Method for operating control unit, computer program product, control unit and motor vehicle comprising said control unit

Abstract

The invention discloses a method for operating a controller, a computer program product, a controller and a motor vehicle comprising the controller. The invention relates to a method for operating a controller (10) for closed-loop control of a controlled system (8), comprising the following steps: (S100) introducing a variable value that affects a frequency response of the controlled system (8); (S200) determining transformation factors (alpha, beta, gamma) according to the imported variable values; and (S300) performing commissioning of the controller (10) for the modified frequency response of the controlled system (8) using the determined transformation factor (alpha, beta, gamma).

Description

Technical Field

[0001] The present invention relates to a method for operating a controller, a computer program product and a controller, and a motor vehicle comprising said controller. Background Technology

[0002] In many systems, resonance must be controlled to ensure system stability. In vibrational systems, there may be a tendency towards underdamped vibration in the resonant frequency region. Examples from the automotive field include chassis and structural vibrations or noise, or even trailer operation under certain conditions. However, similarly, other components (such as pneumatic, hydraulic, or electrical components) may also exhibit a tendency towards underdamped vibration in the resonant frequency region. In many such cases, the resonant frequency depends on one or more parameters that can be measured or estimated. For example, in the case of a chassis, this parameter is the payload. The total mass of a motor vehicle can be continuously estimated and is therefore known. Then, for example, the controller used for a semi-active chassis can be tuned.

[0003] US 5,127,533 A discloses a method for operating a controller, wherein the damping of load oscillation is improved by adjusting the sampling frequency of the controller as the cable length changes. Summary of the Invention

[0004] Therefore, the object of this invention is to demonstrate other ways to tune the controller for changes in resonant frequency caused by variable values.

[0005] The object of this invention is achieved by a method for operating a controller for closed-loop control of a controlled system, the method comprising the following steps: Import variable values ​​that affect the frequency response of the controlled system; The transformation factor is determined based on the imported variable values; and The state-space representation of the controller is discretized using the determined transformation factor to tune the controller for the modified frequency response of the controlled system.

[0006] Therefore, by importing variable values ​​of the frequency response (e.g., the location of one or more resonant frequencies of the controlled system predetermined), a transformation factor is determined, and this transformation factor is used to tune the controller.

[0007] In this way, the controller (especially its bandwidth) can be tuned for the modified frequency response of the controlled system (especially its altered resonant frequency).

[0008] Therefore, the controller itself should be tuned to the modified bandwidth rather than the sampling frequency at which the system's signal waveform is sampled.

[0009] In this way, the controller can be tuned to adjust for changes in resonant frequency caused by variable values.

[0010] Therefore, the controller of a semi-active chassis can be tuned for the modified frequency response of the controlled system. However, this also allows the controller to be tuned for other controlled systems, where the controlled system is any vibration system.

[0011] According to one embodiment, controller debugging includes discretizing the controller's state-space representation. Here, the state-space representation is understood as a system description, such as the description of a linear time-invariant transfer system. All relationships between input variables, output variables, and state variables are represented in matrix and vector form. The state-space representation includes two equations: a first-order state differential equation and an output equation.

[0012] In this way, the controller can be tuned particularly easily for changes in resonant frequency caused by variable values.

[0013] However, the discretization of the controller's state-space representation can be omitted, and instead, for example, transformation factors can be used directly to tune the discrete system matrix and the discrete input matrix.

[0014] This can even further simplify the debugging of controllers for changes in resonant frequency.

[0015] According to another embodiment, the Tustin equation is used to perform discretization. This allows for the use of robust controllers, such as H... ∞ Controller.

[0016] According to another embodiment, the imported value indicates a load. This load can be, for example, the load of a motor vehicle that alters suspension or damping behavior, and thus changes the frequency response of the vehicle's chassis. This load can be determined, for example, during vehicle startup. Startup here refers to starting before the journey begins, such as turning on the ignition switch. This can, for example, improve the damping of the vehicle's chassis.

[0017] The present invention also relates to computer program products and controllers, as well as motor vehicles comprising said controllers. Attached Figure Description

[0018] The invention will now be described with reference to the accompanying drawings, in which: Figure 1 A schematic diagram of a motor vehicle is shown; Figure 2 Show Figure 1 A schematic diagram of the components of a model of a motor vehicle chassis; Figure 3 Show Figure 1A schematic diagram of the components of the control ring of a motor vehicle; Figure 4 The diagram illustrates an operation according to a first exemplary embodiment. Figure 1 A schematic diagram of the process flow for a motor vehicle is shown. Figure 5 The following is illustrated for operation according to a second exemplary embodiment. Figure 1 A schematic diagram of the process flow for a motor vehicle is shown. Figure 6 The third exemplary embodiment is shown for operation Figure 1 The diagram shows a process flow chart for a motor vehicle. Detailed Implementation

[0019] First refer to Figure 1 .

