CN114068051B - Method for calculating temperature flow of nuclear reactor main pipeline coolant based on ultrasonic array - Google Patents
Method for calculating temperature flow of nuclear reactor main pipeline coolant based on ultrasonic array Download PDFInfo
- Publication number
- CN114068051B CN114068051B CN202111251821.9A CN202111251821A CN114068051B CN 114068051 B CN114068051 B CN 114068051B CN 202111251821 A CN202111251821 A CN 202111251821A CN 114068051 B CN114068051 B CN 114068051B
- Authority
- CN
- China
- Prior art keywords
- ultrasonic
- temperature
- model
- coolant
- pipeline
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
- G21C—NUCLEAR REACTORS
- G21C17/00—Monitoring; Testing ; Maintaining
- G21C17/017—Inspection or maintenance of pipe-lines or tubes in nuclear installations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
- G21C—NUCLEAR REACTORS
- G21C17/00—Monitoring; Testing ; Maintaining
- G21C17/02—Devices or arrangements for monitoring coolant or moderator
- G21C17/032—Reactor-coolant flow measuring or monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E30/00—Energy generation of nuclear origin
- Y02E30/30—Nuclear fission reactors
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Plasma & Fusion (AREA)
- High Energy & Nuclear Physics (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Algebra (AREA)
- Computing Systems (AREA)
- Fluid Mechanics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Monitoring And Testing Of Nuclear Reactors (AREA)
Abstract
The invention relates to the field of ultrasonic measurement, in particular to a method for calculating the temperature and flow of a nuclear reactor main pipeline coolant based on an ultrasonic array, which solves the problem of low measurement precision in the process of ultrasonic flow measurement in the prior art. The invention sets up the ultrasonic transducer group outside the pipeline, and connect the remote server through the network module, a model field of temperature and flow rate under the input parameter of a calculation of a model through CFD construction; establishing a relation description model; triplet field measurement and calculation; and the four-way reconstruction field and the model field are used for carrying out feedback setting on the topological structure of the ultrasonic transducer group. According to the invention, the measuring precision of the transducer group is improved by eliminating the influence factors of ultrasonic waves in a nonaqueous medium; and calculating the temperature and the flow velocity under the input parameters through the CFD and the neural network model, performing variance calculation with the temperature and the flow velocity measured by the transducer group, and further improving the measurement accuracy through adjusting the topological structure of the transducer group. The remote server is connected with the transducer group, so that real-time monitoring of measurement is realized.
Description
Technical Field
The invention relates to the field of ultrasonic application, in particular to a method for calculating the temperature and flow of a nuclear reactor main pipeline coolant based on an ultrasonic array.
Background
The nuclear reactor is a complex and efficient nuclear heat conversion device, and the coolant in the hot section of the main pipeline has the characteristics of high temperature, high pressure, high irradiation and high flow rate, and meanwhile, the temperature of the coolant flowing into the hot section of the main pipeline from different fuel channels is different due to different enrichment degrees of various fuel assemblies of the reactor core, so that the phenomenon of obvious temperature stratification exists in the hot section of the main pipeline. The coolant temperature and flow rate of the main nuclear reactor pipeline directly reflect the reactor nuclear power and the reactor core heat conduction capacity, and are one of the important thermodynamic parameters in the safety protection and operation control of the reactor. Accurate measurement and calculation of the temperature and flow of the main coolant is important for reactor safety and economy.
The problems of insufficient reactor output, frequent control rod actions and even unplanned shutdown caused by inaccurate and untimely measurement of the temperature of the coolant of the main pipeline of the reactor become a serious bottleneck which severely restricts the further improvement of the safety and the economy of the reactor. The domestic and foreign research data show that the thermal stratification temperature difference of the reactor core outlet of the pressurized water reactor can reach 15 ℃, the thermal stratification temperature difference of the section of the main pipeline hot section which is 1.5 meters away from the reactor core outlet still reaches 10 ℃, and the layering state is continuously changed along with the operation of the reactor, so that accurate and timely measurement of the representative coolant thermal physical field is very difficult.
