Resistance-capacitance dual-purpose ERT sensor LVF measurement method and system

Abstract

The invention discloses a resistance-capacitance dual-purpose ERT sensor LVF measurement method and system, and the method comprises the steps: employing an ERT sensor, obtaining electrical impedance data through measurement in an ERT mode, and reflecting the conductivity distribution of a fluid; in the capacitance mode, dielectric constant distribution is measured and obtained. And carrying out normalization processing on the electrical impedance and the capacitance value. And constructing a multi-input single-output MISO moisture content calculation model based on electrical impedance and capacitance measurement values, and fitting and solving model parameters through experimental data. The model is used for calculating the moisture content in the gas-water two-phase flow according to actual measurement data. Through the addition of the capacitance measurement module, the dual detection capability of non-conductive and conductive media is realized, and the accuracy of moisture content measurement is improved. A statistical analysis technology is adopted, a combination strategy of measured values is optimized, an MISO model is constructed, and error minimization between the measured values and actual values is achieved. The method is high in adaptability, can be quickly applied to industrial sites, and provides a stable and efficient solution for monitoring the water content of the gas-water two-phase flow in real time.

Description

Technical Field

[0001] The invention relates to the measurement of volumetric water content (LVF) of multiphase flow, and in particular to a LVF measurement method and system of an ERT sensor with both resistance and capacitance. Background Art

[0002] Multiphase flow is a composite fluid system containing two or more different phases. When substances with different physical and chemical properties mix and flow, multiphase flow occurs. Multiphase flow is common in natural observations and industrial production, so the study of its properties helps to better understand the laws of nature and provide technical support for related industrial fields.

[0003] Accurate determination of phase content is a core and challenging task in the field of multiphase flow, which is crucial for calculating the flow rate of each phase. In the process of energy extraction, storage and transportation, accurate measurement of water content is a key factor to ensure production safety and improve production efficiency.

[0004] The working principle of electrical tomography (ET) is to identify the differences in dielectric properties and conductivity properties of various media. Electrical resistance tomography (ERT) technology is based on the resistance-sensitive mechanism to measure the distribution information of multiphase flow, and uses the impedance measurement method to reconstruct the distribution of each phase in the fluid. Electrical capacitance tomography (ECT) technology is similar to ERT technology. It can not only measure the water content, but also reconstruct the medium distribution inside the pipeline.

[0005] ECT sensors are mainly used for the detection of non-conductive media, such as oil-water two-phase flow with oil as the continuous phase or the measurement of parameters related to wet conditions with gas as the continuous phase. ERT sensors are mainly used for the measurement of fluids with conductive continuous phases, such as high-water multiphase flow and the measurement of phase content when water is the continuous phase.

[0006] In the context of gas-water two-phase flow research, when using electrical resistance tomography (ERT) technology to measure volumetric water fraction (LVF), the medium distribution of the gas-liquid two-phase flow is reflected by measuring the change in medium conductivity. However, when the water in the pipeline exists in large quantities in the form of suspended droplets, these dispersed water droplets lack a continuous conductive path, resulting in the inability of the current to effectively penetrate, thereby interfering with the measurement of conductivity. Therefore, in the real multiphase flow water fraction measurement, it is often only applicable to the case of high water content.

[0007] The electrical capacitance tomography (ECT) technology is mainly used for the measurement of non-conductive media. When the water content of multiphase flow is high, the ECT sensor measurement value will be affected by the dielectric constant and conductivity at the same time due to the influence of water conductivity, and may even be in an inoperative state. Therefore, relying solely on ECT or ERT technology is not enough to achieve high-precision, full-range detection of gas-water two-phase fluids in the entire water content range.

[0008] In addition, traditional mathematical models for calculating water content based on electrode measurement values, such as the Maxwell model, Bruggeman model, and Ramu-Rao model, have certain limitations in practical applications, and their applicable scenarios are strictly restricted. These models are usually derived based on idealized assumptions, which are difficult to fully meet in complex real-world environments, thus affecting the universal applicability and prediction accuracy of the models. Therefore, when using traditional mathematical models to estimate water content, it is difficult to ensure the validity and reliability of the results.

[0009] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0010] The main purpose of the present invention is to solve the problems existing in the above-mentioned background technology and to provide a LVF measurement method and system of an ERT sensor with both resistance and capacitance.

[0011] To achieve the above object, the present invention adopts the following technical solutions:

[0012] A method for measuring LVF of an ERT sensor with dual functions of resistance and capacitance, comprising the following steps:

[0013] S1. Using the electrodes arranged on the pipe wall by the ERT sensor, in the ERT mode, the voltage on the electrode pair is measured through current circulation excitation, and the electrical impedance data is obtained according to the voltage to obtain the conductivity distribution of the fluid;

[0014] In the capacitive working mode, the ERT sensor applies a high-frequency, low-amplitude AC voltage signal to the inside of the medium to be tested, and measures the capacitance value on the electrode pair of the ERT sensor to obtain the dielectric constant distribution of the fluid;

[0015] S2. Normalize the electrical impedance and capacitance values ​​and convert them into dimensionless relative values;

[0016] S3. Using the measured values ​​of electrical impedance and capacitance, a multi-input single-output MISO moisture content calculation model that reflects the functional relationship between the electrical impedance and capacitance values ​​and the moisture content is constructed, and the model parameters are solved by fitting the experimental data;

[0017] S4. Apply the water content calculation model to actual measurement data to calculate the water content in the measured gas-water two-phase flow.

