Thermal power generation pipeline vibration mode real-time identification and self-adaptive tuning vibration reduction method

Abstract

The invention discloses a thermal power generation pipeline vibration mode real-time identification and self-adaptive tuning vibration reduction method, system, equipment, medium and program, and belongs to the technical field of energy power engineering. The method comprises the steps of obtaining vibration, stress and temperature data of a thermal power generation pipeline; processing vibration, stress and temperature data of the thermal power generation pipeline by combining a space-time diagram neural network and a digital twinborn model to obtain a real-time identification result of the vibration mode of the thermal power generation pipeline; generating a damping device control signal based on the real-time identification result of the thermal power generation pipeline vibration mode; performing multi-device cooperative vibration reduction simulation on the vibration reduction device control signal to generate a vibration reduction strategy; and the vibration reduction strategy is verified according to the safety threshold value, and after verification is completed, tuning vibration reduction is conducted on the thermal power generation pipeline. According to the method, the problem of insufficient real-time performance or physical authenticity of a single model in vibration mode identification is solved, and the pipeline vibration mode parameters are accurately identified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy power engineering, in particular to a method, system, device, medium and program for real-time identification and self-adaptive tuning and damping of vibration modal of a thermal power generation pipeline. BACKGROUND

[0002] Thermal power generation occupies a pivotal position in the power supply system, and the pipeline system of thermal power generation is like the "blood channel" of the entire power generation process. Its stable operating condition plays a decisive role in power generation efficiency and production safety. In the actual operating environment of thermal power generation, the pipeline is subjected to extremely complex and severe working conditions. High-temperature, high-pressure and high-flow fluid continuously impacts the pipeline, making the pipeline always in a high-load operating state.

[0003] Under such working conditions, various factors can easily induce pipeline vibration. On the one hand, when the fluid flows in the pipeline, its own turbulent characteristics, pressure pulsation, etc. can produce fluid excitation. This excitation, like an invisible "pusher", constantly acts on the pipeline wall, causing pipeline vibration. On the other hand, mechanical equipment in the pipeline system may produce mechanical resonance due to its own imbalance, installation deviation, etc. during operation. This resonance, like an "amplifier", can further exacerbate the vibration amplitude of the pipeline.

[0004] Long-term continuous pipeline vibration is extremely harmful. From a microscopic perspective, it can accelerate the fatigue damage of the pipeline material, causing micro-cracks at the stress concentration points of the pipeline. With the passage of time, these micro-cracks continuously expand and connect, eventually leading to a significant decrease in the strength of the pipeline. From a macroscopic perspective, severe vibration can cause pipeline leakage, resulting in the spouting of high-temperature and high-pressure medium. This not only wastes a large amount of energy and resources, but also poses a serious threat to the safety of surrounding equipment and personnel. In the most extreme cases, the pipeline may even rupture, causing catastrophic accidents such as fires and explosions, resulting in significant economic losses and adverse social impacts for thermal power plants.

[0005] Therefore, it is an urgent need and key technology to realize real-time identification of the vibration modal of the thermal power generation pipeline and to perform self-adaptive tuning and damping based on the identification results to ensure the safe and stable operation of the thermal power unit. Real-time identification of the vibration modal can timely grasp the vibration state of the pipeline and understand key parameters such as the vibration frequency and mode, providing accurate basis for subsequent damping measures. Self-adaptive tuning and damping can automatically adjust the parameters of the damping device based on the real-time identification results, effectively suppressing pipeline vibration under different working conditions. However, in traditional methods for identifying the vibration modal of the thermal power generation pipeline, there is a difficult problem that needs to be solved, i.e., it is difficult to simultaneously consider real-time performance and physical authenticity. This problem seriously affects the effectiveness of subsequent self-adaptive tuning and damping, making it difficult to completely and effectively solve the problem of pipeline vibration. SUMMARY

[0006] In view of the problem in the prior art that the vibration modal identification method of the power generation pipeline is difficult to balance real-time and physical authenticity, and affects the subsequent adaptive tuning and vibration reduction effect, the present application provides a thermal power generation pipeline vibration modal real-time identification and adaptive tuning and vibration reduction method, solves the problem of insufficient real-time or physical authenticity of a single model in vibration modal identification, and achieves accurate identification of pipeline vibration modal parameters.

[0007] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme.

[0008] In the first aspect, the present application provides a thermal power generation pipeline vibration modal real-time identification and adaptive tuning and vibration reduction method, comprising: obtaining vibration, stress and temperature data of the thermal power generation pipeline; processing the vibration, stress and temperature data of the thermal power generation pipeline by combining a space-time graph neural network with a digital twin model to obtain a real-time identification result of the thermal power generation pipeline vibration modal; generating a vibration reduction device control signal based on the real-time identification result of the thermal power generation pipeline vibration modal; performing multi-device collaborative vibration reduction simulation on the vibration reduction device control signal to generate a vibration reduction strategy; verifying the vibration reduction strategy according to a safety threshold, and after verification, tuning and reducing vibration of the thermal power generation pipeline.

[0009] As a further improvement of the present application, the vibration, stress and temperature data of the thermal power generation pipeline are obtained, comprising: obtaining multi-source heterogeneous data of the thermal power pipeline, including collecting global vibration strain and temperature data of the thermal power pipeline through a distributed optical fiber sensor, obtaining surface three-dimensional vibration vector data of the thermal power pipeline through a laser Doppler vibration meter, and collecting stress and strain data of the thermal power pipeline elbow, valve, support point and both sides of the support point through a fiber Bragg grating stress sensor; aligning the collected multi-source heterogeneous data in space and time, and completing data preprocessing by removing noise through a filtering algorithm to obtain the vibration, stress and temperature data of the thermal power generation pipeline.

