Piezoelectric active support rotor damping device and adaptive control method

Abstract

The application discloses a piezoelectric active support rotor damping device and an adaptive control method, and the device comprises a rotor system, a piezoelectric active support, a displacement sensor, a rotating speed sensor, a programmable digital controller and a power amplifier. The piezoelectric active support comprises a frame, a movable bearing sleeve, a rolling bearing and a piezoelectric actuator; the piezoelectric actuator is arranged between the outer wall of the movable bearing sleeve and the inner wall of the frame in the radial direction. The displacement sensor collects vibration displacement signals of the rotor disc in two orthogonal radial directions, the rotating speed sensor collects instantaneous rotating speed and phase signals, the programmable digital controller outputs alternating control voltage according to the signals through an adaptive feedforward algorithm, and the piezoelectric actuator is driven to generate radial active control force through DC bias superposition and power amplification. The piezoelectric actuator can output active control force according to the real-time vibration state of the rotor, and effectively suppresses the one-frequency synchronous vibration response of the rotor in the process of cross-critical variable rotating speed.

Description

Technical Field

[0001] This invention belongs to the field of vibration control technology for aero-engine rotor systems, specifically relating to an active support rotor vibration reduction device and adaptive control method for an engine. Background Technology

[0002] During operation, rotating machinery such as aero engines experiences periodic centrifugal excitation forces due to unavoidable mass imbalances in the rotor system, causing synchronous vibrations at the same frequency as the rotational speed. This is particularly pronounced during startup, shutdown, and operation under varying conditions, as the rotor often traverses first- or even multiple critical speed regions, easily leading to significant vibration responses. Excessive vibration reduces the service life of shafts, bearings, and related components, impacting the overall stability and safety of the machine. Therefore, it is necessary to design efficient vibration control support devices for the rotor system.

[0003] Traditional passive supports rely on fixed stiffness and damping parameters to suppress vibration. However, these parameters are usually determined during the design phase and are difficult to adjust in real time according to changes in rotational speed and operating conditions, resulting in limited vibration suppression effectiveness under variable speed conditions. Active support technology, by introducing sensors, controllers, and actuators at the support location, can output active control forces based on the real-time vibration state of the rotor, thereby improving the system's vibration suppression capability under complex operating conditions. Piezoelectric actuators, due to their advantages of fast response speed, high power density, and compact structure, have good application potential in the active support of rotor systems.

[0004] However, in existing piezoelectric active support schemes, the piezoelectric actuator is usually used only as an independent excitation or actuation element, failing to integrate the piezoelectric actuator as both a load-bearing component and an active control force output unit into the support structure. This results in a complex device structure, large assembly space requirements, and difficulty in achieving effective active support within a compact casing space. Furthermore, existing active support control methods are mostly designed for constant speed conditions, lacking sufficient synchronous vibration tracking performance during rotor variable speed operation, making it difficult to meet the vibration control requirements of the rotor system under transcritical variable speed conditions.

[0005] Therefore, proposing a compact piezoelectric active support rotor vibration reduction device suitable for variable speed operating conditions has significant engineering value. Summary of the Invention

[0006] In view of the above problems, the present invention provides a technical solution for vibration reduction of a piezoelectric active support rotor, which solves the technical problems that the existing passive support parameters are fixed and cannot adapt to variable speed conditions, and the existing active support has insufficient variable speed tracking performance, resulting in poor vibration suppression effect during transcritical operation.

[0007] To achieve the above objectives, in a first aspect, the present invention provides a piezoelectric active support rotor vibration damping device, the device comprising a rotor system, a piezoelectric active support, a displacement sensor, a speed sensor, a programmable digital controller, and a power amplifier; The piezoelectric active support includes a frame, a movable bearing sleeve, a rolling bearing, and a piezoelectric actuator. The frame serves as a fixed base and is rigidly connected to the casing platform of the rotor system. The movable bearing sleeve and the rolling bearing cooperate to form a movable bearing component, and the rolling bearing cooperates with the rotating shaft of the rotor system. The piezoelectric actuator is arranged radially between the outer wall of the movable bearing sleeve and the inner wall of the frame, serving as a load-bearing component to support the movable bearing component and the rotating shaft of the rotor system, and also as an output unit for active control force. The displacement sensor is arranged on the rotor system casing platform, aligned with the rotor disk measuring point of the rotor system, and collects the vibration displacement signals of the rotor disk in two orthogonal radial directions in real time, and inputs them into the programmable digital controller as control error signals in the corresponding directions respectively. The speed sensor is located on the rotor system casing platform to collect the instantaneous speed and phase signals of the rotor system in real time, and inputs them as a synchronization reference signal into the programmable digital controller. The programmable digital controller internally deploys an adaptive feedforward algorithm, which updates the control parameters in two orthogonal radial directions based on the signals from the displacement sensor and the speed sensor, and outputs alternating control voltage signals in the corresponding directions; a DC bias is superimposed on the alternating control voltage signals to form a biased control voltage signal. The power amplifier amplifies the biased control voltage signal output by the programmable digital controller by a corresponding factor, enabling the piezoelectric actuator to generate active control force within the permissible unidirectional voltage range.

[0008] Furthermore, the piezoelectric active support also includes a mechanical preload mechanism, which includes a preload bolt, a preload spring, and a spring guide rail; The preload bolt is fitted with the frame, with one end abutting against the preload spring; the spring guide rail is fitted with the movable bearing sleeve, guiding the preload force provided by the preload spring to be perpendicular to the contact surface of the movable bearing sleeve; the preload spring is arranged between the preload bolt and the spring guide rail, with one end abutting against the preload bolt and the other end abutting against the inner wall of the spring guide rail.

[0009] Furthermore, the piezoelectric active support also includes a force sensor, which is connected in series between the piezoelectric actuator and the frame to monitor the dynamic load borne by the piezoelectric active support in real time.

[0010] Furthermore, the output end of the piezoelectric actuator is provided with a ball head or hinge structure, which abuts against the outer wall of the movable bearing sleeve.

[0011] Furthermore, a radial clearance is provided between the movable bearing sleeve and the frame.

[0012] Furthermore, the spring guide rail is a hollow sleeve structure and is installed on the travel path of the preload spring.

[0013] In a second aspect, the present invention provides an adaptive control method for a piezoelectric active-supported rotor vibration damping device, the method comprising the following steps: S1: Construct a piezoelectric active support rotor vibration damping device as described in the first aspect of the present invention; S2: Before the control closed loop is started, the secondary channel transmission characteristics between the piezoelectric active support and the rotor vibration measurement point are obtained by modeling or testing. The secondary channel transmission characteristics are used to characterize the dynamic transmission relationship of the control signal to the measurement point through the power amplifier, piezoelectric actuator, movable bearing sleeve and rotor system shaft. The secondary channel characteristics are discretized into a digital filter model and deployed in the programmable digital controller. S3: Use a displacement sensor to collect vibration displacement signals in two orthogonal radial directions at the rotor measuring point, and use them as control error signals in the corresponding directions. Use a speed sensor to collect the instantaneous speed and instantaneous phase signals of the rotor, and use them as the basis for constructing the synchronization reference signal. S4: The programmable digital controller executes an adaptive feedforward control algorithm, which calculates and outputs the control voltage signal in real time based on the measured control error signal, the rotor instantaneous phase signal, and the deployed digital filter model. S5: The power amplifier amplifies the biased control voltage signal output by the programmable digital controller by a corresponding factor, driving the piezoelectric actuator to output active control force.

