variable ξ and the power factor. The power factor is in particular a quantity raised to the power of the sliding variable σ, which exponent is advantageously between 0 and 1. For example, the pneumatic control device 1 is designed to calculate the amplification factor k according to the following formula (3):

[0061]

[0062] The exponent α is advantageously used to heuristically influence the speed / aggressiveness of the controller.

[0063] Exemplarily, the amplification factor calculation component 62 is designed to calculate the amplification factor k according to formula (2) or formula (3).

[0064] The pneumatic control device 1 is preferably designed to minimize an objective function in order to calculate the amplification factor k. Preferably, the objective function depends on the sliding variable and / or the control signal u, so that the sliding variable and / or the control signal u are minimized by minimizing the objective function. In particular, the objective function is defined in such a way that the square of the distance to the sliding plane, i.e., for example, the square of the sliding variable, and the square of the control signal u are minimized by minimizing, in particular by adjusting the replacement variable ξ. The square of the control signal u represents a measure of the input energy that should be minimized. The minimization of the objective function is advantageously an optimization problem; the minimization of the objective function will also be referred to as the optimization problem hereinafter.

[0065] For example, the aforementioned minimization is the minimization defined in equation (4):

[0066]

[0067] The objective function to be minimized is The objective function to be minimized comprises, for example, the sum of the square of the substitution variable σ (e.g. multiplied by a penalty coefficient Γ-1) and the square of the control signal u. The penalty coefficient Γ-1 describes the (especially unknown) inverse input uncertainty and results from the mathematical modeling of the uncertain system. In the dynamic solution of the optimization problem discussed below, the penalty coefficient Γ-1 advantageously reduces the non-inverse input uncertainty.

[0068] As already mentioned above, the pneumatic control device 1 is designed to calculate the amplification factor k based on the substitution variable ξ. The pneumatic control device 1 is preferably designed to use the substitution variable ξ as a variable of the objective function. The pneumatic control device 1 is advantageously designed to adjust the substitution variable ξ within the framework of minimizing the objective function (i.e. in order to solve the optimization problem) so that the objective function is minimized.

[0069] The pneumatic control device 1 is preferably designed to solve the optimization problem, i.e., the minimization of the objective function, in a numerical manner, in particular based on the gradient descent method. In particular, the pneumatic control device 1 is designed to dynamically solve the optimization problem so that the objective function converges toward a minimum value. For example, the pneumatic control device 1 is designed to solve the optimization problem according to the following formula (5):

[0070]

[0071] Alternatively, the pneumatic control device 1 may be designed to solve the optimization problem using another method, in particular another numerical method, such as using a momentum gradient descent algorithm.

[0072] By calculating the partial derivatives, the following control law described by formulas (6) and (7) is advantageously obtained. The pneumatic control device 1 is preferably designed to calculate the control signal u based on the control deviation 26 according to this control law, ie according to formulas (6) and (7).

[0073]

[0074] γ is a step size factor, in particular a heuristic parameter. The step size factor γ determines the speed of the algorithm, ie in particular the adaptation speed of ξ. The exponent α is advantageously a heuristic parameter. By means of the parameters α and γ, in particular the aggressiveness of the controller and / or the aggressiveness of the adaptation of ξ and thus the adjustment of the amplification factor k can be adjusted. The control device 1 is preferably designed to automatically select the parameters α and γ. In addition, the parameters α and γ can be input into the control device 1, for example, via the input unit 17 and / or via the communication interface 36.

[0075] Power factor |σ| α Advantageously, and optionally, in particular in formula (6) and / or formula (7), the pneumatic control device 1 can be designed to calculate the replacement variable ξ according to the following formula (8) (in particular as an alternative to formula (7)):

[0076]

[0077] The control unit 52 advantageously comprises substitution variable calculation means 63 designed to calculate the substitution variable ξ, in particular within the framework of the above-mentioned minimization of the objective function and / or according to formula (7) or (8).

[0078] In particular, the pneumatic control device 1 is designed to calculate the value of the replacement variable ξ for each calculation of the value of the control signal u, so as to calculate the value of the control signal u based on the calculated value of the amplification factor ξ (in particular by the amplification factor k).

[0079] The pneumatic control device 1 is preferably designed to calculate the amplification factor k based on the replacement variable ξ and to calculate the current value of the replacement variable ξ based on the previous values ​​of the replacement variable ξ. In particular, each current value of the replacement variable ξ is calculated in this way, i.e. based on the previous values ​​of the replacement variable ξ. The control device 1 advantageously iteratively adjusts the replacement variable ξ, in particular continuously and preferably for the purpose of minimizing the objective function.

[0080] Exemplarily, the output of the substitution variable calculation component 63, ie the calculated substitution variable ξ, is fed back to the input of the substitution variable calculation component 63 so as to be used for calculating the next value of the substitution variable ξ.