[0010] As a further improvement of the present application, the vibration, stress and temperature data of the thermal power generation pipeline are processed by combining a space-time graph neural network with a digital twin model to obtain the vibration modal parameters of the thermal power generation pipeline, comprising: inputting the vibration, stress and temperature data of the thermal power generation pipeline into the space-time graph neural network to obtain a calculation result; inputting the vibration, stress and temperature data of the thermal power generation pipeline into the digital twin model to obtain predicted modal parameters; The calculation result and the predicted modal parameter are weighted and fused to obtain a real-time identification result of the vibration modal of the thermal power generation pipeline.

[0011] As a further improvement of the application, the space-time graph neural network comprises 3 layers of graph convolution layers and 2 layers of gated recurrent units, and is constructed by taking the thermal power generation pipeline structure nodes and sensor data as graph nodes and the physical connection relationship between nodes as edges. The digital twin model is established based on ANSYS finite element analysis software, and integrates the elastic modulus, Poisson's ratio, density parameters of the thermal power generation pipeline material, and fixed support boundary conditions and thermal-structure coupling physical fields.

[0012] As a further improvement of the application, the real-time identification result of the vibration modal of the thermal power generation pipeline is used to generate a vibration reduction device control signal, which comprises: The real-time identification result of the vibration modal of the thermal power generation pipeline is solved in real time by using a lightweight neural network and a model predictive control algorithm to generate a vibration reduction device control signal. The output calculation of the lightweight neural network is as follows:

[0013] Among them, is the output of the lightweight neural network, i.e. the control signal of the piezoelectric ceramic sheet, is the vibration signal feature of the input node, is the number of input nodes, is the weight connected between the output node and the input node, is the bias of the output node, is the activation function.

[0014] As a further improvement of the application, the vibration reduction device control signal is simulated in a multi-device collaborative vibration reduction mode to generate a vibration reduction strategy, which comprises: According to the vibration reduction device control signal, the magnetic rheological liquid damper, the shape memory alloy cable and the piezoelectric ceramic sheet in the thermal power generation pipeline are adjusted to work collaboratively to reduce vibration, and a multi-device collaborative vibration reduction simulation result is obtained. According to the multi-device collaborative vibration reduction simulation result, a vibration reduction strategy is generated.

[0015] As a further improvement of the application, the vibration reduction strategy is verified according to a safety threshold, and after the verification is completed, the thermal power generation pipeline is tuned for vibration reduction, which comprises: A pipeline stress-deformation-safety margin model is established by a physically constrained neural network, and a dynamic safety threshold of the thermal power pipeline is obtained according to the safety margin model:

[0016] wherein, is a dynamic safety threshold, is a maximum allowable stress of the pipeline material, is a maximum allowable strain, is a current temperature, is a current pressure, is a mapping function obtained by training the physically constrained neural network, which comprehensively considers material properties and working condition parameters to determine the safety threshold; extracting real-time stress in the vibration reduction strategy If the real-time stress is greater than or equal to the dynamic safety threshold , the current adjustment sequence of the current of the magneto-rheological fluid damper of the thermal power pipeline and the current adjustment sequence of the heating power of the shape memory alloy cable are recalculated and adjusted. If the real-time stress is less than the dynamic safety threshold , the thermal power pipeline is not adjusted.

[0017] In a second aspect, the present application provides a thermal power pipeline vibration mode real-time identification and adaptive tuning vibration reduction system, comprising: A data acquisition module is used to acquire vibration, stress and temperature data of the thermal power pipeline; An identification result module is used to process the vibration, stress and temperature data of the thermal power pipeline by combining a spatio-temporal graph neural network with a digital twin model to obtain real-time identification results of the thermal power pipeline vibration mode; A signal generation module is used to generate a vibration reduction device control signal based on the real-time identification results of the thermal power pipeline vibration mode; A vibration reduction strategy module is used to perform multi-device collaborative vibration reduction simulation on the vibration reduction device control signal to generate a vibration reduction strategy; A tuning vibration reduction module is used to verify the vibration reduction strategy according to the safety threshold, and after the verification is completed, the thermal power pipeline is tuned and damped.

[0018] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the thermal power pipeline vibration mode real-time identification and adaptive tuning vibration reduction method.

[0019] In a fourth aspect, the present application provides a computer readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for real-time identification of vibration modal of power plant pipeline and adaptive tuning vibration reduction.

[0020] In a fifth aspect, the present application provides a computer program product, characterized by comprising computer instructions, which, when executed by a processor, implement the steps of the method for real-time identification of vibration modal of power plant pipeline and adaptive tuning vibration reduction.

[0021] Compared with the prior art, the present application has the following beneficial effects: The present application ingeniously combines the spatiotemporal graph neural network with the digital twin model to process the vibration, stress and temperature data of the power plant pipeline. The spatiotemporal graph neural network has strong data processing and pattern recognition capabilities, and can quickly capture the dynamic characteristics in the pipeline vibration data, ensuring real-time performance. The digital twin model is based on physical principles and highly restores the actual physical characteristics of the pipeline, ensuring physical authenticity. The two complement each other and can accurately and in real time identify the vibration modal of the power plant pipeline, effectively solving the defects of single model in real-time performance or physical authenticity in vibration modal identification, and laying a solid and reliable foundation for subsequent vibration reduction work. Based on the accurate real-time identification result of the vibration modal, a vibration reduction device control signal is generated, and further simulation of multi-device collaborative vibration reduction is performed on the signal to generate a scientific and reasonable vibration reduction strategy, realizing seamless connection from vibration identification to vibration reduction strategy formulation, and greatly shortening the time cycle of the entire vibration reduction process. At the same time, the multi-device collaborative simulation fully considers the interaction between the vibration reduction devices and the synergistic effect of the overall system, so that the generated vibration reduction strategy is more comprehensive and effective, and can provide personalized solutions for different types of pipeline vibration conditions, significantly improving the vibration reduction effect.