[0018] Furthermore, the moisture content calculation model of the MISO is constructed based on the Maxwell model, the polynomial superposition model or the exponential superposition model.

[0019] The LVF calculation model based on Maxwell is:

[0020]

[0021] The LVF calculation model based on polynomial superposition is:

[0022]

[0023] The LVF calculation model based on exponential function superposition is:

[0024]

[0025] Where α=(α 1 , α 2 , ..., α N ), β=(β 1 , β 2 , ..., β N ) and c are the parameters to be fitted, c is used to eliminate the influence of possible constant noise, x = (x 1 , x 2 , ..., x N ) is the input vector containing the normalized measurements of N channels.

[0026] Furthermore, by using different electrode combination strategies, the voltage measurement values ​​or capacitance measurement values ​​on the electrode pairs are combined in different forms and used as input variables to construct a multi-input single-output MISO moisture content calculation model to fit the moisture content.

[0027] Furthermore, in the resistance mode, three different electrode combination strategies, namely adjacent electrode pairs (Adjacent), electrode pairs separated by 1 (Apart 1) and electrode pairs separated by 2 (Apart 2), are used to obtain voltage measurement values; in the capacitance mode, four different electrode combination strategies, namely adjacent electrode pairs (Adjacent), electrode pairs separated by 1 (Apart 1), electrode pairs separated by 2 (Apart 2) and electrode pairs separated by 3 (Apart 3), are used to obtain capacitance measurement values; wherein Apart 1, Apart 2 and Apart 3 refer to that the number of electrodes separating the electrode pairs is 1, 2 and 3, respectively.

[0028] Furthermore, the eight terminals of the data acquisition unit are simultaneously connected to the eight electrodes of the ERT sensor. In the ERT mode, the two terminals connected to the excitation electrodes simultaneously send current signals with the same amplitude and frequency but opposite phases to the measured area, thereby forming a closed excitation loop and collecting 20 independent voltage or impedance measurement values. In the ECT mode, the terminals connected to the excitation electrodes apply high-frequency, low-amplitude AC voltage signals to the inside of the medium to be measured, forming a series of electric field excitation cycles. In each excitation cycle, 28 independent capacitance values ​​between different plates are collected.

[0029] Furthermore, the collected data is preprocessed as follows:

[0030] According to the starting and ending time points after each operating condition reaches a stable state, the valid data points are cut out from the original data, where the starting time point of the stable state is defined as the moment of the first sampling of each operating condition; the resampling period is set, and the valid data is averaged in the time dimension. The resampled frame of data corresponds to the average value of the output of each channel electrode within the resampling period under the stable condition.

[0031] A resistance and capacitance dual-purpose ERT sensor LVF measurement system, comprising an ERT sensor unit, a data acquisition unit and a data processing unit;

[0032] The ERT sensor unit includes electrodes arranged on the pipe wall. In the ERT mode, the voltage on the electrode pair is measured by the voltage measurement module of the data acquisition unit through current circulation excitation, and electrical impedance data is obtained according to the voltage to obtain the conductivity distribution of the fluid;

[0033] In the capacitive working mode, a high-frequency, low-amplitude AC voltage signal is applied to the inside of the medium to be measured by the ERT sensor unit, and the capacitance measurement module of the data acquisition unit measures the capacitance value on the electrode pair of the ERT sensor to obtain the dielectric constant distribution of the fluid;

[0034] The data processing unit normalizes the electrical impedance and capacitance values ​​and converts them into dimensionless relative values; uses the electrical impedance and capacitance measurement values ​​to construct a multi-input single-output MISO moisture content calculation model that reflects the functional relationship between the electrical impedance and capacitance values ​​and the moisture content, and solves the model parameters by fitting the experimental data; the moisture content calculation model is used for actual measurement data to calculate the moisture content in the measured gas-water two-phase flow.

[0035] Furthermore, it also includes: a temperature controller for accurately controlling the temperature of the fluid in the pipeline; a static mixer for enhancing the uniform mixing of the gas-water mixture during the flow process upstream of the sensor; a circulation pump for providing power for the circulation of the gas-water two-phase flow in the pipeline; a single-phase flow meter located downstream of the control valve for real-time monitoring of parameters such as the temperature, mass flow rate and volume flow rate of the fluid in the pipeline.

[0036] The present invention has the following beneficial effects:

[0037] The present invention proposes a method and system for measuring water content of an ERT sensor with both resistance and capacitance. The system adds a capacitance measurement module to a data acquisition unit without changing the structure of the ERT sensor, so that more fluid information can be obtained using the original sensor. The system is easy to install and has low cost. When the gas phase is used as the continuous phase, the water content is calculated by means of the capacitance measurement value, and in other cases, the water content is calculated by the voltage measurement value. This method combines the respective advantages of ERT and ECT, realizes the full-range measurement of the water content of the gas-water two-phase flow, and improves the measurement accuracy.