[0014] Furthermore, step S4 specifically includes: S41: Based on the rotor instantaneous phase angle Construct an orthogonal synchronization reference signal that is synchronized with the rotor speed and frequency. The calculation formula is as follows: The orthogonal reference signal is used to track the first harmonic synchronous vibration response caused by mass imbalance, where t represents the time quantity of the simulated continuous domain and n represents the discrete sampling time number. S42: Use a secondary channel digital filter to process the quadrature synchronization reference signal. Perform pre-filtering to obtain the filtered reference signal. The calculation formula is as follows: ,in, This represents the vector of discrete coefficients of the secondary channel predictor filter, based on the filter reference signal. and corresponding control error signals The adaptive weights of the narrowband FxLMS algorithm are updated in real time, and a leakage factor is introduced during the weight update process to suppress the continuous accumulation of adaptive weights. The weight update formula is as follows: ,in, This represents the weight at time n+1. This represents the weight at time n. This represents the control error signal at time n. Indicates the leakage factor. Indicates the convergence step size; S43: Calculate the alternating control voltage signal in the corresponding direction based on the updated adaptive weights and the orthogonal synchronization reference signal. The calculation formula is as follows: A DC bias voltage is superimposed inside the programmable digital controller to form a biased control voltage signal. This ensures that the piezoelectric actuator operates within the permissible unidirectional voltage range, wherein, This represents the transpose of the weight at time n.

[0015] Furthermore, in step S4, the convergence step size and leakage factor of the adaptive feedforward control algorithm are determined through the following debugging process, specifically including: Through speed-up tests on a rotor test bench or numerical simulations, combined with the rotor vibration response measured by displacement sensors, the dynamic support load monitored by force sensors, and the actual drive voltage signal output by the power amplifier, the convergence step size and leakage factor are gradually adjusted until the vibration peak of the rotor is significantly suppressed during the process of crossing the critical speed, and the control voltage after bias does not show continuous drift, excessive growth, or exceed the preset physical limit. The corresponding convergence step size and leakage factor at this time are used as the fixed operating parameters of the adaptive feedforward control algorithm in step S4.

[0016] Furthermore, the piezoelectric active support rotor vibration damping device also includes a force sensor; In step S4, the programmable digital controller, while executing the adaptive feedforward control algorithm, also performs online closed-loop correction of the piezoelectric actuator's hysteresis nonlinearity, specifically including the following steps: S401: In the initial stage after the control closed loop is started, the programmable digital controller first applies a set of calibration voltage sequences with increasing amplitude to the piezoelectric actuator in an open-loop manner, and at the same time, it synchronously collects the corresponding actual force response sequence through the force sensor. Based on the two, the voltage-force mapping relationship of the piezoelectric actuator in the voltage range corresponding to the calibration voltage sequence is obtained by online fitting, and the parameters of the compensation filter are initialized according to the mapping relationship. S402: During the closed-loop control process, the actual force signal output by the piezoelectric actuator is collected in real time using a force sensor. Based on the voltage-force mapping relationship obtained in step S401, the alternating control voltage signal output by the adaptive feedforward control algorithm is converted into the corresponding target control force. The force deviation signal between the actual force signal and the target control force is calculated. The force deviation signal is then processed by bandpass filtering and compensation filtering in sequence and superimposed on the alternating control voltage signal to form a corrected control voltage signal. The passband frequency range of the bandpass filter is set according to the current instantaneous rotor speed and synchronized with the rotation frequency to filter out noise interference from non-rotation frequency components. The corrected control voltage signal is superimposed with DC bias and used as the input of the power amplifier. S403: During the closed-loop control process, when the amplitude of the actual force signal collected by the force sensor exceeds the preset safety threshold, the programmable digital controller forcibly reduces the convergence step size of the adaptive feedforward control algorithm or increases the leakage factor to limit the output amplitude of the control voltage; when the programmable digital controller detects that the rotor speed change rate exceeds the set threshold, it suspends the online closed-loop correction, allowing the adaptive feedforward control algorithm to output the control voltage independently, and reactivates the online closed-loop correction after the rotor operating state returns to stability.

[0017] This invention discloses a piezoelectric active support rotor vibration reduction device and adaptive control method, belonging to the field of rotating machinery vibration control technology. The device includes a rotor system, a piezoelectric active support, a displacement sensor, a speed sensor, a programmable digital controller (PDC), and a power amplifier. The piezoelectric active support includes a frame, a movable bearing sleeve, rolling bearings, and a piezoelectric actuator. The piezoelectric actuator is radially arranged between the outer wall of the movable bearing sleeve and the inner wall of the frame, serving as both a load-bearing component and an active control force output unit. The displacement sensor collects vibration displacement signals in two orthogonal radial directions of the rotor disk, and the speed sensor collects instantaneous speed and phase signals. The PDC executes an adaptive feedforward algorithm based on these signals to output an alternating control voltage, which, after DC bias superposition and power amplification, drives the piezoelectric actuator to generate a radial active control force. This invention integrates the piezoelectric actuator into the support structure, resulting in a compact structure, high power density, and the ability to output active control force based on the real-time vibration state of the rotor, effectively suppressing the first-harmonic synchronous vibration response during the rotor's transcritical speed change process.

[0018] The above description of the invention is merely an overview of the technical solution of the present invention. In order to enable those skilled in the art to better understand the technical solution of the present invention and to implement it based on the description and drawings, and to make the above-mentioned objectives and other objectives, features and advantages of the present invention easier to understand, the following description is provided in conjunction with the specific embodiments and drawings of the present invention. Attached Figure Description

[0019] The accompanying drawings are only used to illustrate the principles, implementation methods, applications, features, and effects of specific embodiments of the present invention and other related contents, and should not be considered as limitations on the present invention.

[0020] In the accompanying drawings of the instruction manual: Figure 1 This is a schematic diagram of a piezoelectric active support structure according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the hardware composition and signal transmission of an active control system according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the rotor test bench and piezoelectric active support arrangement according to an embodiment of the present invention; Figure 4 This is a block diagram of a narrowband FxLMS adaptive feedforward control algorithm according to an embodiment of the present invention; Figure 5 This is a curve showing the change in vibration displacement amplitude of a rotor disk measuring point as a function of rotational speed, according to an embodiment of the present invention. Figure 6 This is a time-domain vibration displacement response curve of a rotor disk measuring point when the acceleration rate is 400 rpm / s and the acceleration time is 20s, according to an embodiment of the present invention. Figure 7 This is a flowchart of an adaptive control method for a piezoelectric active support rotor vibration damping device according to an embodiment of the present invention.

[0021] Figure label: 1. Piezoelectric active support; 2. Programmable digital controller; 3. Displacement sensor; 4. Speed ​​sensor; 5. Power amplifier; 6. Rotor system shaft; 11. Preload bolt; 12. Preload spring; 13. Rolling bearing; 14. Movable bearing sleeve; 15. Frame; 16. Spring guide rail; 17. Piezoelectric actuator; 18. Force sensor; Detailed Implementation

[0022] To explain in detail the possible application scenarios, technical principles, specific feasible solutions, and the objectives and effects that can be achieved by this invention, the following detailed description is provided in conjunction with the listed specific embodiments and accompanying drawings. The embodiments described herein are only used to more clearly illustrate the technical solutions of this invention, and are therefore only examples, and should not be used to limit the scope of protection of this invention.