[0022] Further, the vibration reduction strategy is strictly verified by the safety threshold, which adds a solid safety line to the vibration reduction process, ensuring that the vibration reduction strategy adopted meets the vibration reduction demand without causing damage to the pipeline system and other related equipment, and ensuring the safe and stable operation of the power plant system. Through this rigorous verification mechanism, the present application not only improves the reliability of vibration reduction, but also reduces the risk of accidents caused by improper vibration reduction operation, reduces equipment maintenance and replacement costs, and prolongs the service life of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0023] The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present application in any way. In the drawings: Figure 1 The method flowchart of the present application is a method for real-time identification of vibration modal of power plant pipeline and adaptive tuning vibration reduction. Figure 2 A specific method flowchart of a thermal power generation pipeline vibration modal real-time identification and self-adaptive tuning and damping method of the present application; Figure 3 A structural schematic diagram of a thermal power generation pipeline vibration modal real-time identification and self-adaptive tuning and damping system of the present application; Figure 4 An electronic device schematic diagram in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the present application will be clearly and completely described below in combination with the drawings in the present application. The described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the present application.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0026] In view of the problem in the prior art that the vibration modal identification method of the power generation pipeline is difficult to balance real-time and physical authenticity, affecting the subsequent self-adaptive tuning and damping effect, the present application provides a thermal power generation pipeline vibration modal real-time identification and self-adaptive tuning and damping method, as shown in Figure 1 The present application specifically comprises: S100: obtaining vibration, stress and temperature data of the thermal power generation pipeline; S200: processing the vibration, stress and temperature data of the thermal power generation pipeline by combining a space-time graph neural network with a digital twin model to obtain a real-time identification result of the thermal power generation pipeline vibration modal; S300: generating a damping device control signal based on the real-time identification result of the thermal power generation pipeline vibration modal; S400: performing multi-device collaborative damping simulation on the damping device control signal to generate a damping strategy; S500: verifying the damping strategy according to a safety threshold, and after the verification is completed, tuning and damping the thermal power generation pipeline.

[0027] The present method solves the problem of insufficient real-time or physical authenticity in single model vibration modal identification, and achieves accurate identification of pipeline vibration modal parameters.

[0028] The method will be further explained below with reference to the specific accompanying drawings.

[0029] like Figure 2 As shown, the present invention provides a method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines, comprising the following steps: S1: Multi-source data acquisition: Vibration, stress and temperature data of thermal power generation pipelines are acquired through distributed fiber optic sensors, laser Doppler vibration meters, fiber Bragg grating stress sensors and temperature sensors.

[0030] Distributed fiber optic sensors collect vibration, strain, and temperature data across the entire pipeline. A laser Doppler vibrometer acquires three-dimensional vibration vector data on the pipeline surface. Fiber Bragg grating stress sensors collect stress and strain data at pipeline bends, valves, support points, and on both sides of the support points. The collected multi-source heterogeneous data are spatiotemporally aligned, and noise is removed using a filtering algorithm to complete data preprocessing.

[0031] Specifically, by using distributed fiber optic sensors, laser Doppler vibration meters, and fiber Bragg grating stress sensors, the system can collect vibration strain and temperature data across the entire pipeline network, three-dimensional vibration vectors on the surface, and stress and strain data at key locations such as elbows, valves, and support points. This allows for comprehensive monitoring of pipeline operation status, from overall to local conditions and from vibration characteristics to stress states. Distributed fiber optic sensors enable comprehensive pipeline monitoring, capturing overall vibration trends and temperature distribution. Laser Doppler vibration meters focus on three-dimensional vibration vectors on the surface, accurately depicting local vibration details. Fiber Bragg grating stress sensors pinpoint key vulnerable areas, enhancing stress monitoring at critical points. Spatiotemporal alignment uses the sampling time of the distributed fiber optic sensors as a reference, synchronizing data from the laser Doppler vibration meters and fiber Bragg grating stress sensors to the same time series via linear interpolation algorithms. Based on the pipeline's three-dimensional coordinate system, the spatial coordinates of each sensor's monitoring location are bound to data tags to ensure spatial correspondence. Bandpass filtering is used for preprocessing, filtering out low-frequency noise below 2Hz (e.g., slow changes in ambient temperature) and high-frequency noise above 100Hz (e.g., electromagnetic pulse interference) within the common vibration frequency range of thermal power plant pipelines. This retains effective vibration signal frequency bands, providing comprehensive, accurate, and clean raw data support. This ensures the accuracy and reliability of the entire vibration mode identification and adaptive tuning vibration reduction method from the source, allowing subsequent data-driven analysis, control, and optimization operations to be based on a true reflection of the pipeline's condition, achieving effective monitoring and control of pipeline vibration.

[0032] Distributed optical fiber sensor, laser Doppler vibration meter, fiber Bragg grating stress sensor and temperature sensor are used to collect vibration, stress and temperature data of the pipeline. Distributed optical fiber sensor can realize global vibration strain and temperature monitoring of the pipeline, with wide coverage and the ability to capture the overall vibration trend of the pipeline. Laser Doppler vibration meter can accurately obtain the three-dimensional vibration vector of the pipeline surface, providing accurate data for analyzing the local vibration characteristics. Fiber Bragg grating stress sensor focuses on stress and strain collection at specific locations of the pipeline, sensitive to stress changes at key positions, and can help to timely detect stress abnormalities. Temperature sensor collects temperature data, which can assist in correcting the influence of temperature on vibration and stress monitoring results, ensuring that the collected data can fully and accurately reflect the pipeline operating state, laying a reliable data foundation for subsequent vibration modal identification and other steps, and enabling the entire method to have the ability to accurately perceive the pipeline state from the source.

[0033] S2: Real-time vibration modal identification: Use the combination of spatio-temporal graph neural network and digital twin model to process the collected data, identify the pipeline vibration modal parameters, and use the Bayesian optimization algorithm to update the model parameters of the digital twin model online.

[0034] The spatio-temporal graph neural network contains 3 layers of graph convolution layers and 2 layers of gated recurrent units. The pipeline structure nodes and sensor data are used as graph nodes, and the physical connection relationship between nodes is used as edges to construct graph structure data. The digital twin model is established based on ANSYS finite element analysis software, integrating the elastic modulus, Poisson's ratio, density parameters of the pipeline material, fixed support boundary conditions, and thermal-structural coupling physical fields.