The method for measuring the temperature of the coolant of the main reactor pipeline in the prior art is generally a thermal resistance method and an ultrasonic temperature measurement method: the thermal resistance point temperature measurement technology has the following problems: 1. because the dynamic change layering phenomenon exists in the hot-zone coolant and is limited by the installation space of the main pipeline of the reactor, the problems of insufficient representativeness and large measurement error exist, and the typical measurement error reaches 1.95 ℃ (equivalent to 5.5% of full power), so that the operation capability of the reactor is severely limited; 2. because the thermometer has the problem of larger thermal inertia, the response time of the temperature measuring channel exceeds 10s, and the timeliness of safety protection and control is seriously restricted; 3. the inherent safety of the reactor is reduced by the need to open holes at the pressure boundary for the thermometer to be installed. The ultrasonic temperature measurement is adopted as a non-contact temperature measurement method, has the characteristics of timely response and high safety, and can simultaneously form the measured line temperature into the surface temperature and even the body temperature through a reconstruction algorithm, so that compared with the traditional point temperature measurement, the calculation error caused by temperature layering can be greatly reduced, and the measurement accuracy of the average temperature is improved. However, due to the dynamic change characteristics of the temperature, pressure and concentration of the reactor coolant, the composite relationship between the ultrasonic propagation speed and the coolant multi-element is not established at present; there is also a problem that the sub-temperature area is not divided reasonably, so that the error is large.
In the existing pipeline flow measurement, the pipeline flow is calculated in a mode of measuring the pipeline flow speed based on ultrasonic waves, however, the propagation speed of the ultrasonic waves in a medium is influenced by various factors, such as temperature, pressure, medium density and the like. When the influence factors are greatly changed, the calculation accuracy of the existing flowmeter is greatly reduced, and the problems of low robustness, complex structure and the like of the ultrasonic flowmeter exist;
in designing an ultrasonic pipeline flow meter by the principle of the time difference method, the measurement accuracy of the propagation time of ultrasonic waves in the pipeline determines the measurement accuracy of the average flow velocity, and finally, the measurement of the pipeline flow is influenced. However, the actual propagation time measurement includes the propagation time in solid media such as pipe wall and transducer besides the propagation time in the pipe, and the propagation speed of ultrasonic wave in the pipe is disturbed by factors such as temperature, pressure, liquid density, etc., so that the measurement accuracy is difficult to guarantee.
A new temperature and flow calculation method capable of solving the above problems is needed.
Disclosure of Invention
The invention provides a method for calculating the temperature and the flow of a nuclear reactor main pipeline coolant based on an ultrasonic array, which solves the problem of low nuclear reactor energy accounting precision in the prior art and can realize on-line high-precision measurement of average temperature and flow.
The technical scheme of the invention is realized as follows: the method for calculating the temperature flow of the coolant of the nuclear reactor main pipeline based on the ultrasonic array comprises the steps that an ultrasonic transducer group is arranged on the outer side of the pipeline, the transducer group is connected with a remote server through a network module, and a model field is established: establishing a complete continuous main pipeline coolant space-time domain temperature layered diffusion model and a flow velocity distribution model under input parameters through a CFD (computational fluid dynamics) construction model; secondly, establishing a relation description model of three elements of ultrasonic wave propagation speed and coolant; triplet field measurement and calculation: measuring the flying time of each effective acoustic wave propagation path; the network module transmits signals of the transducer group to a remote server through the network module; reconstructing the real-time temperature distribution and flow velocity distribution of the coolant in the pipeline; calculating average temperature and flow; and the four-way reconstruction field and the model field are used for carrying out feedback setting on the topological structure of the ultrasonic transducer group.
The third step comprises ultrasonic transmission time calibration, including simulating an experimental platform and acquiring experimental data; the simulation experiment platform comprises a pipeline capable of controlling temperature and an ultrasonic transducer arranged on the outer side of the pipeline.
The ultrasonic transmission time calibration specifically comprises the following steps:
1 in the empty pipe condition and temperature T 0 Under this condition, τ at this temperature is obtained by changing the ultrasonic propagation path 0 Results of (2);
2 changing the temperature, repeating the experiment to calculate tau at different temperatures 0 Results of (2);
fitting the experimental calculation results to obtain a description model tau between the time delay and the temperature of the ultrasonic wave in the nonaqueous medium 0 ＝F(T)。
The second step is specifically as follows: (1) Establishing a relation between ultrasonic sound velocity and coolant; (2) simulating an experimental platform and acquiring experimental data; (3) Establishing a relation description model between ultrasonic sound velocity and coolant temperature, pressure and concentration by adopting a machine learning method; the simulation experiment platform also comprises a constant temperature tank with a water inlet, a water outlet and a concentration control device and a metal pipeline for controlling pressure; the ultrasonic transducer is connected with the micro-control device.
