A precise monitoring and metering test system for gas-liquid low-temperature two-phase flow

Abstract

The application discloses a kind of gas-liquid low temperature two-phase flow precision monitoring and metering test system, including segmented sensing acquisition module, high-speed signal analysis module, pulsating resonance identification module, active damping control module, phase change intelligent identification module and monitoring metering output module: segmented sensing acquisition module, in low temperature two-phase flow pipeline Multiple groups of separated fluid dynamic response sensors are arranged, respectively located in two-phase mixing inlet section, intermediate stable section and outlet section, realize segmented data acquisition The application has the advantages that: the resonance risk caused by gas-liquid phase change can be accurately identified, the damping structure is real-time regulated, the system resonance and sensor damage are effectively avoided;At the same time, through multi-point flow rate fusion and gas phase mole fraction dynamic correction, the high-precision measurement of two-phase flow volume flow is realized, the measurement stability and system safety are improved, and it is widely used in low-temperature gas-liquid mixed transportation scene.

Description

Technical Field

[0001] This invention relates to the field of measurement technology, and in particular to a precise monitoring and measurement system for low-temperature gas-liquid two-phase flow. Background Technology

[0002] "Accurate Monitoring and Measurement of Cryogenic Two-Phase Flow" refers to the technical process of accurately monitoring and measuring the flow rate of fluids that simultaneously contain both gas and liquid states in a cryogenic environment. Such two-phase fluids are commonly found in the transportation or use of cryogenic media such as liquefied natural gas (LNG), liquid nitrogen, and liquid hydrogen. During transmission, due to temperature or pressure changes, both liquid and gas phases exist simultaneously. Traditional single-phase flow measurement methods struggle to obtain accurate data under such complex flow conditions. Therefore, this technology integrates sensors, signal processing, and electronic control analysis to collect parameters such as fluid pressure, temperature, phase distribution, and flow velocity in real time. Utilizing specialized algorithms or structural designs, it achieves accurate identification of the gas-liquid mixture state and high-precision flow rate measurement, providing reliable support for cryogenic energy management, storage and transportation system safety, and energy efficiency optimization.

[0003] Existing technologies have the following shortcomings: In the precise monitoring and measurement of cryogenic two-phase gas-liquid flow, fluid pulsations caused by gas-liquid phase transitions may generate resonant interference in the measurement system. When a fluid undergoes a drastic phase transition due to temperature and pressure fluctuations, periodic flow pulsations are easily formed, especially in pipe structures, producing an effect similar to "liquid hammer." These pulsations not only cause drastic fluctuations in instantaneous flow velocity and gas-liquid ratio, interfering with measurement stability, but may also excite mechanical resonance in the internal measurement structure of the system, reducing data accuracy. In severe cases, they may even damage sensors or critical metering devices, causing misjudgments or abnormal system shutdowns, affecting the overall safety and reliability of operation. Summary of the Invention

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

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is: a precise monitoring and measurement system for low-temperature gas-liquid two-phase flow, comprising a segmented sensing and 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.

[0006] The segmented sensing and acquisition module deploys multiple sets of separate fluid dynamic response sensors in the low-temperature two-phase flow pipeline, located at the two-phase mixing inlet section, intermediate stable section and outlet section, respectively, to achieve segmented data acquisition.

[0007] The high-speed signal analysis module synchronously inputs the collected instantaneous flow velocity, pressure and temperature signals to the high-speed signal processing module, and uses the spectrum analysis method to extract fluid pulsation characteristic parameters;

[0008] The pulsating resonance identification module establishes a fluid pulsating resonance model and uses the model to calculate and determine whether a fluid excitation frequency close to the natural frequency of the LT-TFM system has occurred.

[0009] The active damping control module, once a potential resonance trend is detected, activates the adaptive damping module to adjust the controllable damping elements inside the flow channel structure and actively reduce the flow disturbance in the resonance frequency band.

[0010] The phase change intelligent identification module, while reducing pulsation interference, uses a phase change feature identification algorithm to dynamically correct the estimated value of the two-phase ratio, thereby improving the accuracy of gas-liquid interface identification.

[0011] The monitoring and metering output module ultimately outputs the corrected flow rate, phase content, and pulsation suppression status to the main control system, achieving accurate monitoring and safe metering of low-temperature two-phase flow.

[0012] Preferably, the multiple sets of separate fluid dynamic response sensors include a miniature piezoelectric pressure sensor, a thermal flow velocity measurement unit, and a miniature thermocouple temperature acquisition module. Each acquisition unit is connected to the data processing center via a high-speed signal bus.

[0013] At the inlet section of the pipeline, sensors are arranged at 20mm intervals 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. At the outlet section, sensors are distributed at an angle in the annular monitoring cavity to collect the final state of the fluid before it leaves the pipe.

[0014] Preferably, the signal processing module includes an embedded FPGA data processing chip, a redundant filter bank, and a fast Fourier transform operation unit;

[0015] All sensors are triggered to sample synchronously, with a data sampling frequency of no less than 100kHz, to ensure the capture of microsecond-level pulsation changes;

[0016] The FFT operation module performs spectrum conversion on the data window collected every second and extracts the top three main frequency components and their amplitudes to analyze the dominant frequency range of fluid disturbance. The redundant filter adopts a bandpass filter structure and adjusts the center frequency according to the main frequency before processing to make the processed signal targeted and avoid background noise interference in analysis and judgment.

