Shaft system self-adaptive vibration suppression system of variable-speed steam turbine

Abstract

The invention relates to the technical field of vibration suppression, in particular to a variable-speed steam turbine shafting self-adaptive vibration suppression system, which establishes a high-precision multi-source vibration data acquisition system through a vibration monitoring module, and provides a reliable data basis for vibration analysis; accurate extraction of vibration characteristics is realized through the signal processing module, and key input is provided for intelligent vibration suppression; an optimal vibration suppression strategy is generated through a self-adaptive vibration suppression control module, so that the accuracy and adaptability of a control effect are ensured; active-passive mixed vibration suppression is achieved through the vibration suppression execution module, and the vibration suppression effect is remarkably improved; and a high-fidelity virtual model is constructed through a digital twinning and health management module, so that the accurate prediction of the health state of the equipment is realized.

Description

Technical Field

[0001] This invention relates to the field of vibration suppression technology, and in particular to an adaptive vibration suppression system for a variable speed steam turbine shaft system. Background Technology

[0002] In the current field of turbine vibration monitoring, traditional systems mainly rely on single vibration sensors for data acquisition, making it difficult to comprehensively capture multi-dimensional vibration characteristics under complex operating conditions. Existing monitoring schemes generally suffer from single data sources and insufficient feature extraction, resulting in insufficient accuracy in identifying low-frequency torsional vibrations and high-frequency harmonics. Meanwhile, traditional vibration suppression control often employs fixed-parameter strategies, failing to adapt to dynamic changes in speed and load, leading to significant lag in vibration suppression response. In terms of health management, reliance is mainly on periodic maintenance and manual experience judgment, lacking accurate life prediction models and hindering early fault warnings. Existing digital twin models often lack synchronization with the physical entity, resulting in significant state prediction deviations. These problems severely restrict the accuracy of turbine vibration control and the level of intelligent equipment management, necessitating the construction of a new monitoring and protection system integrating multi-source sensing, intelligent vibration suppression, and accurate prediction.

[0003] Chinese patent CN105065568A discloses a method for suppressing vibration of rotating shaft systems, belonging to the field of vibration suppression technology. It solves the problems of existing vibration suppression methods for rotating shaft systems rapidly crossing resonant frequencies, such as the inability to eliminate resonance, stringent requirements on drive components, and differences in vibration amplitude at different operating speeds. This method involves placing a piezoelectric adjustment ring on the side of the inner or outer ring of the bearing in the rotating shaft system. When the rotational speed of the rotating shaft approaches a certain resonant frequency, a DC excitation signal is applied to the piezoelectric adjustment ring. Under the inverse piezoelectric effect, the thickness of the piezoelectric adjustment ring increases, thereby increasing the bearing preload and correspondingly increasing the bearing's support stiffness, ultimately raising the resonant frequency of the rotating shaft system. When the rotational speed of the rotating shaft smoothly exceeds the original resonant frequency, the piezoelectric adjustment ring is de-energized, and the resonant frequency of the rotating shaft returns to the original resonant frequency, thus avoiding resonance. This invention is applied to the field of vibration suppression. However, this solution still suffers from problems such as incomplete vibration monitoring, untimely vibration suppression control, and inaccurate equipment status prediction due to limited sensor data, static lag in control strategies, and reliance on manual experience for health management. Summary of the Invention

[0004] To address this, the present invention provides an adaptive vibration suppression system for variable speed turbine shaft systems, which overcomes the problems in the prior art caused by single sensor data, static lag in control strategies, and reliance on human experience for health management, resulting in incomplete vibration monitoring, untimely vibration suppression control, and inaccurate equipment status prediction.

[0005] To achieve the above objectives, the present invention provides an adaptive vibration suppression system for a variable speed steam turbine shaft system, comprising: The vibration monitoring module is used to collect multi-source vibration data; The signal processing module is used to extract features from the multi-source vibration data to obtain vibration feature vectors. An adaptive vibration damping control module is used to generate vibration damping control commands based on the vibration feature vector; The vibration suppression execution module is used to execute vibration suppression control according to the vibration suppression control command; The digital twin and health management module is used to construct a digital twin model of the shaft system and predict the health status of the equipment based on the multi-source vibration data and vibration suppression control commands.

