CN109298070B - Method for detecting water content of crude oil based on ultrasonic sound velocity method - Google Patents

Method for detecting water content of crude oil based on ultrasonic sound velocity method Download PDF

Info

Publication number
CN109298070B
CN109298070B CN201811101335.7A CN201811101335A CN109298070B CN 109298070 B CN109298070 B CN 109298070B CN 201811101335 A CN201811101335 A CN 201811101335A CN 109298070 B CN109298070 B CN 109298070B
Authority
CN
China
Prior art keywords
water
temperature
water content
oil
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201811101335.7A
Other languages
Chinese (zh)
Other versions
CN109298070A (en
Inventor
姜正杰
周洪亮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201811101335.7A priority Critical patent/CN109298070B/en
Publication of CN109298070A publication Critical patent/CN109298070A/en
Application granted granted Critical
Publication of CN109298070B publication Critical patent/CN109298070B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/02—Analysing fluids
    • G01N29/024—Analysing fluids by measuring propagation velocity or propagation time of acoustic waves
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
    • G01N29/449—Statistical methods not provided for in G01N29/4409, e.g. averaging, smoothing and interpolation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00—Indexing codes associated with group G01N29/00
    • G01N2291/02—Indexing codes associated with the analysed material
    • G01N2291/022—Liquids
    • G01N2291/0226—Oils, e.g. engine oils
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00—Indexing codes associated with group G01N29/00
    • G01N2291/02—Indexing codes associated with the analysed material
    • G01N2291/028—Material parameters
    • G01N2291/02845—Humidity, wetness

Landscapes

  • Physics & Mathematics (AREA)
  • Biochemistry (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Acoustics & Sound (AREA)
  • Probability & Statistics with Applications (AREA)
  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)

Abstract

The invention discloses a method for detecting the water content of crude oil based on an ultrasonic sound velocity method. The ultrasonic sound velocity method utilizes the propagation velocity of ultrasonic waves in a water-containing crude oil sample to determine the water content of the tested sample. The ultrasonic sound velocity method is greatly influenced by temperature change, so that temperature compensation is the key for improving detection accuracy. According to the method, a crude oil water content detection mechanism model containing temperature compensation is obtained through derivation; obtaining the propagation speed of ultrasonic waves in the tested sample under different water contents and different temperatures through experiments, and obtaining a preliminary water content prediction result by utilizing a mechanism model; and (4) taking the prediction result and the temperature of the mechanism model as input variables and the true value of the moisture content as output variables, and obtaining the optimized moisture content detection model by adopting a support vector regression method. The invention effectively improves the detection precision of the ultrasonic sound velocity method on the water content of the crude oil at different temperatures by the hybrid modeling method combining mechanism modeling and data-driven modeling.