[0035] Specifically, the combination of spatiotemporal graph neural network and digital twin model is as follows: First, feature extraction is performed on the preprocessed multi-source data using a spatiotemporal graph neural network. This network uses pipeline structural nodes, such as elbows, valves, support points, and sensor monitoring points, as graph nodes, and physical connections between nodes, such as the distance between adjacent nodes along the pipeline axis, as graph edges to construct graph structure data. Spatial correlation of vibration data is extracted through three layers of graph convolutional layers, and time series features are captured through two layers of gated recurrent units, outputting preliminary modal parameters including natural frequencies and mode shapes. Simultaneously, a digital twin model is established based on ANSYS finite element analysis software, inputting pipeline material parameters, fixed support boundary conditions, and thermo-structural coupling physical fields. Real-time temperature and pressure data and preliminary modal parameters are used as excitations to simulate and generate modal prediction values. Finally, the real-time prediction of the neural network and the physical accuracy of the digital twin are fused through a weighted fusion formula to output the final modal parameters. Comprehensive and accurate data are obtained in the multi-source data acquisition step. Based on multi-dimensional data such as pipeline vibration, stress, and temperature, the spatiotemporal graph neural network, an existing technology in the field of vibration mode identification, leverages its advantages in processing graph-structured data. It constructs a graph-structured data system using pipeline structural nodes and sensor data as graph nodes and physical connections between nodes as edges. This system can adapt to the physical topological characteristics of pipelines, fully explore the spatial correlations and temporal evolution patterns between multi-source data, and effectively extract preliminary vibration mode features. Simultaneously, a digital twin model, built using ANSYS finite element analysis software, can reproduce the actual operating scenario of the pipeline based on physical mechanisms. Combined with the accurate data preprocessed from the multi-source data acquisition steps, and in collaboration with the spatiotemporal graph neural network, the algorithm efficiently calculates and quickly captures data features. The physical model ensures the authenticity of the results. By processing the acquired data, the system accurately identifies pipeline vibration mode parameters, ensuring real-time identification and adaptive tuning of vibration modes in the entire thermal power generation pipeline. This allows for precise understanding of the essence of pipeline vibration even under complex operating conditions, achieving effective vibration reduction and safety assurance.

[0036] The Bayesian optimization algorithm updates the parameters of the digital twin model based on the following formula:

[0037] in, For the updated model parameters, For the parameter search space and These are the current parameters. The corresponding mean and standard deviation of the objective function, To weigh the coefficients between exploration and utilization.

[0038] The modal parameters predicted by the digital twin model are weighted and fused with the calculation results of the spatiotemporal graph neural network model to output the real-time identification results of the pipeline vibration modes. The fusion formula is as follows:

[0039] wherein, is the fused modal parameter, is the modal parameter calculated by the spatio-temporal graph neural network, is the modal parameter predicted by the digital twin model, and is the weighted coefficient, and satisfies .

[0040] In the multi-source data acquisition step, comprehensive and pre-processed pipeline data is obtained. On the basis of the spatio-temporal graph neural network and the digital twin model established and integrated with ANSYS, pipeline materials and multi-physical fields, in the vibration modal real-time identification step, the Bayesian optimization algorithm updates the digital twin model parameters according to the formula Taking the pipeline vibration modal identification as an example, assuming that the objective function is the modal frequency prediction error, is the current parameter , the frequency prediction error mean is , the error standard deviation is Balancing the exploration of new parameters and the use of existing optimal parameters, through the formula iteration optimization, for example, the initial parameter corresponds to the error mean , the standard deviation , after calculation and update to , the error is smaller, and the digital twin model is dynamically calibrated to better fit the actual pipeline. Then, through the fusion formula .

[0041] For example, the output of a certain order modal frequency is , the prediction is , and , , then , the spatio-temporal graph neural network calculation result and the digital twin model prediction value are fused, and under the synergistic action, a more accurate pipeline vibration modal real-time identification result is output, providing a reliable basis for the subsequent, improving the precision and dynamic adaptability of the pipeline vibration modal identification, and ensuring more effective vibration control and more in-place safety protection.

[0042] By combining a spatiotemporal graph neural network (SPNN) with a digital twin model to process collected data, pipeline vibration modal parameters are identified. The digital twin model utilizes a Bayesian optimization algorithm to update its parameters online. The SPNN adapts to the topological characteristics of the pipeline structure, effectively uncovering the spatial correlations and temporal evolution patterns of vibration data, and quickly extracting preliminary features of the vibration modes. The digital twin model, built based on physical mechanisms, simulates actual pipeline operating scenarios. Combined with the Bayesian optimization algorithm for online parameter updates, it dynamically calibrates the deviation between the model and the real pipeline, ensuring the model always closely matches the actual pipeline condition. This approach leverages the neural network for efficient computation while relying on the digital twin to guarantee physical accuracy, accurately identifying pipeline vibration modal parameters. This provides a precise basis for subsequent vibration reduction strategy formulation, enabling the system to accurately grasp the essence of pipeline vibration even under complex operating conditions, ensuring that subsequent vibration reduction measures are targeted and effective.

[0043] S3: Adaptive tuning vibration reduction strategy generation: Based on vibration mode parameters and operating condition parameters, the control signal of the vibration reduction device is generated using model predictive control algorithm and lightweight neural network; The model predictive control algorithm module combines the adjustable damping force characteristics of the magnetorheological fluid damper and the stiffness adjustment characteristics of the shape memory alloy cable. Based on the multi-step modal prediction results, the following formulas are used to calculate the current adjustment sequence of the magnetorheological fluid damper and the heating power adjustment sequence of the shape memory alloy cable:

[0044] in, for The control input at any given time is either the current of the magnetorheological fluid damper or the heating power of the shape memory alloy cable. To control the feasible range of inputs. for Reference modal parameters at time 10:00. Based on Time information prediction Modal parameters at time t, To predict the time domain, To control the time domain, To control the incremental penalty coefficient.