[0017] Preferably, the fluid pulsating resonance model is based on the fusion modeling of structural dynamic response analysis and real-time fluid disturbance characteristics;

[0018] A dynamic set of input parameters is established by collecting instantaneous flow rate change, pressure fluctuation amplitude, and fluid composition change trends from sensors.

[0019] The fluid pulsating resonance model considers the physical characteristics of the pipe structure and uses numerical analysis to generate the inherent response characteristics.

[0020] The degree of matching between the perturbation frequency and the structural characteristic frequency is used to comprehensively assess whether structural resonance is induced.

[0021] Preferably, the adaptive damping module uses a liquid adjustable damping element, and the damping characteristics are dynamically adjusted by controlling the response frequency change of the internal throttling structure of the liquid adjustable damping element by a piezoelectric actuator.

[0022] The module uses a high-speed signal as the driving signal and, combined with the main disturbance frequency provided by the resonance model, automatically adjusts the damping range to match the corresponding disturbance band.

[0023] 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, with an adjustment delay of no more than 20ms, to ensure that the fluid disturbance is suppressed in the initial stage and avoid entering the resonance state.

[0024] Preferably, the phase transition feature recognition algorithm uses an improved fuzzy support vector machine to fuse the temperature gradient change rate and pressure hysteresis response features to identify and classify the gas-liquid boundary;

[0025] A multidimensional recognition feature space is constructed using training sample data, and a classification model is established using the first derivative of temperature, the second derivative of pressure, and the amplitude of flow velocity pulsation as input factors.

[0026] Once the actual operational data enters the classification process, it is mapped to the constructed feature space, and the predicted gas-liquid phase ratio is output through SVM.

[0027] Preferably, the resonance risk assessment adopts a dynamic evaluation mechanism based on a frequency response model, and the specific steps are as follows:

[0028] During signal acquisition, a short-time Fourier transform is applied to the inlet section pressure signal to extract the instantaneous dominant frequency from the time-frequency domain. The extraction formula is as follows:

[0029] In the formula, It is the instantaneous main frequency, indicating the frequency at the previous moment. The frequency component with the strongest energy in the pressure fluctuations and in the adjacent time period. It is in time Pressure signals collected at all times It is the half-width of the window for short-time Fourier analysis. These are Fourier basis functions, used to transform time-domain signals to the frequency domain. It is the natural base. It is the imaginary unit. It is the angular frequency multiplied by the time, representing the signal frequency. At that time, the corresponding phase change, It's frequency. It is a time variable. It is the differential symbol;

[0030] Differentiating the pressure curves within the same time window yields the instantaneous pressure change rate, expressed as follows:

[0031] In the formula, It is the instantaneous rate of change of pressure. It is pressure Find the first derivative, which represents the rate at which the pressure changes with time. It is the reference time point for calculating the pressure derivative, i.e., the current sampling time;

[0032] A frequency response function 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:

[0033] In the formula, It is the natural frequency. It is the damping ratio. It is the predicted value of the response amplitude;

[0034] Set the empirical response amplitude threshold If at any time it is satisfied If a resonant excitation risk is detected, the dynamic damping module is immediately triggered, and the current event is marked and archived for subsequent operational strategy adjustments.

[0035] Preferably, the main control system includes a human-machine interface, redundant data acquisition card group, and dual-redundant diagnostic module for security protection;

[0036] All measurement data and flow state parameters are archived in a unified manner and displayed in real time in a graphical visualization interface;

[0037] It has a historical record backtracking function, allowing users to select any time period to query pulsation characteristics, gas-liquid ratio, and flow rate change trends.

[0038] The main control system automatically triggers an emergency stop command after detecting a high-risk frequency that persists for more than 3 seconds, disconnecting the main valve controller to prevent damage.

[0039] Preferably, the gas-liquid phase ratio estimation and volumetric flow rate correction are performed using a multi-parameter coupled joint modeling calculation, with the specific steps as follows:

[0040] Flow velocity data were collected at three locations along the pipeline. The overall average velocity is calculated using the variable weighted average method, and the result is expressed as follows:

[0041] In the formula, It is the average phase velocity. These represent the local instantaneous flow velocities collected at three locations along the pipe. respectively Weighting coefficients;

[0042] The current gas phase mole fraction is estimated using an ideal gas state correction model combined with compressibility factor and density ratio, and the calculation expression is as follows:

[0043] In the formula, It is the gas phase mole fraction. It is the absolute pressure at the measuring point. It is the absolute temperature at the measurement point. It is the universal gas constant. and Both are gas-phase and liquid-phase compressibility factors. and These are the densities of the gas phase and the liquid phase under the current temperature and pressure conditions, respectively.

[0044] Introducing an empirical correction factor, the corrected volumetric flow rate for gas-liquid coupling is calculated using the following expression:

[0045] In the formula, This is the corrected effective volumetric flow rate. It is the cross-sectional area of ​​the pipe. It is a gas phase interference correction factor;

[0046] Continuous sliding calculation If the standard deviation within the five-second time window is less than the threshold... If the current data is not exceeded, it will be marked as a valid measurement result and output to the main control interface; if it exceeds the limit, the refresh will be delayed and a prompt will be displayed indicating that a re-evaluation is needed.