The fourth step is specifically division of subareas: adjusting the topological structure of the ultrasonic transducer: adjusting the topology of the ultrasonic transducer array based on the error distribution of the reconstructed field and the model field; dynamic adjustment based on feedback mechanism: taking reconstruction error as a target variable, and the radius R of the inner circle of the subarea ir And the radial angle theta is used as a control variable to determine the optimal inner ring radius R ir And a radial angle θ.
The ultrasonic transducer array topological structure is broadcast, and particularly the ultrasonic transducers are uniformly arranged on the section of the pipeline to be tested.
The first specific step is as follows: calculating the temperature and flow rate data of the coolant of the typical section of the pipeline under the set parameters by calculating the fluid mechanics CFD; b training CFD simulation results by using the neural network model: training the simulation result in the step a through a neural network model; c prediction result: and c, predicting and obtaining continuous and complete temperature and flow velocity distribution of the coolant in the pipeline by using the neural network model in the step b.
The fitting of the step 3 is a machine learning method combining individual learner learning and individual learner strategies by adopting more than two ultrasonic transmission characteristics.
According to the method for calculating the temperature flow of the nuclear reactor main pipeline coolant based on the ultrasonic array, disclosed by the invention, the measuring accuracy of the transducer group is improved by eliminating the influence factors of ultrasonic waves in a nonaqueous medium; calculating the temperature and the flow rate under the input parameters through the CFD and the neural network model, performing variance calculation with the temperature and the flow rate measured by the transducer group, reconstructing a space-time domain temperature layered diffusion model of the coolant of the main pipeline hot section through ultrasonic measurement, obtaining a relation description model of three elements of the ultrasonic sound velocity coolant through a simulation experiment platform, and further improving the measurement accuracy through adjusting the topological structure of the transducer group. The remote server is connected with the transducer group, so that real-time monitoring of measurement is realized.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1: simulating an experiment platform;
fig. 2: a pipeline subarea division schematic diagram;
fig. 3: establishing a flow chart of a space-time domain temperature layered diffusion description model of the main pipeline hot section coolant;
fig. 4: an ultrasonic transducer array map;
wherein: 1. an ultrasonic transducer; 2. a metal pipe; 3. a micro control device; 4. a constant temperature bath; 5. a pump; 6. Boric acid concentration control means; 7. a water inlet; 8. a water outlet; 9. and a pressure controller.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention discloses a method for calculating the temperature flow of a nuclear reactor main pipeline coolant based on an ultrasonic array, wherein an ultrasonic transducer 1 group is arranged on the outer side of a pipeline, the transducer group is connected with a remote server through a network module, and a model field is established: establishing a complete continuous main pipeline coolant space-time domain temperature layered diffusion model and a flow velocity distribution model under input parameters through a CFD (computational fluid dynamics) construction model; secondly, establishing a relation description model of three elements of ultrasonic wave propagation speed and coolant; triplet field measurement and calculation: measuring the flying time of each effective acoustic wave propagation path; the network module transmits signals of the transducer group to a remote server through the network module; reconstructing the real-time temperature distribution and flow velocity distribution of the coolant in the pipeline; calculating average temperature and flow; specifically, three-dimensional flow velocity field reconstruction and three-dimensional temperature field reconstruction; and the four-way reconstruction field and the model field are used for feedback setting of the topological structure of the ultrasonic transducer 1 group.
The third step comprises ultrasonic transmission time calibration, including simulating an experimental platform and acquiring experimental data; the simulation experiment platform comprises a pipeline with controllable temperature, and an ultrasonic transducer 1 arranged outside the pipeline.