[0023] In the first aspect, such as Figure 1-3 As shown, this application provides a piezoelectric active support rotor vibration reduction device, which includes a rotor system, a piezoelectric active support 1, a displacement sensor 3, a speed sensor 4, a programmable digital controller 2, and a power amplifier 5. The piezoelectric active support 1 includes a frame 15, a movable bearing sleeve 14, a rolling bearing 13, and a piezoelectric actuator 17. The frame serves as a fixed base and is rigidly connected to the casing platform of the rotor system; the movable bearing sleeve and the rolling bearing cooperate to form a movable bearing component, and the rolling bearing cooperates with the rotating shaft of the rotor system; the piezoelectric actuator is arranged radially between the outer wall of the movable bearing sleeve and the inner wall of the frame, serving as a load-bearing component to support the movable bearing component and the rotating shaft of the rotor system, and also serving as an output unit for active control force. The displacement sensor is arranged on the rotor system casing platform, aligned with the rotor disk measuring point of the rotor system, and collects the vibration displacement signals of the rotor disk in two orthogonal radial directions in real time, and inputs them into the programmable digital controller as control error signals in the corresponding directions respectively. The speed sensor is located on the rotor system casing platform to collect the instantaneous speed and phase signals of the rotor system in real time, and inputs them as a synchronization reference signal into the programmable digital controller. The programmable digital controller internally deploys an adaptive feedforward algorithm, which updates the control parameters in two orthogonal radial directions based on the signals from the displacement sensor and the speed sensor, and outputs alternating control voltage signals in the corresponding directions; a DC bias is superimposed on the alternating control voltage signals to form a biased control voltage signal. The power amplifier amplifies the biased control voltage signal output by the programmable digital controller by a corresponding factor, enabling the piezoelectric actuator to generate active control force within the permissible unidirectional voltage range.

[0024] In some embodiments, the piezoelectric active support further includes a mechanical preload mechanism, which includes a preload bolt 11, a preload spring 12, and a spring guide rail 16. The preload bolt is fitted with the frame, with one end abutting against the preload spring; the spring guide rail is fitted with the movable bearing sleeve, guiding the preload force provided by the preload spring to be perpendicular to the contact surface of the movable bearing sleeve; the preload spring is arranged between the preload bolt and the spring guide rail, with one end abutting against the preload bolt and the other end abutting against the inner wall of the spring guide rail.

[0025] In some embodiments, the piezoelectric active support further includes a force sensor 18, which is connected in series between the piezoelectric actuator and the frame to monitor the dynamic load borne by the piezoelectric active support in real time.

[0026] In some embodiments, the output end of the piezoelectric actuator is provided with a ball head or hinge structure, which abuts against the outer wall of the movable bearing sleeve. This reduces the tangential force and additional bending moment experienced by the piezoelectric actuator 17 while transmitting radial control force, thereby lowering the risk of shear damage to the piezoelectric stack inside the actuator 17. The size, stiffness, and output force of the piezoelectric actuator 17 can be selected according to the support load of the rotor system, installation space, and vibration reduction requirements.

[0027] In some embodiments, a radial clearance is provided between the movable bearing sleeve and the frame.

[0028] In some embodiments, the spring guide rail is a hollow sleeve structure and is installed on the travel path of the preload spring.

[0029] In this embodiment, the preload bolt 11 can be an M16×60 cup-head hexagonal bolt, made of 304 stainless steel. The preload bolt 11 mates with the threaded hole on the frame 15, with its head located on the outside of the frame 15 for easy adjustment of the preload force using tools. Its tail extends into the inside of the frame 15 and abuts against one end of the preload spring 12. The other end of the preload spring 12 is installed in the spring guide 16, which is connected to the movable bearing sleeve 14 to transmit the thrust of the preload spring 12 to the movable bearing sleeve 14. The preload spring 12 can be a heavy-duty mold spring, and its stiffness and initial compression are selected according to the preload force requirements of the piezoelectric actuator 17 to ensure that sufficient preload force can be provided during operation, so that the piezoelectric actuator 17 always remains under pressure under the alternating vibration load of the rotor system, avoiding damage due to tension. The rolling bearing 13 can be an NSK deep groove ball bearing, which has low friction torque, is suitable for high-speed rotation, mainly bears radial load, and can also bear a certain bidirectional axial load; the outer ring of the rolling bearing 13 is interference-fitted with the inner hole of the movable bearing sleeve 14, and the axial displacement is constrained by the end cover; the inner ring of the rolling bearing 13 is fitted with the rotor system shaft 6 to transmit radial vibration load; the movable bearing sleeve 14 is installed in the central area of ​​the frame 15, and a radial clearance is reserved between its outer wall and the frame 15.

[0030] In this embodiment, the radial clearance can be set to 1mm to meet the small radial movement requirements of the movable bearing sleeve 14 when the piezoelectric actuator 17 is actuated, while also providing a certain mechanical limiting effect; the spring guide rail 16 is designed as a hollow cylindrical sleeve structure and is installed on the stroke path of the preload spring 12 to constrain the degree of freedom of the preload spring 12, prevent the preload spring 12 from buckling laterally during compression, and ensure that the preload force is transmitted vertically to the movable bearing sleeve 14.

[0031] In this embodiment, a ball-head cylindrical piezoelectric ceramic actuator can be selected, which encapsulates piezoelectric ceramics in a stacked manner and applies mechanical preload to the outside by a stainless steel shell. Its nominal driving voltage range is 0V to 150V, nominal stroke is 57μm, and maximum output force is 7300N. It is suitable for high-load and high-dynamic applications and can serve as the main load-bearing and active control component. The bottom end of the piezoelectric actuator 17 is connected to the top end of the force sensor 18 by bolts. The bottom end of the force sensor 18 is installed inside the frame 15 by bolts for real-time monitoring of the dynamic load borne by the piezoelectric actuator branch.

[0032] During assembly and debugging, an adjustment shim can be added between the force sensor 18 and the frame 15 to eliminate machining tolerances between components and maintain the static geometric alignment of the movable bearing sleeve 14. After alignment is completed, the preload spring 12 is quantitatively compressed by adjusting the screw depth of the preload bolt 11, thereby applying an initial static preload force to the piezoelectric actuator 17. The drive end of the piezoelectric actuator 17 is connected to the output end of the power amplifier 5. like Figure 2 As shown, the active control system provided in this embodiment includes a programmable digital controller (PDC) 2, a displacement sensor 3, a speed sensor 4, a power amplifier 5, and a piezoelectric actuator 17. The PDC 2, as the core computing unit of the active control system, receives the rotor vibration response signal collected by the displacement sensor 3 and the speed phase signal collected by the speed sensor 4, and outputs a biased control voltage signal according to an adaptive feedforward control algorithm. The power amplifier 5 amplifies the biased control voltage signal output by the PDC 2 and drives the piezoelectric actuator 17 to generate an active control force.

[0033] Programmable digital controller 2 can use an MCU, FPGA, or other digital controllers with real-time acquisition and processing capabilities. As an optional embodiment, programmable digital controller 2 uses a NIPXIe-7858R multi-functional reconfigurable I / O module based on the Kintex-7 325T FPGA architecture. This module has multi-channel parallel acquisition and synchronous processing capabilities. Each channel of its analog input is equipped with a dedicated analog-to-digital converter, enabling independent timing and triggering, thereby synchronously acquiring multiple high-frequency vibration signals from the rotor system and ensuring the real-time performance of the control voltage output. Its core logic unit can be configured using the LabVIEW FPGA software environment.

[0034] As an optional implementation, the programmable digital controller 2 can also be connected to a host computer control terminal, which is used for control program download, control parameter setting, operation status monitoring and test data recording, but does not participate in real-time closed-loop control.

