CN-121978929-A - LNG flexible pipeline vibration control method based on multi-physical-field reduced-order model prediction

CN121978929ACN 121978929 ACN121978929 ACN 121978929ACN-121978929-A

Abstract

The invention provides an LNG flexible pipeline vibration control method based on multi-physical-field reduced-order model prediction, which comprises the following steps of collecting temperature distribution data and dynamic strain data along an LNG flexible pipeline in real time, collecting vibration acceleration signals of key nodes of the pipeline by utilizing an acceleration sensor, extracting flow field pressure modes and structural vibration modes under the coupling effect of water hammer and thermal shock by utilizing an intrinsic orthogonal decomposition method, constructing a multi-physical-field reduced-order prediction model capable of being calculated on line in real time, dynamically adjusting allowable vibration amplitude threshold values of all pipeline sections, inputting real-time data into the multi-physical-field reduced-order prediction model, predicting pipeline vibration response and thermal stress evolution tracks in a future limited time domain, solving optimal control input vectors by taking maintenance pipeline response within the allowable vibration amplitude threshold values as constraint, and decomposing the optimal control input vectors into fluid side instructions and structural side instructions. The invention improves the safety of the LNG flexible pipeline under extreme working conditions.

Inventors

WEN MINJIE
ZHANG YIMING
WU JUNTAO
DING PAN
GE XIAONAN
SHEN LULU
GAO ZHIRAN
LI JING
WANG LIXING

Assignees

浙江大学
浙江理工大学

Dates

Publication Date: 20260505
Application Date: 20260121

Claims (9)

1. The LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction is characterized by being implemented in a control system provided with a distributed optical fiber sensing network, an electric control variable damping hydraulic support device and an electrohydraulic servo valve, and comprising the following steps of: s1, acquiring temperature distribution data and dynamic strain data along an LNG flexible pipeline in real time by using a distributed optical fiber sensing network, and acquiring vibration acceleration signals of key nodes of the pipeline by using an acceleration sensor; S2, extracting a flow field pressure mode and a structural vibration mode under the coupling action of water hammer and thermal shock by utilizing an intrinsic orthogonal decomposition method based on pre-acquired high-fidelity finite element simulation data, and constructing a multi-physical-field reduced-order prediction model capable of being calculated on line in real time; S3, dynamically adjusting the allowable vibration amplitude threshold value of each pipe section of the pipeline according to the real-time temperature distribution data acquired in the step S1, and reducing the allowable vibration amplitude threshold value of the pipe section when the fracture toughness of the material is reduced due to the reduction of the temperature of the pipeline; S4, inputting the real-time data of the step S1 into the multi-physical-field reduced-order prediction model of the step S2, predicting pipeline vibration response and thermal stress evolution tracks in a future limited time domain, and solving an optimal control input vector by taking the range of the allowable vibration amplitude threshold for maintaining the pipeline response in the step S3 as a constraint; s5, decomposing the optimal control input vector into a fluid side instruction and a structure side instruction, wherein the fluid side instruction is used for controlling the opening degree of the electrohydraulic servo valve, and the structure side instruction is used for adjusting the opening degree of a throttle valve of the electric control variable damping hydraulic support device.
2. The LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction of claim 1 is characterized in that in the step S2, the construction of the multi-physical-field reduced-order prediction model specifically comprises the following steps: In an off-line stage, calculating pipeline pressure fields and displacement fields under different switching valve speeds and temperature difference working conditions through full physical field simulation to form a snapshot matrix; singular value decomposition is carried out on the snapshot matrix, and a first k-order main mode with the energy ratio exceeding a preset value is selected as a basis function; The Navier-Stokes equation and the structural dynamics equation which control the fluid dynamics are projected onto a low-dimensional subspace which is spanned by the basis functions by using a Galerkin projection method, and the partial differential equation set is converted into a normal differential equation set.
3. The LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction of claim 1 is characterized in that in the step S3, the dynamically adjusting of the allowable vibration amplitude threshold value of each pipe section of the pipeline specifically comprises the following steps: pre-establishing a function relation between fracture toughness and temperature of a pipeline material; defining a positive correlation between critical vibration amplitude and fracture toughness; and in the control period, reading the real-time temperature of the monitoring point, and calculating the current dynamic vibration threshold according to the functional relation and the positive correlation relation, wherein the dynamic vibration threshold is reduced along with the temperature reduction.
4. The LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction is characterized in that in the step S4, a rolling time domain optimization strategy is adopted for solving an optimal control input vector, an established objective function comprises a deviation term between a prediction output and a reference track and a weighting term of a control increment, the deviation term is used for representing a difference value between pipeline vibration and stress response predicted by the multi-physical-field reduced-order model prediction and expected steady-state response, and the control increment comprises an electrohydraulic servo valve opening change rate and an electric control variable damping hydraulic support device throttle valve opening change rate.
5. The LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction is characterized in that the electric control variable damping hydraulic supporting device comprises a hydraulic cylinder connected between an LNG flexible pipeline and a fixed base, a bypass oil way communicated with an upper cavity and a lower cavity of the hydraulic cylinder and a proportional electromagnetic throttle valve arranged on the bypass oil way, and in the step S5, the throttle valve opening of the electric control variable damping hydraulic supporting device is adjusted specifically comprises the following steps: And the control signal input into the proportional electromagnetic throttling valve is changed to adjust the flow resistance of hydraulic oil when the hydraulic oil flows through the bypass oil way, so that the support damping of the pipeline system is changed.
6. The method for controlling vibration of the LNG flexible pipeline based on the multi-physical-field reduced-order model prediction as set forth in claim 1, wherein in the step S5, decomposing the optimal control input vector into a fluid-side instruction and a structure-side instruction specifically includes: Generating a fluid side command when low-frequency large shaking caused by a water hammer pressure wave is predicted; And generating a structure side instruction when the fluid-solid coupling frequency is predicted to be close to the natural frequency of the pipeline to cause resonance trend.
7. The LNG flexible pipeline vibration control method based on the multi-physical field reduced order model prediction as set forth in claim 6, wherein the fluid side instruction specifically comprises: And controlling the electrohydraulic servo valve to execute multistage micro-suspension action, and superposing high-frequency micro reverse action in the closing or opening process of the valve.
8. The LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction of claim 1 is characterized in that in the step S1, the arrangement mode of the distributed optical fiber sensing network and the acceleration sensor specifically comprises the following steps: laying the distributed optical fiber sensors along the axial direction of the LNG flexible pipeline, and acquiring the temperature gradient and average strain along the path; The acceleration sensor is arranged at the elbow, the flange connection and the middle point of the suspended section of the pipeline.
9. The LNG flexible pipeline vibration control method based on the multi-physical field reduced order model prediction as claimed in claim 1, wherein the method further comprises the following steps: Calculating residual errors between predicted values of the multi-physical-field reduced-order prediction model and the actual measurement values acquired in the step S1; And when the residual error exceeds a preset error limit value, updating the basis function coefficient of the multi-physical-field reduced-order prediction model by using current measured data.

