Gas-liquid low-temperature two-phase flow accurate monitoring and metering test system

Abstract

The invention discloses a gas-liquid low-temperature two-phase flow precise monitoring and metering test system, which comprises a segmented sensing acquisition module, a high-speed signal analysis module, a pulsation resonance identification module, an active damping control module, a phase change intelligent identification module and a monitoring metering output module, a plurality of groups of separated fluid dynamic response sensors are arranged in a low-temperature two-phase flow pipeline and are respectively positioned at a two-phase mixing inlet section, a middle stable section and an outlet section, so that segmented data acquisition is realized. The method has the advantages that the resonance risk caused by gas-liquid phase change can be accurately identified, a damping structure is regulated and controlled in real time, and system resonance and sensor damage are effectively avoided; meanwhile, through multi-point flow velocity fusion and gas phase molar fraction dynamic correction, high-precision measurement of the two-phase fluid volume flow is achieved, the measurement stability and the system safety are improved, and the device is widely suitable for low-temperature gas-liquid mixed transportation scenes.

Description

Technical Field

[0001] The present invention relates to the field of measurement technology, and in particular to a gas-liquid low-temperature two-phase flow precision monitoring and measurement testing system. Background Art

[0002] "Precise monitoring and metering test of gas-liquid low-temperature two-phase flow" refers to the technical process of accurately monitoring and measuring the flow of fluids that contain both gas and liquid states in real time under low-temperature environments. This two-phase fluid is commonly found in the transportation or use of cryogenic media such as liquefied natural gas (LNG), liquid nitrogen, and liquid hydrogen. During the transmission process, due to temperature or pressure changes, liquid and gas phases will coexist. Traditional single-phase flow measurement methods have difficulty obtaining accurate data under such complex flow states. Therefore, this technology uses integrated sensors, signal processing, electronic control analysis and other means to collect parameters such as fluid pressure, temperature, phase distribution and flow rate in real time. Using specialized algorithms or structural designs, it can achieve accurate identification of the state of gas-liquid mixed fluids and high-precision flow measurement, providing reliable support for cryogenic energy management, storage and transportation system safety, and energy efficiency optimization.

[0003] The existing technology has the following shortcomings: In the existing precise monitoring and metering tests of low-temperature gas-liquid two-phase flows, the fluid pulsation caused by the gas-liquid phase change may cause resonant interference to the measurement system. When the fluid undergoes a drastic phase change due to temperature and pressure fluctuations, it is easy to form periodic flow pulsations, especially in the pipeline structure, which will produce a similar "liquid hammer" effect. This pulsation not only causes drastic fluctuations in instantaneous flow rate and gas-liquid ratio, interfering with measurement stability, but also may stimulate mechanical resonance of the system's internal measurement structure, reducing data accuracy. In severe cases, it may even damage sensors or key metering devices, causing misjudgment or abnormal system shutdown, affecting the safety and reliability of the overall operation. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a gas-liquid low-temperature two-phase flow accurate monitoring and metering test system.

[0005] To solve the above technical problems, the technical solution of the present invention is: a gas-liquid low-temperature two-phase flow precision monitoring and measurement test system, including a segmented sensing acquisition module, a high-speed signal analysis module, a pulsation resonance identification module, an active damping control module, a phase change intelligent identification module, and a monitoring and measurement output module: The segmented sensing acquisition module deploys multiple sets of separate fluid dynamic response sensors in the low-temperature two-phase flow pipeline, located in the two-phase mixing inlet section, the intermediate stable section, and the outlet section, to achieve segmented data acquisition; The high-speed signal analysis module synchronously inputs the collected instantaneous flow velocity, pressure and temperature signals into the high-speed signal processing module and uses the spectrum analysis method to extract the fluid pulsation characteristic parameters; The pulsation resonance identification module establishes a fluid pulsation resonance model and uses the model calculation to determine whether a fluid excitation frequency close to the natural frequency of the LT-TFM system occurs; The active damping control module, upon detecting a potential resonance trend, activates the adaptive damping module to adjust the controllable damping elements within the flow channel structure to actively weaken the flow disturbance in the resonant frequency band; The phase change intelligent identification module uses a phase change feature recognition algorithm to dynamically correct the two-phase ratio estimate while reducing pulsation interference, thereby improving the accuracy of gas-liquid interface identification. The monitoring and metering output module ultimately outputs the corrected flow rate, phase holdup, and pulsation suppression status to the main control system, enabling accurate monitoring and safe metering of low-temperature two-phase flow.

[0006] Preferably, the multiple sets of separate fluid dynamic response sensors include a miniature piezoelectric pressure sensor, a thermal flow rate measurement unit and a miniature thermocouple temperature acquisition module, and each acquisition unit is connected to the data processing center via a high-speed signal bus; At the inlet section of the pipeline, sensors are arranged at a spacing of 20mm to form a high-density sampling array, which is used to capture local fluctuations of the fluid in the initial mixing state; in the intermediate stable section, the sensor spacing is increased to 50mm to analyze the changes in macroscopic parameters during the stable transmission of the mixture; in the outlet section, sensors are distributed at an angle in the annular monitoring cavity to collect the final state of the fluid before leaving the pipe.

[0007] Preferably, the signal processing module includes an embedded FPGA data processing chip, a redundant filter bank, and a fast Fourier transform operation unit; All sensors trigger sampling synchronously, and the data sampling frequency is no less than 100kHz to ensure the capture of microsecond-level pulsation changes; The FFT operation module performs spectrum conversion on the data window collected every second, extracts the first three main frequency components and their amplitudes, and analyzes the dominant frequency range of the fluid disturbance; the redundant filter adopts a bandpass filter structure and adjusts the center frequency according to the main frequency before processing, so that the processed signal is targeted and avoids background noise interference in analysis and judgment.