[0045] Specifically, based on the accurate pipeline vibration modal parameters output in the real-time vibration modal identification step, the model predictive control algorithm module, combined with the adjustable damping force of the magnetorheological fluid damper and the adjustable stiffness of the shape memory alloy cable, uses multi-step modal prediction results and formulas to... Calculate the control sequence, taking the frequency of a certain vibration mode of a pipeline needing to be adjusted from 20Hz to 18Hz as an example. Let the prediction time domain be... Control Time Domain Penalty coefficient Reference modal parameters 、 、 , predicting based on modal parameters identified at time k 、 、 , substituting into the formula to solve for the magnetorheological fluid damper current, such as adjusting from an initial 2A to a sequence of 2.2A, etc., and adjusting the shape memory alloy cable heating power sequence, so that the control input is within the feasible range , such as current 0-5A, heating power 0-200W, both gradually approaching the reference value of the predicted modal parameters, and avoiding large fluctuations in the control quantity through the penalty term, generating a precise control sequence that adapts to the dynamic vibration of the pipeline using the model predictive control algorithm, providing instructions for the collaborative control steps of multiple vibration reduction devices, adjusting the magnetorheological fluid damper, shape memory alloy cable, etc. to achieve adaptive tuning of the pipeline vibration, effectively suppressing the vibration amplitude and tracking the modal frequency drift, improving the accuracy and adaptability of the vibration reduction strategy, and ensuring the stable operation of the pipeline.

[0046] In the adaptive tuning vibration reduction strategy, a lightweight neural network is used on the edge computing node to real-time solve the vibration signals sensed by the piezoelectric ceramic sheet, generating active and dynamic control signals for the piezoelectric ceramic sheet. The output calculation of the lightweight neural network is as follows:

[0047] Wherein, is the output of the lightweight neural network, i.e. the control signal of the piezoelectric ceramic sheet, is the vibration signal feature of the th input node, is the number of input nodes, is the weight connecting the th output node and the th input node, is the bias of the th output node, is the activation function.

[0048] Specifically, in the adaptive tuning vibration reduction strategy generation process, relying on the accurate pipeline vibration modal parameters output by the vibration modal real-time identification step, the lightweight neural network plays a role on the edge computing node. Taking the high-frequency vibration of the pipeline due to fluid pulsation as an example, after the piezoelectric ceramic sheet senses the vibration signal, it will extract features such as the vibration frequency of 25Hz and the amplitude of at a certain time, etc. The corresponding input node features , , the number of input nodes , etc. are input into the lightweight neural network, and the calculation is performed according to the formula , assuming that the connection weight , , bias , activation function is the ReLU function, and the calculation is as follows , the active control signal of the piezoelectric ceramic sheet is generated, and the vibration signal sensed by the piezoelectric ceramic sheet is solved in real time at the edge node by means of a lightweight neural network, which can efficiently respond to high-frequency and transient vibrations of the pipeline, such as vibrations caused by rapid opening and closing of the valve, making up for the shortcomings of model predictive control algorithm in extremely short time scale control. Collaborate with the control strategies of magnetorheological fluid damper and shape memory alloy cable to make the piezoelectric ceramic sheet output reverse control force in time, suppress the resonance of the pipeline, and improve the processing capacity of the vibration reduction system to complex and high-frequency vibrations, enhance the comprehensiveness and real-time performance of the power generation pipeline vibration control, and ensure the stability of the pipeline operation.

[0049] Based on the vibration modal parameters and working condition parameters, the model predictive control algorithm and the lightweight neural network are used to generate the control signal of the vibration reduction device. The model predictive control algorithm has predictability and optimality, which can plan the vibration reduction action in advance combined with the working condition parameters, and adapt to the dynamically changing operating conditions of the pipeline; the lightweight neural network focuses on real-time signal solving, which can quickly respond to high-frequency vibrations and other transient changes. The model predictive control algorithm optimizes the vibration reduction strategy from the macro working condition level, and the lightweight neural network ensures the real-time performance of the control from the micro signal level. The generated control signal can accurately drive the vibration reduction device, so that the vibration reduction strategy can dynamically adapt to the changes of the pipeline vibration, improve the adaptability and response speed of the vibration reduction system, and provide accurate instructions for effective vibration reduction.

[0050] S4: Multi-vibration reduction device collaborative control: adjust the magnetorheological fluid damper, shape memory alloy cable and piezoelectric ceramic sheet to work collaboratively for vibration reduction according to the control signal; Adjust the magnetorheological fluid damper, shape memory alloy cable and piezoelectric ceramic sheet to work collaboratively for vibration reduction according to the control signal. The magnetorheological fluid damper can quickly adjust the damping to effectively suppress high-frequency vibrations; the shape memory alloy cable can adapt to the drift of the pipeline vibration mode by changing its own stiffness; the piezoelectric ceramic sheet can output a reverse force to directly cancel out the resonance of the pipeline. Collaborative cooperation covers a wide range of vibration scenarios, accurately applies force according to different vibration characteristics, makes up for the functional limitations of single vibration reduction device, forms composite vibration reduction capacity, and comprehensively and stably reduces pipeline vibration, reducing the adverse effects of vibration on the safety and efficiency of power generation pipeline equipment; S5: Safety boundary monitoring and control optimization: establish a safety margin model based on stress and strain data, and optimize the vibration reduction strategy and verify it according to the safety threshold.

[0051] A pipeline stress-deformation-safety margin model is established by a physically constrained neural network, and the dynamic safety threshold under the current working condition is calculated according to the safety margin model using the following formula:

[0052] wherein, is a dynamic safety threshold, is the maximum allowable stress of the pipe material, is the maximum allowable strain, is the current temperature, is the current pressure, is a mapping function trained by a physically constrained neural network, which comprehensively considers material properties and working condition parameters to dynamically determine the safety threshold.