[0006] Furthermore, the vibration monitoring module collects multi-source vibration data including: Vibration signals are collected from key points of the turbine bearing housing and rotor to obtain raw vibration data; the raw vibration data is demodulated to obtain shaft torsional vibration characteristic data; the shaft torsional vibration characteristic data is fused from multiple sources to obtain a preliminary vibration dataset; the preliminary vibration dataset is then subjected to quality verification and format standardization to obtain multi-source vibration data.

[0007] Furthermore, the process of acquiring vibration signals from key points of the turbine bearing housing and rotor to obtain raw vibration data specifically involves: acquiring vibration acceleration, displacement, and phase signals using a non-contact laser vibration meter, and acquiring temperature signals using a fiber optic Bragg grating sensor to form raw vibration data containing vibration and temperature parameters. The specific steps for demodulating the original vibration data to obtain shaft torsional vibration characteristic data are as follows: the original vibration data is processed using rotational speed pulse signal demodulation technology, the shaft torsional vibration mode is analyzed in real time, and characteristic data including torsional vibration frequency and amplitude are obtained. The step of fusing multi-source data to obtain a preliminary vibration dataset for the torsional vibration characteristic data of the shaft system specifically involves: performing spatiotemporal registration of the torsional vibration characteristic data and temperature signals, eliminating measurement differences between sensors through a data fusion algorithm, and forming a preliminary vibration dataset. The process of performing quality verification and format standardization on the preliminary vibration dataset to obtain multi-source vibration data specifically involves: performing outlier detection and data integrity verification on the preliminary vibration dataset, standardizing it according to a unified data format, and outputting multi-source vibration data containing timestamps, device identifiers, and measurement values.

[0008] Furthermore, the signal processing module performs feature extraction based on the multi-source vibration data to obtain a vibration feature vector including: The multi-source vibration data is processed by wavelet packet transform to obtain time-frequency domain features; the time-frequency domain features are processed by modal decomposition to obtain vibration modal parameters; the vibration modal parameters are processed by interference cancellation to obtain clean feature data; and the clean feature data is used to construct feature vectors to obtain vibration feature vectors.

[0009] Furthermore, the step of performing wavelet packet transform processing on the multi-source vibration data to obtain time-frequency domain features specifically involves: using the wavelet packet transform algorithm to perform frequency band analysis of the multi-source vibration data from 0.1 to 500 Hz, extracting the energy distribution features of each frequency band, and obtaining time-frequency domain features containing low-frequency oscillations and high-frequency harmonic components. The modal decomposition of the time-frequency domain features to obtain vibration modal parameters specifically involves: performing modal decomposition on the time-frequency domain features based on the improved antlion algorithm, identifying the main vibration modes, and obtaining vibration modal parameters including modal frequencies, damping ratios, and mode shape parameters; The process of eliminating interference in the vibration modal parameters to obtain pure feature data specifically involves: using frequency domain blind source separation technology to eliminate the noise influence of electromagnetic interference, extracting essential vibration features through independent component analysis, and obtaining pure feature data containing the features of the main vibration sources. The process of constructing a vibration feature vector from the pure feature data specifically involves extracting key feature parameters such as vibration energy entropy, modal damping ratio, principal mode frequency, frequency center, root mean square frequency, kurtosis, skewness, and energy proportion of a specific frequency band from the pure feature data, performing feature dimensionality reduction through principal component analysis, and finally forming an 8-dimensional vibration feature vector.

[0010] Furthermore, the adaptive vibration suppression control module generates vibration suppression control commands based on the vibration feature vector, including: The vibration characteristic vector is used to calculate control quantities to obtain basic control parameters; disturbance compensation is performed on the basic control parameters to obtain disturbance rejection control parameters; the disturbance rejection control parameters are optimized online to obtain optimized control parameters; and the optimized control parameters are encapsulated into instructions to obtain the final vibration suppression control instruction.

[0011] Furthermore, the calculation of the control quantity for the vibration characteristic vector to obtain the basic control parameters specifically involves: inputting the 8-dimensional vibration characteristic vector into the rotational speed-guide vane dual-variable active disturbance rejection controller, and calculating the basic damping force and guide vane adjustment amount through the nonlinear state error feedback law to form the basic control parameters; The disturbance compensation of the basic control parameters to obtain the disturbance rejection control parameters is specifically performed as follows: a fractional-order disturbance observer is used to estimate the unmodeled dynamics and external disturbances of the system in real time, and an extended state observer is used to perform feedforward compensation of the total disturbance to generate the disturbance rejection control parameters. The online optimization of the anti-disturbance control parameters to obtain optimized control parameters specifically involves: online tuning of the control parameters based on the improved antlion algorithm to adapt to the dynamic adjustment requirements of ±15° guide angle changes and 0.1~0.5mm sealing gap, and outputting optimized control parameters; The process of encapsulating the optimized control parameters into instructions to obtain the final vibration suppression control instructions specifically involves converting the optimized control parameters into control signals that the actuator can recognize, including magnetorheological damper current instructions and hydraulic actuator displacement instructions, thereby completing the final generation of the vibration suppression control instructions.