[0008] Preferably, the fluid pulsation resonance model is based on the fusion modeling of structural dynamic response analysis and real-time fluid disturbance characteristics; A dynamic input parameter set is established through the instantaneous flow rate change rate, pressure fluctuation amplitude and fluid component change trend collected by the sensor; The fluid pulsation resonance model considers the physical characteristics of the pipeline structure and uses numerical analysis to generate inherent response characteristics; The degree of matching between the disturbance frequency and the structural characteristic frequency is used to comprehensively evaluate whether there is any excitation of structural resonance.

[0009] Preferably, the adaptive damping module adopts a liquid adjustable damping element, and the liquid adjustable damping element is controlled by a piezoelectric driver to achieve dynamic adjustment of the damping characteristics by responding to frequency changes of its internal throttling structure; The module uses a high-speed signal as the driving signal and automatically adjusts the damping range to match the corresponding disturbance band in combination with the main disturbance frequency provided by the resonance model. When the disturbance frequency range changes, the piezoelectric control signal is updated through closed-loop control to achieve real-time response and dynamic compensation; at the same time, the module has a fast response capability, and the adjustment delay does not exceed 20ms, ensuring that the fluid disturbance is suppressed in the early stage and avoids entering a resonance state.

[0010] Preferably, the phase change feature recognition algorithm uses an improved fuzzy support vector machine to fuse the temperature gradient change rate and the pressure hysteresis response characteristics to identify and classify the gas-liquid boundary; A multi-dimensional recognition feature space is constructed by training sample data, and a classification model is established with the first-order derivative of temperature, the second-order derivative of pressure, and the flow velocity pulsation amplitude as input factors; When the actual operation data enters the classification process, it is mapped to the constructed feature space, and the predicted value of the gas-liquid phase ratio is output through SVM.

[0011] Preferably, the resonance risk judgment adopts a dynamic evaluation mechanism based on a frequency response model, and the specific steps are as follows: During the signal acquisition process, short-time Fourier transform is applied to the inlet pressure signal to extract the instantaneous dominant frequency from the time-frequency domain. The extraction formula is as follows: , where is the instantaneous main frequency, indicating the previous moment The frequency component with the strongest energy in the pressure fluctuation in its adjacent time period, It's in time The pressure signal collected at all times, is the half-width of the window for short-time Fourier analysis, is the Fourier basis function, which is used to transform the time domain signal into the frequency domain. is the natural base, is an imaginary unit, is the angular frequency multiplied by time, which means the signal is at frequency When , the corresponding phase change is, is the frequency, is the time variable, is the differential symbol; The instantaneous pressure change rate is obtained by taking the derivative of the pressure curve in the simultaneous time window. The calculation expression is as follows: , where is the instantaneous rate of pressure change, It's pressure Find the first derivative, which represents how quickly pressure changes with time, is the reference time point for calculating the pressure derivative, i.e. the current sampling moment; A frequency response function model for a single degree of freedom system is established to calculate the response amplitude in the current state. The calculation expression is as follows: , where is the natural frequency, is the damping ratio, is the predicted value of the response amplitude; Set the empirical response amplitude threshold , if at any moment , it is determined that there is a risk of resonant excitation, the dynamic damping module is triggered immediately, and the current event mark is archived for subsequent operation strategy adjustment.

[0012] Preferably, the main control system includes a human-computer interaction interface, a redundant data acquisition card set and a dual-redundant diagnostic module for safety protection; All measurement data and flow state parameters are archived uniformly and displayed in real time on a visual interface in a graphical manner; With historical record backtracking function, users can select any time period to query pulsation characteristics, gas-liquid ratio and flow rate change trend; The main control system automatically triggers an emergency stop command after detecting that a high-risk frequency exists continuously for more than 3 seconds, disconnecting the main valve controller to prevent damage.

[0013] Preferably, the gas-liquid phase ratio estimation and volume flow correction are calculated by multi-parameter coupled joint modeling, and the specific steps are as follows: Set up three sections of the pipeline to collect flow velocity data , the overall average flow velocity is solved using the variable weight weighted average method, and the calculation expression is as follows: , where is the average phase velocity, Represent the local instantaneous flow velocities collected at the three sections of the pipeline, respectively The weighting coefficient of The current gas phase mole fraction is estimated by combining the compressibility factor and density ratio using the ideal gas state correction model. The calculation expression is as follows: , where is the gas phase mole fraction, is the absolute pressure at the measuring point, is the absolute temperature of the measurement point, is the universal gas constant, and are the gas and liquid phase compressibility factors, and are the densities of the gas phase and liquid phase under the current temperature and pressure conditions; An empirical correction factor is introduced to calculate the corrected volume flow rate of gas-liquid coupling. The calculation expression is as follows: , where is the corrected effective volume flow rate, is the cross-sectional area of ​​the pipe, is the gas phase interference correction factor; Continuous sliding calculation If the standard deviation within the five-second time window is less than the threshold , the current data will be marked as a valid measurement result and output to the main control interface; if it exceeds, the refresh will be delayed and a prompt will be given that re-judgment is required.

[0014] Beneficial effects of the present invention: By constructing a resonance determination model that combines a frequency response function with the dynamic characteristics of the fluid, this invention achieves rapid identification and precise response control of periodic pulsations caused by gas-liquid phase transitions in low-temperature two-phase flows. Compared to traditional methods that rely solely on pressure thresholds or static models, this solution can capture and quantify pulsation excitation energy in real time and dynamically adjust damping elements to actively eliminate frequency bands that may cause resonance. This significantly improves the system's operational stability under complex operating conditions, reduces sensor structural fatigue and false alarm rates, and enhances the structural safety and service life of the monitoring and metering system.

