Adaptive real-time gain scheduling control law for damper test system

Abstract

A damper testing system and a method for operating the testing system are disclosed. The testing system includes an actuator for operating a test specimen having a damping constant and / or a spring constant. The damper testing system and method estimate the impedance of the test specimen in real time; and based on the estimated impedance, adjust one or more parameters in a controller for the actuator in real time.

Description

[0001] Cross-reference to related applications

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 799,732, filed May 4, 2025, entitled “ADAPTIVE REAL-TIME GAIN SCHEDULING CONTROL LAW FOR DAMPER TEST SYSTEMS,” the contents of which are incorporated herein by reference. Background Technology

[0003] When precise control of motion on a test specimen is desired, the test system used to apply controlled forces or displacements to the test specimen is inherently complex. An example of such a test system is a damper test system used to test vehicle shock absorbers, which primarily consists of a damper or vehicle strut, such as a MacPherson strut, which includes a damper and a spring. This damper test system supports the damper or strut, with one end connected to an actuator and the opposite end connected to a reactive structure. Users or operators of damper test systems rarely have a background in control science. Therefore, an ideal damper test system should not require any "tuning," which refers to adjustments made to the user or operator portion of the damper test system. "Tuning" is the process of adjusting the gain in the controller's control laws or control logic. A common control law is the PID (Proportional-Integral-Derivative) controller.

[0004] Some common damper testing systems use hydraulic actuators or rotary electric actuators. Rotary electric crank dynamometers connect a conventional rotary motor to a linear actuator via a crankshaft or crank-connecting rod. In the case of a rotary electric system, the operator does not perform any adjustments. In the case of a hydraulically actuated system, the operator makes minor (but not real-time) adjustments to the PID control, or none at all.

[0005] In damper testing systems, the variable to be controlled is motion, such as displacement or velocity. Using hydraulic actuators or crank dynamometers to control motion is highly unaffected by external forces generated by the damper specimen. In a control sense, this actuator technology is very "stiff." The actuator is inherently unaffected by external force disturbances from the specimen. The control structures of hydraulic actuators and crank dynamometers are relatively simple and well-known. In the case of hydraulic actuators, a single PID loop is typically used to provide a servo valve signal based on the displacement feedback response. In the case of crank dynamometers, conventional constant angular velocity motor control laws are used to control the motor to a fixed, predefined angular velocity.

[0006] The relatively new actuators used in damper testing systems employ linear electric motors (such as permanent magnet synchronous motors). These are sometimes referred to as EMAs (Electric Mechanical Actuators). Some advantages of EMAs over hydraulic actuators are energy efficiency, the absence of oil, and the elimination of the need for a hydraulic pump.

[0007] The testing profile of an EMA (Electronic Dynamic Mechanism) is superior to that of a crank dynamometer because it can be arbitrary (i.e., a road curve, or any composite repeating shape), and is not limited to a fixed-frequency sine curve. Some disadvantages include the complexity of the control laws and the actuator's inability to reject external disturbances caused by damping forces. The most typical control law of an EMA involves three cascaded PID controllers. (Reference) Figure 1 The inner loop 1 controls the current to one or more motors. The force from the motor is proportional to this current. Looking outward, the next loop 2 controls the speed. Finally, the outermost loop 3 controls the displacement. The complexity of this control structure makes commissioning extremely difficult. Therefore, it is unreasonable to expect operators of typical damper testing systems to have the background or time to commission such systems.

[0008] To achieve the desired control (reference command tracking), finding a fixed set of tuning parameters (PID gain) that can be adequately executed across a wide range of specimens is very difficult (and often impractical). Trade-offs must be made to achieve acceptable reference command tracking and robust stability margins. In the specific case of damper testing, the damping constant of the specimen can be drastically varied from one specimen to the next. For EMA actuators, the generated force is proportional to the current, and the desired variable to be controlled is motion. Because the actuator generates force, and the specimen's "resistance" generates a resistive force, the closed-loop control action of the EMA control law is crucial for generating the desired motion in the presence of external force disturbances from the damper specimen. Compared to historical servo hydraulic actuators or crank dynamometer actuators, this EMA actuator is not "rigid" in terms of motion control. The EMA actuator is less affected by such external force disturbances from the damper specimen. In damper testing, any progress in tuning the PID parameters (multiple) in the damper test system will be advantageous. Summary of the Invention

[0009] A method of operating a test system having an actuator for testing a specimen having a damping constant and / or a spring constant, the method comprising: estimating the impedance of the specimen in real time; and adjusting one or more parameters in a controller for the actuator in real time based on the estimated impedance.