[0053] Specifically, the safety margin model is established based on a physically constrained neural network. In the safety boundary monitoring and control optimization step based on the regulation of the pipe vibration, a pipe stress-deformation-safety margin model is established by the physically constrained neural network, and the dynamic safety threshold is calculated according to the formula , , Taking Q345B steel as an example for the power generation pipeline, the maximum allowable stress and the maximum allowable strain , if the current working condition temperature and the pressure , the physically constrained neural network is trained to learn the correlation between material properties and working condition parameters, and a mapping function is constructed. The above parameters are substituted into the formula, and it is assumed that is calculated by the network. This dynamic safety threshold comprehensively considers the inherent properties of the material and the real-time working condition, and is more practical than the fixed threshold, providing accurate basis for subsequent judgment of whether the pipe stress is safe. When the real-time stress monitored by the fiber Bragg grating stress sensor in the multi-source data acquisition step approaches or exceeds the threshold, feedback can be triggered, and the control is optimized in the links such as real-time identification of vibration mode, generation of self-adaptive tuning damping strategy, etc. For example, the current of the magnetorheological fluid damper calculated by the model predictive control algorithm, the heating power of the shape memory alloy cable, etc. are adjusted, or the modified damping strategy is pre- simulated in the digital twin model, and the safety is verified based on the multi-physical field coupling equation to avoid deformation, cracking and other failures of the pipeline due to stress exceeding the limit, to ensure the safe operation of the power generation pipeline structure and the system stability, and to achieve the dynamic balance of vibration control and safety protection.

[0054] If the real-time stress approaches or exceeds the dynamic safety threshold , the information is fed back to the model predictive control algorithm module, and the current adjustment sequence of the magnetorheological fluid damper and the heating power adjustment sequence of the shape memory alloy cable are recalculated and adjusted. The adjustment process is based on the model predictive control algorithm formula, and the real-time data transmission between steps is realized through the OPCUA protocol interface.

[0055] When the real-time stress Approaching or exceeding the safety threshold The information is fed back to the model predictive control algorithm module, which generates formulas for calculating the magnetorheological fluid damper current adjustment sequence and the shape memory alloy cable heating power adjustment sequence in the previous adaptive tuning damping strategy generation step, re-substitutes the updated stress constraint condition, re-calculates and adjusts the magnetorheological fluid damper current adjustment sequence and the shape memory alloy cable heating power adjustment sequence, ensures that the damping operation ensures the vibration suppression effect of the pipeline while not breaking the safety boundary of the pipeline, and realizes real-time data transmission between steps through the OPCUA protocol interface, ensures efficient linkage from safety boundary monitoring to control strategy adjustment, forms a closed loop for the entire thermal power pipeline vibration control system, dynamically adapts to the pipeline safety and vibration suppression requirements in the process of collaborative control of multiple damping devices such as magnetorheological fluid dampers and shape memory alloy cables, effectively avoids the risk of pipeline structure safety caused by damping operation, and further strengthens the safety line under the premise of ensuring the function of the vibration modal real-time identification and adaptive tuning damping method.

[0056] The adjusted damping strategy is pre-played in the pipeline digital twin, simulating the vibration response and stress distribution under different working conditions, and verifying the safety of the control parameters through real-time closed-loop simulation. The simulation process is based on the multi-physical field coupling equation of the pipeline:

[0057] wherein, is the stress tensor, is the material density, is the acceleration, is the elastic matrix, is the strain tensor, is the thermal strain, is the displacement vector.

[0058] When the dynamic safety threshold is calculated by the physical constraint neural network, and the real-time stress triggers feedback adjustment of the damping strategy, the adjusted damping strategy is pre-played in the pipeline digital twin, and the multi-physical field coupling equation

[0059] can simulate the vibration response and stress distribution under different working conditions. Taking the thermal power pipeline material as steel, the thermal strain is affected by the current temperature , for example, assuming that the pipeline vibration control strategy is changed by adjusting the magnetorheological fluid damper current and the shape memory alloy cable heating power, the new strategy parameters are input into the digital twin, and substituted into the coupling equation, in which the stress tensor , the strain tensor The physical quantities are correlated with each other, and through numerical calculation, such as discrete solution by using the finite element method, the vibration displacement of the pipeline under the new vibration reduction strategy and different working conditions, such as pressure fluctuation caused by load change, can be obtained , acceleration and stress distribution. For example, the simulation shows that the maximum stress of a certain section of the pipeline under the adjusted strategy is , which is lower than the dynamic safety threshold . Combined with the maximum allowable stress of the steel material , the current temperature and pressure, etc., it is verified that the control parameter is safe and can be applied to actual pipeline vibration reduction. If the stress is out of limit, the strategy is adjusted again and simulated again to verify the safety. Through the digital twin and the multi-physical field coupling equation, the safety of the vibration reduction strategy is verified in the virtual environment in advance to avoid the risk of pipeline stress caused by the simultaneous adjustment of multiple vibration reduction devices such as magnetorheological fluid damper, shape memory alloy cable and piezoelectric ceramic sheet, to ensure the structural safety of the power generation pipeline in the vibration control process, to improve the reliability of the real-time identification and adaptive tuning of the vibration mode and the vibration reduction scheme, and to ensure the stable operation of the pipeline.

[0060] A safety margin model is established based on stress and strain data, and the vibration reduction strategy is optimized and verified according to the safety threshold. The stress and strain data are directly related to the safety of the pipeline structure, and the safety margin model is established based on the stress and strain data, which can dynamically define the safe operation range of the pipeline. Combined with the safety threshold, the vibration reduction strategy can suppress vibration while avoiding the risk of stress exceeding the limit due to vibration reduction operation. Through the verification link, the execution effect of the vibration reduction strategy under the safety boundary condition can be reproduced to ensure that the vibration reduction measures are balanced with the safety of the pipeline structure. This step builds a strong defense line for the safe operation of the pipeline, ensures the safety of the pipeline structure while efficiently reducing vibration, and improves the reliability of the power generation system operation; Through multi-dimensional data acquisition, accurate modal identification, adaptive vibration reduction strategy generation, multi-device collaborative vibration reduction and safety boundary control, from data acquisition to strategy execution to safety protection, the problem of accurate identification and effective vibration reduction of power generation pipeline vibration is solved, realizing real-time monitoring, accurate identification and adaptive vibration reduction of pipeline vibration, effectively improving the stability and safety of power generation pipeline operation, and ensuring the reliable operation of the power generation system.