[0015] The present invention also combines real-time data acquisition with a multi-point velocity fusion algorithm to construct a gas-liquid phase ratio identification and volume flow correction model with dynamic self-adaptation capabilities. Through weighted fusion of multi-point velocity data, estimation of the gas phase mole fraction, and dynamic correction formulas, the system makes the overall flow assessment of the two-phase flow closer to the actual transportation state, avoiding the significant errors caused by single-point measurement under uneven flow or disturbance conditions. Therefore, in low-temperature gas-liquid mixed transmission pipelines, LNG equipment, and aerospace propulsion systems, this solution can significantly improve fluid metering accuracy, providing key technical support for improving energy utilization efficiency and safe system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a module diagram of a gas-liquid low-temperature two-phase flow precision monitoring and measurement test system. DETAILED DESCRIPTION

[0017] The following is a further description of specific embodiments of the present invention in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is intended to facilitate understanding of the present invention and does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0018] Working principle of the present invention: A gas-liquid low-temperature two-phase flow precision monitoring and measurement test system, including a segmented sensing acquisition module, a high-speed signal analysis module, a pulsation resonance identification module, an active damping control module, a phase change intelligent identification module, and a monitoring and measurement output module: The segmented sensing acquisition module deploys multiple sets of separate fluid dynamic response sensors in the low-temperature two-phase flow pipeline, located in the two-phase mixing inlet section, the intermediate stable section, and the outlet section, to achieve segmented data acquisition; The multiple sets of separate fluid dynamic response sensors include miniature piezoelectric pressure sensors, thermal flow rate measurement units, and miniature thermocouple temperature acquisition modules. Each acquisition unit is connected to the data processing center via a high-speed signal bus. At the inlet section of the pipeline, sensors are arranged at a spacing of 20mm to form a high-density sampling array, which is used to capture local fluctuations of the fluid in the initial mixing state; in the intermediate stable section, the sensor spacing is increased to 50mm to analyze the changes in macroscopic parameters during the stable transmission of the mixture; in the outlet section, sensors are distributed at an angle in the annular monitoring cavity to collect the final state of the fluid before leaving the pipe.

[0019] This distribution design ensures that the system can capture dynamic physical characteristics with high spatial resolution in real time from the entire process of fluid entry, evolution to output, providing a sufficient basis for subsequent resonance judgment and dynamic correction.

[0020] This step ensures the system can achieve high-precision, continuous dynamic monitoring of the entire two-phase flow path, avoiding omissions in the spatial distribution of critical flow states. By subdividing the functions of different pipe sections and configuring multiple sensor groups with corresponding densities, the system can effectively detect the fluid evolution trajectory throughout the gas-liquid mixing, transmission, and discharge processes, providing a detailed data foundation for subsequent signal analysis and dynamic control.

[0021] This step achieves monitoring coverage by deploying sensor networks of different types and densities at three key locations: the inlet, middle section, and outlet. A high-density array of sensors is used at the inlet to capture the initial disturbance behavior of the fluid entering the system; sparse sensors are deployed in the middle section to track the homogeneous changes in the flow process; and a ring-shaped sensor distribution is used at the outlet to grasp the final performance of the fluid state. All sensors use a unified time base to ensure that data collection is completed with minimal time differences, and sampling synchronization control is achieved through a central processing node. This system layout constitutes a full-process, segmented, and high-frequency physical quantity perception system.

[0022] The high-speed signal analysis module synchronously inputs the collected instantaneous flow velocity, pressure and temperature signals into the high-speed signal processing module and uses the spectrum analysis method to extract the fluid pulsation characteristic parameters; The signal processing module includes an embedded FPGA data processing chip, a redundant filter bank, and a fast Fourier transform (FFT) operation unit; All sensors trigger sampling synchronously, and the data sampling frequency is no less than 100kHz to ensure the capture of microsecond-level pulsation changes; The FFT operation module performs spectrum conversion on the data window collected every second, extracts the first three main frequency components and their amplitudes, and analyzes the dominant frequency range of the fluid disturbance; the redundant filter adopts a bandpass filter structure and adjusts the center frequency according to the main frequency before processing, so that the processed signal is targeted and avoids background noise interference in analysis and judgment.

[0023] In addition, the system warns of possible resonance states based on the main frequency drift trend, providing basic spectrum data for the next step of model judgment.

[0024] This step transforms the collected fluid physical signals into data representations with time and frequency domain characteristics, thereby identifying potential periodic disturbance signatures. Spectral analysis not only determines flow stability but also identifies the dominant disturbance frequencies that influence system stability, providing the prerequisites for predicting resonance and implementing control interventions.

[0025] By sampling data synchronously at high frequencies, the system can respond to rapid disturbances occurring within a short period of time. The signals are then fed into an analysis module, where a conversion method is used to identify the primary frequency segments of periodic fluctuations in the fluid state. The data is further filtered and classified to ensure that dominant frequencies of engineering significance are identified, rather than random noise signals. The system also utilizes adjustable signal filtering to focus analysis on the frequency range of interest, effectively suppressing interference from non-target noise on the frequency identification results.