[0012] Furthermore, the vibration suppression execution module performs vibration suppression control according to the vibration suppression control command, including: The vibration suppression control commands are parsed and allocated to obtain the control parameters of each execution unit; the current of the magnetorheological damper control parameters is adjusted to achieve rapid adjustment of the damping characteristics; the fluid injection control of the solid-liquid coupling device control parameters is implemented to achieve three-dimensional flow field vibration suppression; and the execution units are coordinated and their status is monitored to form a closed-loop vibration suppression system.

[0013] Furthermore, the specific steps of parsing and allocating the vibration suppression control command to obtain the control parameters of each execution unit are as follows: parsing the magnetorheological damper current command and the hydraulic actuator displacement command in the vibration suppression control command, and allocating them to the corresponding magnetorheological elastomer intelligent bearing unit and solid-liquid two-phase coupling vibration suppression device unit. The method of adjusting the current of the control parameters of the magnetorheological damper to achieve rapid adjustment of the damping characteristics is as follows: according to the current control command of 0-5A, the electromagnetic coil is driven by the power amplifier, so that the magnetorheological elastomer completes the viscoelastic adjustment within ≤0.1s, and achieves continuous damping control in the range of 10-1000kN·s / m. The fluid injection control of the solid-liquid coupling device control parameters to achieve three-dimensional flow field vibration suppression specifically involves: adjusting the opening of the non-Newtonian fluid injection valve according to the displacement control command to form a stable three-dimensional flow field within the 0.2mm narrow gap of the rotor, effectively suppressing the rotor vortex phenomenon; The coordinated control and status monitoring of each execution unit to form a closed-loop vibration suppression system specifically involves: real-time acquisition of the working status of each execution unit, and dynamic adjustment through a hybrid active-passive vibration suppression system composed of spring isolators to ensure the design target of reducing vibration transmission rate by 20%.

[0014] Furthermore, the digital twin and health management module constructs a shaft system digital twin model and predicts equipment health status based on the multi-source vibration data and vibration suppression control commands, including: The multi-source vibration data and vibration suppression control commands are fused to obtain modeling input data; a multibody dynamics model of the shaft system is constructed based on the modeling input data to obtain an initial digital twin; the initial digital twin is subjected to parameter correction and model verification to obtain a high-precision digital twin model; and a health status prediction is performed based on the high-precision digital twin model to obtain equipment health assessment results.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: the system establishes a high-precision multi-source vibration data acquisition system through a vibration monitoring module, providing a reliable data foundation for vibration analysis; the system also achieves accurate extraction of vibration characteristics through a signal processing module, providing key input for intelligent vibration suppression; the system also generates the optimal vibration suppression strategy through an adaptive vibration suppression control module, ensuring the accuracy and adaptability of the control effect; the system also achieves active-passive hybrid vibration suppression through a vibration suppression execution module, significantly improving the vibration suppression effect; and the system also constructs a high-fidelity virtual model through a digital twin and health management module, enabling accurate prediction of equipment health status. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the adaptive vibration suppression system for the variable speed turbine shaft system in this embodiment. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 The diagram shown is a structural schematic of the adaptive vibration suppression system for the variable speed turbine shaft system in this embodiment. The system includes: The vibration monitoring module is used to collect multi-source vibration data; A signal processing module is used to extract features from the multi-source vibration data to obtain vibration feature vectors. The signal processing module is connected to the vibration monitoring module. An adaptive vibration damping control module is used to generate vibration damping control commands based on the vibration feature vector. The adaptive vibration damping control module is connected to the signal processing module. A vibration damping execution module is used to execute vibration damping control according to the vibration damping control command. The vibration damping execution module is connected to the adaptive vibration damping control module. The digital twin and health management module is used to construct a digital twin model of the shaft system and predict the health status of the equipment based on the multi-source vibration data and vibration suppression control commands. The digital twin and health management module is connected to the vibration suppression execution module.