[0010] Further embodiments may include one or more of the following features.

[0011] The controller may include one or more PID controllers, and wherein regulation includes adjusting one or more gains of the one or more PID controllers, such that the one or more controllers may include cascaded PID control loops. If desired, the cascaded PID control loops may include three cascaded control loops: a force loop, a velocity loop, and a displacement loop.

[0012] The adjustment may include: adjusting the integral gain in the speed PID controller in the speed loop, and / or the adjustment may include: adjusting the proportional gain in the speed PID controller in the speed loop.

[0013] In one embodiment, the controller may include a backfeedforward model, and tuning includes adjusting one or more parameters in the backfeedforward model.

[0014] Estimating impedance may include estimating the damping constant and / or spring constant of the test specimen. In one embodiment, estimating impedance may include estimating the impedance of the actuator and the test specimen. For example, estimation may include obtaining an estimate of the force used to move the actuator and the force applied to the test specimen. For example, obtaining the estimate of the force may include measuring parameters that characterize the force provided by the actuator.

[0015] Estimating impedance may include estimating the force applied to the specimen rather than the force used to move the actuator. For example, estimating the force applied to the specimen rather than the force used to move the actuator may include acquiring a signal from a force sensor coupled to the specimen in the force transmission path.

[0016] On the other hand, there is a damper testing system comprising: a frame; an actuator, wherein the frame and the actuator are configured to apply force to a specimen; and a controller configured to: control the actuator; estimate the impedance of the specimen in real time; and adjust one or more parameters in the controller for the actuator in real time based on the estimated impedance.

[0017] Further embodiments may include one or more of the following features.

[0018] The controller may include one or more PID controllers, and wherein the controller is configured to adjust one or more gains of the one or more PID controllers, such that the one or more controllers may include cascaded PID control loops. If desired, the cascaded PID control loops may include three cascaded control loops: a force loop, a velocity loop, and a displacement loop.

[0019] The controller can be configured to: adjust the integral gain in the speed PID controller in the speed loop, and / or be configured to: adjust the proportional gain in the speed PID controller in the speed loop.

[0020] The controller may include a backfeedforward model, and the controller is configured to adjust one or more parameters in the backfeedforward model.

[0021] The controller may be configured to estimate the damping constant and / or spring constant of the specimen. For example, the controller may be configured to estimate the impedance of the actuator and the specimen, or to estimate the force used to move the actuator and the estimated force applied to the specimen.

[0022] The controller may be configured to measure parameters characterizing the force provided by the actuator, and / or the controller may be configured to estimate the force applied to the specimen, rather than the force used to move the actuator.

[0023] A force sensor can be coupled to the specimen in the force transmission path, wherein the controller is coupled to receive signals from the force sensor. Attached Figure Description

[0024] Figure 1 It is a schematic diagram of the control logic or control law used in a damper testing system.

[0025] Figure 2 It is a graphical representation of the damper testing system.

[0026] Figure 3 This is a schematic diagram of the control law or control logic of the first embodiment of the present invention used for damper testing machinery.

[0027] Figure 4 This is a schematic diagram of the control law or control logic of the second embodiment of the present invention used for damper testing machinery.

[0028] Figure 5 It is a block diagram showing some components of a computing device. Detailed Implementation

[0029] exist Figure 2 The diagram illustrates a test system 10 for applying a load to a test specimen 12, and the test system 10 includes a base 14. A pair of vertical columns 18 are mounted to the base 14, and the vertical columns 14 extend vertically to support a crossbeam 16 above the base 14. A first specimen support 20 is configured to engage a first end of the test specimen 12. A second specimen support 22 is configured to engage a second end of the test specimen 12. In the illustrated embodiment, the test specimen 12 is a damper (e.g., a vehicle shock absorber), and the test specimen may be or may include a spring 12A, which, if present, is typically referred to as a strut.