[0038] Based on the analysis of the variation law of voltage and capacitance measurement values ​​with water content, the present invention provides the effective working range of resistance and capacitance measurement modes. Aiming at the problem of large error in water content calculation of traditional mathematical models, the present invention optimizes the water content LVF calculation model from the algorithm level by analyzing the distribution characteristics of normalized measurement values, and adopts statistical analysis methods to combine electrode measurement values ​​(voltage or capacitance) in different forms and use them as input variables, construct a multiple inputs and single output (MISO) function model, and fit the water content, thereby minimizing the error between the LVF calculation value and the reference value. The scheme is easy to implement, and has high flexibility and practicality, and can achieve a rapid transition from a laboratory environment to a field production scene, providing a reliable solution for monitoring the water content of gas-water two-phase flow.

[0039] In summary, the present invention combines ERT sensing with capacitance measurement, proposes a dual-purpose capacitance and resistance moisture content detection method and system, breaks the measurement range limitation of traditional ERT sensors, and uses electrical impedance and capacitance measurement values ​​to construct a multi-input single-output MISO moisture content calculation model that reflects the functional relationship between electrical impedance and capacitance values ​​and moisture content. Based on the improved LVF calculation model, the precision and accuracy of moisture content measurement are improved.

[0040] The core of the present invention is to improve the accuracy and reliability of water content measurement by integrating the technical advantages of ERT and ECT, while expanding the application scope of measurement technology. Compared with traditional technologies, the main advantages of the present invention are:

[0041] (1) While maintaining the original structure of the ERT sensor, the sensor is integrated with capacitance measurement to achieve dual detection capabilities for non-conductive and conductive media. Therefore, in gas-water two-phase flow conditions, accurate water content measurement results can be obtained regardless of whether the continuous phase is gas or water.

[0042] (2) By analyzing the changes in resistance and capacitance measurement values ​​with water content, the optimal working range of the resistance and capacitance modes can be determined. Using statistical analysis technology, the voltage and capacitance measurement values ​​are input into the MISO model with the optimal combination strategy to minimize the error between the water content measurement value and the actual value, thereby greatly improving the measurement accuracy.

[0043] The technical solution of the present invention has excellent field adaptability and can be quickly transitioned to an industrial production environment, providing a stable and efficient solution for real-time monitoring of the water content of a complex multiphase fluid system.

[0044] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of a gas-water two-phase flow measurement pipeline according to an embodiment of the present invention.

[0046] Figure 2 Schematic diagram of an ERT system according to an embodiment of the present invention.

[0047] Figure 3 Schematic diagram of data collection principle of an embodiment of the present invention.

[0048] Figure 4 1 is a distribution diagram of the resistance mode measurement values ​​of the ERT sensor according to an embodiment of the present invention, (a) normalized measurement values ​​of all channels; (b) normalized measurement values ​​after averaging using different combination strategies.

[0049] Figure 5 1. Distribution of capacitance mode measurement values ​​of the ERT sensor according to an embodiment of the present invention: (a) normalized measurement values ​​of all channels; (b) normalized measurement values ​​after averaging using different combination strategies.

[0050] Figure 6 LVF calculation results of the ERT based on the optimization model in the resistance mode of the embodiment of the present invention.

[0051] Figure 7 It is the relative error of the ERT based on the optimization model in the resistance mode of the embodiment of the present invention.

[0052] Figure 8 LVF calculation result of the ERT based on the optimization model in the capacitive mode according to the embodiment of the present invention.

[0053] Fig. 9 Graph 1 is the relative error of the ERT based on the optimization model in the capacitive mode according to the embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0055] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0056] See also Figure 1 The embodiment of the present invention provides a method for measuring LVF of an ERT sensor with both resistance and capacitance, comprising the following steps:

[0057] S1. Using the electrodes arranged on the pipe wall by the ERT sensor, in the ERT mode, the voltage on the electrode pair is measured through current circulation excitation, and the electrical impedance data is obtained according to the voltage to obtain the conductivity distribution of the fluid;

[0058] In the capacitive working mode, the ERT sensor applies a high-frequency, low-amplitude AC voltage signal to the inside of the medium to be tested, and measures the capacitance value on the electrode pair of the ERT sensor to obtain the dielectric constant distribution of the fluid;

[0059] S2. Normalize the electrical impedance and capacitance values ​​and convert them into dimensionless relative values;

[0060] S3. Using the measured values ​​of electrical impedance and capacitance, a multi-input single-output MISO moisture content calculation model that reflects the functional relationship between the electrical impedance and capacitance values ​​and the moisture content is constructed, and the model parameters are solved by fitting the experimental data;

[0061] S4. Apply the water content calculation model to actual measurement data to calculate the water content in the measured gas-water two-phase flow.

[0062] In some embodiments, the moisture content calculation model of the MISO is constructed based on the Maxwell model, the polynomial superposition model or the exponential superposition model.

[0063] In some embodiments, different electrode combination strategies are used to combine the voltage measurement values ​​or capacitance measurement values ​​on the electrode pairs in different forms and use them as input variables to construct a multi-input single-output MISO moisture content calculation model to fit the moisture content.