[0022] Specifically, the system is applied to an adaptive vibration suppression terminal for variable-speed steam turbine shaft systems. The system establishes a high-precision multi-source vibration data acquisition system through a vibration monitoring module, providing a reliable data foundation for vibration analysis. The system also achieves accurate extraction of vibration characteristics through a signal processing module, providing key input for intelligent vibration suppression. Furthermore, the system generates optimal vibration suppression strategies through an adaptive vibration suppression control module, ensuring the accuracy and adaptability of the control effect. The system also implements a hybrid active-passive vibration suppression through a vibration suppression execution module, significantly improving the vibration suppression effect. Finally, the system constructs a high-fidelity virtual model through a digital twin and health management module, enabling accurate prediction of equipment health status.

[0023] Specifically, the vibration monitoring module collects multi-source vibration data including: Vibration signals are collected from key points of the turbine bearing housing and rotor to obtain raw vibration data; the raw vibration data is demodulated to obtain shaft torsional vibration characteristic data; the shaft torsional vibration characteristic data is fused from multiple sources to obtain a preliminary vibration dataset; the preliminary vibration dataset is then subjected to quality verification and format standardization to obtain multi-source vibration data.

[0024] Specifically, vibration signals are collected from key points of the turbine bearing housing and rotor to obtain raw vibration data. Specifically, vibration acceleration, displacement and phase signals are collected by a non-contact laser vibration meter, and temperature signals are collected by a fiber Bragg grating sensor to form raw vibration data containing vibration and temperature parameters. The original vibration data is demodulated to obtain shaft torsional vibration characteristic data. Specifically, the original vibration data is processed using rotational speed pulse signal demodulation technology to analyze the shaft torsional vibration mode in real time and obtain characteristic data including torsional vibration frequency and amplitude. The preliminary vibration dataset is obtained by multi-source data fusion of the torsional vibration characteristic data of the shaft system. Specifically, the torsional vibration characteristic data and temperature signal are spatiotemporally registered, and the measurement differences between sensors are eliminated by data fusion algorithm to form a preliminary vibration dataset. The preliminary vibration dataset is subjected to quality verification and format standardization to obtain multi-source vibration data. Specifically, the preliminary vibration dataset is subjected to outlier detection and data integrity verification, and standardized according to a unified data format to output multi-source vibration data containing timestamps, device identifiers and measurement values.

[0025] Specifically, the signal processing module performs feature extraction based on the multi-source vibration data to obtain a vibration feature vector, including: The multi-source vibration data is processed by wavelet packet transform to obtain time-frequency domain features; the time-frequency domain features are processed by modal decomposition to obtain vibration modal parameters; the vibration modal parameters are processed by interference cancellation to obtain clean feature data; and the clean feature data is used to construct feature vectors to obtain vibration feature vectors.

[0026] Specifically, the wavelet packet transform processing of the multi-source vibration data to obtain time-frequency domain features is as follows: the wavelet packet transform algorithm is used to perform frequency band analysis of the multi-source vibration data from 0.1 to 500 Hz, extract the energy distribution features of each frequency band, and obtain time-frequency domain features containing low-frequency oscillations and high-frequency harmonic components. Modal decomposition of the time-frequency domain features to obtain vibration modal parameters specifically involves: performing modal decomposition of the time-frequency domain features based on the improved antlion algorithm, identifying the main vibration modes, and obtaining vibration modal parameters including modal frequencies, damping ratios, and mode shape parameters; The interference elimination process for the vibration modal parameters to obtain pure feature data is specifically as follows: frequency domain blind source separation technology is used to eliminate the noise influence of electromagnetic interference, and essential vibration features are extracted through independent component analysis to obtain pure feature data containing the features of the main vibration sources. The vibration feature vector is constructed by extracting key feature parameters such as vibration energy entropy, modal damping ratio, principal mode frequency, frequency center, root mean square frequency, kurtosis, skewness, and energy proportion of a specific frequency band from the pure feature data. Principal component analysis is used to reduce the dimensionality of the features, and finally an 8-dimensional vibration feature vector is formed.

[0027] Specifically, the adaptive vibration damping control module generates vibration damping control commands based on the vibration feature vector, including: The vibration characteristic vector is used to calculate control quantities to obtain basic control parameters; disturbance compensation is performed on the basic control parameters to obtain disturbance rejection control parameters; the disturbance rejection control parameters are optimized online to obtain optimized control parameters; and the optimized control parameters are encapsulated into instructions to obtain the final vibration suppression control instruction.