[0030] An actuator 40 (e.g., a hydraulic or electric actuator) is mounted to the base 14 and connected to the specimen support 22. A displacement sensor (such as, but not limited to, a linear encoder, LVDT39, etc.) is operatively coupled to the actuator 40 to measure and provide a displacement signal characterizing the displacement of an end of the actuator 40 and the displacement of the specimen support connected to that end. In this embodiment, a force sensor 42 is mounted to the crossbeam 16 and connected to the second specimen support 20.

[0031] Although the test system 10 is specifically designed to test damper specimens (such as in...) Figure 2 The damper specimen (schematically represented by 12) has opposite ends connected to the first damper support 20 and the second damper support 22, but the test system 10 can be configured to test an elastomer device, wherein the concepts described herein can be used to control the operation of the actuator 40.

[0032] Actuator 40 is controlled by system controller 41 according to the test to be performed, and the appropriate force signal is obtained from force sensor 42, while the motion signal can be obtained from displacement sensor, velocity sensor, or acceleration sensor. In the illustrated embodiment, displacement sensor or encoder 39 is used to measure the motion of actuator 40. If desired, a signal characterizing the force can also be obtained by measuring the current supplied to actuator 40, if actuator 40 is implemented as a linear electric motor by a suitable electric drive (not shown), or if actuator 40 is implemented as a hydraulic actuator by a servo hydraulic system (not shown). It should be noted that in the case of a hydraulic actuator, the fluid pressure measurement can also provide a characterization of the force applied to test specimen 12.

[0033] exist Figure 3 The control law or control logic of controller 41 is shown at 80. The desired command displacement signal for actuator 40 is provided at 82. Feedback displacement 84 is received by adder 86 to generate a displacement error at 88. This displacement error is provided to displacement PID controller 90 to generate speed command 92, which is then provided to adder 94; adder 94 is added to speed feedback 96 to provide speed error 98. Speed ​​error 98 is provided to speed PID controller 100 to generate current command 102; current command 102 is then provided to adder 104; adder 104 is added to current feedback 106 to provide current error 108. Current error 108 is provided to current PID controller 110 to generate a voltage for motor 114. Motor 118 has a torque or force constant 116, and the output of motor 118 is force 120.

[0034] One aspect of the invention proposes a control law that estimates the specimen impedance in real time and adjusts (also in real time) the PID gain of one or more controllers 80 to optimize the control law for a specific test specimen installed in the test system.

[0035] At 120, the force of the motor is

[0036]

[0037] in:

[0038] (Primarily the quality of the motor, but to a lesser extent the quality of the test specimen's movement, such as the damper.)

[0039] (Mainly the test specimen damper)

[0040] (Mainly the test specimen damper)

[0041]

[0042]

[0043]

[0044] Alternatively, in the Laplace domain, the transfer function of the system impedance (actuator and test specimen damper) is:

[0045]

[0046] It should be noted that the force of a motor includes inertial force ( This inertial force is used to drive the total mass of the actuator 40's moving mass and, to a lesser extent, the moving mass of the test specimen. However, there is also a force from the test specimen. External forces or disturbances. Generally, the moving mass (primarily the moving mass of actuator 41) is constant, and the controller 80 can be tuned to that moving mass. However, each test specimen is different, and therefore, the forces from the test specimen... The interference force will change, and in order to achieve better performance, the test machine will be able to adjust the controller 80 to take into account the changes in b and k.

[0047] Return to reference Figure 3 At 122, force 120 is applied to the transfer function of the forced response to produce the amount of motion (x, ) characterized at 124. (Outside the control loop, motion 124 enters the test specimen indicated at 126.) In the equation, at 128, motion 124 is measured as force by force sensor 42, or alternatively, the force of the motor can be estimated at 129 based on current feedback at 106 and torque constant or force constant at 116.