[0035] like Figure 3As shown, the rotor system shaft 6 is the controlled object in this embodiment. It can be a chrome-plated straight steel shaft with two rotor disks symmetrically arranged. The rotor system shaft 6 is mounted on two piezoelectric active supports 1 via rolling bearings 13 to simulate the bending vibration characteristics of a high-pressure rotor in an aero-engine. The two piezoelectric active supports 1 are respectively arranged at the two end support positions of the rotor system shaft 6 to apply active control force to the rotor system while bearing the rotor support load.

[0036] In this embodiment, eddy current displacement sensors are selected as displacement sensors 3. Two displacement sensors 3 are respectively fixed to the base of the rotor test bench using dedicated brackets. Their probes are respectively aligned with the two orthogonal radial directions of the rotor disk measuring point in the rotor system, used to collect vibration displacement signals in the two orthogonal radial directions at the rotor disk measuring point. The measurement directions of the two displacement sensors 3 are consistent with the actuation directions of the corresponding piezoelectric actuators 17, and the collected radial vibration displacement signals are respectively input as control error signals in the corresponding directions to the programmable digital controller 2.

[0037] In this embodiment, a photoelectric speed sensor 4 is selected. The speed sensor 4 is fixed to the base of the rotor test bench via a magnetic base, with its probe facing the outer edge of the coupling on the rotor system shaft 6. A ring of non-reflective black tape is pasted around the outer edge of the coupling, and a strip of highly reflective sticker is pasted at the corresponding measurement position. The speed sensor 4 is used to acquire the instantaneous speed and instantaneous phase signals of the rotor system shaft 6 in real time, and uses these signals as the basis for constructing the synchronization reference signal in the programmable digital controller 2.

[0038] In this embodiment, the power amplifier 5 is a multi-channel piezoelectric controller, and each channel can independently output a drive voltage. As one implementation, the voltage amplification factor of the power amplifier 5 is 15 times, which is used to amplify the biased control voltage output by the programmable digital controller 2 to a drive voltage in the range of 0V to 150V, thereby driving the piezoelectric actuator 17 to output active control force.

[0039] like Figure 4 As shown, this embodiment employs a narrowband FxLMS adaptive feedforward control algorithm to actively suppress the rotor's first-harmonic synchronous vibration. The speed sensor 4 acquires the instantaneous phase signal of the rotor system's shaft 6. The programmable digital controller 2 constructs an orthogonal reference signal synchronized with the frequency shift based on the instantaneous phase signal. The orthogonal reference signal Secondary channels estimated The pre-filtered reference signal is obtained and the control error signal measured by displacement sensor 3 Together they are used for adaptive weight update with leakage factor to obtain weight vector. .

[0040] Programmable digital controller 2 based on the updated weight vector and orthogonal synchronization reference signal Calculate alternating control voltage Subsequently, a DC bias is superimposed on the alternating control voltage within the programmable digital controller 2 to form the biased control voltage. The force is amplified by power amplifier 5 and then applied to piezoelectric actuator 17. The active control force output by piezoelectric actuator 17 is transmitted through the actual secondary channel. Acting on the rotor system, it reduces the first-harmonic synchronous vibration response at the rotor measuring point.

[0041] The working principle of the device of the present invention is given below: The device of this invention is installed between the stator casing platform and the rotor system shaft 6 of rotating machinery. During rotor system operation, periodic centrifugal excitation forces caused by mass imbalance act on the rotor system shaft 6, causing the rotor to generate a first-harmonic vibration response synchronized with the rotational frequency. This synchronized vibration persists during acceleration, deceleration, and variable operating conditions, and its excitation frequency changes with the rotational speed. When the rotor system crosses the critical speed range during startup, shutdown, or variable operating conditions, the vibration response typically increases significantly. To suppress this synchronized vibration response, the piezoelectric active support 1 in the device of this invention serves two purposes: firstly, as a support structure for the rotor system bearing radial loads; and secondly, as an active control actuator, it applies an active control force to the movable bearing sleeve 14 via the piezoelectric actuator 17, thereby adjusting the vibration response of the rotor system shaft 6.

[0042] When the rotor system shaft 6 generates radial vibration, the vibration load is first transmitted to the rolling bearing 13, and then from the rolling bearing 13 to the movable bearing sleeve 14. Under the radial load, the movable bearing sleeve 14 generates a small radial movement, transmitting the load to the piezoelectric actuation branch and the mechanical preload branch arranged between the movable bearing sleeve 14 and the frame 15. Specifically, the piezoelectric actuation branch containing the piezoelectric actuator 17 bears part of the radial support load and outputs active control force, while the mechanical preload branch containing the preload spring 12 bears part of the radial support load and provides continuous preload force. Under the action of the driving voltage, the piezoelectric actuator 17 generates active expansion and contraction deformation, applying an active radial control force to the rolling bearing 13 and the rotor system shaft 6 via the movable bearing sleeve 14, thereby actively adjusting the radial force at the support position.

[0043] To ensure the reliability of the piezoelectric actuator 17 during operation, a mechanical preload mechanism is formed by the preload bolt 11, the preload spring 12, and the spring guide rail 16. The preload bolt 11 provides an initial static preload to the movable bearing sleeve 14 and the piezoelectric actuator 17 by adjusting the compression of the preload spring 12. This ensures that the piezoelectric actuator 17 remains under pressure during alternating vibration loads, preventing damage to the piezoelectric ceramic stack due to reverse tension. The spring guide rail 16 constrains the movement direction of the preload spring 12, ensuring that the preload is transmitted to the movable bearing sleeve 14 in a set direction, thereby improving the stability of the preload transmission.

[0044] During active control, displacement sensor 3 is used to collect vibration displacement signals in two orthogonal radial directions at the rotor measuring point, and these signals are used as control error signals in the corresponding directions. Input programmable digital controller 2; speed sensor 4 collects the instantaneous speed and instantaneous phase signals of rotor system shaft 6 in real time, and uses them as the basis for constructing synchronization reference signals.

[0045] In the specific control process, the programmable digital controller 2 internally executes a narrowband FxLMS adaptive feedforward control algorithm, updating the adaptive weights in real time based on the filtered reference signal and the control error signal input from the displacement sensor 3, and calculating the alternating control voltage signal. To ensure that the piezoelectric actuator 17 operates within the allowable unidirectional voltage range, the programmable digital controller 2 internally superimposes a DC bias on the alternating control voltage signal to form a biased control voltage signal. This biased control voltage signal is amplified by the power amplifier 5 and applied to the piezoelectric actuator 17, causing the piezoelectric actuator 17 to output an active control force.

[0046] Through the above control process, the programmable digital controller 2 adjusts the amplitude and phase of the control voltage in real time according to the instantaneous phase of the rotor and the vibration error signal at the measuring point. This allows the active control force output by the piezoelectric actuator 17 to be transmitted through the support structure and the rotor system, thereby reducing the first-harmonic synchronous vibration component caused by mass imbalance at the rotor measuring point. When the rotor shaft 6 of the rotor system crosses the critical speed region, this active control action can reduce the resonance peak and transient vibration response caused by synchronous excitation, and improve the cross-critical operation stability of the rotor system.

[0047] Force sensor 18 monitors the supporting force or driving force in the piezoelectric actuation branch in real time and inputs the force signal into programmable digital controller 2 or data acquisition system to record the force state of the piezoelectric active support 1 during operation and observe the changes in support load during the test. In this embodiment, the force sensor 18 is not used as a control error signal for the adaptive feedforward control algorithm.