[0026] The pulsation resonance identification module establishes a fluid pulsation resonance model and uses model calculations to determine whether a fluid excitation frequency close to the natural frequency of the LT-TFM system (gas-liquid low-temperature two-phase flow precision monitoring and measurement test system) occurs. The fluid pulsation resonance model is based on the fusion modeling of structural dynamic response analysis and real-time fluid disturbance characteristics; A dynamic input parameter set is established through the instantaneous flow rate change rate, pressure fluctuation amplitude and fluid component change trend collected by the sensor; The model considers the physical characteristics of the pipeline structure, such as its geometry, material parameters, and boundary conditions, and uses numerical analysis to generate the inherent response characteristics of the system; The degree of matching between the disturbance frequency and the structural characteristic frequency is used to comprehensively evaluate whether there is any excitation of structural resonance.

[0027] This model not only determines the presence of resonance risk in real time but also possesses online learning capabilities, automatically correcting model parameters based on historical operating data to improve adaptability under various flow conditions. When the system detects an increasing trend in the matching between disturbance signatures and structural responses, it generates a risk score and identifies high-risk periods. This is then used to trigger vibration suppression control mechanisms and alarms, significantly improving the stability and structural safety of the two-phase flow measurement system.

[0028] This step models the interaction between fluid disturbance characteristics and the system's structural response to determine whether the system is currently in a dangerous range that could trigger mechanical resonance. This model establishes an early warning mechanism, enabling the system to detect potential risks at an early stage, without relying on feedback from actual structural damage to trigger a response.

[0029] The implementation process involves comprehensive modeling of the mechanical properties of the test pipeline structure and, in combination with the fluid disturbance frequencies detected by the sensors, establishing a judgment model. This model compares the current fluid state with the structural response characteristics, identifying any frequency overlap or similarity, and thus predicting the possibility of mechanical resonance. The model can also be revised and updated based on data trends and historical operation records to adapt to model drift caused by changes in fluid state, ensuring the real-time and accurate judgment results.

[0030] The active damping control module, upon detecting a potential resonance trend, activates the adaptive damping module to adjust the controllable damping elements within the flow channel structure to actively weaken the flow disturbance in the resonant frequency band; The adaptive damping module uses a liquid adjustable damping element. The piezoelectric driver controls the internal throttling structure of the liquid adjustable damping element to respond to frequency changes and achieve dynamic adjustment of the damping characteristics. The module uses a high-speed signal as the driving signal and automatically adjusts the damping range to match the corresponding disturbance band in combination with the main disturbance frequency provided by the resonance model. When the disturbance frequency range changes, the piezoelectric control signal is updated through closed-loop control to achieve real-time response and dynamic compensation; at the same time, the module has a fast response capability, and the adjustment delay does not exceed 20ms, ensuring that the fluid disturbance is suppressed in the early stage and avoids entering a resonance state.

[0031] This damping control strategy has higher sensitivity and stability than traditional passive buffer devices.

[0032] This step provides an active response mechanism. When the system identifies a resonance risk, it instantly adjusts the damping device's operating state, mitigating or blocking the energy propagation path of the disturbance at its source and preventing the system from entering an unstable operating state. This damping mechanism overcomes the limitations of traditional passive design and enables intelligent energy control.

[0033] The control module dynamically adjusts the damping device's internal structure, enabling it to respond appropriately to disturbances of varying frequencies. Based on the dominant frequency of the disturbance detected by the system, the device rapidly alters the resistance or energy absorption characteristics of the fluid flow path, thereby absorbing or converting the disturbance energy and rapidly reducing the disturbance amplitude. This entire process is automated, offering fast response times and high accuracy, making it particularly suitable for frequency-sensitive, low-temperature two-phase flow systems.

[0034] The phase change intelligent identification module uses a phase change feature recognition algorithm to dynamically correct the two-phase ratio estimate while reducing pulsation interference, thereby improving the accuracy of gas-liquid interface identification. The phase change feature recognition algorithm uses an improved fuzzy support vector machine (F-SVM) to fuse the temperature gradient change rate and pressure hysteresis response characteristics to identify and classify the gas-liquid boundary; A multi-dimensional recognition feature space is constructed by training sample data, and a classification model is established with the first-order derivative of temperature, the second-order derivative of pressure, and the flow velocity pulsation amplitude as input factors; When the actual operation data enters the classification process, it is mapped to the constructed feature space, and the predicted value of the gas-liquid phase ratio is output through SVM.

[0035] This algorithm can adapt to complex phase change processes under different working conditions, effectively reduce the misjudgment rate, and ensure the accuracy of flow separation calculation and subsequent measurement.

[0036] This step accurately identifies the presence and changing trends of the gas-liquid interface, providing a foundation for subsequent flow measurement and phase holdup determination. During two-phase flow, the gas-liquid interface frequently shifts with changes in the environment and flow regime, making traditional identification methods prone to misjudgment. Intelligent identification methods can effectively address this challenge.

[0037] The system collects multiple, real-time fluid characteristic signals and analyzes them jointly using intelligent algorithms. During the training phase, the algorithm establishes recognition patterns for different gas-liquid states. Once actual measurements enter the recognition process, the system compares them with known patterns to determine the current gas-liquid ratio. When transitional states or blurred boundaries exist, the system assigns appropriate weights to avoid extreme discrimination results, improving recognition stability and anti-interference capabilities. This recognition capability persists throughout the system's operation, ensuring that subsequent parameter calculations are based on sufficient and accurate data.