[0047] The beneficial effects of this invention are:

[0048] This invention achieves rapid identification and precise response control of periodic pulsations caused by gas-liquid phase transitions in low-temperature two-phase flow by constructing a resonance determination model that combines frequency response functions with fluid dynamic characteristics. Compared to traditional methods that rely solely on pressure thresholds or static models, this approach can capture and quantify the 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 conditions, reduces sensor structural fatigue and false alarm rates, and enhances the structural safety and service life of the monitoring and measurement system.

[0049] This invention also combines real-time data acquisition with a multi-point velocity fusion algorithm to construct a dynamically adaptive gas-liquid phase ratio identification and volumetric flow rate correction model. 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 rate assessment of the two-phase flow more closely resemble the actual transport conditions, avoiding significant errors caused by single-point measurements under uneven or disturbed flow conditions. Therefore, in cryogenic gas-liquid mixed transport pipelines, LNG plants, and aerospace propulsion systems, this solution can significantly improve fluid metering accuracy, providing key technical support for improving energy utilization efficiency and ensuring safe system operation. Attached Figure Description

[0050] Figure 1 This is a module diagram of a precise monitoring and measurement system for low-temperature gas-liquid two-phase flow. Detailed Implementation

[0051] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0052] The working principle of this invention: A precise monitoring and measurement system for low-temperature gas-liquid two-phase flow includes a segmented sensing and acquisition module, a high-speed signal analysis module, a pulsating resonance identification module, an active damping control module, a phase change intelligent identification module, and a monitoring and measurement output module.

[0053] The segmented sensing and acquisition module deploys multiple sets of separate fluid dynamic response sensors in the low-temperature two-phase flow pipeline, located at the two-phase mixing inlet section, intermediate stable section and outlet section, respectively, to achieve segmented data acquisition.

[0054] The deployed set of separate fluid dynamic response sensors includes miniature piezoelectric pressure sensors, thermal velocity measurement units, and miniature thermocouple temperature acquisition modules. Each acquisition unit is connected to the data processing center via a high-speed signal bus.

[0055] At the inlet section of the pipeline, sensors are arranged at 20mm intervals 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. At the outlet section, sensors are distributed at an angle in the annular monitoring cavity to collect the final state of the fluid before it leaves the pipe.

[0056] This distributed design ensures that the system can capture dynamic physical characteristics with high spatial resolution in real time throughout the entire process of fluid entry, evolution, and output, providing sufficient basis for subsequent resonance judgment and dynamic correction.

[0057] The purpose of this step is to ensure that the system can achieve high-precision, continuous dynamic monitoring throughout the entire two-phase flow transport path, avoiding omissions of key flow states in spatial distribution. By subdividing the functions of different pipe sections and configuring multiple sets of sensors with corresponding densities, the fluid evolution trajectory throughout the entire process of gas-liquid mixing, transport, and discharge can be effectively perceived, thereby providing a detailed data foundation for subsequent signal analysis and dynamic control.

[0058] 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 at the inlet captures the initial disturbances in the fluid entering the system; sparse sensors in the middle section track homogeneous changes in the flow process; and a ring-shaped sensor distribution at the outlet monitors the final state of the fluid. All sensors use a unified time reference to ensure data acquisition is completed with minimal time differences, and sampling synchronization is achieved through a central processing node. This system deployment constitutes a comprehensive, segmented, and high-frequency physical quantity sensing system.

[0059] The high-speed signal analysis module synchronously inputs the collected instantaneous flow velocity, pressure and temperature signals to the high-speed signal processing module, and uses the spectrum analysis method to extract fluid pulsation characteristic parameters;

[0060] The signal processing module includes an embedded FPGA data processing chip, a redundant filter bank, and a Fast Fourier Transform (FFT) operation unit;

[0061] All sensors are triggered to sample synchronously, with a data sampling frequency of no less than 100kHz, to ensure the capture of microsecond-level pulsation changes;

[0062] The FFT operation module performs spectrum conversion on the data window collected every second and extracts the top three main frequency components and their amplitudes to analyze the dominant frequency range of fluid disturbance. The redundant filter adopts a bandpass filter structure and adjusts the center frequency according to the main frequency before processing to make the processed signal targeted and avoid background noise interference in analysis and judgment.

[0063] In addition, the system provides early warnings of possible resonance states based on the main frequency drift trend, providing basic spectral data for the next step of model judgment.

[0064] The main function of this step is to transform the acquired fluid physics signals into a data representation with time and frequency domain characteristics, thereby identifying potential periodic disturbance features. Spectral analysis can not only be used to determine whether the flow state is stable, but also to further identify the dominant disturbance frequencies affecting system stability, providing a prerequisite for predicting resonance and implementing control interventions.

[0065] By sampling data synchronously at high frequency, the system can respond to rapid disturbances occurring within a short period. The signal is then fed into the analysis module, where a conversion method identifies the main frequency ranges of periodic fluctuations in the fluid state. The data is further filtered and classified to ensure that dominant frequencies with engineering significance are identified, rather than random noise signals. Simultaneously, the system utilizes an adjustable signal filter to focus on the frequency range of interest, effectively suppressing interference from non-target noise on the frequency identification results.

[0066] The pulsation resonance identification module establishes a fluid pulsation resonance model and uses the model to calculate and determine whether a fluid excitation frequency close to the natural frequency of the LT-TFM system (precision monitoring and measurement system for low-temperature gas-liquid two-phase flow) occurs.