[0048] The amount of motion (x, System identification is performed using the force applied to the test specimen (characterized at 124) and the force (measured herein by force sensor 42). One form of system identification is achieved by using estimator 130. In other words, the force on the test specimen can be estimated from the measurements of force sensor 42, and the force can be estimated from the actuator displacement measurements and their derivatives (x, The motion (displacement and higher-order derivatives) is estimated using [a method / mechanism]. It is important to note that in this context, measurements are considered a form of estimation. The dynamic parameters representing the system impedance are m, b, and k. More sophisticated specimen impedance models can be used, and nonlinear specimens can be employed. These model parameters can be estimated using various techniques, such as Kalman filters, or another form of estimator or observer, or least-squares optimization algorithms, or other forms of system identification.

[0049] By estimating the key parameters b and k via means such as an observer, these values ​​can be used to change one or more gains in controller 80. It has been found that the gains in the speed PID controller 100 can have a significant impact on system performance. In particular, an estimate of the damping b can be used to adjust the gain of the integral element in the speed PID controller 100, and / or, an estimate of the spring constant k can be used to adjust the gains of the proportional and integral elements of the speed PID controller 100. However, this should not be considered limiting, as other gains in controller 80 can be adjusted. A convenient way to determine the appropriate gain adjustment for b and / or k is to use a lookup table (embodied herein in scheduler 110), although other techniques for relating b to integral gain adjustment and / or other techniques for relating k to proportional and integral gain adjustment (such as through linear or nonlinear representations) can be employed.

[0050] The preceding section described a method for adjusting one or more PID tuning parameters based on real-time estimates of the specimen impedance. Similar to adjusting PID tuning parameters, another technique or method can be used to estimate the dynamic parameters in a feedforward model control scheme. (Reference) Figure 4 The inverse feedforward model 150 can be a test specimen with a linear or nonlinear representation. The inverse feedforward model 150 (e.g., a digital filter, a nonlinear model, an analog model, etc.) will receive the command displacement signal 82. Furthermore, the feedforward correction can also be implemented as a pre-filter (applied at the front end of the summation point), where the input to the filter is a reference command, and the output from the filter can be a modified reference command to the PID summation point.

[0051] It should be noted that if the test specimen consists only of a damper, then only the damping b can be the only parameter (in the integral gain, the only parameter estimated by the corresponding adjustment is used), for example, in one embodiment, in Figure 3 The speed controller 100 in the embodiment is the only parameter. However, if the test specimen includes a spring or a damper with spring 12A, the spring constant k can also be estimated, for example, in one embodiment, by utilizing corresponding adjustments of the proportional and integral gains in the speed controller 100. Additionally, any other PID gain in the three cascaded loops (displacement, velocity, or current) can be adjusted based on the estimated dynamic parameters. The operator can use a user-operated input selector (such as a button) to identify the test specimen as including a damper and / or a spring. Depending on the selection, different estimation algorithms can be used.

[0052] In one embodiment, the rate at which the estimator 130 identifies the specimen impedance parameters and the rate at which the PID controller gain changes will be adjustable. It is desirable that the convergence or adaptive rate of these terms is slower than the PID correction rate (e.g., several orders of magnitude slower) to avoid interaction between the two feedback mechanisms.

[0053] If desired, limits can be set for the permissible parameter estimates (e.g., b and k) and the permissible adjustable controller gain.

[0054] In one embodiment, the estimator 130 may be allowed to make estimations only when exposed to an input range such as a specific range of motion or a specific range of damping forces. This may be desirable to prevent the estimator 130 from making incorrect estimates from an excitation that may be too large or too small and that is susceptible to errors due to signal-to-noise ratio and nonlinearity.

[0055] In one embodiment, prior to actual testing, the user may be advised to perform some initial learning loops to allow the estimator and subsequent PID gain tuning to learn the characteristics of the test specimen. Similarly, it may be desirable that when a new test specimen is installed, or after the installation (e.g., when the test machinery is powered on, or when the test machinery is otherwise enabled), the gain reverts to a set of preset values ​​that may be suitable for many (if not all) test specimens. This is helpful for new test specimens to be tested, so as to avoid having inappropriate gains. Therefore, the increase in gain via estimator 130 and thus the decrease in stability will only occur after an estimation of the current specimen has already occurred.