[0028] Specifically, the control quantity calculation for the vibration characteristic vector to obtain the basic control parameters is as follows: the 8-dimensional vibration characteristic vector is input into the rotational speed-guide vane dual-variable active disturbance rejection controller, and the basic damping force and guide vane adjustment amount are calculated through the nonlinear state error feedback law to form the basic control parameters. The disturbance compensation for the basic control parameters is performed to obtain the disturbance rejection control parameters. Specifically, the unmodeled dynamics and external disturbances of the system are estimated in real time using a fractional-order disturbance observer, and the total disturbance is fed forward to compensate by an extended state observer to generate the disturbance rejection control parameters. The disturbance rejection control parameters are optimized online to obtain optimized control parameters. Specifically, the control parameters are tuned online based on the improved antlion algorithm to adapt to the dynamic adjustment requirements of ±15° guide angle change and 0.1~0.5mm sealing gap, and the optimized control parameters are output. The optimized control parameters are encapsulated into instructions to obtain the final vibration suppression control instructions. Specifically, the optimized control parameters are converted into control signals that the actuator can recognize, including magnetorheological damper current instructions (0-5A) and hydraulic actuator displacement instructions, thus completing the final generation of the vibration suppression control instructions.

[0029] Specifically, the vibration suppression execution module performs vibration suppression control according to the vibration suppression control command, including: The vibration suppression control commands are parsed and allocated to obtain the control parameters of each execution unit; the current of the magnetorheological damper control parameters is adjusted to achieve rapid adjustment of the damping characteristics; the fluid injection control of the solid-liquid coupling device control parameters is implemented to achieve three-dimensional flow field vibration suppression; and the execution units are coordinated and their status is monitored to form a closed-loop vibration suppression system.

[0030] Specifically, parsing and allocating the vibration suppression control command to obtain the control parameters of each execution unit is as follows: parsing the magnetorheological damper current command and hydraulic actuator displacement command in the vibration suppression control command and allocating them to the corresponding magnetorheological elastomer intelligent bearing unit and solid-liquid two-phase coupling vibration suppression device unit. The control parameters of the magnetorheological damper are adjusted by current to achieve rapid adjustment of damping characteristics. Specifically, according to the current control command of 0-5A, the electromagnetic coil is driven by the power amplifier so that the magnetorheological elastomer completes viscoelastic adjustment within ≤0.1s, and achieves continuous damping control in the range of 10-1000kN·s / m. The control parameters of the solid-liquid coupling device are controlled by fluid injection to achieve three-dimensional flow field vibration suppression. Specifically, the opening of the non-Newtonian fluid injection valve is adjusted according to the displacement control command to form a stable three-dimensional flow field within the 0.2mm narrow gap of the rotor, which effectively suppresses the rotor vortex phenomenon. The coordinated control and status monitoring of each execution unit forms a closed-loop vibration suppression system. Specifically, the working status of each execution unit is collected in real time, and dynamic adjustments are made through a hybrid active-passive vibration suppression system composed of spring isolators to ensure the design target of reducing vibration transmission rate by 20%.

[0031] Specifically, the digital twin and health management module constructs a shaft system digital twin model and predicts the equipment health status based on the multi-source vibration data and vibration suppression control commands, including: The multi-source vibration data and vibration suppression control commands are fused to obtain modeling input data; a multibody dynamics model of the shaft system is constructed based on the modeling input data to obtain an initial digital twin; the initial digital twin is subjected to parameter correction and model verification to obtain a high-precision digital twin model; and a health status prediction is performed based on the high-precision digital twin model to obtain equipment health assessment results.

[0032] Specifically, the multi-source vibration data and vibration suppression control command are fused to obtain the modeling input data. Specifically, the vibration acceleration, displacement, phase and temperature signals in the multi-source vibration data are spatiotemporally registered with the damping parameters and control current in the vibration suppression control command. The Kalman filter algorithm is used to eliminate data inconsistencies and form the modeling input data. The initial digital twin is obtained by constructing a multibody dynamics model of the shaft system based on the modeling input data. Specifically, the ANSYS multibody dynamics simulation platform is used to establish a dynamic model of the rotor-bearing-foundation coupled system based on the modeling input data, and the thermo-mechanical coupling analysis module is integrated to form the initial digital twin. The initial digital twin is subjected to parameter correction and model verification to obtain a high-precision digital twin model. Specifically, the output of the digital twin model is compared with the actual monitoring data by using a residual analysis algorithm, and the model parameters are corrected by using a cyclic plastic damage accumulation algorithm to ensure that the microcrack identification accuracy is ≥90%, thus completing the model verification. Based on the high-precision digital twin model, the health status prediction is performed to obtain the equipment health assessment results as follows: combining the thermo-mechanical coupling model of the gradient porous metal blade, the fatigue life evolution trend of the blade is predicted, and when the life damage is detected to reach the threshold, an early warning signal is generated, and a health assessment report containing the remaining life and maintenance suggestions is output.