[0067] The fluid pulsating resonance model is based on the fusion modeling of structural dynamic response analysis and real-time fluid disturbance characteristics;

[0068] A dynamic set of input parameters is established by collecting instantaneous flow rate change, pressure fluctuation amplitude, and fluid composition change trends from sensors.

[0069] The model considers the physical properties of the pipeline structure, such as geometry, material parameters, and boundary conditions, and uses numerical analysis to generate the inherent response characteristics of the system.

[0070] The degree of matching between the perturbation frequency and the structural characteristic frequency is used to comprehensively assess whether structural resonance is induced.

[0071] This model not only assesses the presence of resonance risks in real time but also possesses online learning capabilities, automatically adjusting 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 characteristics and structural response, it generates a risk score and marks high-risk periods for subsequent triggering of vibration suppression control mechanisms and alarm prompts, significantly enhancing the stability and structural safety of the two-phase flow measurement system.

[0072] The purpose of this step is to perform a linked modeling of fluid disturbance characteristics and system structural response characteristics to determine whether the system is currently in a dangerous range that may trigger mechanical resonance. This model constructs an early warning mechanism, enabling the system to detect potential risks at an early stage without relying on feedback from actual structural failure to trigger a response.

[0073] The implementation process involves comprehensively modeling the mechanical properties of the test pipeline structure and establishing a judgment model based on the fluid disturbance frequencies collected by sensors. This model compares the current fluid state with the structural response characteristics to identify whether there are frequency overlaps or approximations, thereby predicting the possibility of mechanical resonance. The model can also be corrected and updated based on data change trends and historical operating records to adapt to model shifts caused by changes in fluid state, ensuring the real-time nature and accuracy of the judgment results.

[0074] The active damping control module, once a potential resonance trend is detected, activates the adaptive damping module to adjust the controllable damping elements inside the flow channel structure and actively reduce the flow disturbance in the resonance frequency band.

[0075] The adaptive damping module uses a liquid adjustable damping element, which is dynamically adjusted in terms of damping characteristics by controlling the frequency change of its internal throttling structure with a piezoelectric actuator.

[0076] The module uses a high-speed signal as the driving signal and, combined with the main disturbance frequency provided by the resonance model, automatically adjusts the damping range to match the corresponding disturbance band.

[0077] 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, with an adjustment delay of no more than 20ms, to ensure that the fluid disturbance is suppressed in the initial stage and avoid entering the resonance state.

[0078] This damping control strategy offers higher sensitivity and stability compared to traditional passive buffer devices.

[0079] The purpose of this step is to provide an active response mechanism that, when the system identifies a risk of resonance, can immediately adjust the operating state of the damping device to reduce or block the energy propagation path of the disturbance at its source, preventing the system from entering an unstable operating state. This damping mechanism breaks through the limitations of traditional passive design and achieves intelligent energy control.

[0080] The internal structure of the damping device is dynamically adjusted by the control module, enabling it to respond appropriately to disturbances at different frequencies. The device can rapidly change the path resistance or energy absorption characteristics of the fluid flow based on the dominant frequency of the disturbance detected by the system, thereby absorbing or converting the disturbance energy and rapidly reducing the disturbance amplitude. The entire process is completed automatically by the system, with fast response time and high precision, making it particularly suitable for applications in frequency-sensitive cryogenic two-phase flow systems.

[0081] The phase change intelligent identification module, while reducing pulsation interference, uses a phase change feature identification algorithm to dynamically correct the estimated value of the two-phase ratio, thereby improving the accuracy of gas-liquid interface identification.

[0082] The phase transition feature recognition algorithm uses an improved fuzzy support vector machine (F-SVM) to fuse the temperature gradient change rate and pressure hysteresis response features to identify and classify the gas-liquid boundary;

[0083] A multidimensional recognition feature space is constructed using training sample data, and a classification model is established using the first derivative of temperature, the second derivative of pressure, and the amplitude of flow velocity pulsation as input factors.

[0084] Once the actual operational data enters the classification process, it is mapped to the constructed feature space, and the predicted gas-liquid phase ratio is output through SVM.

[0085] This algorithm can adapt to complex phase change processes under different operating conditions, effectively reduce the misjudgment rate, and ensure the accuracy of flow separation calculation and subsequent measurement.

[0086] The purpose of this step is to accurately identify the existence and changing trends of the gas-liquid interface, providing a fundamental basis for subsequent flow measurement and phase holdup determination. During two-phase flow, the gas-liquid interface moves frequently with changes in the environment and flow regime, making traditional identification methods prone to misjudgment. Intelligent identification methods can effectively solve this problem.

[0087] The system collects various real-time changing fluid characteristic signals and performs joint analysis through 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 ambiguous boundaries exist, the system assigns appropriate membership weights to avoid extreme judgment results, improving the stability and anti-interference capability of the recognition. This recognition capability is maintained throughout the entire system operation, ensuring that subsequent parameter calculations are based on sufficient and accurate data.