[0048] To illustrate the vibration reduction performance of the proposed piezoelectric active support device and its adaptive control method, this paper focuses on... Figure 3 The rotor system shown is established with a rotor dynamics simulation model, and numerical simulation analysis of the transcritical acceleration process is performed. During the simulation, the rotor system shaft 6 accelerates at a set acceleration rate, and the radial vibration displacement response is extracted at the rotor disk measuring point. The vibration response of the rotor disk measuring point is analyzed in two cases: (1) without active control, the piezoelectric active support 1 only participates in the rotor system dynamic response as a support structure, and the piezoelectric actuator 17 does not output active control force; (2) with active control, that is, the narrowband FxLMS adaptive feedforward control algorithm described in this invention is enabled, the programmable digital controller 2 outputs control voltage, and the piezoelectric actuator 17 is driven by the power amplifier 5 to output active control force.

[0049] like Figure 5 As shown, the two colored lines represent the vibration displacement amplitude of the rotor disk measuring point within the range of 0–8000 rpm under the two conditions described above. Figure 5 As can be seen, without active control, the rotor system exhibits significant vibration displacement peaks near both the first and second critical speeds. After implementing narrowband FxLMS algorithm control, the vibration displacement peaks near both critical speeds are significantly reduced. For example, the displacement peak near the first critical speed decreases from approximately 0.3 mm to approximately 0.03 mm without control, and the displacement peak near the second critical speed decreases from approximately 0.8 mm to approximately 0.2 mm without control, indicating that this method can effectively reduce the synchronous vibration peaks during the rotor's transcritical acceleration process.

[0050] like Figure 6 The figure shows the time-domain vibration displacement response curves of the rotor disk measuring points when the acceleration rate is 400 rpm / s and the acceleration time is 20 s. From... Figure 6 As can be seen, without active control, the rotor exhibits significant transient vibration responses when traversing the first and second critical speed regions. After incorporating the narrowband FxLMS algorithm control, the time-domain vibration amplitudes corresponding to both critical speed regions are significantly suppressed. For example, the peak time-domain displacement near the first critical speed decreases from approximately 0.07 mm without control to approximately 0.02 mm, and the peak time-domain displacement near the second critical speed decreases from approximately 0.24 mm without control to approximately 0.12 mm. These results demonstrate that the piezoelectric active support device and its adaptive control method described in this invention can effectively reduce resonance peaks and transient vibration responses near critical speeds during cross-critical acceleration.

[0051] In the second aspect, such as Figure 7 As shown, the present invention also provides an adaptive control method for a piezoelectric active-supported rotor vibration damping device, the method comprising the following steps: S1: Construct a piezoelectric active support rotor vibration damping device as described in the first aspect of the present invention; S2: Before the control closed loop is started, the secondary channel transmission characteristics between the piezoelectric active support and the rotor vibration measurement point are obtained by modeling or testing. The secondary channel transmission characteristics are used to characterize the dynamic transmission relationship of the control signal to the measurement point through the power amplifier, piezoelectric actuator, movable bearing sleeve and rotor system shaft. The secondary channel characteristics are discretized into a digital filter model and deployed in the programmable digital controller. S3: Use a displacement sensor to collect vibration displacement signals in two orthogonal radial directions at the rotor measuring point, and use them as control error signals in the corresponding directions. Use a speed sensor to collect the instantaneous speed and instantaneous phase signals of the rotor, and use them as the basis for constructing the synchronization reference signal. S4: The programmable digital controller executes an adaptive feedforward control algorithm, which calculates and outputs the control voltage signal in real time based on the measured control error signal, the rotor instantaneous phase signal, and the deployed digital filter model. S5: The power amplifier amplifies the biased control voltage signal output by the programmable digital controller by a corresponding factor, driving the piezoelectric actuator to output active control force.

[0052] In this embodiment, step S4 specifically includes: S41: Based on the rotor instantaneous phase angle Construct an orthogonal synchronization reference signal that is synchronized with the rotor speed and frequency. The calculation formula is as follows: The orthogonal reference signal is used to track the first harmonic synchronous vibration response caused by mass imbalance, where t represents the time quantity of the simulated continuous domain and n represents the discrete sampling time number. S42: Use a secondary channel digital filter to process the quadrature synchronization reference signal. Perform pre-filtering to obtain the filtered reference signal. The calculation formula is as follows: ,in, This represents the vector of discrete coefficients of the secondary channel predictor filter, based on the filter reference signal. and corresponding control error signals The adaptive weights of the narrowband FxLMS algorithm are updated in real time, and a leakage factor is introduced during the weight update process to suppress the continuous accumulation of adaptive weights. The weight update formula is as follows: ,in, This represents the weight at time n+1. This represents the weight at time n. This represents the control error signal at time n. Indicates the leakage factor. Indicates the convergence step size; S43: Calculate the alternating control voltage signal in the corresponding direction based on the updated adaptive weights and the orthogonal synchronization reference signal. The calculation formula is as follows: A DC bias voltage is superimposed inside the programmable digital controller to form a biased control voltage signal. This ensures that the piezoelectric actuator operates within the permissible unidirectional voltage range, wherein, This represents the transpose of the weight at time n.

[0053] In step S1, a vibration reduction device scheme for the rotor system is designed, including a rotor system, a piezoelectric active support 1, a programmable digital controller 2, a displacement sensor 3, a speed sensor 4, a power amplifier 5, and a rotor system shaft 6. The piezoelectric active support 1 is positioned at the support location of the rotor system shaft 6 and connected to the stator casing platform; the displacement sensor 3 is positioned near the rotor disk measuring point to collect the rotor's radial vibration displacement signal; the speed sensor 4 is positioned near the rotor system shaft 6 or the coupling to collect the rotor's instantaneous speed and instantaneous phase signal; the programmable digital controller 2 is connected to the displacement sensor 3, the speed sensor 4, and the power amplifier 5; the power amplifier 5 is connected to the piezoelectric actuator 17 to amplify the control voltage and drive the piezoelectric actuator 17 to output active control force. In the overall scheme design, the rotor system structure, support location, speed range, critical speed distribution, and expected vibration reduction effect can be comprehensively considered to design the layout of the positions and connections of each component of the device. Then, based on the radial load requirements of the rotor system and the support installation space, the structural parameters of the piezoelectric active support 1 are designed. These structural parameters include the dimensions, stiffness, stroke, and maximum output of the piezoelectric actuator 17; the radial clearance between the movable bearing sleeve 14 and the frame 15; the stiffness and initial compression of the preload spring 12; and the arrangement direction of the piezoelectric actuation branch and the mechanical preload branch. Through rotor dynamics simulation, the rotor critical speed, support load, and vibration response are simulated and analyzed under different piezoelectric actuator stiffness, output magnitude, preload spring stiffness, and support structure parameters. Combined with the actual installation space and the allowable working range of the piezoelectric actuator, the structural design scheme of the piezoelectric active support 1 is determined. During the design, it should be ensured that the piezoelectric actuator 17 always maintains a compressed working state under alternating vibration loads, and the influence of tangential force and additional bending moment on the piezoelectric actuator 17 is reduced through ball joint or hinge structure and preload mechanism. In step S2, before the control closed loop is started, the secondary channel transmission characteristics between the piezoelectric active support and the rotor vibration measurement point are obtained through modeling or testing. These secondary channel transmission characteristics characterize the dynamic transmission relationship of the control signal through the power amplifier 5, piezoelectric actuator 17, movable bearing sleeve 14, and rotor system shaft 6 to the measurement point. In this embodiment, a small-amplitude sweep frequency excitation signal can be output from the programmable digital controller 2 or an external signal source near the DC bias operating point of the piezoelectric actuator 17. After being amplified by the power amplifier 5, this signal drives the piezoelectric actuator 17 to generate an actuation response. Simultaneously, the radial vibration displacement response at the rotor disk measurement point is collected by the displacement sensor 3. Based on the frequency response relationship between the sweep frequency excitation signal and the measurement point displacement response, the secondary channel transmission characteristics from the alternating control voltage input to the rotor measurement point displacement response output are obtained. These secondary channel characteristics are then discretized into a digital filter model. And deployed in a programmable digital controller, the corresponding coefficient sequence is denoted as This is used for pre-filtering of subsequent reference signals; the digital filter model It can be represented as: ; in, For discrete coefficient sequence The One coefficient, Let the filter order be . for Complex variables in change Indicates delay One sampling period; In step S3, two displacement sensors 3 are respectively arranged at the two orthogonal radial directions of the rotor disk measuring point, with their measuring directions consistent with the actuation direction of the piezoelectric actuator 17 in the corresponding directions. They are used to collect vibration displacement signals in the two orthogonal radial directions and serve as control error signals in the corresponding directions. The instantaneous rotational speed of the rotor is collected using speed sensor 4. and instantaneous phase signal This serves as the basis for constructing the synchronization reference signal.