[0064] In a preferred embodiment, in the resistance mode, three different electrode combination strategies, namely adjacent electrode pairs (Adjacent), electrode pairs separated by 1 (Apart 1) and electrode pairs separated by 2 (Apart 2), are used to obtain voltage measurement values; in the capacitance mode, four different electrode combination strategies, namely adjacent electrode pairs (Adjacent), electrode pairs separated by 1 (Apart 1), electrode pairs separated by 2 (Apart 2) and electrode pairs separated by 3 (Apart 3), are used to obtain capacitance measurement values; wherein Apart 1, Apart 2 and Apart 3 refer to that the number of electrodes separating the electrode pairs is 1, 2 and 3, respectively.

[0065] In one embodiment, the eight terminals of the data acquisition unit are simultaneously connected to the eight electrodes of the ERT sensor. In the ERT mode, the two terminals connected to the excitation electrodes simultaneously send current signals with the same amplitude and frequency but opposite phases to the measured area, thereby forming a closed excitation loop and collecting 20 independent voltage or impedance measurement values. In the ECT mode, the terminals connected to the excitation electrodes apply high-frequency, low-amplitude AC voltage signals to the inside of the medium to be measured, forming a series of electric field excitation cycles. In each excitation cycle, 28 independent capacitance values ​​between different plates are collected.

[0066] The embodiment of the present invention also provides a dual-purpose ERT sensor LVF measurement system for resistance and capacitance, comprising an ERT sensor unit, a data acquisition unit and a data processing unit; wherein the ERT sensor unit comprises electrodes arranged on the pipe wall, and in the ERT mode, the voltage on the electrode pair is measured by the voltage measurement module of the data acquisition unit through current circulation excitation, and the impedance data is obtained according to the voltage to obtain the conductivity distribution of the fluid; in the capacitance working mode, a high-frequency, low-amplitude AC voltage signal is applied to the inside of the medium to be measured by the ERT sensor unit, and the capacitance value on the electrode pair of the ERT sensor is measured by the capacitance measurement module of the data acquisition unit to obtain the dielectric constant distribution of the fluid; the data processing unit normalizes the impedance and capacitance values ​​and converts them into dimensionless relative values; the impedance and capacitance measurement values ​​are used to construct a multi-input single-output MISO water content calculation model reflecting the functional relationship between the impedance and capacitance values ​​and the water content, and the model parameters are solved by fitting the experimental data; the water content calculation model is used for actual measurement data to calculate the water content in the measured gas-water two-phase flow.

[0067] The LVF measurement method and system of the dual-purpose ERT sensor for resistance and capacitance provided by the present invention expand the function of the ERT sensor to capacitance measurement without making any changes to the structure of the existing ERT sensor. This improvement enables the sensor to adapt to a wider range of application scenarios, especially in the measurement of water content in gas-water two-phase flow. By integrating the capacitance measurement module in the data acquisition unit, the present invention can achieve accurate measurement of the dielectric constant of the fluid at a low cost and a simple installation process, thereby providing reliable water content data when the gas phase or liquid phase is used as the continuous phase.

[0068] At the same time, the present invention can determine the optimal working range under different measurement modes by analyzing the variation of the impedance and capacitance measurement values ​​with the water content. Using statistical analysis methods, the present invention further optimizes the calculation model of the water content, and significantly improves the accuracy and reliability of the water content measurement by constructing a multi-input single output (MISO) function model. The improved model can effectively combine the measurement values ​​from different electrode pairs and use them as input variables to minimize the error between the LVF calculation value and the reference value.

[0069] Another significant advantage of the present invention is its excellent field adaptability. It can quickly transition from a laboratory environment to a field production environment, providing a stable and efficient solution for real-time monitoring of the water content of complex multiphase fluid systems. This flexibility and practicality make the present invention highly valuable in practical applications. In short, by integrating the technical advantages of ERT and ECT, the present invention not only improves the accuracy and reliability of water content measurement, but also expands the application scope of the measurement technology, providing strong technical support for related industrial fields.

[0070] The specific embodiments of the present invention are further described below.

[0071] Design of LVF Measurement System of ERT Sensor with Capacitance and Resistance

[0072] Figure 1 The pipes, sensors and other equipment included in the measurement system are presented. In the experimental setup, the gas phase and liquid phase are generated by an air compressor and a water pump respectively. When flowing through a specific test pipe section, the ERT electrode measurement value is collected by the data acquisition board and uploaded to the computer, and finally flows out from the outlet. In the entire experimental device, the temperature controller is used to accurately control the temperature of the fluid in the pipeline; in order to enhance the uniform mixing of the gas-water mixture during the flow process, the experiment is equipped with a static mixer upstream of the sensor; the circulation pump is used to provide power for the circulation of the gas-water two-phase flow in the pipeline; the single-phase flowmeter located downstream of the control valve is used to monitor the key parameters such as the temperature, mass flow rate and volume flow rate of the fluid in the pipeline in real time, providing a solid foundation for experimental data analysis.

[0073] ERT system such as Figure 2 As shown in the figure, it mainly consists of three parts: ERT sensor unit, data acquisition unit and data processing unit. The electrodes are evenly attached to the inner wall of the pipe, and the electrode measurement values ​​reflect the dynamic characteristics of multiphase fluid flow and the instantaneous changes of process parameters on the cross section of the pipe. The data acquisition board is one of the key components of the system and has dual responsibilities: on the one hand, it provides the necessary excitation signal to the sensor; on the other hand, it converts the raw data collected by the sensor into digital signals, which are then transmitted to the data processing unit for the calculation of process parameters such as water content.