Description

LNG flexible pipeline vibration control method based on multi-physical-field reduced-order model prediction Technical Field The invention relates to an LNG flexible pipeline vibration control method, in particular to an LNG flexible pipeline vibration control method based on multi-physical-field reduced-order model prediction, and belongs to the technical field of ocean engineering and fluid pipeline vibration control. Background Liquefied Natural Gas (LNG) is widely applied to marine transportation and exploitation, the running environment of an LNG flexible pipeline is extremely severe, and the LNG flexible pipeline bears alternating load caused by deep sea storm current for a long time. During LNG transfer, the opening and closing operations of the valve are unavoidable. However, rapid on-off valve operation can cause severe fluctuations in fluid pressure within the pipe, resulting in substantial vibration of the pipe. At the same time, the introduction or cutting off of the low-temperature fluid can generate strong impact on the pipeline, so that the temperature of the pipeline wall changes sharply. The coupling effect of the water hammer and the thermal shock is extremely destructive, on one hand, the vibration excited by the water hammer can cause fatigue failure of the pipeline, and on the other hand, the pipeline material becomes brittle at low temperature, the fracture toughness is reduced, and the pipeline is more easily subjected to brittle fracture under the same vibration amplitude. In the prior art, there is little active control on the vibration of the pipeline, and most of the existing control methods are based on feedback control, namely, the control is performed after the vibration is detected. However, the water hammer wave propagation speed is extremely fast, and when the sensor detects a vibration peak, damage may have occurred. The existing vibration control threshold is typically a fixed constant. However, in LNG pipelines, as the temperature decreases, the cracking resistance of the material decreases significantly. If the vibration safety threshold at normal temperature is still used under the extremely low temperature working condition, the low temperature brittle failure accident is extremely easy to be caused. The prior art either only relieves the water hammer by slowing down the valve closing speed or only dissipates energy by a passive damper, lacks a cooperative control mechanism combining fluid side control and structural side control, and is difficult to cope with the violent coupling vibration of the LNG flexible pipeline under severe working conditions. Disclosure of Invention Based on the background, the invention aims to provide an LNG flexible pipeline vibration control method based on multi-physical-field reduced-order model prediction, which realizes millisecond-level response prediction by constructing an eigenvoice-based reduced-order prediction model and solves the technical problems that LNG flexible pipeline vibration control is lagged and low-temperature brittleness risks are easy to ignore in the prior art by combining dynamic threshold setting considering temperature-toughness coupling and a fluid-structure double-side cooperative execution mechanism. In order to achieve the above object, the present invention provides the following technical solutions: the LNG flexible pipeline vibration control method based on the multi-physical-field reduced-order model prediction is implemented in a control system with a distributed optical fiber sensing network, an electric control variable damping hydraulic support device and an electrohydraulic servo valve, and comprises the following steps: s1, acquiring temperature distribution data and dynamic strain data along an LNG flexible pipeline in real time by using a distributed optical fiber sensing network, and acquiring vibration acceleration signals of key nodes of the pipeline by using an acceleration sensor; S2, extracting a flow field pressure mode and a structural vibration mode under the coupling action of water hammer and thermal shock by utilizing an intrinsic orthogonal decomposition method based on pre-acquired high-fidelity finite element simulation data, and constructing a multi-physical-field reduced-order prediction model capable of being calculated on line in real time; S3, dynamically adjusting the allowable vibration amplitude threshold value of each pipe section of the pipeline according to the real-time temperature distribution data acquired in the step S1, and reducing the allowable vibration amplitude threshold value of the pipe section when the fracture toughness of the material is reduced due to the reduction of the temperature of the pipeline; S4, inputting the real-time data of the step S1 into the multi-physical-field reduced-order prediction model of the step S2, predicting pipeline vibration response and thermal stress evolution tracks in a future limited time domain, and solving an optimal co