The ultrasonic transmission time calibration specifically comprises the following steps:
1 in the empty pipe condition and temperature T 0 Under this condition, τ at this temperature is obtained by changing the ultrasonic propagation path 0 Results of (2);
2 changing the temperature, repeating the experiment to calculate tau at different temperatures 0 Results of (2);
fitting the experimental calculation results to obtain a description model tau between the time delay and the temperature of the ultrasonic wave in the nonaqueous medium 0 ＝F(T)。
The second step is specifically as follows: (1) Establishing a relation between ultrasonic sound velocity and coolant; (2) simulating an experimental platform and acquiring experimental data; (3) Establishing a relation description model between ultrasonic sound velocity and coolant temperature, pressure and concentration by adopting a machine learning method; the simulation experiment platform also comprises a constant temperature tank 4 with a water inlet 7, a water outlet 8 and a concentration control device, and a metal pipeline 2 for controlling pressure; the ultrasonic transducer 1 is connected with a micro-control device 3.
The first specific step is as follows: calculating the temperature and flow rate data of the coolant of the typical section of the pipeline under the set parameters by calculating the fluid mechanics CFD; b training CFD simulation results by using the neural network model: training the simulation result in the step a through a neural network model; c prediction result: and c, predicting and obtaining continuous and complete temperature and flow velocity distribution of the coolant in the pipeline by using the neural network model in the step b.
The fitting of the step 3 is a machine learning method combining individual learner learning and individual learner strategies by adopting more than two ultrasonic transmission characteristics.
The temperature data prediction using ELM takes place in the following manner: the section data obtained by CFD analysis is used as training samples, and the training sample format is { loc ] mn ,T mn M=1, …, M is the total number of temperature measurement sections calculated by CFD software, n=1, …, N is the total number of grids divided on the temperature measurement sections, and the gridding schematic diagram of the main pipeline and the temperature measurement sections is shown in fig. 3. loc mn ＝{[(x mn ,y mn ),d m ,t m ]}, where d m Distance t is the distance from the m section to the core outlet m For the m-th section coolant relative movement time, (x) mn ,y mn ) Is the center point coordinate of the nth region on the mth section, T mn The temperature value corresponding to the nth region on the mth section. The ELM regression model for an activation function g (x) containing R hidden layer nodes is expressed as:
wherein omega is r ＝[ω r1 ,ω r2 ,L,ω rm ] T Representing weights between the input node and the r hidden layer node; beta r ＝[β r1 ,β r2 ,L,β rm ] T Representing weights between the output node and the r-th hidden layer node; b r Is the i-th hidden layer threshold; r is the number of hidden layer nodes.
To improve model predictive power, g (x) may be chosen to train samples in a zero error approximation, i.e
Then there is
The above-mentioned rewriting into matrix form can be expressed as
Gβ＝T
Wherein G is an hidden layer output matrix, and the weight omega is input r Hidden layer threshold b r All are randomly assigned values, so that the hidden layer output matrix G is known, and training is completed after the output weight matrix beta is obtained. Let the number of predicted sections be H, let the coordinate information loc of the H predicted sections be H hn Inputting the temperature value into a trained ELM model to correspondingly output a predicted temperature value T hn And the establishment of a space-time domain temperature layered diffusion description model of the main pipeline hot section coolant is realized. The complete process flow of the method is shown in FIG. 3;
establishing a relation description model of three elements of ultrasonic wave propagation speed and coolant;
(1) And establishing a relation between ultrasonic sound velocity and coolant. According to the wave equation, the propagation velocity of ultrasonic waves in a liquid medium is related to the adiabatic compression coefficient K and the density ρ of the liquid, which are expressed as follows:
wherein, the adiabatic compression coefficient K and the density ρ are related to temperature and pressure. The hot-section coolant of the main reactor pipeline is boric acid solution, and the temperature, pressure and boron concentration of the boric acid solution can influence the propagation speed of ultrasonic waves. The association relationship between the ultrasonic propagation speed and the coolant is expressed as follows:
w＝f(T,P,S) (X)
where w is the ultrasonic sound velocity, T is the coolant temperature, P is the coolant pressure, and S is the coolant boron concentration. Referring to the ultrasonic transmission characteristics in the seawater, an ultrasonic sound velocity expression under the multi-element compound influence in the coolant is given:
w＝w 0 +w T +w P +w S +w TPS (XI)
w in 0 Is constant, w T Is the relation between the ultrasonic sound velocity and the coolant temperature, w P Is the relation between the ultrasonic sound velocity and the pressure of the coolant, w S Is the relation between the ultrasonic sound velocity and the boron concentration of the coolant, w TPS Is the relationship between the ultrasonic sound velocity and the temperature, pressure and boron concentration of the coolant.