[0074] Data collection principle Figure 3 As shown. The core component of the data acquisition unit is a data acquisition board based on AD conversion. The 8 terminals of the board are simultaneously connected to the 8 electrodes of the ERT sensor. When the system works in ERT mode, the two terminals connected to the excitation electrodes simultaneously send current signals with the same amplitude and frequency and opposite phases to the measured area, thereby forming a closed excitation loop. At the same time, the system records 20 independent voltage (or resistance) measurements. When the system works in ECT mode, the terminals connected to the excitation electrodes apply high-frequency, low-amplitude AC voltage signals to the inside of the medium to be measured, forming a series of short electric field excitation cycles. In each excitation cycle, 28 independent capacitance values ​​between different plates are collected for subsequent data processing.

[0075] Optimization Design of Moisture Content Calculation Model

[0076] Through the analysis of the original measurement value (voltage) of the ERT sensor, we can see that as the normalized measurement value increases, the water content reference value shows a decreasing trend, and the data points are basically gathered around a curve where the dependent variable decreases as the independent variable increases, that is, there is a negative correlation between the LVF reference value and the ERT measurement value (voltage). Therefore, when constructing the LVF calculation model, consider using a monotonically decreasing function as the basis. Since the Maxwell model not only has good theoretical explanatory power, but also its function characteristics show a monotonically decreasing trend, we give priority to constructing the LVF calculation model based on the Maxwell model structure, and solve the model parameters based on the experimental data fitting:

[0077]

[0078] Where θ=(θ 1 ,θ 2 ,θ 3 ) is a parameter vector, and x is the electrode measurement value or the average value of electrode measurement data based on different combination strategies.

[0079] As for the electrode measurement values ​​in the capacitive working mode, the overall trend is that the LVF increases. Therefore, when constructing the LVF calculation model, a monotonically increasing function is considered as the basis, and the model parameters are fitted based on the experimental data:

[0080] LVF(x,θ 1 ,θ 2 ,θ 3 )=θ 1 ·x 2 +θ 2 .x+θ 3

[0081] Where θ=(θ 1 ,θ 2 ,θ 3 ) is a parameter vector, and x is the electrode measurement value or the average value of electrode measurement data based on different combination strategies.

[0082] In addition, based on further analysis of the experimental data, it can be seen that the normalized measurement values ​​of some channels have no obvious functional mapping relationship with the reference LVF. The possible reason for this phenomenon is that the LVF weights "observed" by the measurement channels of the ERT electrode at different positions will change with the changes in LVF and the gas-water mixing conditions. Therefore, it is considered to use multi-channel data under different combination strategies as input to construct a water content calculation model in the form of MISO, and to construct an LVF calculation model with the Maxwell model, polynomial superposition model and exponential superposition model as the basic structure.

[0083] LVF calculation model based on Maxwell:

[0084]

[0085] LVF calculation model based on polynomial superposition:

[0086]

[0087] LVF calculation model based on exponential function superposition:

[0088]

[0089] Where α=(α 1 , α 2 , ..., α N ), β=(β 1 , β 2 , ..., β N ) and c are the parameters to be fitted, and c is used to eliminate the influence of possible constant noise. 1 , x 2 , …, x N) is an input vector containing normalized measurement values ​​of N channels. In order to avoid inconsistency with actual calculation results, when the calculated value is greater than 1 or less than 0, the output of the model will be limited to 1 and 0 respectively.

[0090] All three models have unique mathematical properties and can adapt to complex and changing measurement environments. This solution not only fully considers the limitations of ERT sensors in multiphase flow measurement, but also enhances the model's predictive ability and adaptability to complex working conditions by introducing a multivariate data fusion strategy, providing a more accurate calculation model for accurately estimating water content.

[0091] Examples

[0092] A. Dynamic Experimental Process

[0093] The operating flow settings for the online gas-water two-phase flow detection experiment are shown in Table 1.

[0094] Table 1 Air-water two-phase flow dynamic experimental settings

[0095]

[0096] During the experiment, the temperature of the fluid was controlled at about 24.5°C. The experimental water was local tap water with a conductivity of 0.01S / at room temperature and a density of 996.5kg / m 3 Different LVFs can be obtained by changing the volume ratio of gas and water added to the pipeline. The LVF obtained in this way is the proportional LVF. The LVF obtained by the single-phase flow meter during the experiment is the reference LVF, which is calculated as follows:

[0097]

[0098]

[0099] Among them, Q ng，meter is the gas phase volume flow rate under standard conditions, which needs to be further converted into the gas phase volume flow rate under working conditions Q g，meter , T is the working temperature, in degrees Celsius, P is the working pressure, P atm is the actual atmospheric pressure, the unit is KPa, Q l，meter It is the liquid phase volume flow rate measured by the single-phase flow meter.

[0100] The complete test process is as follows:

[0101] 1) First, start the air compressor to establish a certain amount of static pressure in the entire system. At the same time, turn on the temperature control device and set the target temperature to 24.5℃, waiting for the temperature to reach the set value.