[0038] Resonance risk judgment adopts a dynamic assessment mechanism based on the frequency response model. The specific steps are as follows: During the signal acquisition process, short-time Fourier transform (STFT) is applied to the inlet pressure signal to extract the instantaneous dominant frequency from the time-frequency domain. The extraction formula is as follows: , where is the instantaneous main frequency (unit: Hz), indicating the The frequency component with the strongest energy in the pressure fluctuation in the adjacent time period is the key indicator for judging the resonance trend. It's in time The pressure signal (unit: Pa) collected at every moment represents the instantaneous pressure of the two-phase flow in the pipeline. is the half-width of the short-time Fourier analysis window (unit: s), which determines the time span of the analysis interval, usually in milliseconds. is the Fourier basis function, which is used to transform the time domain signal into the frequency domain. is the natural base, is an imaginary unit, satisfying , is the angular frequency multiplied by time, which means the signal is at frequency When , the corresponding phase change is, is the frequency, is the time variable, It is a differential symbol, used in integral expressions to represent "infinitesimal accumulation with time as the variable";

[0039] This formula identifies the main disturbance frequency in the current system in the form of a sliding time window, ensuring that the effective main component of the spectrum can be extracted even in unsteady flow conditions.

[0040] The instantaneous pressure change rate is obtained by taking the derivative of the pressure curve in the simultaneous time window. The calculation expression is as follows: , where It is the instantaneous pressure change rate (unit: Pa / s), which is used to measure the intensity of pressure over time and is a measurement parameter of the excitation source strength. It's pressure Find the first derivative, which represents how quickly pressure changes with time, is the reference time point for calculating the pressure derivative, i.e. the current sampling moment; This rate of change reflects whether there is a drastic energy input in the fluid system and serves as the core excitation parameter in the subsequent resonant excitation calculation.

[0041] A frequency response function (FRF) model for a single degree of freedom system is established to calculate the response amplitude under the current state. The calculation expression is as follows: , where is the natural frequency (unit: Hz), which is determined by the structural characteristics (such as length, stiffness, boundary conditions) and is the critical frequency at which resonance occurs. is the damping ratio (unitless), which measures the system's ability to dissipate energy. The larger the value, the less likely resonance will occur. Typical values ​​are 0.05–0.2. is the predicted value of the response amplitude (unit: arbitrary amplitude unit, can be normalized), indicating the instantaneous main frequency Under this condition, will the system have an amplification effect? The larger the response amplitude, the closer the system is to the resonance point.

[0042] Set the empirical response amplitude threshold , if at any moment , it is determined that there is a risk of resonant excitation, the dynamic damping module is triggered immediately, and the current event mark is archived for subsequent operation strategy adjustment.

[0043] This method combines signal spectrum analysis with structural dynamics modeling to achieve real-time prediction and suppression of implicit resonance problems under complex flow backgrounds, effectively improving the robustness and service life of the test system.

[0044] The core function of this step is to provide the system with a means to evaluate and judge in real time whether there is a risk of structural resonance caused by fluid pulsation in the gas-liquid low-temperature two-phase flow. The key to this function is to identify and analyze the possible excitation effects of fluid disturbances on the system structure, to give early warnings and to intervene proactively. Due to the unevenness of gas-liquid mixing and the suddenness of local phase changes, low-temperature two-phase fluids can easily cause periodic fluctuating pressure or flow rate changes. If the frequency of these disturbances is close to the natural frequency of the system structure, they may excite mechanical resonance, causing increased vibration of the equipment and even fatigue damage to components. Therefore, the content described in this dependent claim essentially provides the entire monitoring system with the ability of "active safety prediction", so that the system has the intelligent features of self-perception, self-identification, and self-intervention, and is one of the indispensable core technologies for achieving safe and reliable measurement.

[0045] During implementation, the system must first possess high-resolution data acquisition capabilities, capable of capturing instantaneous pressure fluctuations in low-temperature two-phase flows in real time. By deploying highly sensitive pressure sensors at the pipeline inlet, the system can accurately sense the disturbance characteristics of the fluid entering the pipeline and record continuous pressure curves at microsecond sampling speeds. These curves are then processed in real time by a signal analysis module, extracting the frequency components with the highest energy concentration within each time window and identifying the primary excitation frequency currently affecting the system.

[0046] The system then calculates the rate of pressure change during this time period to determine the severity of the energy input. This pressure change rate is an important basis for determining the intensity of fluid excitation and is the first indicator for identifying sudden disturbances. By combining frequency information with pressure variation characteristics, the system can determine whether the "frequency-intensity" characteristics of the disturbance pose a potential excitation risk.

[0047] Based on this, the system constructs a dynamic model to predict the risk of resonant excitation. Using the structure's natural frequency, damping ratio, and current disturbance frequency as key parameters, this model calculates the theoretical response amplitude of the structure under a given disturbance. If this response amplitude exceeds a set threshold, the system automatically marks the period as a "potential resonance risk zone" and triggers the dynamic damping module to implement response control.

[0048] It's worth noting that this risk assessment process doesn't rely on a single instantaneous value, but rather combines multiple assessment results within a sliding time window to comprehensively determine whether there is persistent risk. In this way, the system can not only identify occasional disturbances, but also capture the precursors of resonance with a certain degree of persistence and trend, thus achieving early warning capabilities.

[0049] This step also automatically transmits the judgment results to the main control system, providing real-time risk warnings in the human-machine interface. Data for each stimulus event is archived for historical maintenance analysis and future algorithm optimization. This closed-loop "perception-modeling-judgment-response" mechanism enables the low-temperature two-phase flow test system to evolve from passive defense to active safety, and is a key component in achieving high-reliability operation for the entire solution.

[0050] The monitoring and metering output module ultimately outputs the corrected flow rate, phase holdup, and pulsation suppression status to the main control system, enabling accurate monitoring and safe metering of low-temperature two-phase flows. The main control system includes a human-machine interface, redundant data acquisition card sets and dual-redundant diagnostic modules for safety protection; All measurement data and flow state parameters are archived uniformly and displayed in real time on a visual interface in a graphical manner; With historical record backtracking function, users can select any time period to query pulsation characteristics, gas-liquid ratio and flow rate change trend; The main control system automatically triggers an emergency stop command after detecting that a high-risk frequency exists continuously for more than 3 seconds, disconnecting the main valve controller to prevent system damage.