[0054] In step S4, the programmable digital controller 2 executes an adaptive feedforward control algorithm based on the measured control error signal. Rotor instantaneous phase signal The adaptive feedforward control algorithm, which utilizes a deployed digital filter model and calculates the control outputs in two orthogonal directions in real time, can employ a narrowband FxLMS algorithm. The specific process includes: When the primary control objective of the rotor system is the first-harmonic synchronous vibration caused by mass imbalance, its synchronous excitation force It can be represented as: ; in, Indicates the equivalent unbalance quantity. Indicates the instantaneous angular velocity of the rotor. Indicates the instantaneous phase angle of the rotor; Since the mass imbalance excitation is synchronized with the rotor rotation phase, an orthogonal reference signal synchronized with the rotor speed and frequency is constructed based on the rotor's instantaneous phase angle. , can be represented as: in, It is a discrete-time index, representing the first... Each sampling time; Then, the secondary channel digital filter established in the previous step is used to process the orthogonal synchronization reference signal. Perform pre-filtering to obtain the filtered reference signal. To account for the propagation effect of the secondary channel during the adaptive weight update process, compensate for its phase lag, and improve control stability, it can be expressed as: ; in, This represents the discrete convolution operation; The weight vector of the algorithm As the control process is continuously updated, the weight update formula can be expressed as: ;in, To converge the step size, Leakage factor This is used to suppress the continuous accumulation of adaptive weights and reduce the risk of control voltage exceeding limits; The alternating control voltage signal output by the programmable digital controller according to the algorithm is: , can be represented as: ; To ensure that the piezoelectric actuator operates within the permissible unidirectional voltage range, the programmable digital controller internally superimposes a DC bias voltage onto the alternating control voltage signal. This forms the biased control voltage signal. , can be represented as: ; After completing the deployment of the narrowband FxLMS algorithm, the convergence step size is determined based on the vibration reduction requirements and system stability requirements of the rotor system. With leakage factor The value of is determined by the following: Specifically, through rotor speed-up tests or numerical simulations, combined with the rotor vibration response measured by displacement sensors, the dynamic support load monitored by force sensors, and the actual drive voltage signal output by the power amplifier, the convergence step size of the narrowband FxLMS algorithm is gradually adjusted. With leakage factor When the vibration peak of the rotor is significantly suppressed during the process of passing the critical speed, and the control voltage after bias does not show continuous drift, excessive increase or exceed the preset physical limit, it is determined to be a suitable control parameter.

[0055] In step S5, the power amplifier 5 amplifies the biased control voltage signal output by the programmable digital controller 2 by a corresponding factor, driving the piezoelectric actuators 17 in the corresponding directions to output active control force. The active control force generated by the piezoelectric actuators 17 acts on the rotor system through the movable bearing sleeve 14, the rolling bearing 13, and the rotor system shaft 6.

[0056] After designing the piezoelectric active support device, modeling the secondary channel, and adjusting the control parameters for the rotor system according to the above method, the first-harmonic synchronous vibration can be effectively suppressed during the rotor's transcritical variable speed operation, reducing the resonance peak and transient vibration response near the critical speed.

[0057] This invention, by employing the above technical solutions, has the following advantages and positive effects compared with existing rotor system support vibration reduction technologies: 1) Using a piezoelectric actuator as the active control force output element, compared with traditional passive supports with fixed stiffness and damping parameters, it can output active control force according to the real-time vibration response of the rotor, improving the adaptability of the support device to variable speed conditions; 2) Through a mechanical preload protection strategy, the piezoelectric actuator is always kept under pressure, reducing the risk of tensile or shear failure under alternating loads, and improving the reliability of the device under complex vibration conditions; 3) Utilizing a speed sensor and a programmable digital controller, an orthogonal reference signal synchronized with the rotational frequency is constructed based on the instantaneous phase of the rotor, which can detect mass imbalances. 4) The first harmonic synchronous vibration response is tracked in real time, overcoming the problem that existing control methods for constant speed conditions are difficult to adapt to the continuous change of synchronous excitation frequency; 5) The narrowband FxLMS adaptive feedforward control algorithm is adopted, and the secondary channel digital filter is deployed in the programmable digital controller to pre-filter the orthogonal synchronous reference signal, so that the weight update process can take into account the influence of secondary channel transmission, compensate for phase lag, and improve the stability of active control; 6) The leakage factor is introduced in the weight update process, which can suppress the continuous accumulation of adaptive weights, reduce the risk of continuous increase of control voltage or exceeding physical limits, thereby reducing the resonance peak and transient vibration response when the rotor crosses the critical speed region, and improving the operating stability of the rotor system.

[0058] In some embodiments, in step S4, the programmable digital controller performs online closed-loop correction of the piezoelectric actuator hysteresis nonlinearity while executing the adaptive feedforward control algorithm, specifically including the following steps: S401: In the initial stage after the control closed loop is started, the programmable digital controller first applies a set of calibration voltage sequences with increasing amplitude to the piezoelectric actuator in an open-loop manner, and at the same time, it synchronously collects the corresponding actual force response sequence through the force sensor. Based on the two, the voltage-force mapping relationship of the piezoelectric actuator in the voltage range corresponding to the calibration voltage sequence is obtained by online fitting, and the parameters of the compensation filter are initialized according to the mapping relationship. S402: During the closed-loop control process, the actual force signal output by the piezoelectric actuator is collected in real time using a force sensor. Based on the voltage-force mapping relationship obtained in step S401, the alternating control voltage signal output by the adaptive feedforward control algorithm is converted into the corresponding target control force. The force deviation signal between the actual force signal and the target control force is calculated. The force deviation signal is then processed by bandpass filtering and compensation filtering in sequence and superimposed on the alternating control voltage signal to form a corrected control voltage signal. The passband frequency range of the bandpass filter is set according to the current instantaneous rotor speed and synchronized with the rotation frequency to filter out noise interference from non-rotation frequency components. The corrected control voltage signal is superimposed with DC bias and used as the input of the power amplifier. S403: During the closed-loop control process, when the amplitude of the actual force signal collected by the force sensor exceeds the preset safety threshold, the programmable digital controller forcibly reduces the convergence step size of the adaptive feedforward control algorithm or increases the leakage factor to limit the output amplitude of the control voltage; when the programmable digital controller detects that the rotor speed change rate exceeds the set threshold, it suspends the online closed-loop correction, allowing the adaptive feedforward control algorithm to output the control voltage independently, and reactivates the online closed-loop correction after the rotor operating state returns to stability.