[0081] For example, the replacement variable calculation component 63 calculates the corresponding current value of the replacement variable ξ based on the previous calculated value of the replacement variable ξ and formula (7) (or alternatively formula (8)) (calculates the amplification factor k based on this value for calculating the control signal u). For example, the replacement variable calculation component 63 substitutes the previously calculated value of the replacement variable ξ into formula (7) (or alternatively formula (8)) to calculate For example, the replacement variable calculation component 63 is based on In particular, by The current value of the substitution variable ξ is calculated by adding it to the previously calculated value of the substitution variable ξ.

[0082] The pneumatic control device 1 is preferably designed to calculate the current value of the replacement variable ξ based on the previous value of the replacement variable ξ and based on the sliding variable σ, ie, for example, based on the control deviation 26. Figure 2 As shown, the control deviation 26 is provided to the replacement variable calculation component 63, and the replacement variable calculation component 63 calculates the current value of the replacement variable ξ based on the control deviation 26, particularly using formula (7) or formula (8).

[0083] As described above, the substitution variable ξ may advantageously be adjusted iteratively.

[0084] The pneumatic control device 1 is advantageously designed to carry out a plurality of iterative steps in order to minimize the objective function and to calculate in each iterative step the corresponding current value of the substitution variable ξ, in particular in such a way that each iterative step aims at minimizing the objective function (in particular in the sense that the objective function is further reduced relative to the previous iterative step).

[0085] The pneumatic control device 1 is advantageously designed to calculate the corresponding current value of the substitution variable ξ based on the value of the substitution variable ξ calculated in the previous iteration step and based on the change of the substitution variable ξ calculated in particular to minimize the objective function, for example as the sum of the value of the substitution variable ξ calculated in the previous iteration step and the change of the substitution variable ξ. For example, the change is The pneumatic control device 1 is designed to calculate the change (of the replacement variable ξ) in each iteration step based on the sliding variable σ and / or based on the replacement variable ξ, in particular according to formula (7) or formula (8).

[0086] The pneumatic control device 1 is advantageously designed to calculate in each iteration step a respective current value of the control signal u based on the respective current value of the replacement variable ξ calculated in the respective iteration step, and thus to control the valve device 12. The pneumatic control device 1 is advantageously designed to calculate a control deviation 26 resulting from the control, and in the next iteration step to calculate the respective next current value of the replacement variable ξ based on the control deviation 26, in particular according to formula (7) or formula (8).

[0087] Exemplarily, by iterative adjustment of the substitution variable ξ, a time curve of the calculated amplification factor k (based on the substitution variable ξ) can be obtained, wherein the amplification factor initially increases until the amplification factor k reaches a threshold value that is large enough to compensate for the uncertainty of the system and converge to the sliding plane. During the convergence of the system to the sliding plane, the adaptive dynamics is increasingly simplified to a differential formula that converges to zero. This means that the amplification factor k decreases again. If it falls below the threshold again, it will start to rise again, and so on. This leads to the convergence of the amplification factor k to the threshold. The threshold value advantageously represents a solution to the optimization problem; for example, the threshold value represents the minimum allowed value of the amplification factor k, i.e., in particular, the minimum value at which the amplification factor k no longer decreases further. If the disturbance / parameter of the system changes, the threshold value changes and the threshold value converges to this new threshold value.

[0088] refer to Figure 3 , a method for iteratively adjusting the replacement variable ξ (in order to thereby adjust the amplification factor k) which is carried out in particular by the control device 1 will be described below.

[0089] The method comprises an optional first step S1 in which a first value of the control signal u is obtained, for example by setting an initial value for the substitution variable ξ or the amplification factor k and calculating the first value of the control signal u based thereon, or setting the initial value as the first value of the control signal u.

[0090] The method comprises a second step S2 in which the valve device 12 is controlled using a first value of the control signal u such that the valve device 12 provides a pneumatic actuating signal 21 as a function of the control signal u in order to influence the actuating variable 18 .

[0091] The method comprises a third step S3 in which a current value of the sliding variable σ, ie, for example, the control deviation 26 , is calculated based on the affected manipulated variable 18 .

[0092] The method comprises a fourth step S4, in which a current value of the replacement variable ξ is calculated based on the current value of the sliding variable σ calculated in step S3, in particular based on a previously obtained value (eg an initial value) of the replacement variable ξ and preferably using formula (7) or (8).

[0093] The method comprises a fifth step S5 in which a current value of the amplification factor k is calculated based on the replacement variable ξ calculated in step S4, in particular based on formula (2) or (3).

[0094] The method comprises a sixth step S6 , in which a current value of the control signal u is calculated based on the amplification factor k calculated in step S5 , in particular based on formula (1).

[0095] The method then returns to step S2 and the valve device 2 is controlled using the current value of the control signal u calculated in step S6 so that the valve device provides the pneumatic actuating signal 21 in accordance with the control signal u in order to influence the actuating variable 18 .