[0061] In summary, firstly, this invention utilizes a combination of spatiotemporal graph neural networks and digital twin models to process collected data, updates the digital twin model parameters using a Bayesian optimization algorithm, and weightedly fuses the results of both. This solves the problem of insufficient real-time performance or physical realism in vibration mode identification by a single model, achieving accurate identification of pipeline vibration mode parameters and improving mode identification accuracy and dynamic adaptability. Secondly, based on vibration mode parameters and operating condition parameters, this invention uses model predictive control algorithms and lightweight neural networks to generate control signals for vibration damping devices, and adjusts multiple vibration damping devices, such as magnetorheological fluid dampers, to work collaboratively. This solves the problem that traditional vibration damping methods are unable to adapt to dynamic changes in pipeline vibration and have limited damping effects, achieving the generation of precise control signals and effective suppression of wideband vibration, thus improving the adaptability and response speed of the vibration damping system. Third, this invention establishes a safety margin model through a physical constraint neural network, optimizes vibration reduction strategies based on safety thresholds, and pre-verifies the model in a pipeline digital twin. This solves the problems of potential pipeline stress exceeding limits and insufficient safety assurance during vibration reduction, achieving the effects of dynamically monitoring safety boundaries, optimizing vibration reduction strategies, and verifying their safety, thus ensuring the structural safety of thermal power generation pipelines and the stable operation of the system. Fourth, this invention uses multiple types of sensors, such as distributed fiber optic sensors and laser Doppler vibration meters, to collect multi-source data. It also performs spatiotemporal alignment and filtering preprocessing on the data, solving the problems of incomplete monitoring by traditional single sensors and the presence of noise and spatiotemporal bias in the data. This achieves comprehensive and accurate acquisition of pipeline vibration, stress, and temperature data, providing a reliable data foundation for subsequent vibration mode identification and vibration reduction control.

[0062] The second objective of this invention is to propose a real-time vibration mode identification and adaptive tuning vibration reduction system for thermal power generation pipelines, such as... Figure 3 As shown, it includes: Data acquisition module 100: used to acquire vibration, stress and temperature data of thermal power generation pipelines; Recognition Result Module 200: Used to process the vibration, stress and temperature data of thermal power generation pipelines by combining spatiotemporal graph neural networks and digital twin models to obtain real-time recognition results of the vibration modes of thermal power generation pipelines; Signal generation module 300: used to generate control signals for vibration reduction devices based on the real-time identification results of vibration modes of thermal power generation pipelines; Vibration reduction strategy module 400: used to simulate multi-device coordinated vibration reduction based on the control signals of the vibration reduction device, and generate a vibration reduction strategy; Tuned vibration reduction module 500: Used to verify the vibration reduction strategy according to the safety threshold. After verification, the thermal power generation pipeline is tuned for vibration reduction.

[0063] like Figure 4As shown, the third object of the present application is to provide an electronic device, which comprises a processor 601, a memory 602 and a display screen 603. Wherein the memory 602 and the display screen 603 are connected with the processor 601, such as through a bus 604. Optionally, the electronic device can further comprise a transceiver 605. It should be noted that the transceiver 605 is not limited to one in actual application, and the structure of the electronic device does not constitute a limitation to the embodiments of the present application.

[0064] The processor 601 can be a CPU (Central Processing Unit, central processor), a general-purpose processor, a DSP (Digital Signal Processor, data signal processor), an ASIC (Application Specific Integrated Circuit, application specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure of the present application. The processor 601 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0065] The bus 604 can include a channel for transmitting information between the above-mentioned components. The bus 604 can be a PCI (Peripheral Component Interconnect, peripheral component interconnect) bus or an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus, etc. The bus 604 can be divided into an address bus, a data bus, a control bus, etc.

[0066] The memory 602 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions; a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions; an EEPROM (Electrically Erasable Programmable Read-Only Memory), a CD-ROM (Compact Disc Read-Only Memory) or other optical disc storage, a magnetic disk storage or other magnetic storage devices or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0067] The memory 602 is configured to store application program codes for implementing the solutions of the present application, and the processor 601 is configured to control the execution. The processor 601 is configured to execute the application program codes stored in the memory 602 to implement the content shown in the foregoing method embodiments.

[0068] Figure 4 The electronic device shown is merely an example, and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0069] A fourth object of the present application is to provide a computer readable storage medium storing a computer program, which stores a computer program, and the program is executed by a processor to implement various processes of the method embodiments as described above. Figure 1 The memory including instructions executable by a processor of an electronic device to perform the methods described above.

[0070] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium can be, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the above. Specifically, the computer readable storage medium can be a portable computer diskette, a hard disk, a USB (Universal Serial Bus) flash disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, or any combination of the above.

[0071] A fifth objective of this invention is to provide a computer program product, including computer instructions that, when executed by a processor, implement the above-described... Figure 1 The various processes of the method embodiments shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0072] Many embodiments and applications beyond the examples provided will be apparent to those skilled in the art upon reading the foregoing description. Therefore, the scope of this teaching should not be determined by reference to the foregoing description, but rather by reference to the foregoing claims and the full scope of their equivalents. For purposes of completeness, all articles and references, including patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein in the foregoing claims is not intended as a waiver of that subject matter, nor should it be construed as an indication that the applicant has not considered that subject matter as part of the disclosed inventive subject matter.

[0073] The above content provides a further detailed description of the present invention. It should not be construed that the specific embodiments of the present invention are limited to this. For those skilled in the art, several simple deductions or substitutions can be made without departing from the concept of the present invention, and all such deductions or substitutions should be considered as falling within the scope of protection of the present invention as defined by the submitted claims.