Description

Method for detecting water content of crude oil based on ultrasonic sound velocity method
Technical Field
The invention belongs to the technical field of detection of water content of crude oil, and particularly relates to a method for detecting water content of crude oil based on an ultrasonic sound velocity method.
Background
In the petroleum industry, the water content of crude oil is a very important index, and the requirement is put forward on the detection of the water content of the crude oil in the processes of mining, gathering, transportation, metering, refining and the like. The water content determines the quality of the crude oil to a great extent, so that the accurate measurement of the water content in the crude oil has very important significance for the petroleum industry.
The existing methods for detecting the water content of crude oil are generally divided into two categories. The first type is an off-line measurement method, including distillation, electrodeionization, karl fischer titration, and the like. The distillation method and the electrodeionization method are based on oil-water separation, and the karl fischer titration method is based on the correlation between the amount of karl fischer reagent consumed during neutralization titration and the moisture content in a sample to be measured. The method generally has the advantage of higher measurement accuracy, but needs more complex experimental equipment and professional experimental operation, needs longer time and cannot meet the requirement of on-site rapid detection. The second type is an online detection method, which mainly comprises the following steps:
(1) the conductivity method has the advantages that in an oil-water mixture, the conductivity difference of oil and water is large, pure oil is almost non-conductive, and the water content plays a decisive role in the conductivity of the oil-water mixture. When the water is a continuous phase, the water content can be calculated through a Maxwell formula, but the water in the crude oil product with the lower water content is a discrete phase, and the method cannot be used at the moment. In addition, the method is greatly influenced by the ion concentration and the mineralization degree of the medium.
(2) The capacitance method is based on the fact that the dielectric constant difference between oil and water is large, and the dielectric constant of an oil-water mixture is very sensitive to the change of the water content, so that the water content can be measured by the equivalent capacitance when the measuring medium is the oil-water mixture. The capacitance method has a good measuring effect on an oil-water mixture with water as a discrete phase, but when the water content is high, water mostly starts to be communicated in crude oil, and the capacitance method cannot detect the water content, so that the capacitance method is only suitable for detecting low water content.
(3) The electromagnetic wave method is a non-contact measuring method, which transmits and receives electromagnetic waves with fixed energy through an oil-water mixture based on the great difference of the absorption capacities of oil and water to the electromagnetic waves, and determines the water content of the oil-water mixture by detecting the intensity of the received electromagnetic waves. The short wave method and the microwave method are branches of the method, but the short wave method has higher requirement on the temperature of the environment, cannot have larger temperature fluctuation, and can normally measure the temperature only by generating stable oscillation frequency by an oscillator, while the microwave method is more complex, needs good anti-interference equipment and has higher cost.
(4) Ray method is also a non-contact measurement method, and the basic working principle is that the absorption capacity of oil and water to rays is greatly different. The ray method is usually used for metering between metering stations of oil wells and oil fields, and has the biggest defects of low safety, high cost of protective measures and high daily maintenance cost due to the fact that radiation sources are contained in instruments.
The ultrasonic method is a non-intrusive online detection method, and two solutions applied to detection of the water content of crude oil are generally an attenuation method and a sound velocity method. The ultrasonic attenuation method is used for measuring based on different acoustic characteristics of oil and water phases, the excitation probe excites ultrasonic waves with certain frequency and intensity, the ultrasonic waves pass through crude oil and are attenuated by various mechanisms of the oil and water phases, and finally the ultrasonic waves reach the receiving probe. Under the condition that the excitation energy is fixed, the intensity of the ultrasonic signal at the receiving end can be used for representing the water content in the crude oil. However, attenuation mechanisms of ultrasonic waves in oil-water mixtures are complex and affected by many factors, and inaccurate measurement is easily caused. The ultrasonic sound velocity law is based on the difference of the propagation velocities of ultrasonic waves in oil-water two-phase substances, and the sound velocity is obtained by measuring the conversion of the transit time of the ultrasonic waves passing through crude oilThereby obtaining the water content. The propagation speed of an ultrasonic wave in a two-phase mixture is often not only dependent on its individual propagation speed in each phase, but may also be related to the density of the substance and some thermodynamic parameters. Many researches have been made on the mixing mode of the propagation velocity of the ultrasonic wave in the two-phase mixture, and common sound velocity mixing models include: time-averaging model, modified time-averaging model, Urick model, modified Urick model, and Kuster-
Figure GDA0001981988530000021