[0059] In this embodiment, piezoelectric hysteresis nonlinearity refers to the inherent characteristics of piezoelectric ceramic materials. During the rise and fall of the same driving voltage, the actual force output by the piezoelectric material does not coincide, resulting in hysteresis loop. This will cause deviations in the amplitude and phase of the control force, reducing the vibration reduction accuracy across critical speeds.

[0060] The voltage-force mapping relationship refers to the curve and polynomial model obtained by fitting the stepped calibration voltage sequence with the corresponding measured force, which characterizes the correspondence between the piezoelectric input voltage and the actual output force, and is used to correct the control force error caused by hysteresis.

[0061] Force deviation signal refers to the difference between the target control force calculated based on the algorithm and the actual piezoelectric force measured by the force sensor. It is the core feedback basis for hysteresis closed-loop correction.

[0062] Bandpass filtering refers to the process where the passband frequency is adjusted synchronously with the instantaneous rotational speed of the rotor, retaining only the force deviation component corresponding to the first harmonic vibration of the rotor, and filtering out irrelevant frequency noise such as gear, fluid, and motor noise, as well as other harmonic disturbances in the aero-engine rotor system.

[0063] The speed change rate threshold refers to the upper limit of the rotor speed increase or decrease per unit time. When the speed changes abruptly, the rotor dynamic state fluctuates violently. Pausing hysteresis correction can avoid the correction logic interfering with the basic adaptive vibration reduction algorithm.

[0064] This embodiment adds online closed-loop correction logic for piezoelectric hysteresis nonlinearity to the basic FxLMS adaptive control, and is equipped with a dual protection strategy for overload and sudden speed change. The execution process consists of three levels: open-loop calibration modeling, closed-loop real-time correction, and adaptive protection under operating conditions. The entire correction closed-loop is achieved based on the load signal measured by the force sensor. The specific working principle is as follows: In step S401, the open-loop calibration initialization compensation parameters are first performed, as follows: In the initial stage of the closed-loop vibration reduction start-up of the device, the controller first switches to the open-loop output mode and outputs a set of calibration voltage sequences with progressively increasing amplitude to the piezoelectric actuator; simultaneously, the actual piezoelectric force response corresponding to each calibration voltage is collected through force sensors arranged in series, and a continuous mapping relationship is obtained by fitting multiple sets of voltage-force discrete data points, generating the initial parameters of the hysteresis compensation filter and storing them in the controller, thus establishing a basic correction model for the hysteresis characteristics of the piezoelectric material.

[0065] In step S402, closed-loop real-time hysteresis correction is then performed, specifically including: after the entire machine switches to closed-loop vibration reduction mode, the adaptive feedforward algorithm first calculates the target control force required to counteract vibration; the controller retrieves the voltage-force mapping relationship obtained in S401 and converts the alternating control voltage into the theoretical target control force; comparing the actual piezoelectric force transmitted back by the force sensor in real time, the difference between the two is calculated to obtain the force deviation signal. The force deviation signal is first sent to a bandpass filter synchronized with the rotor frequency to filter out non-first harmonic noise and retain only the effective deviation component corresponding to the unbalanced vibration; the filtered deviation signal is input to a compensation filter to generate a correction voltage compensation amount, which is superimposed on the original alternating control voltage to obtain the corrected control voltage waveform; finally, a fixed DC bias is superimposed to generate the final control signal driving the power amplifier. This correction process compensates for the control force error caused by piezoelectric hysteresis in real time, significantly improving the vibration suppression accuracy under variable speed transcritical conditions.

[0066] In step S403, the overload and speed change adaptive protection logic is then executed: In the first protection scenario, when the amplitude of the piezoelectric dynamic force collected by the force sensor exceeds the preset safety threshold, it is determined that there is an overload risk in the piezoelectric branch. The controller forcibly reduces the convergence step size of the adaptive algorithm and synchronously increases the leakage factor, directly limiting the amplitude of the control voltage output and reducing the output thrust of the piezoelectric actuator to achieve overload hardware protection. In the second protection scenario, the rotor speed change rate is calculated in real time. When the speed rises or falls rapidly and the change rate exceeds the set threshold, the rotor dynamic state fluctuates violently, and the hysteresis correction logic will introduce additional interference. The controller temporarily suspends the online closed-loop correction and only runs the basic narrowband FxLMS adaptive control to ensure the stability of the basic vibration reduction function. After the rotor speed tends to stabilize and the change rate falls back to within the threshold, the hysteresis correction process is automatically reactivated to restore high-precision compensation vibration reduction.

[0067] This embodiment has the following beneficial effects: First, online closed-loop correction of piezoelectric hysteresis nonlinearity significantly improves vibration reduction control accuracy. A correction closed loop is constructed based on the measured force from a force sensor, which in real time offsets the amplitude and phase deviations of the control force caused by the inherent hysteresis of the piezoelectric ceramic, further enhancing the vibration suppression effect of the rotor across the critical resonance range.

[0068] Second, the bandpass filter synchronously follows the rotor speed, exhibiting strong resistance to clutter interference. The calibration process extracts only the first harmonic force deviation component, filters out various types of clutter noise during equipment operation, and ensures that the calibration logic is not interfered with by irrelevant vibration signals, resulting in stable calibration without additional oscillations.

[0069] Third, a dual-condition adaptive protection mechanism balances calibration accuracy and system stability. When the load exceeds limits, it actively limits the output force to protect the piezoelectric element; when the speed changes abruptly, it temporarily shuts down the calibration to ensure the foundation vibration damping does not fail, thus balancing high-precision calibration with reliability under extreme changing conditions.

[0070] Fourth, open-loop calibration enables rapid model initialization, adapting to different piezoelectric specifications. Voltage-force mapping calibration is automatically completed each time the device is started, eliminating the need for manual modification of calibration parameters when changing piezoelectric actuators with different output specifications, thus enhancing the device's versatility.

[0071] Finally, it should be noted that although the above embodiments have been described in the description and drawings of this invention, this should not limit the scope of patent protection of this invention. Any technical solutions that are based on the essential concept of this invention, utilize the content described in the description and drawings of this invention to make equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this invention.

Claims

1. A piezoelectric active-support rotor vibration damping device, characterized in that, The device includes a rotor system, a piezoelectric active support, a displacement sensor, a speed sensor, a programmable digital controller, and a power amplifier. The piezoelectric active support includes a frame, a movable bearing sleeve, a rolling bearing, and a piezoelectric actuator. The frame serves as a fixed base and is rigidly connected to the casing platform of the rotor system. The movable bearing sleeve and the rolling bearing cooperate to form a movable bearing component, and the rolling bearing cooperates with the rotating shaft of the rotor system. The piezoelectric actuator is arranged radially between the outer wall of the movable bearing sleeve and the inner wall of the frame, serving as a load-bearing component to support the movable bearing component and the rotating shaft of the rotor system, and also as an output unit for active control force. The displacement sensor is arranged on the rotor system casing platform, aligned with the rotor disk measuring point of the rotor system, and collects the vibration displacement signals of the rotor disk in two orthogonal radial directions in real time, and inputs them into the programmable digital controller as control error signals in the corresponding directions respectively. The speed sensor is located on the rotor system casing platform to collect the instantaneous speed and phase signals of the rotor system in real time, and inputs them as a synchronization reference signal into the programmable digital controller. The programmable digital controller internally deploys an adaptive feedforward algorithm, which updates the control parameters in two orthogonal radial directions based on the signals from the displacement sensor and the speed sensor, and outputs alternating control voltage signals in the corresponding directions; a DC bias is superimposed on the alternating control voltage signals to form a biased control voltage signal. The power amplifier amplifies the biased control voltage signal output by the programmable digital controller by a corresponding factor, enabling the piezoelectric actuator to generate active control force within the permissible unidirectional voltage range.