[0056] For example, the gain can be set to a value that makes the damping b equal to zero, which would be the case without any test specimen. However, although performance may be lacking, the test mechanism is in a very stable state. Once the test specimen is installed and the test mechanism is turned on, the estimator 130 will determine an estimate of the impedance parameters. The speed at which it determines this estimate may depend on the convergence rate of the estimator 130, the value of the Kalman filter gain, or other parameters of the estimator being used.

[0057] Normally, estimator 130 will always be running in the background, but this is not required.

[0058] The controller 41 can be implemented on a digital computer and / or an analog computer. Figure 5 The accompanying discussion provides a brief general description of a suitable computing environment in which controller 41 can be implemented. While not strictly necessary, the controller is described at least in part within the general context of computer-executable instructions (such as program modules and block diagrams) executed by computer 219. Typically, program modules include routines, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. Those skilled in the art can implement the above description and / or block diagrams as computer-executable instructions capable of being stored on a computer-readable medium. Furthermore, those skilled in the art will understand that the invention can be practiced using other computer system configurations, including multiprocessor systems, networked personal computers, minicomputers, mainframe computers, etc. Various aspects of the invention can also be practiced in distributed computing environments, where tasks are performed by remote processing devices linked via a communication network. In a distributed computing environment, program modules can reside in both local and remote memory storage devices.

[0059] Figure 5The computer 219 shown includes a conventional computer having a central processing unit (CPU) 227, memory 233, and a system bus 235 that connects various system components, including memory 233, to the CPU 227. The system bus 235 can be any one or more of several types of bus architectures, including a memory bus or memory controller, a peripheral bus, and a local bus using any one or more of various bus architectures. Memory 233 includes read-only memory (ROM) and random access memory (RAM). A basic input / output system (BIOS) is stored in the ROM, containing basic routines (such as during startup) that facilitate the transfer of information between components within the computer 219. Storage devices 237 (such as hard disks, disk drives, optical disk drives, etc.) are connected to the system bus 235 and used to store programs and data. Those skilled in the art will understand that other types of computer-readable media accessible by a computer (such as magnetic tape cassettes, flash memory cards, digital video disks, random access memory, read-only memory, etc.) can also be used as storage devices. Generally, a program is loaded into memory 233 from at least one of the storage devices 237, which may or may not have accompanying data.

[0060] Input devices (such as keyboard 241 and pointing device (mouse) 243) allow the user to issue commands to computer 219. Screen 245 or other types of output devices are also connected to system bus 235 via a suitable interface and provide feedback to the user. If screen 245 is a touchscreen, the pointing device 243 can be combined with it. Screen 245 and the input pointing device 243 (such as a mouse), together with corresponding software drivers, form a graphical user interface (GUI) 247 for computer 219, which is particularly useful for the aspects described below.

[0061] Interface 249 on system controller 41 allows communication. Interface 249 also represents circuitry for sending or receiving signals as described above, and accepting other parameters from the physical system (such as the state of the lock, the state of the door, the state of the identifier, whether power is applied, etc.). Typically, such circuitry includes digital-to-analog converters (D / A) and analog-to-digital converters (A / D) known in the art. As is known in the art, controller 41 may also include an analog controller, which may or may not have digital supervision. The functionality of controller 41 can be combined into a computer system. In another computing environment, controller 41 is a single-board computer capable of operating on a network bus of another computer or another supervising computer. Figure 5The diagrams in the diagrams are intended to represent, in general, the computers used in these computing environments and other suitable computing environments.

[0062] Detailed embodiments of the invention have been disclosed herein; however, it should be understood that the disclosed embodiments are merely examples of the invention and that the invention may be implemented in various forms. Therefore, the specific structural and functional details disclosed herein should not be construed as limiting, but rather as the basis for interpretation of the claims and as a teaching for those skilled in the art to adopt the representative basis of the invention differently in any suitable detailed structure contemplated. In particular, features presented and described in individual dependent claims may be applied in combinations, and any advantageous combination of such claims disclosed herein may be applied to features presented and described in individual dependent claims.