And (4) modeling. The work on the comparison of these models is described in detail in the paper Composition measurements of crop oil and process water applications using a viscosity-film ultrasound transducers. The paper concluded that the Urick model and Kuster-
Figure GDA0001981988530000022
The models are two models which are more consistent with experimental results. According to the experimental verification of the inventor, the Urick model is more consistent with the experimental result.
The ultrasonic sound velocity method has the advantages of wide water content measurement range, simple equipment, lower price and easy maintenance, and can be conveniently applied to on-line monitoring as a non-intrusive detection method to realize the real-time measurement of the water content of the crude oil. However, the ultrasonic sound velocity method has a serious disadvantage that the measurement accuracy is greatly influenced by temperature.
Disclosure of Invention
The invention aims to provide a method for detecting the water content of crude oil based on an ultrasonic sound velocity method, which aims to realize a crude oil water content measurement model by using a hybrid modeling mode combining a mechanism model and a data driving model and improve the water content detection precision of the ultrasonic sound velocity method.
The technical implementation scheme of the invention is as follows:
a crude oil water content detection method based on an ultrasonic sound velocity method. The method comprises the following steps:
1) according to the Newton-Laplace sound velocity equation, whether the oil-water mixture or one of the substances has the following relation:
v2＝(kρ)-1(1)
wherein v represents the propagation velocity of ultrasonic waves in a substance in m/s, and ρ represents the density of the substance in kg/m3K represents the adiabatic coefficient of compression of a substance, and k can be generally solved by v and ρ by equation (1). According to the Urick model, the adiabatic compressibility and density of a mixture are related to the adiabatic compressibility, density, and two-phase ratio of the two-phase medium comprising the mixture. For oil-water mixtures, the adiabatic compressibility k and the density ρ satisfy the relation:
Figure GDA0001981988530000031
wherein α represents the water content in the oil-water mixture, kw，koAdiabatic coefficient of compression, rho, for water and oil, respectivelyw，ρoSubstituting equation (2) into equation (1) and appropriately rewriting, a one-dimensional quadratic equation for water cut α can be obtained:
A1α2+A2α+A3-v-2＝0 (3)
wherein A is1，A2，A3Is a coefficient related to the adiabatic compressibility and density of oil and water:
Figure GDA0001981988530000032
solving the quadratic equation of the unary about the water content alpha can obtain the analytic expression of the water content alpha as follows:
Figure GDA0001981988530000033
in the formula, B1，B2，B3Is a and A1，A2，A3The relevant parameters are:
Figure GDA0001981988530000034
because in practical application, A1The value of (A) is always negative2Is always negative, then B1The value of (d) is always negative. And the right half part of the formula (5) is a root formula, the value is determined to be positive, and the minus sign of the formula (5) can be removed by taking the actual physical meaning of the calculation result (the water content cannot be negative), so that the formula (5) is modified as follows:
Figure GDA0001981988530000035
the expression is an expression of the water cut. It can be seen from the derivation process of the expression that the water content is related to the density of water and oil, and the wave velocity of ultrasonic waves in water and oil, and the physical parameters are influenced by the temperature.
2) According to a historical empirical formula and an experimental method, the propagation velocity v of the ultrasonic wave in the water can be obtainedwDensity of water ρwAnd propagation velocity v of ultrasonic wave in crude oiloAnd density ρ of crude oiloTemperature T. Wherein the propagation velocity v of the ultrasonic wave in waterwThe relation of the temperature variation can be obtained by literature data, and the relation is as follows:
vw(T)＝∑kiTi(8)
wherein the coefficient kiThe values of (A) can be referred to the following table:
TABLE 1 temperature coefficient ki
Figure GDA0001981988530000041
Density of water ρwThe relationship with temperature variation can also be obtained from literature and is expressed as:
Figure GDA0001981988530000042
propagation velocity v of ultrasonic waves in crude oiloThe relation of the temperature change is obtained by an experimental method, and the relation is as follows:
vo(T)＝-0.0195T2-2.5296T+1529 (10)
density of crude oil ρoThe relationship with temperature change is also obtained by an experimental method, and the relationship is as follows:
ρo(T)＝-0.6T+0.9264 (11)
substituting equations (8) to (11) into parameter B in step 1)1～B3B can be obtained by calculating a formula and carrying out appropriate simplification processing1～B3The relationship to temperature T is:
B1(T)＝-1.847×10-4T3+2.198×10-2T2-0.9164T+10.36 (12)
B2(T)＝9921T3-1.193×106T2+5.098×107T-8.467×108(13)
B3(T)＝-4.136×10-3T3+0.4902T2-20.5T+350.6 (14)
equations (12) to (14) will be used instead of B in equation (7)1～B3Then, the following can be obtained:
Figure GDA0001981988530000043
the expression is a water cut detection mechanism model including temperature compensation.
3) And carrying out ultrasonic sound velocity method water content detection experiments, including a water content change experiment under constant temperature and a temperature change experiment under constant water content, recording the transit time and the sample temperature under different water content conditions, and calculating the ultrasonic wave velocity in the sample.
The propagation velocity of the ultrasonic wave in the oil-water mixture sample can be obtained by measuring the transit time between the ultrasonic probes, and the conversion method is that the measured transit time t and the propagation velocity v of the ultrasonic wave in the sample are generally considered to have the following relation:
Figure GDA0001981988530000044
where l is the actual propagation distance of the ultrasonic wave in the sample, and Δ t is a common mode deviation amount, which is usually caused by the propagation time of the ultrasonic wave in the mounting plane of the ultrasonic probe and some delay of the hardware circuit, and can be considered as a constant. In general, l and Δ t in formula (16) can be obtained by measuring the transit time of two substances (e.g. water and absolute ethyl alcohol) with known ultrasonic propagation velocities at a specific temperature, and l can be accurately measured by a vernier caliper or the like in the case that the installation plane of the ultrasonic probe is fixed and parallel, so that the actual expression of formula (16), namely the conversion relationship between the measured transit time t and the propagation velocity v of the ultrasonic wave in the sample, can be obtained by measuring the transit time of only one substance with known ultrasonic propagation velocities.
4) The sound velocity v of the ultrasonic wave propagating in the sample obtained by conversion and the temperature T recorded at the same time are substituted into the formula (15), and a preliminary water content prediction result alpha' based on the mechanism model can be obtained.
5) And (4) optimizing the water content prediction result of the formula (13) by considering a method of adding support vector regression. And (3) taking the preliminary water content prediction result alpha' and the temperature as two-dimensional input variables for supporting vector regression analysis, taking the actual water content alpha of each group of experiments as a one-dimensional output variable, constructing a data set, and dividing the data set into a training set and a testing set according to the proportion of 2: 1-3: 1.
6) And training the training set by adopting support vector regression, setting the kernel function as a radial basis kernel function, and continuously correcting a penalty factor c in the support vector regression until the obtained model achieves the best prediction effect on the test set sample, wherein the prediction effect is measured by the root mean square error of the moisture content prediction result, and the smaller the root mean square error is, the better the model prediction effect is. When the root mean square error between the prediction result of the model obtained by training on the test set and the expected water content output result of the test set is the minimum, the iterative training is stopped, and at the moment, a regression model with the optimal prediction effect on the test set can be obtained, wherein the regression model is a crude oil water content detection temperature compensation mixed model obtained based on a mixed modeling method.
The invention has the advantages that:
1. the method is based on mechanism modeling, considers the relation between each physical parameter and temperature in the water content and the ultrasonic propagation speed (namely formula (7)), and deduces a water content detection mechanism model (namely formula (15)) based on an Urick model from the mechanism, wherein the water content detection mechanism model is quite consistent with the physical law of the ultrasonic wave propagation in an oil-water mixture under the real condition to a certain extent. The use of the method reduces the number of experiments required in the ultrasonic sound velocity method modeling process in the temperature compensation process and the required modeling sample amount.
2. The method combines the thought of data modeling, adopts a modeling method of support vector regression, and establishes a regression model based on the mechanism model by the method described in the steps 5) to 6) by considering that certain deviation still exists between the deduced temperature compensation mechanism model based on the Urick model and the propagation rule of the actual ultrasonic wave in the oil-water mixture, so that the deviation between the prediction result and the true value of the original mechanism model (formula (15)) can be effectively reduced by the regression model, and the prediction precision of the mechanism model is improved.
Drawings
FIG. 1 is an illustration of the steps of a method for temperature compensation of crude oil water cut detection based on ultrasonic sound velocity method;
FIG. 2 is a block diagram of an experimental apparatus in the examples;
FIG. 3 is a result of a sound velocity detection experiment of changes in water content at a constant temperature of 25 ℃;
FIG. 4 is a result of an experiment for detecting the speed of sound at a constant water content for a temperature change;
fig. 5 is a comparison graph of the moisture content prediction results of the mechanism model and the mixture model.
FIG. 6 is a comparison graph of water content prediction reference errors of a mechanism model and a mixed model.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
In this embodiment, the method for detecting the water content of crude oil based on the ultrasonic sound velocity method is divided into 3 steps, as shown in fig. 1: 1) deducing a temperature compensation mechanism model based on a Urick model according to the propagation mechanism of ultrasonic waves in the oil-water mixture and the influence mechanism of temperature on related physical parameters; 2) carrying out ultrasonic sound velocity method water content detection experiments, including different water content sample experiments at constant temperature and temperature change experiments in constant water content samples, and obtaining related experiment data; 3) and establishing a regression model by adopting a support vector regression method based on the mechanism model and the experimental data obtained in the first two steps to obtain a final crude oil water content detection model.
Specific embodiments of these three steps will be described in detail below.