[0088] The resonance risk assessment adopts a dynamic evaluation mechanism based on the frequency response model, and the specific steps are as follows:

[0089] During signal acquisition, a short-time Fourier transform (STFT) is applied to the inlet section pressure signal to extract the instantaneous dominant frequency from the time-frequency domain. The extraction formula is as follows:

[0090] In the formula, It is the instantaneous main frequency (unit: Hz), indicating the frequency at the previous moment. The frequency component with the strongest energy in pressure fluctuations within and around the same time period is a key indicator for determining the resonance trend. It is in time The pressure signal (unit: Pa) collected at all times represents the instantaneous pressure of the two-phase flow in the pipeline. It is the half-width of the window (in seconds) for short-time Fourier analysis, which determines the time span of the analysis interval, typically on the order of milliseconds. These are Fourier basis functions, used to transform time-domain signals to the frequency domain. It is the natural base. It is the imaginary unit, satisfying , It is the angular frequency multiplied by the time, representing the signal frequency. At that time, the corresponding phase change, It's frequency. It is a time variable. It is the differential symbol, used in integral expressions to represent "the accumulation of infinitesimals with time as the variable";

[0091] This formula identifies the main disturbance frequency in the current system using a sliding time window, ensuring that effective main spectral components can be extracted even under non-steady-state flow conditions.

[0092] Differentiating the pressure curves within the same time window yields the instantaneous pressure change rate, expressed as follows:

[0093] In the formula, It is the instantaneous rate of change of pressure (unit: Pa / s), used to measure the degree of pressure change over time, and is a parameter for measuring the intensity of the excitation source. It is pressure Find the first derivative, which represents the rate at which the pressure changes with time. It is the reference time point for calculating the pressure derivative, i.e., the current sampling time;

[0094] This rate of change reflects the presence of drastic energy input in the fluid system and serves as the core excitation parameter in subsequent resonant excitation calculations.

[0095] A frequency response function (FRF) 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:

[0096] In the formula, It is the natural frequency (unit: Hz), determined by structural characteristics (such as length, stiffness, boundary conditions), and is the critical frequency at which resonance occurs. This is the damping ratio (dimensionless), which measures a system's ability to dissipate energy. A higher value indicates a lower likelihood of resonance; typical values ​​range from 0.05 to 0.2. It is the predicted amplitude of the response (unit: arbitrary amplitude unit, normalizable), representing the instantaneous dominant frequency. Will the system exhibit a magnification effect?

[0097] A larger response amplitude indicates that the system is closer to the resonance point.

[0098] Set the empirical response amplitude threshold If at any time it is satisfied If a resonant excitation risk is detected, the dynamic damping module is immediately triggered, and the current event is marked and archived for subsequent operational strategy adjustments.

[0099] This method combines signal spectrum analysis with structural dynamics modeling to achieve real-time prediction and suppression of implicit resonance problems in complex flow backgrounds, effectively improving the robustness and service life of the test system.

[0100] The core function of this step is to provide the system with a means to assess and determine in real time whether there is a risk of structural resonance caused by fluid pulsation in cryogenic gas-liquid two-phase flow. The key to this function lies in identifying and analyzing the potential excitation effects of fluid disturbances on the system structure, providing early warnings, and proactively intervening. Due to the non-uniformity of gas-liquid mixing and the suddenness of local phase transitions, cryogenic two-phase fluids are prone to periodic fluctuations in pressure or velocity. If the frequency of these disturbances is close to the natural frequency of the system structure, they may induce mechanical resonance, leading to increased equipment vibration or even fatigue damage to components. Therefore, the content described in this dependent claim essentially provides the entire monitoring system with the capability of "proactive safety prediction," enabling the system to possess intelligent characteristics of self-sensing, self-identification, and self-intervention, making it one of the indispensable core technologies for achieving safe and reliable measurement.

[0101] In its implementation, the system first needs to possess high temporal resolution data acquisition capabilities, enabling it to capture instantaneous pressure fluctuation signals in cryogenic two-phase flow in real time. By deploying highly sensitive pressure sensors at the pipe inlet section, the system can accurately sense the disturbance characteristics during the fluid entry phase and record continuous pressure curves at microsecond-level sampling rates. These curves are then processed in real time by a signal analysis module to extract the frequency components with the highest energy concentration within each time window, identifying the dominant excitation frequency currently experienced by the system.

[0102] Next, the system calculates the pressure change rate over that time period to determine the intensity of the energy input. This pressure change rate is a crucial indicator for assessing the intensity of fluid excitation and serves as the primary response indicator for identifying sudden disturbances. By combining frequency information with pressure change characteristics, the system can determine whether the "frequency-intensity" characteristics of the disturbance pose a potential excitation risk.

[0103] Based on this, a dynamic model for predicting the risk of resonant excitation was constructed. This model uses the structure's natural frequency, damping ratio, and the current disturbance frequency as basic parameters to calculate the theoretical response amplitude of the structure under the disturbance. If this response amplitude exceeds a set threshold, the system automatically marks that period as a "potential resonance risk region" and triggers the dynamic damping module for response control.

[0104] It is worth mentioning that this risk assessment process does not rely on a single instantaneous value, but rather combines multiple assessment results within a sliding time window to comprehensively determine whether there is a persistent risk. In this way, the system can not only identify occasional disturbances, but also capture resonance precursors with a certain degree of persistence and trend, thus achieving the function of early warning.

[0105] Furthermore, this step automatically transmits the judgment results to the main control system, provides real-time risk alerts on the human-machine interface, and archives data for each stimulus event for historical analysis and future algorithm optimization. This closed-loop mechanism of "perception-modeling-judgment-response" provides the cryogenic two-phase flow testing system with the ability to evolve from passive defense to proactive safety, and is a key supplement to the entire solution for achieving high-reliability operation.

[0106] The monitoring and metering output module ultimately outputs the corrected flow rate, phase content, and pulsation suppression status to the main control system, thereby achieving accurate monitoring and safe metering of low-temperature two-phase flow.