[0063] Furthermore, the terms and phrases used herein are not intended to be limiting, but rather to provide an understandable description of the invention. As used herein, the term “a” or “an” is defined as one or more. As used herein, the term “a plurality” is defined as two or more. As used herein, the term “another” is defined as at least a second or more. As used herein, the terms “containing” and / or “having” are defined as including (i.e., open-ended language). As used herein, the terms “connected” or “coupled” are defined as being connected, although not necessarily directly connected.

Claims

1. A method of operating a test system, the test system having an actuator, the test system being used to test a specimen having a damping constant and / or a spring constant, the method comprising: The impedance of the specimen is estimated in real time; and One or more parameters in the controller for the actuator are adjusted in real time based on the estimated impedance.

2. The method as described in claim 1, wherein, The controller includes one or more PID controllers, and the adjustment includes adjusting one or more gains of the one or more PID controllers.

3. The method as described in claim 2, wherein, The one or more controllers include cascaded PID control loops.

4. The method of claim 3, wherein, The cascaded PID control loop includes three cascaded control loops: a force loop, a velocity loop, and a displacement loop.

5. The method of claim 4, wherein, The adjustment includes adjusting the integral gain in the speed PID controller within the speed loop.

6. The method of claim 4, wherein, The adjustment includes adjusting the proportional gain in the speed PID controller within the speed loop.

7. The method of claim 1, wherein, The controller includes a backfeedforward model, and the adjustment includes adjusting one or more parameters in the backfeedforward model.

8. The method according to any one of claims 1 to 7, wherein, Estimating the impedance includes estimating the damping constant and / or spring constant of the specimen.

9. The method according to any one of claims 1 to 8, wherein, Estimating impedance includes estimating the impedance of the actuator and the test piece.

10. The method of claim 9, wherein, The estimation includes obtaining the force required to move the actuator and an estimated value for applying the force to the specimen.

11. The method of claim 10, wherein, The force estimate is obtained by measuring parameters that characterize the force provided by the actuator.

12. The method according to any one of claims 9 to 11, wherein, Estimating impedance includes estimating the force applied to the specimen, rather than the force used to move the actuator.

13. The method of claim 12, wherein, Estimating the force applied to the specimen, rather than the force used to move the actuator, includes acquiring a signal from a force sensor coupled to the specimen in the force transmission path.

14. A damper testing system, comprising: frame; An actuator, wherein the frame and the actuator are configured to apply force to a specimen; The controller is configured to: Control the actuator; The impedance of the specimen is estimated in real time; and One or more parameters in the controller for the actuator are adjusted in real time based on the estimated impedance.

15. The system of claim 14, wherein, The controller includes one or more PID controllers, and wherein the controller is configured to adjust one or more gains of the one or more PID controllers.

16. The system of claim 15, wherein, The one or more controllers include cascaded PID control loops.

17. The system of claim 16, wherein, The cascaded PID control loop includes three cascaded control loops: a force loop, a velocity loop, and a displacement loop.

18. The system of claim 17, wherein, The controller is configured to adjust the integral gain of the speed PID controller in the speed loop.

19. The system of claim 17, wherein, The controller is configured to adjust the proportional gain in the speed PID controller within the speed loop.

20. The system of claim 14, wherein, The controller includes a backfeedforward model, and the controller is configured to adjust one or more parameters in the backfeedforward model.

21. The system as claimed in any one of claims 14 to 20, wherein, The controller is configured to estimate the damping constant and / or spring constant of the specimen.

22. The system as claimed in any one of claims 14 to 21, wherein, The controller is configured to estimate the impedance of the actuator and the test piece.

23. The system of claim 22, wherein, The controller is configured to estimate the force required to move the actuator and apply it to the specimen.

24. The system of claim 23, wherein, The controller is configured to measure parameters that characterize the force provided by the actuator.

25. The system as claimed in any one of claims 22 to 24, wherein, The controller is configured to estimate the force applied to the specimen, rather than the force used to move the actuator.

26. The system of claim 25, further comprising a force sensor coupled to the specimen in the force transmission path, wherein, The controller is coupled to receive signals from the force sensor.

Images