Step 1):
the detailed derivation process of the temperature compensation mechanism model based on the uri model in step 1) is the same as the derivation process described in steps 1) to 2) in the summary of the invention, and a description thereof is not repeated.
Step 2):
the experimental setup for performing the ultrasonic sonic method water content detection experiment in this example is shown in fig. 2. The MCU adopts an MSP430 singlechip. After the MCU controls the TDC to generate 2MHz ultrasonic excitation pulse, the excitation probe can generate 2MHz ultrasonic signals and the ultrasonic signals pass through the oil-water mixture sample and reach the receiving probe, the ultrasonic signals received by the receiving probe are transmitted to the TDC after being filtered and amplified, the TDC calculates the time difference between the excitation signals and the receiving signals and sends the time difference to the MCU, and the MCU properly processes the values to obtain the measurement result of the transit time.
While recording the transit time, the MCU reads the PT100 temperature sensor. The PT100 is a common temperature sensor, can quickly respond to the change of temperature to change the resistance value of a thermal resistor, uploads temperature data to an MCU (microprogrammed control unit) through proper signal conditioning and analog-to-digital conversion, and the MCU properly processes the numerical value to obtain a temperature measurement result.
This embodiment has carried out two experiments to this, and the moisture content change experiment under 25 ℃ constant temperature condition was measured to every moisture content point and is got the average more than the cubic. The water content detection range is limited to 0-60% due to the limitation of sample configuration conditions. The second experiment is an experiment of temperature change under the condition that the moisture content is kept unchanged, and the second experiment is carried out on six moisture content points of 0%, 7.12%, 20%, 22.44%, 37.5% and 52.5%. Due to the limitation of experimental conditions, the temperature variation range is 25-40 ℃.
The experimentally measured transit time then needs to be converted to the propagation velocity of the ultrasonic waves in the oil-water mixture sample according to equation (16) in the summary of the invention. The container for containing the oil-water mixture sample adopted in this embodiment is a rectangular parallelepiped container having parallel surfaces, and therefore, the distance measurement is performed on the planar inner wall of the container mounting probe by using a vernier caliper, and the measurement result is l 69.50 mm. This example was then measured at 25 ℃ using pure water as a sample, and the transit time t was determined to be 49.702 μ s. From the table lookup, when the ultrasonic propagation velocity v in pure water at 25 ℃ is 1496.704m/s, Δ t in equation (16) can be calculated to be 49.702 μ s. Substituting the values of l and Δ t into equation (14) and taking the conversion of the units into consideration, obtaining a calculation equation (17) of the wave velocity v,
Figure GDA0001981988530000071
wherein t is in units of μ s and v is in units of m/s.
Fig. 3 and 4 are obtained by converting the transit time measured in the above two experiments into the sound velocity v by the formula (17).
Step 3):
firstly, substituting the sound velocity v and the temperature T obtained in the step 2) into the formula (17), and calculating to obtain a prediction result alpha' of each experimental data point formula (17), wherein the average value of the reference error between the prediction result and the actual water content is 2.96%.
Then, according to the description of step 5) in the summary of the invention, the present embodiment uses the predicted result α' and the temperature T as two-dimensional inputs for support vector regression analysis, uses the actual water content of each set of experiments as one-dimensional output to construct a data set, and divides the data set into a training set (82 sets) and a test set (28 sets) according to a proportion.
Through comparative analysis, the radial basis kernel function is selected as the kernel function of the support vector regression in the embodiment. The training process of the support vector regression mainly corrects the penalty factor c continuously to pursue that the trained regression model achieves the best prediction effect on the test set. In this embodiment, when the value of c is 60, the root mean square error between the water content result predicted by the trained regression model and the actual water content result reaches the minimum value 0.000202973, and at this time, the trained regression model is the final temperature compensation model for detecting the water content of the crude oil obtained in this embodiment. In this embodiment, the prediction result pair of the two models is, for example, as shown in fig. 5, the average value of the reference error between the prediction result of each test point in the test set and the actual moisture content of the temperature compensation mixed model obtained by the mixed modeling by the method is 0.949%, while the average value of the reference error between the prediction result of each test point in the same test set and the actual moisture content of the temperature compensation mechanism model based on the uri model (formula (15)) is 2.497%, and the average value of the reference error is reduced by 61.99% in the mixed model after the support vector regression is added compared with the temperature compensation mechanism model before the addition. Reference error pairs for both models for each data point of the test set are shown in FIG. 6, where multiple sets of values at the same water cut represent data points at different temperatures. From both data and graphical perspectives, this example demonstrates that the addition of support vector regression analysis helps to obtain a more accurate temperature compensation model.