[0033] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An adaptive vibration suppression system for a variable speed steam turbine shaft system, characterized in that, include: The vibration monitoring module is used to collect multi-source vibration data; The signal processing module is used to extract features from the multi-source vibration data to obtain vibration feature vectors. An adaptive vibration damping control module is used to generate vibration damping control commands based on the vibration feature vector; The vibration suppression execution module is used to execute vibration suppression control according to the vibration suppression control command; The digital twin and health management module is used to construct a digital twin model of the shaft system and predict the health status of the equipment based on the multi-source vibration data and vibration suppression control commands.

2. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 1, characterized in that, The vibration monitoring module collects multi-source vibration data, including: Vibration signals are collected from key points of the turbine bearing housing and rotor to obtain raw vibration data; the raw vibration data is demodulated to obtain shaft torsional vibration characteristic data; the shaft torsional vibration characteristic data is fused from multiple sources to obtain a preliminary vibration dataset; the preliminary vibration dataset is then subjected to quality verification and format standardization to obtain multi-source vibration data.

3. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 2, characterized in that, The process of collecting vibration signals from key points of the turbine bearing housing and rotor to obtain raw vibration data specifically involves: collecting vibration acceleration, displacement, and phase signals using a non-contact laser vibration meter, and collecting temperature signals using a fiber optic Bragg grating sensor to form raw vibration data containing vibration and temperature parameters. The specific steps for demodulating the original vibration data to obtain shaft torsional vibration characteristic data are as follows: the original vibration data is processed using rotational speed pulse signal demodulation technology, the shaft torsional vibration mode is analyzed in real time, and characteristic data including torsional vibration frequency and amplitude are obtained. The step of fusing multi-source data to obtain a preliminary vibration dataset for the torsional vibration characteristic data of the shaft system specifically involves: performing spatiotemporal registration of the torsional vibration characteristic data and temperature signals, eliminating measurement differences between sensors through a data fusion algorithm, and forming a preliminary vibration dataset. The process of performing quality verification and format standardization on the preliminary vibration dataset to obtain multi-source vibration data specifically involves: performing outlier detection and data integrity verification on the preliminary vibration dataset, standardizing it according to a unified data format, and outputting multi-source vibration data containing timestamps, device identifiers, and measurement values.

4. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 1, characterized in that, The signal processing module performs feature extraction based on the multi-source vibration data to obtain a vibration feature vector including: The multi-source vibration data is processed by wavelet packet transform to obtain time-frequency domain features; the time-frequency domain features are processed by modal decomposition to obtain vibration modal parameters; the vibration modal parameters are processed by interference cancellation to obtain clean feature data; and the clean feature data is used to construct feature vectors to obtain vibration feature vectors.

5. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 4, characterized in that, The process of performing wavelet packet transform on the multi-source vibration data to obtain time-frequency domain features is as follows: the wavelet packet transform algorithm is used to perform frequency band analysis of the multi-source vibration data from 0.1 to 500 Hz, extract the energy distribution features of each frequency band, and obtain time-frequency domain features containing low-frequency oscillations and high-frequency harmonic components. The modal decomposition of the time-frequency domain features to obtain vibration modal parameters specifically involves: performing modal decomposition on the time-frequency domain features based on the improved antlion algorithm, identifying the main vibration modes, and obtaining vibration modal parameters including modal frequencies, damping ratios, and mode shape parameters; The process of eliminating interference in the vibration modal parameters to obtain pure feature data specifically involves: using frequency domain blind source separation technology to eliminate the noise influence of electromagnetic interference, extracting essential vibration features through independent component analysis, and obtaining pure feature data containing the features of the main vibration sources. The process of constructing a vibration feature vector from the pure feature data specifically involves extracting key feature parameters such as vibration energy entropy, modal damping ratio, principal mode frequency, frequency center, root mean square frequency, kurtosis, skewness, and energy proportion of a specific frequency band from the pure feature data, performing feature dimensionality reduction through principal component analysis, and finally forming an 8-dimensional vibration feature vector.

6. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 1, characterized in that, The adaptive vibration damping control module generates vibration damping control commands based on the vibration feature vector, including: The vibration characteristic vector is used to calculate control quantities to obtain basic control parameters; disturbance compensation is performed on the basic control parameters to obtain disturbance rejection control parameters; the disturbance rejection control parameters are optimized online to obtain optimized control parameters; and the optimized control parameters are encapsulated into instructions to obtain the final vibration suppression control instruction.

7. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 6, characterized in that, The calculation of the control quantity of the vibration feature vector to obtain the basic control parameters is specifically as follows: the 8-dimensional vibration feature vector is input into the speed-guide vane dual-variable active disturbance rejection controller, and the basic damping force and guide vane adjustment amount are calculated through the nonlinear state error feedback law to form the basic control parameters. The disturbance compensation of the basic control parameters to obtain the disturbance rejection control parameters is specifically performed as follows: a fractional-order disturbance observer is used to estimate the unmodeled dynamics and external disturbances of the system in real time, and an extended state observer is used to perform feedforward compensation of the total disturbance to generate the disturbance rejection control parameters. The online optimization of the anti-disturbance control parameters to obtain optimized control parameters specifically involves: online tuning of the control parameters based on the improved antlion algorithm to adapt to the dynamic adjustment requirements of ±15° guide angle changes and 0.1~0.5mm sealing gap, and outputting optimized control parameters; The process of encapsulating the optimized control parameters into instructions to obtain the final vibration suppression control instructions specifically involves converting the optimized control parameters into control signals that the actuator can recognize, including magnetorheological damper current instructions and hydraulic actuator displacement instructions, thereby completing the final generation of the vibration suppression control instructions.

8. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 1, characterized in that, The vibration suppression execution module performs vibration suppression control according to the vibration suppression control command, including: The vibration suppression control commands are parsed and allocated to obtain the control parameters of each execution unit; the current of the magnetorheological damper control parameters is adjusted to achieve rapid adjustment of the damping characteristics; the fluid injection control of the solid-liquid coupling device control parameters is implemented to achieve three-dimensional flow field vibration suppression; and the execution units are coordinated and their status is monitored to form a closed-loop vibration suppression system.

9. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 8, characterized in that, The specific steps for parsing and allocating the vibration suppression control command to obtain the control parameters of each execution unit are as follows: parsing the magnetorheological damper current command and the hydraulic actuator displacement command in the vibration suppression control command, and allocating them to the corresponding magnetorheological elastomer intelligent bearing unit and solid-liquid two-phase coupling vibration suppression device unit. The method of adjusting the current of the control parameters of the magnetorheological damper to achieve rapid adjustment of the damping characteristics is as follows: according to the current control command of 0-5A, the electromagnetic coil is driven by the power amplifier, so that the magnetorheological elastomer completes the viscoelastic adjustment within ≤0.1s, and achieves continuous damping control in the range of 10-1000kN·s / m. The fluid injection control of the solid-liquid coupling device control parameters to achieve three-dimensional flow field vibration suppression specifically involves: adjusting the opening of the non-Newtonian fluid injection valve according to the displacement control command to form a stable three-dimensional flow field within the 0.2mm narrow gap of the rotor, effectively suppressing the rotor vortex phenomenon; The coordinated control and status monitoring of each execution unit to form a closed-loop vibration suppression system specifically involves: real-time acquisition of the working status of each execution unit, and dynamic adjustment through a hybrid active-passive vibration suppression system composed of spring isolators to ensure the design target of reducing vibration transmission rate by 20%.

10. The adaptive vibration suppression system for a variable-speed steam turbine shaft system according to claim 1, characterized in that, The digital twin and health management module constructs a shaft system digital twin model based on the multi-source vibration data and vibration suppression control commands, and predicts the equipment health status, including: The multi-source vibration data and vibration suppression control commands are fused to obtain modeling input data; a multibody dynamics model of the shaft system is constructed based on the modeling input data to obtain an initial digital twin; the initial digital twin is subjected to parameter correction and model verification to obtain a high-precision digital twin model; and a health status prediction is performed based on the high-precision digital twin model to obtain equipment health assessment results.

Images