(2) Simulating an experiment platform and acquiring experiment data; and a machine learning method is adopted to establish a relation description model between ultrasonic sound velocity and coolant temperature, pressure and concentration. And a multi-element ultrasonic transmission characteristic experimental platform is established in a simulated real environment to study the association relationship between three elements of ultrasonic sound velocity and coolant, and a schematic diagram of the experimental platform is shown in fig. 1. The experiment platform takes boric acid solution (coolant) as an ultrasonic wave propagation medium, and the device can control and adjust the temperature, pressure and concentration of the coolant. The experiment adopts a controlled variable method to respectively study the influence of each element of the coolant on the ultrasonic wave propagation speed and acquire a large amount of experimental data.
Further, the simulation experiment platform comprises a constant temperature tank 4 with a water inlet 7, a water outlet 8 and a concentration control device, and a metal pipeline 2 for controlling pressure, wherein the metal pipeline 2 is also provided with an ultrasonic transducer 1; the ultrasonic transducer 1 is connected with the micro-control device 3.
Preferably, the method of machine learning employs more than two ultrasound transmission characteristics individual learner learning and individual learner combining strategies.
(1) Determining the topological structure and the effective acoustic path of the ultrasonic transducer 1; a radial-axial array of several ultrasonic transducers is shown in fig. 4 (a), with 20 ultrasonic transducers 1 being exemplified, the transducers being uniformly mounted around the pipe, as in the black dots in fig. 4 (a). These transducers are considered transceivers that are controlled to transmit and detect ultrasonic signals at different times. The ultrasonic signal propagates from one transducer to the other, producing an effective acoustic path as shown in fig. 4 (a). Theoretically, an ultrasonic path exists between every two transducers. However, since the ultrasonic paths on the edges or surfaces do not contribute much to the reconstruction of the internal temperature, they are not necessary.
(2) Reconstructing a three-dimensional temperature field;
after the flying time on each effective acoustic path is obtained, the three-dimensional cylindrical temperature field is inverted by using a radial base temperature field reconstruction algorithm by utilizing the mapping relation between the ultrasonic wave propagation speed and the coolant. The number of ultrasonic transducers 1 generates I effective ultrasonic paths between every two ultrasonic transducers, and divides the effective ultrasonic paths into J sub-temperature areas, as shown in figure 2. The obtained sonic flight time on each ultrasonic propagation path can be expressed as:
where a (x, y, z) represents the inverse of the sound velocity of the ultrasonic wave, and when radial basis functions are used, a (x, y, z) is represented as a linear combination of J radial basis functions:
wherein: epsilon j Is a coefficient to be determined; (x) j ,y j ,z j ) Center point coordinates of the jth sub-temperature region; alpha is a radial basis function phi j Shape parameters of (x, y, z) it is necessary to determine the appropriate shape parameters during a specific experiment.
Order the
Then it can be obtained
Rewriting formula (XVI) into matrix form
t＝F·E (XVII)
Wherein: t= [ t ] 1 ,t 2 ,...,t I ] T ，F＝(f kj ) k＝1,2,...I；j＝1,2,...J ，E＝[ε 1 ,ε 2 ,...,ε J ] T ；
Using the singular value decomposition of matrix F and Tikhonov regularization technique, the regularization solution of equation (XVII) is:
wherein: sigma is the non-zero singular value of the coefficient matrix F, J is the total number of non-zero singular values; u (u) j 、v j Left and right singular vectors of F respectively; μ is a regularization parameter. After the ultrasonic receiving and transmitting array position is determined, the shape parameter alpha of the radial basis function is given, and the coefficient matrix F and the singular value thereof can be obtained. When each acoustic wave path is obtainedAfter the transition time matrix t of (c), the parameter vector epsilon can be determined according to formula (XVIII). After the sound velocity reciprocal a (x, y, z) of the ultrasonic wave is obtained, the sound velocity reciprocal a is taken into the acquired experimental data to establish a mapping relation between the ultrasonic wave propagation velocity and the coolant three elements:
T＝f -1 (w,P,S) (XIX)
the three-dimensional temperature distribution of the coolant can be reconstructed from the formula (XIX).