[0096] The method continues with steps S3 , S4 , S5 and S6 , wherein in step S4 a current value of the substitution variable ξ is calculated based on the value of the substitution variable ξ calculated in the previously performed step S4 .

[0097] Steps S2 to S6 are repeated continuously in order to control the manipulated variable 18 and, during this, to continuously adjust the replacement variable ξ (and thus the amplification factor k), in particular in such a way that the objective function is minimized. Execution of steps S2 to S6 advantageously represents the iteration step explained above.

[0098] Preferably, the pneumatic control device 1 is designed to perform a diagnosis based on the amplification factor k and / or the replacement variable ξ. The pneumatic control device 1 is preferably designed to provide a diagnostic function based on the change of the amplification factor k and / or the replacement variable ξ, advantageously taking into account the control signal u, for example in order to detect a leak or an increased wear of the actuator 2, in particular in the absence of obvious signs in the position or control. The diagnostic function advantageously includes providing a warning to the user.

[0099] For example, the pneumatic control device 1 is designed to detect a leakage and optionally quantify the extent of the leakage based on an increase in the amplification factor k and / or a deviation of the control signal u (for example compared to a stored reference control signal).

[0100] Optionally, the controller unit 52 is designed as a high-order sliding mode controller, which acts on high-order derivatives of the sliding variable. Such a high-order sliding mode controller is advantageously designed to calculate the control signal u based on the sliding variable, one or more of their derivatives and an amplification factor. Optionally, the controller unit 52 may also include a super-warp algorithm.

Claims

1. A pneumatic control device (1), in particular a position controller, for controlling a control variable (18) to a setpoint value (19) by providing at least one pneumatic control signal (21), wherein: The control device (1) is designed to perform sliding mode control using a sliding variable and to calculate a control signal (u) for controlling a valve device (12) within the framework of the sliding mode control based on the product of an amplification factor (k) and a sign function acting on the sliding variable, so as to provide the at least one pneumatic control signal (21) using the valve device (12).

2. The pneumatic control device (1) according to claim 1, wherein: The pneumatic control device (1) is designed to adjust the amplification factor (k) during the sliding mode control.

3. The pneumatic control device (1) according to claim 1 or 2, wherein: The pneumatic control device (1) is designed to continuously adjust the amplification factor (k).

4. The pneumatic control device (1) according to any one of the preceding claims, wherein: The pneumatic control device (1) is designed to calculate the amplification factor (k) based on the square of the substitution variable (ξ).

5. The pneumatic control device (1) according to any one of the preceding claims, wherein: The pneumatic control device (1) is designed to calculate an amplification factor (k) based on one or more substitution variables (ξ) and to calculate a current value of the substitution variable based on previous values ​​of the substitution variable.

6. The pneumatic control device (1) according to claim 5, wherein: The pneumatic control device (1) is designed to calculate a current value of the replacement variable (ξ) based on a previous value of the replacement variable (ξ) and based on the sliding variable.

7. Pneumatic control device (1) according to any one of the preceding claims, wherein: The pneumatic control device (1) is designed to perform a minimization of an objective function in order to calculate the amplification factor (k).

8. The pneumatic control device (1) according to claim 7, wherein: The objective function depends on the sliding variable and / or the control signal (u), such that the sliding variable and / or the control signal (u) are minimized by minimizing the objective function.

9. The pneumatic control device (1) according to claim 7 or 8, wherein: The pneumatic control device (1) is designed to calculate the amplification factor (k) based on one or more substitution variables (ξ) and use the substitution variables as variables of the objective function.

10. The pneumatic control device (1) according to claim 9, wherein: The pneumatic control device (1) is designed to perform a plurality of iteration steps in order to minimize the objective function and to calculate in each iteration step the corresponding current value of the substitution variable (ξ).

11. The pneumatic control device (1) according to claim 10, wherein: The pneumatic control device (1) is designed to calculate, in each iteration step, a respective current value of the control signal (u) based on the respective current value of the replacement variable (ξ) calculated in the respective iteration step and to control the valve device (12) accordingly.

12. Pneumatic control device (1) according to any one of the preceding claims, wherein: The pneumatic control device (1) is designed to perform a diagnosis based on the amplification factor (k) and / or one or more substitution variables (ξ), the amplification factor (k) being based on the substitution variable.

13. A process valve unit comprising a position controller designed according to the control device (1) as claimed in any one of claims 1 to 10 and a process valve having a drive (38) and a fitting (39), wherein: The position controller is designed to control the process valve.

14. A method for operating a pneumatic control device (1) according to any one of claims 1 to 12 or a process valve unit according to claim 13, comprising the following steps: The control signal (u) is calculated based on the product of the amplification factor (k) and the sign function.

Images