Claims (2)

1. A method for detecting the water content of crude oil based on an ultrasonic sound velocity method is characterized by comprising the following steps:
1) according to a Newton-Laplace sound velocity equation, an Urick model and the relationship between the sound velocity of ultrasonic waves in two media of crude oil and water, the density of the medium and the temperature, a mechanism model of the relationship between the water content and the propagation velocity and the temperature of the ultrasonic waves in the oil-water mixture is deduced:
Figure FDA0002513875410000011
wherein α represents the water content of the oil-water mixture, v represents the propagation velocity of ultrasonic wave in the oil-water mixture, and T represents the temperature of the oil-water mixture, and B represents the temperature1,B2,B3Parameters related to the adiabatic compression coefficients of oil and water, the wave velocity of ultrasonic waves in the two media, and the temperature;
2) measuring the propagation time T of the ultrasonic wave in the measured oil-water mixture sample and the temperature T of the oil-water mixture sample through experiments, and calculating the propagation speed v of the ultrasonic wave in the measured sample by using the propagation time T under the condition of constant propagation distance;
3) substituting the propagation velocity v and the temperature T into the formula (1) to obtain a preliminary prediction result alpha' of the water content;
4) using the preliminary prediction result alpha' and the temperature T as input variables, using the corresponding true moisture content value as an output variable, modeling by adopting a Support Vector Regression (SVR) method, and obtaining an optimal regression model of moisture content prediction by adjusting parameters, wherein the model is a final moisture content detection temperature compensation mixed model;
4.1) taking the preliminary prediction result alpha' and the temperature T as two-dimensional input variables, taking the corresponding true moisture content value as a one-dimensional output variable to construct a data set, and dividing the data set into a training set and a test set;
4.2) carrying out iterative training on the training set by adopting a support vector regression method, setting a kernel function as a radial basis kernel function, setting a penalty factor and the pursuit fitting precision, stopping iterative training when the fitting precision of the training model reaches a set value or reaches the upper limit of the default iteration times, carrying out moisture content prediction on the test set by using the regression model obtained at the moment, and recording the root mean square error of a moisture content prediction result;
4.3) continuously modifying the penalty factors in the step 4.2), repeating the iterative training in the step 3), and observing the root mean square error of the model obtained under different penalty factors on the water content prediction result of the test set; when the root mean square error reaches a minimum value, the prediction effect of the model can be considered to be the best, and the support vector regression model obtained at the moment is a final crude oil water content detection temperature compensation mixed model;
5) and substituting the propagation velocity v and the temperature T of the ultrasonic wave in the oil-water mixture sample to be detected into the mixing model to obtain the final detection result of the water content.
2. The method for detecting the water content of crude oil based on the ultrasonic sound velocity method as claimed in claim 1, wherein a water content detection mechanism model including temperature compensation is obtained through derivation based on a Urick model; the method comprises the following specific steps:
1) in the Urick model, the propagation velocity of ultrasonic waves in a two-phase mixture is related to the adiabatic coefficient of compression and the density according to the Newton-Laplace equation:
v2＝(kρ)-1(2)
in the formula: v is the propagation velocity of the ultrasonic waves in the mixture, and ρ is the density of the mixture in kg/m3K represents the adiabatic compressibility of the mixture; the adiabatic compressibility and density of the mixture, in turn, are related to the adiabatic compressibility, density, and two-phase ratio of the two-phase medium comprising the mixture, and for oil-water mixtures, the adiabatic compressibility and density of the mixture have the following relationships:
Figure FDA0002513875410000021
kw，koadiabatic coefficient of compression, rho, for water and oil, respectivelyw，ρoWater and oil density, α, water content of the mixture, subscripts w and o, respectively, water and oil, and the adiabatic compressibility and density of the mixture, respectively, substituted with the adiabatic compressibility and density of oil and water, respectively, to yield a one-dimensional quadratic equation for water content α:
A1α2+A2α+A3-v-2＝0 (4)
wherein A is1,A2,A3Is a coefficient related to the adiabatic compressibility and density of oil and water:
Figure FDA0002513875410000022
solving equation (4) and considering that the water content can not be negative, obtaining an expression of the water content as follows:
Figure FDA0002513875410000023
wherein, B1,B2,B3Is a and A1,A2,A3The relevant parameters are:
Figure FDA0002513875410000024
as can be seen from the expression (6) of the water content, the water content is related to the density of water and oil and the wave velocity of ultrasonic waves in the water and the oil, and the physical parameters are influenced by the temperature;
2) according to a historical empirical formula and an experimental method, the relation between the relevant physical parameters of the crude oil and the water and the temperature T can be obtained, and the relation is specifically as follows:
variation of sound velocity of water with temperature vw(T) temperature dependence of the water densityw(T) relationship of change of sound velocity of crude oil with temperature vo(T) temperature dependence of crude oil density [ rho ]o(T)；vw(T) and ρw(T) is available from literature, vo(T) and ρo(T) is obtained by experimental measurement and data fitting;
substituting the variation relationship of the four physical parameters with temperature into the parameter B in equation (7)1～B3The calculation formula of (A) is simplified to obtain B1～B3The relationship with temperature T is:
Figure FDA0002513875410000031
and substituting the formula (8) into the formula (6) to obtain the moisture content detection mechanism model with temperature compensation.
CN201811101335.7A 2018-09-20 2018-09-20 Method for detecting water content of crude oil based on ultrasonic sound velocity method Expired - Fee Related CN109298070B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811101335.7A CN109298070B (en) 2018-09-20 2018-09-20 Method for detecting water content of crude oil based on ultrasonic sound velocity method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811101335.7A CN109298070B (en) 2018-09-20 2018-09-20 Method for detecting water content of crude oil based on ultrasonic sound velocity method