[0107] The main control system includes a human-machine interface, redundant data acquisition cards, and a dual-redundant diagnostic module for security protection;

[0108] All measurement data and flow state parameters are archived in a unified manner and displayed in real time in a graphical visualization interface;

[0109] It has a historical record backtracking function, allowing users to select any time period to query pulsation characteristics, gas-liquid ratio, and flow rate change trends.

[0110] The main control system automatically triggers an emergency stop command after detecting a high-risk frequency that persists for more than 3 seconds, disconnecting the main valve controller to prevent system damage.

[0111] The entire control logic is based on fault-tolerant design, ensuring that the backup path can quickly take over in the event of failure of the primary path, thus achieving high reliability operation.

[0112] The purpose of this step is to take over the system's measurement, judgment, and execution tasks through an integrated main control system, forming a control closed loop oriented towards safety, real-time performance, and user visibility. Its ultimate goal is to improve the system's automated handling capabilities in response to emergencies and failures, reduce human intervention, and ensure continuous and stable system operation.

[0113] The main control system integrates multiple sub-modules, including a data storage and traceability module, a status graphical display module, and an automatic shutdown protection module. The data module centrally records various sensor and processing data, supporting later review and analysis; the display module presents the operating status in real-time via a graphical interface, facilitating operator understanding of system dynamics; and the protection module, upon determining that the system has entered a high-risk state, immediately disconnects the main circuit power supply through built-in logic control, ensuring the system is not damaged due to delayed operation. These modules work collaboratively to form a highly intelligent monitoring and response platform.

[0114] The estimation of the gas-liquid phase ratio and the correction of the volumetric flow rate are performed using a multi-parameter coupled joint modeling calculation. The specific steps are as follows:

[0115] Flow velocity data were collected at three locations along the pipeline. The overall average velocity is calculated using the variable weighted average method, and the result is expressed as follows:

[0116] In the formula, It is the average phase velocity, a weighted estimate of the actual fluid flow velocity across the entire pipe section, used for subsequent volumetric flow rate calculations. These represent the local instantaneous flow velocities collected at three pipe sections (e.g., inlet, middle, and outlet), typically measured using a thermal or laser Doppler velocimeter. respectively The weighting coefficients are dynamically adjusted based on the disturbance intensity or historical stability index of each segment. They are often generated by the standard deviation or the reciprocal of the variance of the flow velocity in that segment, which reduces the weight of segments with high noise and improves the accuracy of the average value; dimensionless.

[0117] The current gas phase mole fraction is estimated using an ideal gas state correction model combined with compressibility factor and density ratio, and the calculation expression is as follows:

[0118] In the formula, This is the gas phase mole fraction, representing the proportion of gaseous molecules per unit volume, used for subsequent volumetric flow correction. The value range is between 0 and 1, and it is dimensionless. It measures the absolute pressure at the measurement point, typically acquired using a high-precision cryogenic pressure sensor. It is the absolute temperature of the measurement point, obtained by a low-temperature thermocouple or platinum resistance thermometer. It is a universal gas constant used to convert temperature and pressure into molar concentration. and These are both gas-phase and liquid-phase compressibility factors, used to correct the equation of state under non-ideal conditions. and These are the densities of the gas and liquid phases under the current temperature and pressure conditions, respectively, in kg / m³. They are used to evaluate the mass distribution ratio of the two phases.

[0119] Introducing an empirical correction factor, the corrected volumetric flow rate for gas-liquid coupling is calculated using the following expression:

[0120] In the formula, It is the corrected effective volumetric flow rate, representing the net output volumetric rate of the actual gas-liquid mixture passing through the cross-section. It is the cross-sectional area of ​​the pipe. It is the gas phase interference correction factor, an empirically obtained adjustment coefficient used to describe the nonlinear weakening effect of the increase in the gas phase ratio on the superposition of the effective area and velocity of the overall flow. It is usually taken in the range of 0.1–0.5.

[0121] Continuous sliding calculation If the standard deviation within the five-second time window is less than the threshold... , This is the set stability threshold, representing 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 displayed indicating that a re-evaluation is needed.

[0122] This algorithm significantly improves the robustness of two-phase mixed flow calculation, especially at phase transition points where it has the ability to quickly identify and dynamically correct, providing a mathematical foundation and computational guarantee for accurate system metering.

[0123] This step enables accurate estimation of flow rate in gas-liquid two-phase systems at low temperatures, particularly by real-time identification of instantaneous phase ratios and correction of the overall volumetric flow rate output. In practical applications, low-temperature two-phase flows are characterized by non-uniform flow velocities, complex phase distributions, and frequent gas-liquid transitions, rendering traditional single-phase fluid measurement methods ineffective in obtaining accurate flow data. The level of gas phase content directly affects the effectiveness and measurement error of the volumetric flow rate; without correction, this can ultimately distort the entire measurement system, impacting energy metering, transmission efficiency assessment, and system safety analysis. Therefore, this dependent claim integrates multi-point velocity information with real-time phase ratio estimation logic to construct a refined volumetric flow rate calculation system for gas-liquid mixtures, thereby providing higher accuracy, stronger robustness, and more practical metering support for the entire system.