Claims

1. A method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines, characterized in that, include: Obtain vibration, stress, and temperature data for thermal power generation pipelines; By combining spatiotemporal graph neural networks with digital twin models to process vibration, stress, and temperature data of thermal power generation pipelines, real-time identification results of vibration modes of thermal power generation pipelines are obtained. Based on the real-time identification results of the vibration modes of thermal power generation pipelines, control signals for vibration reduction devices are generated. Multi-device coordinated vibration reduction simulation is performed on the control signals of the vibration reduction device to generate a vibration reduction strategy; The vibration reduction strategy was verified based on the safety threshold. After verification, the thermal power generation pipeline was tuned for vibration reduction.

2. The method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines according to claim 1, characterized in that, The acquisition of vibration, stress, and temperature data for thermal power generation pipelines includes: Acquiring multi-source heterogeneous data of thermal power pipelines includes collecting full-domain vibration, strain, and temperature data of thermal power pipelines through distributed optical fiber sensors, acquiring three-dimensional vibration vector data of the surface of thermal power pipelines through laser Doppler vibration meters, and collecting stress and strain data of thermal power pipeline elbows, valves, support points, and the distance to both sides of the support points through fiber Bragg grating stress sensors. The process involves spatiotemporal alignment of the collected multi-source heterogeneous data and noise removal through a filtering algorithm to complete data preprocessing, thereby obtaining vibration, stress, and temperature data of thermal power generation pipelines.

3. The method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines according to claim 1, characterized in that, The method utilizes a combination of spatiotemporal graph neural network and digital twin model to process vibration, stress, and temperature data of thermal power generation pipelines, obtaining vibration modal parameters of the thermal power generation pipelines, including: Vibration, stress, and temperature data of thermal power generation pipelines are input into a spatiotemporal neural network to obtain calculation results; Vibration, stress, and temperature data of thermal power generation pipelines are input into a digital twin model to obtain predicted modal parameters; The calculated results and predicted modal parameters are weighted and fused to obtain the real-time identification results of the vibration modes of thermal power generation pipelines.

4. The method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines according to claim 1, characterized in that, The spatiotemporal graph neural network contains 3 layers of graph convolutional layers and 2 layers of gated recurrent units. It constructs graph structure data using thermal power generation pipeline structural nodes and sensor data as graph nodes and physical connection relationships between nodes as edges. The digital twin model was built using ANSYS finite element analysis software, integrating the elastic modulus, Poisson's ratio, and density parameters of the thermal power generation pipeline material, as well as the fixed support boundary conditions and the thermo-structural coupled physical field.

5. The method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines according to claim 1, characterized in that, The real-time identification results based on the vibration modes of thermal power generation pipelines generate control signals for vibration reduction devices, including: By utilizing lightweight neural networks and model predictive control algorithms, the real-time identification results of vibration modes of thermal power generation pipelines are calculated in real time to generate control signals for vibration reduction devices. The output of the lightweight neural network is calculated as follows: in, The output of the lightweight neural network is the control signal for the piezoelectric ceramic sheet. For the first Vibration signal characteristics of each input node, The number of input nodes. To connect the first The output node and the first The weights of each input node, For the first The bias of each output node. This is the activation function.

6. The method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines according to claim 1, characterized in that, The process of performing multi-device coordinated vibration reduction simulation on the control signal of the vibration reduction device to generate a vibration reduction strategy includes: Based on the control signal of the vibration reduction device, the magnetorheological fluid damper, shape memory alloy cable and piezoelectric ceramic sheet in the thermal power generation pipeline are adjusted to work together to reduce vibration, and the simulation results of multi-device coordinated vibration reduction are obtained. Vibration reduction strategies are generated based on the simulation results of multi-device collaborative vibration reduction.

7. The method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines according to claim 1, characterized in that, The process of verifying the vibration reduction strategy based on a safety threshold, followed by tuning and vibration reduction of the thermal power generation pipeline after verification, includes: A stress-deformation-safety margin model for pipelines is established using a physical constraint neural network, and the dynamic safety threshold for thermal power generation pipelines is obtained based on the safety margin model. in, For dynamic security thresholds, This represents the maximum allowable stress of the pipe material. For the maximum allowable strain, The current temperature. Due to current pressure, The safety threshold is determined by comprehensively considering material properties and working parameters, using the mapping function obtained through training of a physical constraint neural network. Extracting real-time stress from vibration reduction strategies If real-time stress Greater than or equal to the dynamic security threshold The current regulation sequence of the magnetorheological fluid damper and the heating power regulation sequence of the shape memory alloy cable of the thermal power generation tube were recalculated and adjusted. If real-time stress Less than the dynamic security threshold If so, no adjustments will be made to the thermal power generation pipelines.

8. A real-time vibration mode identification and adaptive tuning vibration reduction system for thermal power generation pipelines, characterized in that, include: Data acquisition module: used to acquire vibration, stress and temperature data of thermal power generation pipelines; Recognition Result Module: This module uses a combination of spatiotemporal graph neural network and digital twin model to process vibration, stress and temperature data of thermal power generation pipelines, and obtains real-time recognition results of vibration modes of thermal power generation pipelines. Signal generation module: used to generate control signals for vibration reduction devices based on the real-time identification results of vibration modes of thermal power generation pipelines; Vibration reduction strategy module: used to simulate multi-device coordinated vibration reduction based on the control signals of vibration reduction devices, and generate vibration reduction strategies; Tuned vibration reduction module: Used to verify the vibration reduction strategy according to the safety threshold. After verification, the thermal power generation pipeline is tuned for vibration reduction.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method for real-time identification and adaptive tuning vibration reduction of vibration modes of thermal power generation pipelines as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method for real-time identification and adaptive tuning vibration reduction of vibration modes in thermal power generation pipelines as described in any one of claims 1-7.

Images