[0102] 2) Open the upper computer software and record the electrode capacitance measurement data in the empty pipe state in the pipeline for low calibration of capacitance value in empty pipe state.

[0103] 3) Open the water valve and start the water pump. Stop the water pump when the fluid in the pipeline is completely filled with water. Collect the ERT voltage measurement value of the full water condition in a static state and calibrate it with the voltage value of full water low.

[0104] 4) According to the preset liquid and gas volume flow rates in the dynamic experimental scheme, adjust the output frequency of the frequency regulator connected to the circulation pump, and adjust the flow rates of each phase in the pipeline by changing the speed of the circulation pump. In order to simplify the experimental operation, the host computer software is always in the data recording state. The original data contains the data sampling time (accurate to seconds). During the experiment, the start and end time nodes of each operating point after stabilization are recorded to facilitate the subsequent interception of effective measurement data under each operating condition. Adjust to the next operating point and repeat step 4.

[0105] B. Data preprocessing and calculation process

[0106] In order to improve the experimental efficiency, a continuous sampling method is used to collect data during the experiment. This method will cause valid data to be mixed with a large amount of redundant data, so it is necessary to filter out valid data from the original data.

[0107] First, according to the starting and ending time points after each operating condition reaches a stable state, the valid data points are extracted from the original data. The starting time point of the stable state is defined as the moment when each operating condition is first sampled. This can effectively eliminate a large amount of redundant and invalid data recorded during the experiment.

[0108] Secondly, in order to reduce the impact of data fluctuations on subsequent calculations, based on data rescreening, the valid data will be averaged in the time dimension with a resampling period of 2 minutes. A frame of data after resampling corresponds to the average value of the electrode output of each channel within two minutes under stable working conditions. Subsequent data analysis and LVF calculation are based on this data.

[0109] The electrode measurement values ​​under different combination strategies are used as the input features of the model. The LVF obtained by single-phase flow meter detection is used as the reference required for the model construction process. Three types of MISO calculation models based on polynomial superposition, exponential superposition and Maxwell superposition are used to compare and analyze the errors between the LVF calculated values ​​and the reference values.

[0110] Experimental results analysis

[0111] A. Distribution characteristics of normalized measurements

[0112] In order to select a suitable moisture content calculation model, it is first necessary to observe and analyze the normalized electrode measurement data.

[0113] Figure 4 The trend of the normalized measurement value of ERT in resistance working mode changing with the LVF reference value is shown. Figure 4 Each data point in (a) corresponds to the normalized measurement value of a channel, Figure 4 Each data point in (b) is the average value of the normalized measurement value of the corresponding channel under a specific combination strategy. For example, each data point corresponding to the Adjacent strategy represents the average value of the normalized measurement value of the output channel of 8 adjacent electrode pairs. As can be seen from the figure, under the three combination strategies, as the LVF reference value increases, the normalized measurement value shows a downward trend and is basically gathered around a decreasing smooth curve. Therefore, in the ERT resistance working mode, an LVF calculation model based on a monotonically decreasing function is a better choice.

[0114] Figure 5 The trend of the normalized measurement value of ERT in capacitive working mode changing with the LVF reference value is shown. Figure 5 Each data point in (a) corresponds to the normalized measurement value of a channel, Figure 5 Each data point in (b) is the average value of the normalized measurement value of the corresponding channel under a specific combination strategy. It can be seen from the figure that under the four combination strategies, as the LVF reference value increases, the normalized measurement value shows a downward trend. Therefore, for the capacitive working mode, the LVF calculation model based on the monotonically increasing function is given priority. In addition, by comparing the normalized measurement values ​​of the four different combination strategies, it can be seen that the increasing trend of the normalized measurement value distribution of the Adjacent and Apart 3 electrode pairs is not obvious, and the correlation with the LVF reference value is weak, making it difficult to construct a model for characterizing the functional relationship between the electrode measurement value and the LVF reference value.

[0115] B. Analysis of LVF calculation model results based on MISO function form

[0116] Figure 6 and Figure 7 The LVF calculated values ​​obtained by the optimized MISO model based on three electrode combination strategies (adjacent electrode pair "Adjacent" combination strategy, every other electrode pair "Apart 1" combination strategy, every other electrode pair "Apart 2" combination strategy) in the resistance working mode of ERT and their relative errors with the LVF reference values ​​are presented respectively.

[0117] Figure 6The calculated LVF values ​​and corresponding reference values ​​based on different electrode combination strategies and optimization models are presented. The red dotted line is the ±10% relative error band, and the red solid line is the ideal case where the calculated value is equal to the reference value.

[0118] Figure 7 The relative errors of LVF based on different electrode combination strategies and optimization models are presented. Figure 6 and Figure 7 It can be seen that the calculation accuracy of LVF increases with the increase of LVF. In the non-humid part (LVF>5%), the relative errors of the calculated values ​​are all controlled within the ±10% error band, and the relative errors of most data are between -5% and 5%. The ERT resistance working mode has reliable working performance in non-humid gas-liquid two-phase flow detection. In the wet range (LVF≤5%), the LVF calculation accuracy is significantly reduced. This phenomenon confirms the defect that when the water content of the pipeline is low, ERT is difficult to achieve high-precision measurement due to the lack of a continuous conductive path. Therefore, it is necessary to introduce a capacitive ERT in the wet range (LVF≤5%) to achieve full-range measurement of gas-liquid two-phase flow.