[0051] The entire control logic is based on a fault-tolerant design, ensuring that the backup path quickly takes over in the event of a primary path failure, thereby ensuring high-reliability operation.

[0052] This step aims to fully manage the system's measurement, judgment, and execution tasks through an integrated master control system, forming a closed control loop focused on safety, real-time performance, and user visualization. The ultimate goal is to enhance the system's automated response capabilities to emergencies and failures, reduce human intervention, and ensure continuous and stable system operation.

[0053] The main control system integrates multiple submodules, including a data storage and backtracking module, a status graphical display module, and an automatic shutdown protection module. The data module centrally records various sensor and processing data, enabling later review and analysis. The display module presents operating status in real time through a graphical interface, facilitating operator understanding of system dynamics. If the protection module determines that the system has entered a high-risk state, it immediately disconnects the main circuit power supply through built-in logic control to ensure that the system is not damaged by delayed operation. These modules work together to form a highly intelligent monitoring and response platform.

[0054] The gas-liquid phase ratio estimation and volume flow correction are calculated using a multi-parameter coupled joint modeling method. The specific steps are as follows: Set up three sections of the pipeline to collect flow velocity data , the overall average flow velocity is solved using the variable weight weighted average method, and the calculation expression is as follows: , where is the average phase velocity, which is a weighted estimate of the true flow velocity of the fluid in the entire pipe section and is used for subsequent volume flow calculations. They represent the local instantaneous flow velocities collected at three sections of the pipeline (such as the inlet section, middle section, and outlet section), usually measured by a thermal or laser Doppler flowmeter. respectively The weighting coefficient is dynamically adjusted based on the disturbance intensity or historical stability index of each section. It is often generated by the standard deviation or inverse variance of the flow velocity in that section, which reduces the weight of the noisy section and improves the accuracy of the average value. It is dimensionless. The current gas phase mole fraction is estimated by combining the compressibility factor and density ratio using the ideal gas state correction model. The calculation expression is as follows: , where It is the gas phase mole fraction, which indicates the proportion of molecules in the gas phase per unit volume and is used for subsequent volume flow correction. The value range is between 0 and 1 and has no unit. It is the absolute pressure at the measuring point, usually collected using a high-precision low-temperature pressure sensor. It is the absolute temperature of the measuring point, measured by a low-temperature thermocouple or platinum resistance thermometer. is the universal gas constant, used to convert temperature and pressure to molar concentration, and are the gas and liquid phase compressibility factors, which are used to correct the state equation under non-ideal conditions. and The density of the gas phase and liquid phase at the current temperature and pressure conditions, in kg / m³. Used to evaluate the mass distribution ratio of the two phases; An empirical correction factor is introduced to calculate the corrected volume flow rate of gas-liquid coupling. The calculation expression is as follows: , where is the corrected effective volume flow rate, representing the net output volume rate of the mixed gas-liquid fluid that actually passes through the cross section. is the cross-sectional area of ​​the pipe, is the gas phase interference correction factor, which is an empirically obtained adjustment coefficient used to describe the nonlinear weakening effect of the increase in the gas phase proportion on the overall flow effective area and velocity superposition, and is usually in the range of 0.1–0.5; Continuous sliding calculation If the standard deviation within the five-second time window is less than the threshold , It is the set stability threshold, which represents the maximum allowable output fluctuation. If it exceeds the threshold, the estimation is considered unstable, and the current data is marked as a valid measurement result and output to the main control interface. If it exceeds the threshold, the refresh is delayed and a prompt is given that re-judgment is required.

[0055] This algorithm significantly improves the robustness of two-phase mixed flow calculations, especially with the ability to quickly identify and dynamically correct phase mutation points, providing a mathematical basis and computational guarantee for accurate measurement of the system.

[0056] This step realizes the accurate estimation of the flow rate of gas-liquid two-phase in a low-temperature environment, especially the real-time identification of the instantaneous phase ratio and the correction of the overall volume flow output. Since in the actual working process, the low-temperature two-phase flow has the characteristics of uneven flow velocity, complex phase distribution, frequent gas-liquid conversion, etc., the traditional single-phase fluid measurement method cannot obtain accurate flow data. The level of gas phase content directly affects the validity and measurement error of the volume flow. If it is not corrected, it may eventually cause the entire measurement system to be distorted, affecting energy metering, transmission efficiency evaluation and system safety analysis. Therefore, this dependent claim constructs a set of refined volume flow calculation system for gas-liquid mixtures by integrating multi-point flow velocity information and real-time phase ratio estimation logic, thereby providing higher precision, stronger robustness and more practical measurement support for the entire system.

[0057] In terms of implementation, the system first collects flow velocity information from the inlet, middle and outlet sections by deploying high-response flow velocity sensors at multiple key sections in the low-temperature two-phase flow pipeline. The locations of these sampling points are selected based on the empirical flow model, which can represent the typical state of the fluid from unmixed to fully mixed and then to the outlet stage. In order to avoid data distortion due to local flow anomalies, the system uses an adjustable weighting algorithm to dynamically assign different weights according to the fluctuation of the data collected by each sensor to achieve a more robust average flow velocity estimation. This weighting process can be calculated inversely proportional to the fluctuation amplitude of historical data, so that the locations with large fluctuations automatically reduce their influence in the total flow velocity calculation.