Publications (2)

Publication Number Publication Date
CN109298070A CN109298070A (en) 2019-02-01
CN109298070B true CN109298070B (en) 2020-10-02

Family

ID=65163819

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811101335.7A Expired - Fee Related CN109298070B (en) 2018-09-20 2018-09-20 Method for detecting water content of crude oil based on ultrasonic sound velocity method

Country Status (1)

Country Link
CN (1) CN109298070B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110596238A (en) * 2019-10-15 2019-12-20 合肥精特仪表有限公司 Ultrasonic detection method for water content of crude oil and detector using same
WO2022272059A1 (en) * 2021-06-24 2022-12-29 Triad National Security, Llc Method and system of moisture content detection
CN113567558B (en) * 2021-06-30 2022-11-22 广州大学 Soil moisture content ultrasonic measuring device and measuring method
CN113640320B (en) * 2021-08-25 2022-06-10 江苏麦赫物联网科技有限公司 Method for detecting water content of crude oil
CN114280155A (en) * 2021-12-27 2022-04-05 南方英特空调有限公司 Refrigerant oil content measuring method for refrigerant circulating system of automobile air conditioner

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3892127A (en) * 1973-07-27 1975-07-01 Nusonics Oil detection system
US3973430A (en) * 1973-07-27 1976-08-10 Process And Pollution Controls Company Water pollution monitor
JPS52122187A (en) * 1976-04-07 1977-10-14 Nippon Kokan Kk Method of measuring water mixed in oil
US4059987A (en) * 1976-10-20 1977-11-29 Texaco Inc. Apparatus and method for measuring the water content of oil flowing in a pipe
EP0012160A1 (en) * 1978-12-11 1980-06-25 Conoco Phillips Company Ultrasonic method and apparatus for determining quantitative and volumetric data of an oil-water mixture
CN107238658A (en) * 2016-03-28 2017-10-10 中国石油化工股份有限公司 Ultrasonic measurement system and method
CN107255672A (en) * 2017-07-31 2017-10-17 中国石油大学(华东) Based on ultrasonic measurement oil compressibility experimental system and method
CN107462630A (en) * 2017-08-16 2017-12-12 中国石油化工股份有限公司 A kind of polymer concentration detection method and equipment based on sound velocity method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7954362B2 (en) * 2007-07-27 2011-06-07 Multiphase Flow International Llc Ultrasound multiphase fraction meter and method for determining phase fractions in a multiphase fluid
US8880363B2 (en) * 2009-03-24 2014-11-04 Cameron International Corporation Method and apparatus for the measurement of the mass fraction of water in oil-water mixtures

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3892127A (en) * 1973-07-27 1975-07-01 Nusonics Oil detection system
US3973430A (en) * 1973-07-27 1976-08-10 Process And Pollution Controls Company Water pollution monitor
JPS52122187A (en) * 1976-04-07 1977-10-14 Nippon Kokan Kk Method of measuring water mixed in oil
US4059987A (en) * 1976-10-20 1977-11-29 Texaco Inc. Apparatus and method for measuring the water content of oil flowing in a pipe
EP0012160A1 (en) * 1978-12-11 1980-06-25 Conoco Phillips Company Ultrasonic method and apparatus for determining quantitative and volumetric data of an oil-water mixture
CN107238658A (en) * 2016-03-28 2017-10-10 中国石油化工股份有限公司 Ultrasonic measurement system and method
CN107255672A (en) * 2017-07-31 2017-10-17 中国石油大学(华东) Based on ultrasonic measurement oil compressibility experimental system and method
CN107462630A (en) * 2017-08-16 2017-12-12 中国石油化工股份有限公司 A kind of polymer concentration detection method and equipment based on sound velocity method

Non-Patent Citations (12)