[0119] Figure 8 and Fig. 9 The LVF calculated values ​​obtained by the optimized MISO model based on four electrode combination strategies (adjacent electrode pair "Adjacent" combination strategy, every other electrode pair "Apart 1" combination strategy, every other electrode pair "Apart 2" combination strategy, every other electrode pair "Apart 3" combination strategy) in ERT capacitive working mode and their relative errors with the LVF reference values ​​are presented respectively.

[0120] Figure 8 The calculated LVF values ​​and corresponding reference values ​​based on different electrode combination strategies and optimization models are presented. The red dotted line is the ±10% relative error band, and the red solid line is the ideal case where the calculated value is equal to the reference value.

[0121] Fig. 9 The relative errors of LVF based on different electrode combination strategies and optimization models are presented. Figure 8 and Fig. 9It can be seen that the ERT sensor working in the capacitive mode can achieve accurate tracking of LVF within the humid range through a specific electrode combination strategy and an optimized calculation model. Among the three types of LVF calculation models, the LVF calculation model based on polynomial superposition can achieve relatively accurate and stable performance under the four electrode combination strategies, while the optimization models based on exponential superposition and modified Maxwell superposition have poor calculation effects under certain electrode combination strategies. For example, when the modified Maxwell superposition model is applied to adjacent electrode pairs and the exponential superposition model is applied to every other electrode pair, the calculated value completely deviates from the reference value. Among the four types of electrode combination strategies, the model can obtain better calculation accuracy than other combination strategies by optimizing the measurement values ​​based on every other two electrode pairs.

[0122] C. Comparison of evaluation indicators based on different electrode combination strategies and LVF calculation models

[0123] In order to quantitatively analyze and compare the performance of the optimized LVF calculation model, this paper introduces two evaluation indicators, Mean Absolute Error (MAE) and the confidence probability that the prediction result falls within the 0.5% absolute error band (Acc_0.5), to evaluate the performance of the model. The calculation formula of the evaluation indicator is as follows:

[0124]

[0125]

[0126] Where I{·} is a judgment function, and its value is 1 when the inequality in the brackets holds, otherwise it is 0.

[0127] Table 2 shows the performance indicators of the MISO calculation model on the data collected in the ERT resistance working mode within the non-humidity range (LVF>5%). The three columns "Adjacent", "Apart 1" and "Apart 2" represent the electrode combination strategies used in the calculation process. In order to verify the effect of the MISO model on improving the LVF calculation accuracy, the table also gives the results of the Maxwell model on the same test data.

[0128] Table 2 Statistics of calculation results of ERT resistance working mode in non-humidity range

[0129]

[0130] The MISO model has a significant effect on improving the LVF calculation accuracy in the non-humid gas range, especially the optimization model based on polynomial superposition, which can control more than 85% of the calculation results within the absolute error range of 0.5%. Among the three electrode combination strategies, the every-other-electrode combination strategy also has a slight advantage in calculation accuracy compared with the other two strategies.

[0131] Table 3 shows the performance indicators of the MISO calculation model on the data collected in the ERT capacitor working mode in the non-humidity range (LVF≤5%). The three columns "Adjacent", "Apart 1", "Apart 2" and "Apart3" respectively represent the electrode combination strategies used in the calculation process. Similarly, in order to compare the LVF calculation accuracy of the MISO model and the Maxwell model, the table also gives the results of the Maxwell model on the same test data.

[0132] Table 3 Statistics of calculation results of ECT resistance working mode in humidity range

[0133]

[0134] Different from the non-humidity range, the error of the calculation results of the traditional Maxwell model is small after correction, and the corrected results can be controlled within the absolute error range of 0.5%. The calculation accuracy of the optimization model based on polynomial superposition is significantly better than the other two optimization models and the Maxwell model, and the follow-up of the LVF reference value is relatively stable. Among the four electrode combination strategies, the combination strategies of 2 electrode pairs and 3 electrode pairs are higher in accuracy and stability.

[0135] The above content is a further detailed description of the present invention in combination with specific / preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, without departing from the concept of the present invention, it can also make several substitutions or modifications to these described embodiments, and these substitutions or modifications should be regarded as belonging to the protection scope of the present invention. In the description of this specification, the description of the reference terms "an embodiment", "some embodiments", "preferred embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In the absence of mutual contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A method for measuring LVF of an ERT sensor with dual functions of resistance and capacitance, characterized in that: The steps include: S1. Using the electrodes arranged on the pipe wall by the ERT sensor, in the ERT mode, the voltage on the electrode pair is measured through current circulation excitation, and the electrical impedance data is obtained according to the voltage to obtain the conductivity distribution of the fluid; In the capacitive working mode, the ERT sensor applies a high-frequency, low-amplitude AC voltage signal to the inside of the medium to be tested, and measures the capacitance value on the electrode pair of the ERT sensor to obtain the dielectric constant distribution of the fluid; S2. Normalize the electrical impedance and capacitance values ​​and convert them into dimensionless relative values; S3. Using the measured values ​​of electrical impedance and capacitance, a multi-input single-output MISO moisture content calculation model that reflects the functional relationship between the electrical impedance and capacitance values ​​and the moisture content is constructed, and the model parameters are solved by fitting the experimental data; S4. Apply the water content calculation model to actual measurement data to calculate the water content in the measured gas-water two-phase flow.