[0020] It shows motor vehicle 2.

[0021] In this exemplary embodiment, motor vehicle 2 is a passenger car. As a variation of this exemplary embodiment, motor vehicle 2 may also be other land vehicles, such as commercial vehicles, like trucks or buses.

[0022] The motor vehicle 2 has a chassis 4. Here, chassis 4 refers to all components in the motor vehicle 2 that are connected to the road via the wheels 6a and 6b of the motor vehicle 2.

[0023] Now also refer to Figure 2 .

[0024] It shows parts of a 1 / 4 scale model 12 of chassis 4.

[0025] These components are the body mass m of motor vehicle 2. Body The mass m of one of wheels 6a and 6b wheel The spring constant K of the springs associated with specific wheels 6a and 6b susp And the spring constant K of one of the tires of wheels 6a and 6b. Tire Furthermore, there is the damping coefficient b of the damper associated with specific wheels 6a and 6b. Susp .

[0026] The resonant frequency f of model 1 / 4 without effective load r0 With effective load m Payload The resonant frequency f decreases with increasing load, while the resonant frequency f under effective load conditions decreases. r Then it is given by the following formula: (Equation 1).

[0027] Now also refer to Figure 3 .

[0028] The diagram shows components of a control loop 14, which includes a controlled system 8 and a controller 10, which in this exemplary embodiment is the controller of a semi-active closed-loop damping control system for the chassis 4. Sensors detect the actual driving conditions and then actively intervene in the chassis tuning of the chassis 4 of the motor vehicle 2. This can improve the suspension and / or damping in terms of improving driving comfort and / or enhancing driving safety.

[0029] In this exemplary embodiment, the controlled system 8 is based on a 1 / 4 model of the chassis 4, and the controller 10 in this exemplary embodiment is a robust controller, such as H... ∞ Controller.

[0030] In this exemplary embodiment, the controller 10 is described by the following state-space representation, which includes a first-order state differential equation (Equation 2) and an output equation (Equation 3).

[0031] (Equation 2) (Equation 3).

[0032] According to a first exemplary embodiment, the two time-continuous state equations are discretized using the Tustin equation.

[0033] (Equation 4).

[0034] Then, the system matrix A, input matrix B, output matrix C, and feedforward matrix D, discretized in this way, are shown below: (Equations 5a-5d) And α = 2 / T s0 Where α is the transformation factor, and T s0 That is the sampling time.

[0035] Therefore, sampling time T s0 Keep constant, but according to the following equation for the effective load m payload To adjust the transformation factor α: (Equation 6).

[0036] The controller 10 may have appropriately designed hardware and / or software components for this purpose.

[0037] As a variation of this exemplary embodiment, when the controlled system 8 is any vibration system other than the chassis 4 or the 1 / 4 model 12, the controller 10 described above can be tuned for the modified frequency response of the controlled system 8.

[0038] According to another exemplary embodiment, the dynamic behavior of the time-continuous controller is approximated by a time-discrete signal as follows: (Equation 7) against x k+1 Solving equation 7 yields: (Equation 8).

[0039] Discrete system matrix A adapt and discrete input matrix B adapt The corresponding relationship can be directly deduced from equation 8: ; (Equations 10a-10b) However, the corresponding output matrix C and feedforward matrix D remain unchanged.

[0040] The further transformation factor β obtained here is shown below: (Equation 11).

[0041] The controller 10 may have appropriately designed hardware and / or software components for this purpose.

[0042] Therefore, this eliminates the need for discretization of the state-space representation of controller 10. Instead, the transformation factor β is used directly to tune the discrete system matrix A. adapt and discrete input matrix B adapt .

[0043] As a variation of this exemplary embodiment, when the controlled system 8 is any vibration system other than the chassis 4 or the 1 / 4 model 12, the controller 10 described above can be tuned for the modified frequency response of the controlled system 8.

[0044] According to another exemplary embodiment, for the payload m payoad To debug the sampling time T of controller 10 s0 The transformation factor γ is determined as follows: (Equation 12) The modified sampling time T is then determined using the transformation factor γ. s : (Equation 13).

[0045] The controller 10 may have appropriately designed hardware and / or software components for this purpose.

[0046] As a variation of this exemplary embodiment, when the controlled system 8 is any vibration system other than the chassis 4 or the 1 / 4 model 12, the controller 10 described above can be tuned for the modified frequency response of the controlled system 8.