[0124] In its implementation, the system first deploys high-response velocity sensors at multiple key cross-sections within the cryogenic two-phase flow pipeline to collect velocity information from the inlet, middle, and outlet sections. The selection of these sampling points is based on an empirical flow model, representing typical fluid states from unmixed to fully mixed and then to the outlet. To avoid data distortion due to localized flow anomalies, the system uses an adjustable weighting algorithm, dynamically assigning different weights based on the fluctuations in the data collected by each sensor, achieving a more robust average velocity estimation. This weighting process is calculated inversely to the historical data fluctuation amplitude, automatically reducing the influence of locations with large fluctuations in the overall velocity calculation.

[0125] Next, the system synchronously collects temperature and pressure data of the fluid in this segment and matches them with a pre-stored database of gas-liquid physical properties. Using gas-liquid state parameters such as compressibility factor, density ratio, and molar mass, the system can deduce the current proportion of the gas phase based on the thermodynamic state. This phase proportion estimation is not a simple proportional conversion, but rather a result derived from thermophysical modeling and flow field information, exhibiting strong adaptability and the ability to handle nonlinear changes in proportion caused by abrupt phase transitions.

[0126] After obtaining the average flow velocity and gas phase ratio, the system inputs both into the flow correction module. This module introduces a nonlinear correction model for the gas-liquid phase coupling characteristics, dynamically adjusting the theoretical flow rate based on the actual gas phase ratio. This ensures that the final output volumetric flow rate reflects both the influence of gas phase changes and the measurement disturbances caused by uneven two-phase distribution. Compared to traditional correction methods based on average density, this method offers superior instantaneous response and nonlinear compensation capabilities, and is particularly effective in reflecting the true physical flow characteristics in application scenarios with drastic changes in operating conditions.

[0127] Finally, to ensure the long-term stability and engineering applicability of the output results, the system introduces a sliding time window standard deviation analysis mechanism to evaluate the volatility of the corrected volumetric flow rate over continuous periods. When the volatility is too large, the system automatically delays data output and re-executes the phase inclusion judgment logic to prevent erroneous measurement information from flowing into the control system; when the volatility is within an acceptable range, the system outputs the data after confirming its stability and uses the data for updating dynamic model parameters.

[0128] Through the aforementioned modeling, data acquisition, identification, and correction process, high-precision, real-time output of the actual volumetric 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.

[0129] This invention achieves rapid identification and precise response control of periodic pulsations caused by gas-liquid phase transitions in low-temperature two-phase flow by constructing a resonance determination model that combines frequency response functions with fluid dynamic characteristics. Compared to traditional methods that rely solely on pressure thresholds or static models, this approach can capture and quantify the 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 conditions, reduces sensor structural fatigue and false alarm rates, and enhances the structural safety and service life of the monitoring and measurement system.

[0130] This invention also combines real-time data acquisition with a multi-point velocity fusion algorithm to construct a dynamically adaptive gas-liquid phase ratio identification and volumetric flow rate correction model. 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 rate assessment of the two-phase flow more closely resemble the actual transport conditions, avoiding significant errors caused by single-point measurements under uneven or disturbed flow conditions. Therefore, in cryogenic gas-liquid mixed transport pipelines, LNG plants, and aerospace propulsion systems, this solution can significantly improve fluid metering accuracy, providing key technical support for improving energy utilization efficiency and ensuring safe system operation.

[0131] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. A precise monitoring and measurement system for low-temperature gas-liquid two-phase flow, characterized in that, It includes a segmented sensing and acquisition module, a high-speed signal analysis module, a pulsating resonance identification module, an active damping control module, a phase change intelligent identification module, and a monitoring and measurement output module. The segmented sensing and acquisition module deploys multiple sets of separate fluid dynamic response sensors in the low-temperature two-phase flow pipeline, located at the two-phase mixing inlet section, intermediate stable section and outlet section, respectively, to achieve segmented data acquisition. The high-speed signal analysis module synchronously inputs the collected instantaneous flow velocity, pressure and temperature signals to the high-speed signal processing module, and uses the spectrum analysis method to extract fluid pulsation characteristic parameters; The pulsating resonance identification module establishes a fluid pulsating resonance model and uses the model to calculate and determine whether a fluid excitation frequency close to the natural frequency of the LT-TFM system has occurred. The active damping control module, once a potential resonance trend is detected, activates the adaptive damping module to adjust the controllable damping elements inside the flow channel structure and actively reduce the flow disturbance in the resonance frequency band. The phase change intelligent identification module, while reducing pulsation interference, uses a phase change feature identification algorithm to dynamically correct the estimated value of the two-phase ratio, thereby improving the accuracy of gas-liquid interface identification. The monitoring and metering output module ultimately outputs the corrected flow rate, phase content, and pulsation suppression status to the main control system, thereby achieving accurate monitoring and safe metering of low-temperature two-phase flow. The resonance risk assessment adopts a dynamic evaluation mechanism based on the frequency response model, and the specific steps are as follows: During signal acquisition, a short-time Fourier transform is applied to the inlet section pressure signal to extract the instantaneous dominant frequency from the time-frequency domain. The extraction formula is as follows: In the formula, It is the instantaneous main frequency, indicating the frequency at the previous moment. The frequency component with the strongest energy in the pressure fluctuations and in the adjacent time period. It is in time Pressure signals collected at all times It is the half-width of the window for short-time Fourier analysis. These are Fourier basis functions, used to transform time-domain signals to the frequency domain. It is the natural base. It is the imaginary unit. It is the angular frequency multiplied by the time, representing the signal frequency. At that time, the corresponding phase change, It's frequency. It is a time variable. It is the differential symbol; Differentiating the pressure curves within the same time window yields the instantaneous pressure change rate, expressed as follows: In the formula, It is the instantaneous rate of change of pressure. It is pressure Find the first derivative, which represents the rate at which the pressure changes with time. It is the reference time point for calculating the pressure derivative, i.e., the current sampling time; A frequency response function 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: In the formula, It is the natural frequency. It is the damping ratio. It is the predicted value of the response amplitude; Set the empirical response amplitude threshold If at any time it is satisfied If a resonant excitation risk is detected, the dynamic damping module is immediately triggered, and the current event is marked and archived for subsequent operational strategy adjustments.