2. The piezoelectric active-support rotor vibration damping device as described in claim 1, characterized in that, The piezoelectric active support also includes a mechanical preload mechanism, which includes a preload bolt, a preload spring, and a spring guide rail. The preload bolt is fitted with the frame, with one end abutting against the preload spring; the spring guide rail is fitted with the movable bearing sleeve, guiding the preload force provided by the preload spring to be perpendicular to the contact surface of the movable bearing sleeve; the preload spring is arranged between the preload bolt and the spring guide rail, with one end abutting against the preload bolt and the other end abutting against the inner wall of the spring guide rail.

3. The piezoelectric active-support rotor vibration damping device as described in claim 1, characterized in that, The piezoelectric active support also includes a force sensor, which is connected in series between the piezoelectric actuator and the frame to monitor the dynamic load borne by the piezoelectric active support in real time.

4. The piezoelectric active-support rotor vibration damping device as described in claim 1, characterized in that, The output end of the piezoelectric actuator is provided with a ball head or hinge structure, which abuts against the outer wall of the movable bearing sleeve.

5. The piezoelectric active-support rotor vibration damping device as described in claim 1, characterized in that, A radial clearance is provided between the movable bearing sleeve and the frame.

6. The piezoelectric active-support rotor vibration damping device as described in claim 2, characterized in that, The spring guide rail is a hollow sleeve structure and is installed on the travel path of the preloaded spring.

7. An adaptive control method for a piezoelectric active-supported rotor vibration damping device, characterized in that, The method includes the following steps: S1: Construct a piezoelectric active support rotor vibration damping device as described in any one of claims 1 to 6; S2: Before the control closed loop is started, the secondary channel transmission characteristics between the piezoelectric active support and the rotor vibration measurement point are obtained by modeling or testing. The secondary channel transmission characteristics are used to characterize the dynamic transmission relationship of the control signal to the measurement point through the power amplifier, piezoelectric actuator, movable bearing sleeve and rotor system shaft. The secondary channel characteristics are discretized into a digital filter model and deployed in the programmable digital controller. S3: Use a displacement sensor to collect vibration displacement signals in two orthogonal radial directions at the rotor measuring point, and use them as control error signals in the corresponding directions. Use a speed sensor to collect the instantaneous speed and instantaneous phase signals of the rotor, and use them as the basis for constructing the synchronization reference signal. S4: The programmable digital controller executes an adaptive feedforward control algorithm, which calculates and outputs the control voltage signal in real time based on the measured control error signal, the rotor instantaneous phase signal, and the deployed digital filter model. S5: The power amplifier amplifies the biased control voltage signal output by the programmable digital controller by a corresponding factor, driving the piezoelectric actuator to output active control force.

8. The adaptive control method for the piezoelectric active-supported rotor vibration damping device as described in claim 7, characterized in that, Step S4 specifically includes: S41: Based on the rotor instantaneous phase angle Construct an orthogonal synchronization reference signal that is synchronized with the rotor speed and frequency. The calculation formula is as follows: The orthogonal reference signal is used to track the first harmonic synchronous vibration response caused by mass imbalance, where t represents the time quantity of the simulated continuous domain and n represents the discrete sampling time number. S42: Use a secondary channel digital filter to process the quadrature synchronization reference signal. Perform pre-filtering to obtain the filtered reference signal. The calculation formula is as follows: ,in, This represents the vector of discrete coefficients of the secondary channel predictor filter, based on the filter reference signal. and corresponding control error signals The adaptive weights of the narrowband FxLMS algorithm are updated in real time, and a leakage factor is introduced during the weight update process to suppress the continuous accumulation of adaptive weights. The weight update formula is as follows: ,in, This represents the weight at time n+1. This represents the weight at time n. This represents the control error signal at time n. Indicates the leakage factor. Indicates the convergence step size; S43: Calculate the alternating control voltage signal in the corresponding direction based on the updated adaptive weights and the orthogonal synchronization reference signal. The calculation formula is as follows: A DC bias voltage is superimposed inside the programmable digital controller to form a biased control voltage signal. This ensures that the piezoelectric actuator operates within the permissible unidirectional voltage range, wherein, This represents the transpose of the weight at time n.

9. The adaptive control method for the piezoelectric active-supported rotor vibration damping device as described in claim 7, characterized in that, In step S4, the convergence step size and leakage factor of the adaptive feedforward control algorithm are determined through the following debugging process, specifically including: Through speed-up tests on a rotor test bench or numerical simulations, combined with the rotor vibration response measured by displacement sensors, the dynamic support load monitored by force sensors, and the actual drive voltage signal output by the power amplifier, the convergence step size and leakage factor are gradually adjusted until the vibration peak of the rotor is significantly suppressed during the process of crossing the critical speed, and the control voltage after bias does not show continuous drift, excessive growth, or exceed the preset physical limit. The corresponding convergence step size and leakage factor at this time are used as the fixed operating parameters of the adaptive feedforward control algorithm in step S4.

10. The adaptive control method for the piezoelectric active-supported rotor vibration damping device as described in claim 7, characterized in that, The piezoelectric active support rotor vibration damping device also includes a force sensor; In step S4, the programmable digital controller, while executing the adaptive feedforward control algorithm, also performs online closed-loop correction of the piezoelectric actuator's hysteresis nonlinearity, specifically including the following steps: S401: In the initial stage after the control closed loop is started, the programmable digital controller first applies a set of calibration voltage sequences with increasing amplitude to the piezoelectric actuator in an open-loop manner, and at the same time, it synchronously collects the corresponding actual force response sequence through the force sensor. Based on the two, the voltage-force mapping relationship of the piezoelectric actuator in the voltage range corresponding to the calibration voltage sequence is obtained by online fitting, and the parameters of the compensation filter are initialized according to the mapping relationship. S402: During the closed-loop control process, the actual force signal output by the piezoelectric actuator is collected in real time using a force sensor. Based on the voltage-force mapping relationship obtained in step S401, the alternating control voltage signal output by the adaptive feedforward control algorithm is converted into the corresponding target control force. The force deviation signal between the actual force signal and the target control force is calculated. The force deviation signal is then processed by bandpass filtering and compensation filtering in sequence and superimposed on the alternating control voltage signal to form a corrected control voltage signal. The passband frequency range of the bandpass filter is set according to the current instantaneous rotor speed and synchronized with the rotation frequency to filter out noise interference from non-rotation frequency components. The corrected control voltage signal is superimposed with DC bias and used as the input of the power amplifier. S403: During the closed-loop control process, when the amplitude of the actual force signal collected by the force sensor exceeds the preset safety threshold, the programmable digital controller forcibly reduces the convergence step size of the adaptive feedforward control algorithm or increases the leakage factor to limit the output amplitude of the control voltage; when the programmable digital controller detects that the rotor speed change rate exceeds the set threshold, it suspends the online closed-loop correction, allowing the adaptive feedforward control algorithm to output the control voltage independently, and reactivates the online closed-loop correction after the rotor operating state returns to stability.

Images