* Cited by examiner, † Cited by third party
Title
"Acoustic velocities in petroleum fluids: Measurement and prediction";Hameeda H.Dashti等;《Journal of Petroleum Science and Engineering》;20141231;第94-104页 *
"Composition measurements of crude oil and process water emulsions using thick-film ultrasonic transducers";Guangtian Meng等;《Chemical Engineering and Processing》;20060531;第45卷(第5期);第383-391页 *
"Effect of temperature on the ultrasonic properties of oil-in-water emulsions";Ratjika Chanamai等;《Colloids and Surfaces》;19981231;第241-250页 *
"LIBSVM：A Library for Support Vector Machines";CHIH-CHUNG CHANG等;《ACM Transactions on Intelligent Systems and Technology(TIST)》;20110531;第1-27页 *
"Speed of sound in pure water as a function of temperature";Nykolai Bilaniuk等;《The Journal of the Acoustical Society of America》;19930331;第1609-1612页 *
"Speed of Sound in Pure Water";Del Grosso等;《The Journal of the Acoustical Society of America》;19721231;第1442-1446页 *
"支持向量回归机研究及其应用";梁伟锋;《中国优秀硕士学位论文全文数据库 信息科技辑》;20070415;第1-57页 *
"最新温标纯水密度表";周湄生;《计量技术》;20001231;第40-42页 *
"温度对原油密度测量的影响";唐蒙等;《石油与天然气化工》;20001231;第147-148页 *
"用超声波声速法测定油品中含水率";阎向宏;《石油大学学报(自然科学版)》;19980228;第22卷(第1期);第103-104页 *
"超声波检测原油含水率研究";郑磊等;《研究与开发》;20180228;第44卷(第2期);第143页 *
"辽河油区原油密度与温度关系的统计方程";康志勇;《石油勘探与开发》;19990831;第107-109页 *

Also Published As

Publication number Publication date
CN109298070A (en) 2019-02-01

Similar Documents

Publication Publication Date Title
CN109298070B (en) Method for detecting water content of crude oil based on ultrasonic sound velocity method
JPH04505368A (en) How to measure particle size distribution and concentration in suspensions using ultrasound
CN103278561B (en) Universal ultrasonic liquid concentration detection apparatus
CN109406444B (en) Transformer oil water content-absorption coefficient standard curve fitting method, water content detection method and device and storage medium
Henning et al. Process monitoring using ultrasonic sensor systems
US20090093977A1 (en) Method and device for acoustic measurement of the specific density of a gaseous or liquid medium
Norisuye et al. Simultaneous evaluation of ultrasound velocity, attenuation and density of polymer solutions observed by multi-echo ultrasound spectroscopy
CN108535158A (en) A kind of sound-electric compound dust concentration detecting system and detection method
CN112129832A (en) Sweep frequency ultrasonic attenuation measurement method for detecting content of dispersed phase
Kerdtongmee et al. Quantifying dry rubber content in latex solution using an ultrasonic pulse
Krause et al. Ultrasonic sensor for predicting sugar concentration using multivariate calibration
CN102980941A (en) Method and device for detecting ultrasound gas relative humidity through utilizing sound speed variations
Guo et al. Water Level Sensing in a Steel Vessel Using A0 and Quasi‐Scholte Waves
Geier et al. Effects of yeast and maltose concentration on ultrasonic velocity and attenuation coefficient and its application for process monitoring
Greenwood et al. Long-path measurements of ultrasonic attenuation and velocity for very dilute slurries and liquids and detection of contaminates
Zhao et al. Concentration measurement of carbon dioxide based on ultrasonic principle
Cegla et al. Fluid bulk velocity and attenuation measurements in non-Newtonian liquids using a dipstick sensor
Zhou et al. Ultrasound velocity-based phase content measurement of oil sludge in storage tank
Martinsson et al. Model-based phase velocity and attenuation estimation in wideband ultrasonic measurement systems
Gao et al. A novel NaCl concentration detection method based on ultrasonic impedance method
RU2227320C2 (en) Method for measuring quality characteristics of oil products
Hussein et al. Robust spectral estimation for speed of sound with phase shift correction applied online in yeast fermentation processes
Oblak et al. Applications and limitations of a four-microphone impedance tube in the liquid media in the low-frequency range
Gushchina et al. Development of the experimental equipment for measuring the velocity of ultrasonic waves with high accuracy
Gao et al. Temperature compensation and amplitude prediction in ultrasonic measurement based on BP neural network mode

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20201002

Termination date: 20210920