And calculating the error between the reconstruction field and the radial axial direction of the model field according to the established pipeline space-time domain temperature distribution model. And (3) bringing the central coordinates in the dividing sub-temperature regions into the central coordinates to obtain the temperature value of the central coordinates of each sub-temperature region, and solving the temperature value of the central coordinate point of each sub-temperature region of the temperature field after reconstruction and calculating a model to obtain the maximum absolute error, the minimum absolute error, the average relative error and the root mean square error of the temperature value. The expression is as follows:
E max ＝max|T(x i ,y i ,z i )-T m (x i ,y i ,z i )| (XX)
E min ＝min|T(x i ,y i ,z i )-T m (x i ,y i ,z i )| (XXI)
in the formulas (XX) - (XXIII), n is the number of all central coordinate temperature values in the area to be detected; t (T) a The average temperature of the simulated field; t (x) i ,y i ,z i ) For model fields at coordinates (x i ,y i ,z i ) Temperature value of (2); t (T) m (x i ,y i ,z i ) In coordinates (x i ,y i ,z i ) Is a temperature value of (a).
(3) Reconstructing a three-dimensional flow velocity field;
aiming at complex flow states of different positions of a reactor pipeline, the existence of turbulent eddies in the pipeline flow field is considered, and compared with a two-dimensional pipeline section average flow velocity calculation scheme, the flow calculation method based on three-dimensional cylindrical flow field distribution is more accurate.
And installing an ultrasonic detection array on the reactor pipeline, selecting a proper ultrasonic signal emission period, measuring the flying time of the acoustic wave path, calculating the flow velocity according to the flying time difference between the countercurrent flow and the downstream flow after the flying time is obtained, and inverting the three-dimensional flow velocity field.
The propagation time t of ultrasonic wave from downstream to upstream u And upstream to downstream propagation time t d Can be expressed as:
the relationship between ultrasonic flight time and flow rate can be obtained from equation (1):
thus, after the time of flight on each effective acoustic path is obtained, the three-dimensional flow velocity field is inverted using a radial basis based reconstruction algorithm using the relationship between the elapsed ultrasonic time of flight and the flow velocity.
The number of ultrasonic transducers 1 generates I effective ultrasonic paths between every two, and divides the effective ultrasonic paths into J sub-areas, as shown in fig. 2. The subareas are divided into inner circle radius R ir And the radial angle theta is two key parameters, the radius of the inner ring is different, the radial angle is different, and the divided subareas are changed accordingly. The relationship between the flight time of an ultrasonic wave along a specific propagation path in a region to be measured and the flow rate can be expressed as a line integral as follows:
wherein a (x, y, z)) Represents the reciprocal of the flow rate, l k Is an ultrasonic propagation path. When radial basis functions are used, a (x, y, z) is expressed as a linear combination of J radial basis functions:
wherein: epsilon j Is a coefficient to be determined; (x) j ,y j ,z j ) Center point coordinates of the j-th sub-region; alpha is a radial basis function phi j Shape parameters of (x, y, z) it is necessary to determine the appropriate shape parameters during a specific experiment.
Order the
Then it can be obtained
Writing formula (6) into matrix form
Δt＝F·E (8)
Wherein: Δt= [ Δt ] 1 ,Δt 2 ,...,Δt I ] T ，F＝(f kj ) k＝1,2,...I；j＝1,2,...J ，E＝[ε 1 ,ε 2 ,...,ε J ] T ；
Using singular value decomposition of matrix F and Tikhonov regularization technique, the regularization solution of equation (27) is:
wherein: non-zero singular with sigma matrix FThe value, J, is the total number of non-zero singular values; u (u) j 、v j Left and right singular vectors of F respectively; μ is a regularization parameter. After the ultrasonic receiving and transmitting array position is determined, the shape parameter alpha of the radial basis function is given, and the matrix F and the singular value thereof can be obtained. Δt can also be determined by actually measuring the sound wave flight time and the sound channel angle θ, and the parameter vector ε can be determined according to the formula (28) and substituted into the formula (23) to obtain the distribution function of the flow velocity reciprocal, so that the distribution function of the flow velocity is obtained through the following formula, and the reconstruction of the flow velocity field is realized.