[0047] Now also refer to Figure 4 To explain the method flow according to the first exemplary embodiment.

[0048] In the first step S100 of this exemplary embodiment, the controller 10 imports a variable value, which in this exemplary embodiment indicates the payload m. Payload The value of .

[0049] In a further step S200 of this exemplary embodiment, the controller 10 uses Equation 6 to determine the transformation factor α, which is the imported payload m. Payload A function of (and other factors).

[0050] In a further step S300 of this exemplary embodiment, the controller 10 uses equations 5a-5d and the determined transformation factor α (and other factors) to discretize the state-space representation given by equations 2 and 3, so as to thereby discretize the state-space representation given by the controlled system 8 for the payload m. payload Modify the frequency response to debug controller 10.

[0051] Now also refer to Figure 5 To explain the method flow according to the second exemplary embodiment.

[0052] The first step S100 according to the second exemplary embodiment corresponds to the first step S100 according to the first exemplary embodiment.

[0053] In a further step S200 of this exemplary embodiment, the controller 10 uses Equation 11 to determine the transformation factor β, which is the imported payload m Payload A function of (and other factors).

[0054] In a further step S300 of this exemplary embodiment, the controller 10 is tuned using equations 10a and 10b and the determined transformation factor β, so as to adjust the load m of the controlled system 8 accordingly. payload Modify the frequency response to debug controller 10.

[0055] Now also refer to Figure 6 To explain the method flow according to the third exemplary embodiment.

[0056] The first step S100 according to the third exemplary embodiment corresponds to the first step S100 according to the first exemplary embodiment.

[0057] In a further step S200 of this exemplary embodiment, the controller 10 uses equation 12 or 13 to determine the transformation factor γ, which is the imported payload m. Payload A function of (and other factors).

[0058] In a further step S300 of this exemplary embodiment, the controller 10 is tuned using the determined transformation factor γ so as to adjust the load m of the control system 8 accordingly. payload Modify the frequency response to debug controller 10.

[0059] As an alternative to this exemplary embodiment, the order of the steps may also be different. Furthermore, multiple steps may be performed simultaneously or synchronously. Additionally, as an alternative to this exemplary embodiment, individual steps may be skipped or omitted.

[0060] In this way, the controller 10 can be tuned for changes in resonant frequency caused by variable values ​​affecting the frequency response of the controlled system 8. The controlled system 8 can be the chassis 4, the 1 / 4 scale model 12, or any vibration system.

[0061] List of reference numerals 2 motor vehicles 4 chassis 6a wheels 6b wheels 8 Controlled Systems 10 controllers 12 1 / 4 model 14 control loops System A matrix A adapt System Matrix B Input Matrix B adapt Input matrix b Susp Damping coefficient C output matrix D feedforward matrix f r Resonance frequency with effective load f r0 Resonance frequency without effective load K Susp Spring constant K Tire Spring constant m Body Vehicle body quality m wheel Wheel mass m Payload Payload T s Modified sampling time T s0 Sampling time α transformation factor β transformation factor γ transformation factor S100 steps S200 Steps S300 steps.

Claims

1. A method for operating a controller (10) for closed-loop control of a controlled system (8), the method comprising the steps of: (S100) Introduce variable values ​​that affect the frequency response of the controlled system (8); (S200) Determine the transformation factors (α, β, γ) based on the imported variable values; and (S300) The controller (10) is tuned for the modified frequency response of the controlled system (8) using the determined transformation factors (α, β, γ).

2. The method according to claim 1, wherein the debugging of the controller (10) includes the discretization of the state space representation of the controller (10).

3. The method of claim 2, wherein the discretization is performed using the Tustin equation.

4. The method according to any one of claims 1-3, wherein the imported value indicates the effective load (m Payload ).

5. A computer program product designed to perform the method according to any one of claims 1 to 4.

6. A controller (10) for closed-loop control of a controlled system (8), wherein the controller (10) is designed to introduce variable values ​​that affect the frequency response of the controlled system (8), determine transformation factors (α, β, γ) based on the introduced variable values, and perform tuning of the controller (10) for the modified frequency response of the controlled system (8) using the determined transformation factors (α, β, γ).

7. The controller (10) according to claim 6, wherein the debugging of the controller (10) includes the discretization of the state space representation of the controller (10).

8. The controller (10) according to claim 7, wherein the controller (10) is designed to perform the discretization using the Tustin equation.

9. The controller (10) according to any one of claims 6 to 8, wherein the imported value indicates the payload (m Payload ).

10. A motor vehicle (2) having a controller (10) according to any one of claims 6 to 9.

Images