2. The gas-liquid low-temperature two-phase flow precise monitoring and measurement system according to claim 1, characterized in that, The deployed set of separate fluid dynamic response sensors includes miniature piezoelectric pressure sensors, thermal velocity 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 20mm intervals 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. At the outlet section, sensors are distributed at an angle in the annular monitoring cavity to collect the final state of the fluid before it leaves the pipe.

3. The precise monitoring and measurement system for low-temperature gas-liquid two-phase flow according to claim 1, 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 are triggered to sample synchronously, with a data sampling frequency of 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 and extracts the top three main frequency components and their amplitudes to analyze the dominant frequency range of fluid disturbance. The redundant filter adopts a bandpass filter structure and adjusts the center frequency according to the main frequency before processing to make the processed signal targeted and avoid background noise interference in analysis and judgment.

4. The precise monitoring and measurement system for low-temperature gas-liquid two-phase flow according to claim 1, characterized in that, The fluid pulsating resonance model is based on the fusion modeling of structural dynamic response analysis and real-time fluid disturbance characteristics; A dynamic set of input parameters is established by collecting instantaneous flow rate change, pressure fluctuation amplitude, and fluid composition change trends from sensors. The fluid pulsating resonance model considers the physical characteristics of the pipe structure and uses numerical analysis to generate the inherent response characteristics. The degree of matching between the perturbation frequency and the structural characteristic frequency is used to comprehensively assess whether structural resonance is induced.

5. The precise monitoring and measurement system for low-temperature gas-liquid two-phase flow according to claim 1, characterized in that, The adaptive damping module uses a liquid adjustable damping element, which is dynamically adjusted in terms of damping characteristics by controlling the frequency change of its internal throttling structure with a piezoelectric actuator. The module uses a high-speed signal as the driving signal and, combined with the main disturbance frequency provided by the resonance model, automatically adjusts the damping range to match the corresponding disturbance band. 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, with an adjustment delay of no more than 20ms, to ensure that the fluid disturbance is suppressed in the initial stage and avoid entering the resonance state.

6. The precise monitoring and measurement system for low-temperature gas-liquid two-phase flow according to claim 1, characterized in that, The phase transition feature recognition algorithm uses an improved fuzzy support vector machine to fuse temperature gradient change rate and pressure hysteresis response features to identify and classify gas-liquid boundaries; A multidimensional recognition feature space is constructed using training sample data, and a classification model is established using the first derivative of temperature, the second derivative of pressure, and the amplitude of flow velocity pulsation as input factors. Once the actual operational data enters the classification process, it is mapped to the constructed feature space, and the predicted gas-liquid phase ratio is output through SVM.

7. The precise monitoring and measurement system for low-temperature gas-liquid two-phase flow according to claim 1, characterized in that, The main control system includes a human-machine interface, redundant data acquisition cards, and a dual-redundant diagnostic module for security protection; All measurement data and flow state parameters are archived in a unified manner and displayed in real time in a graphical visualization interface; It has a historical record backtracking function, allowing users to select any time period to query pulsation characteristics, gas-liquid ratio, and flow rate change trends. The main control system automatically triggers an emergency stop command after detecting a high-risk frequency that persists for more than 3 seconds, disconnecting the main valve controller to prevent damage.

8. The precise monitoring and measurement system for low-temperature gas-liquid two-phase flow according to claim 1, characterized in that, The estimation of the gas-liquid phase ratio and the correction of the volumetric flow rate are performed using a multi-parameter coupled joint modeling calculation. The specific steps are as follows: Flow velocity data were collected at three locations along the pipeline. The overall average velocity is calculated using the variable weighted average method, and the result is expressed as follows: In the formula, It is the average phase velocity. These represent the local instantaneous flow velocities collected at three locations along the pipe. respectively Weighting coefficients; The current gas phase mole fraction is estimated using an ideal gas state correction model combined with compressibility factor and density ratio, and the calculation expression is as follows: In the formula, It is the gas phase mole fraction. It is the absolute pressure at the measuring point. It is the absolute temperature at the measurement point. It is the universal gas constant. and Both are gas-phase and liquid-phase compressibility factors. and These are the densities of the gas phase and the liquid phase under the current temperature and pressure conditions, respectively. Introducing an empirical correction factor, the corrected volumetric flow rate for gas-liquid coupling is calculated using the following expression: In the formula, This is the corrected effective volumetric flow rate. It is the cross-sectional area of ​​the pipe. It is a gas phase interference correction factor; Continuous sliding calculation If the standard deviation within the five-second time window is less than the threshold... If the current data is not exceeded, it will be marked as a valid measurement result and output to the main control interface; if it exceeds the limit, the refresh will be delayed and a prompt will be displayed indicating that a re-evaluation is needed.

Images