[0058] Next, the system simultaneously collects temperature and pressure data for that section of fluid and matches it with a pre-stored database of gas-liquid physical properties. Using gas-liquid state parameters such as compressibility, density ratio, and molar mass, the system derives the current gas phase proportion based on the thermodynamic state. This phase proportion estimation is not a simple conversion; rather, it is the result of a combination of thermophysical property modeling and flow field information inference. It is highly adaptable and can handle nonlinear changes in proportion caused by sudden phase transitions.

[0059] After obtaining the average flow rate and gas phase ratio, the system inputs both into the flow correction module. This module incorporates a nonlinear correction model tailored to the gas-liquid coupling characteristics. It dynamically adjusts the theoretical flow rate based on the actual gas phase ratio, ensuring that the final output volume flow rate reflects the effects of gas phase variations while smoothing out measurement disturbances caused by uneven distribution of the two phases. Compared to traditional correction methods based on average density, this method offers greater instantaneous response and nonlinear compensation capabilities, particularly in scenarios with drastic operating conditions.

[0060] Finally, to ensure the long-term stability and engineering applicability of the output results, the system incorporates a sliding time window standard deviation analysis mechanism to evaluate the fluctuation of the corrected volume flow rate over consecutive time periods. If the fluctuation is excessive, the system automatically delays data output and re-executes the phase inclusion decision logic to prevent erroneous measurement information from entering the control system. If the fluctuation is within an acceptable range, the system confirms the data is stable before outputting it and uses the data to update the dynamic model parameters.

[0061] Through the modeling, acquisition, identification, and correction process described above, a high-precision, real-time output of the actual volume flow rate in low-temperature two-phase flow is achieved. This dependent claim enhances the measurement system's adaptability to complex phase fluids, reduces measurement errors, and provides strong data support for energy management and system control.

[0062] By constructing a resonance determination model that combines a frequency response function with the dynamic characteristics of the fluid, this invention achieves rapid identification and precise response control of periodic pulsations caused by gas-liquid phase transitions in low-temperature two-phase flows. Compared to traditional methods that rely solely on pressure thresholds or static models, this solution can capture and quantify pulsation excitation energy in real time and dynamically adjust damping elements to actively eliminate frequency bands that may cause resonance. This significantly improves the system's operational stability under complex operating conditions, reduces sensor structural fatigue and false alarm rates, and enhances the structural safety and service life of the monitoring and metering system.

[0063] The present invention also combines real-time data acquisition with a multi-point velocity fusion algorithm to construct a gas-liquid phase ratio identification and volume flow correction model with dynamic self-adaptation capabilities. Through weighted fusion of multi-point velocity data, estimation of the gas phase mole fraction, and dynamic correction formulas, the system makes the overall flow assessment of the two-phase flow closer to the actual transportation state, avoiding the significant errors caused by single-point measurement under uneven flow or disturbance conditions. Therefore, in low-temperature gas-liquid mixed transmission pipelines, LNG equipment, and aerospace propulsion systems, this solution can significantly improve fluid metering accuracy, providing key technical support for improving energy utilization efficiency and safe system operation.

[0064] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It is apparent to those skilled in the art that various changes, modifications, substitutions, and variations to these embodiments may be made without departing from the principles and spirit of the present invention, and these changes and modifications still fall within the scope of protection of the present invention.

Claims

1. A gas-liquid low-temperature two-phase flow precision monitoring and measurement test system, characterized in that: It includes segmented sensing acquisition module, high-speed signal analysis module, pulsation resonance identification module, active damping control module, phase change intelligent identification module and monitoring and metering output module: The segmented sensing acquisition module deploys multiple sets of separate fluid dynamic response sensors in the low-temperature two-phase flow pipeline, located in the two-phase mixing inlet section, the intermediate stable section, and the outlet section, to achieve segmented data acquisition; The high-speed signal analysis module synchronously inputs the collected instantaneous flow velocity, pressure and temperature signals into the high-speed signal processing module and uses the spectrum analysis method to extract the fluid pulsation characteristic parameters; The pulsation resonance identification module establishes a fluid pulsation resonance model and uses the model calculation to determine whether a fluid excitation frequency close to the natural frequency of the LT-TFM system occurs; The active damping control module, upon detecting a potential resonance trend, activates the adaptive damping module to adjust the controllable damping elements within the flow channel structure to actively weaken the flow disturbance in the resonant frequency band; The phase change intelligent identification module uses a phase change feature recognition algorithm to dynamically correct the two-phase ratio estimate while reducing pulsation interference, thereby improving the accuracy of gas-liquid interface identification. The monitoring and metering output module ultimately outputs the corrected flow rate, phase holdup, and pulsation suppression status to the main control system, enabling accurate monitoring and safe metering of low-temperature two-phase flow.

2. A gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1, characterized in that: The multiple sets of separate fluid dynamic response sensors include miniature piezoelectric pressure sensors, thermal flow rate measurement units, and miniature thermocouple temperature acquisition modules. Each acquisition unit is connected to the data processing center via a high-speed signal bus. At the inlet section of the pipeline, sensors are arranged at a spacing of 20mm to form a high-density sampling array, which is used to capture local fluctuations of the fluid in the initial mixing state; in the intermediate stable section, the sensor spacing is increased to 50mm to analyze the changes in macroscopic parameters during the stable transmission of the mixture; in the outlet section, sensors are distributed at an angle in the annular monitoring cavity to collect the final state of the fluid before leaving the pipe.

3. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1 is characterized in that: The signal processing module includes an embedded FPGA data processing chip, a redundant filter bank, and a fast Fourier transform operation unit; All sensors trigger sampling synchronously, and the data sampling frequency is no less than 100kHz to ensure the capture of microsecond-level pulsation changes; The FFT operation module performs spectrum conversion on the data window collected every second, extracts the first three main frequency components and their amplitudes, and analyzes the dominant frequency range of the fluid disturbance; the redundant filter adopts a bandpass filter structure and adjusts the center frequency according to the main frequency before processing, so that the processed signal is targeted and avoids background noise interference in analysis and judgment.

4. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1 is characterized in that: The fluid pulsation resonance model is based on the fusion modeling of structural dynamic response analysis and real-time fluid disturbance characteristics; A dynamic input parameter set is established through the instantaneous flow rate change rate, pressure fluctuation amplitude and fluid component change trend collected by the sensor; The fluid pulsation resonance model considers the physical characteristics of the pipeline structure and uses numerical analysis to generate inherent response characteristics; The degree of matching between the disturbance frequency and the structural characteristic frequency is used to comprehensively evaluate whether there is any excitation of structural resonance.

5. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1 is characterized in that: The adaptive damping module uses a liquid adjustable damping element. The piezoelectric driver controls the internal throttling structure of the liquid adjustable damping element to respond to frequency changes and achieve dynamic adjustment of the damping characteristics. The module uses a high-speed signal as the driving signal and automatically adjusts the damping range to match the corresponding disturbance band in combination with the main disturbance frequency provided by the resonance model. When the disturbance frequency range changes, the piezoelectric control signal is updated through closed-loop control to achieve real-time response and dynamic compensation; at the same time, the module has a fast response capability, and the adjustment delay does not exceed 20ms, ensuring that the fluid disturbance is suppressed in the early stage and avoids entering a resonance state.

6. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1 is characterized in that: The phase change feature recognition algorithm uses an improved fuzzy support vector machine to fuse the temperature gradient change rate and pressure hysteresis response characteristics to identify and classify the gas-liquid boundary; A multi-dimensional recognition feature space is constructed by training sample data, and a classification model is established with the first-order derivative of temperature, the second-order derivative of pressure, and the flow velocity pulsation amplitude as input factors; When the actual operation data enters the classification process, it is mapped to the constructed feature space, and the predicted value of the gas-liquid phase ratio is output through SVM.

7. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1 is characterized in that: Resonance risk judgment adopts a dynamic assessment mechanism based on the frequency response model. The specific steps are as follows: During the signal acquisition process, short-time Fourier transform is applied to the inlet pressure signal to extract the instantaneous dominant frequency from the time-frequency domain. The extraction formula is as follows: , where is the instantaneous main frequency, indicating the previous moment The frequency component with the strongest energy in the pressure fluctuation in its adjacent time period, It's in time The pressure signal collected at all times, is the half-width of the window for short-time Fourier analysis, is the Fourier basis function, which is used to transform the time domain signal into the frequency domain. is the natural base, is an imaginary unit, is the angular frequency multiplied by time, which means the signal is at frequency When , the corresponding phase change is, is the frequency, is the time variable, is the differential symbol; The instantaneous pressure change rate is obtained by taking the derivative of the pressure curve in the simultaneous time window. The calculation expression is as follows: , where is the instantaneous rate of pressure change, It's pressure Find the first derivative, which represents how quickly pressure changes with time, is the reference time point for calculating the pressure derivative, i.e. the current sampling moment; A frequency response function model for a single degree of freedom system is established to calculate the response amplitude in the current state. The calculation expression is as follows: , where is the natural frequency, is the damping ratio, is the predicted value of the response amplitude; Set the empirical response amplitude threshold , if at any moment , it is determined that there is a risk of resonant excitation, the dynamic damping module is triggered immediately, and the current event mark is archived for subsequent operation strategy adjustment.

8. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1 is characterized in that: The main control system includes a human-machine interface, redundant data acquisition card sets and dual-redundant diagnostic modules for safety protection; All measurement data and flow state parameters are archived uniformly and displayed in real time on a visual interface in a graphical manner; With historical record backtracking function, users can select any time period to query pulsation characteristics, gas-liquid ratio and flow rate change trend; The main control system automatically triggers an emergency stop command after detecting that a high-risk frequency exists continuously for more than 3 seconds, disconnecting the main valve controller to prevent damage.

9. The gas-liquid low-temperature two-phase flow precision monitoring and measurement test system according to claim 1, characterized in that: The gas-liquid phase ratio estimation and volume flow correction are calculated using a multi-parameter coupled joint modeling method. The specific steps are as follows: Set up three sections of the pipeline to collect flow velocity data , the overall average flow velocity is solved using the variable weight weighted average method, and the calculation expression is as follows: , where is the average phase velocity, Represent the local instantaneous flow velocities collected at the three sections of the pipeline, respectively The weighting coefficient of The current gas phase mole fraction is estimated by combining the compressibility factor and density ratio using the ideal gas state correction model. The calculation expression is as follows: , where is the gas phase mole fraction, is the absolute pressure at the measuring point, is the absolute temperature of the measurement point, is the universal gas constant, and are the gas and liquid phase compressibility factors, and are the densities of the gas phase and liquid phase under the current temperature and pressure conditions; An empirical correction factor is introduced to calculate the corrected volume flow rate of gas-liquid coupling. The calculation expression is as follows: , where is the corrected effective volume flow rate, is the cross-sectional area of ​​the pipe, is the gas phase interference correction factor; Continuous sliding calculation If the standard deviation within the five-second time window is less than the threshold , the current data will be marked as a valid measurement result and output to the main control interface; if it exceeds, the refresh will be delayed and a prompt will be given that re-judgment is required.

Images