2. The LVF measurement method of the ERT sensor with dual resistance and capacitance as claimed in claim 1, characterized in that: The moisture content calculation model of MISO is constructed based on the Maxwell model, the polynomial superposition model or the exponential superposition model.

3. The LVF measurement method of the ERT sensor with dual resistance and capacitance as claimed in claim 2, characterized in that: The LVF calculation model based on Maxwell is: The LVF calculation model based on polynomial superposition is: The LVF calculation model based on exponential function superposition is: Among them, α = (α1, α2,..., α N ), β=(β1, β2,...,β N ) and c are parameters to be fitted, c is used to eliminate the influence of possible constant noise, and is an input vector containing normalized measurement values ​​of N channels.

4. The LVF measurement method of the ERT sensor with dual resistance and capacitance as claimed in any one of claims 1 to 3, characterized in that: By using different electrode combination strategies, the voltage measurement values ​​or capacitance measurement values ​​on the electrode pairs are combined in different forms and used as input variables to construct a multi-input single-output (MISO) moisture content calculation model and fit the moisture content.

5. The LVF measurement method of the ERT sensor with dual resistance and capacitance as claimed in claim 4, characterized in that: In the resistance mode, three different electrode combination strategies, namely adjacent electrode pairs (Adjacent), electrode pairs separated by 1 (Apart 1) and electrode pairs separated by 2 (Apart 2), are used to obtain voltage measurement values; in the capacitance mode, four different electrode combination strategies, namely adjacent electrode pairs (Adjacent), electrode pairs separated by 1 (Apart 1), electrode pairs separated by 2 (Apart 2) and electrode pairs separated by 3 (Apart 3), are used to obtain capacitance measurement values; Apart 1, Apart 2 and Apart 3 refer to the number of electrodes separating the electrode pairs as 1, 2 and 3, respectively.

6. The LVF measurement method of the ERT sensor with dual resistance and capacitance as claimed in any one of claims 1 to 5, characterized in that: The eight terminals of the data acquisition unit are simultaneously connected to the eight electrodes of the ERT sensor. In the ERT mode, the two terminals connected to the excitation electrodes simultaneously send current signals with the same amplitude and frequency but opposite phases to the measured area, thereby forming a closed excitation loop and collecting 20 independent voltage or impedance measurement values. In the ECT mode, the terminals connected to the excitation electrodes apply high-frequency, low-amplitude AC voltage signals to the inside of the medium to be measured, forming a series of electric field excitation cycles. In each excitation cycle, 28 independent capacitance values ​​between different plates are collected.

7. The LVF measurement method of the ERT sensor with dual resistance and capacitance according to any one of claims 1 to 6, characterized in that: Step S2 also includes the following preprocessing of the collected data: according to the starting and ending time points after each operating point reaches a stable state, the valid data points are cut out from the original data, where the starting time point of the stable state is defined as the moment of the first sampling of each operating condition; the resampling period is set, and the valid data is averaged in the time dimension, and a frame of data after resampling corresponds to the average value of the output of each channel electrode within the resampling period under the stable working condition.

8. A resistance and capacitance dual-purpose ERT sensor LVF measurement system, characterized in that: It includes an ERT sensor unit, a data acquisition unit and a data processing unit; The ERT sensor unit includes electrodes arranged on the pipe wall. In the ERT mode, the voltage on the electrode pair is measured by the voltage measurement module of the data acquisition unit through current circulation excitation, and electrical impedance data is obtained according to the voltage to obtain the conductivity distribution of the fluid; In the capacitive working mode, a high-frequency, low-amplitude AC voltage signal is applied to the inside of the medium to be measured by the ERT sensor unit, and the capacitance measurement module of the data acquisition unit measures the capacitance value on the electrode pair of the ERT sensor to obtain the dielectric constant distribution of the fluid; The data processing unit normalizes the electrical impedance and capacitance values ​​and converts them into dimensionless relative values; uses the electrical impedance and capacitance measurement values ​​to construct a multi-input single-output MISO moisture content calculation model that reflects the functional relationship between the electrical impedance and capacitance values ​​and the moisture content, and solves the model parameters by fitting the experimental data; the moisture content calculation model is used for actual measurement data to calculate the moisture content in the measured gas-water two-phase flow.

9. The resistance and capacitance dual-purpose ERT sensor LVF measurement system as claimed in claim 8, characterized in that: Also includes: The temperature controller is used to accurately control the temperature of the fluid in the pipeline; the single-phase flow meter is used to monitor the mass flow and / or volume flow of the fluid in the pipeline in real time.

10. The resistance and capacitance dual-purpose ERT sensor LVF measurement system according to claim 8 or 9, characterized in that: Also includes: Static mixers to enhance the homogeneous mixing of the air-water mixture during flow upstream of the sensor; A circulation pump is used to provide power for the circulation of gas-water two-phase flow in a pipeline.

Images