According to the established pipeline space-time domain flow velocity distribution model, calculating an error between a reconstruction field and a model field, bringing the center coordinates in the divided subareas into the center coordinates to obtain a flow velocity value of the center coordinates of each subarea, and solving the flow velocity value of the center coordinate point of each subarea of the flow velocity field after reconstruction and calculating the maximum absolute error, the minimum absolute error, the average relative error and the root mean square error of the flow velocity value by the model. The expression is as follows:
E max ＝max|v(x i ,y i ,z i )-v m (x i ,y i ,z i )| (11)
E min ＝min|v(x i ,y i ,z i )-v m (x i ,y i ,z i )| (12)
in the formulae (11) - (14):
n: number of flow velocity values of all center coordinates in the region to be measured
v a : analog fieldAverage flow velocity of (2)
v(x i ,y i ,z i ): model field is at coordinates (x i ,y i ,z i ) Flow velocity value of (2)
v m (x i ,y i ,z i ): the flow velocity field is at coordinates (x i ,y i ,z i ) Flow velocity value of (2)
Further, the third step further includes: a, calculating the average temperature in the measurement area according to the reconstructed temperature field; and b, calculating the average flow in the measurement area according to the reconstructed flow velocity field.
As shown in the modeling flow chart of the three-dimensional cylindrical pipeline of the reactor in fig. 3, specifically: the modeling of the space-time domain temperature and flow velocity distribution model is carried out on the cylindrical pipeline of the reactor, and the modeling can be carried out according to single-phase turbulence because the fluid does not generate phase change in the pipeline. The fluid flow process in a pipe can thus be described by establishing the following time-averaged single-phase continuity equation, momentum equation, and energy equation.
Continuity equation:
momentum equation:
wherein mu eff ＝μ+μ turb 。
Energy equation:
wherein k is eff ＝k+Pr t (μ t /ρα)。
Since turbulence affects both mass and energy transfer, it is necessary to build a suitable turbulence model to quantify turbulence viscosity. Analysis may be aided based on a standard k-epsilon turbulence model that solves the equation for turbulence momentum k and its dissipation rate epsilon, expressed as follows:
wherein turbulent viscosity is defined as μ t ＝ρC μ (k 2 ∈/epsilon). The turbulent planter number was also introduced in the above formulas (4) and (5) to quantify the turbulent energy dissipation coefficient of turbulent viscosity.
As shown in the grid division diagram of the cylindrical pipeline modeling of the reactor in fig. 4, the calculation of the equation is converted into the calculation of a discrete equation, so that the grid division of the modeling area is performed by utilizing the CFD software to realize discretization, and the temperature and flow velocity distribution of the pipeline flow measurement section is obtained.
In order to establish a space-time domain temperature distribution model and a flow velocity distribution model of the whole three-dimensional cylindrical pipeline flow measurement section of the reactor, data prediction is required on the basis of existing data. The radial basis function RBF neural network is adopted to predict flow velocity data, and the extreme learning machine ELM is adopted to predict temperature data.
The flow rate data prediction using the RBF neural network may be performed in the following manner: the flow velocity data obtained by CFD analysis is used as training samples, and the training sample format is { X, V }, wherein X= [ X ] 1 ,x 2 ,x 3 ,x 4 ] T ，x 1 Is the abscissa of the model grid position, x 2 Is the ordinate, x, of the grid position of the model 3 Is the vertical coordinate of the position of the model grid, x 4 The relative movement time t of water in the reactor pipeline; v is the one-dimensional output speed obtained by CFD simulation; the RBF network output is Y= [ Y ]]The method comprises the steps of carrying out a first treatment on the surface of the p is the number of hidden layer nodes (p>4) N is the total number of input samples; for the iteration termination accuracy. The output of the RBF neural network can be expressed as:
the radial basis neuron structure is shown in fig. 4, and the formula is as follows:
the purpose of model training is to obtain three parameters W in the formula j 、C j And D j And obtaining a space-time domain flow velocity distribution model. The training of RBF neural network is mainly divided into two steps, firstly, the parameter C between the input layer and the hidden layer is determined by performing unsupervised learning j And D j Then, supervised learning is carried out to determine the weight W between the hidden layer and the output layer j 。
First, for the connection weight w= [ W ] of the hidden layer to the output layer 1 ,w 2 ,…,w p ] T Center parameter C of each node of hidden layer j ＝[c j1 ,c j2 ,c j3 ,c j4 ] T Width vector D of each node of hidden layer j ＝[d j1 ,d j2 ,d j3 ,d j4 ] T (j=1, 2, …, p) is initialized, and the initialization formulas are respectively:
in the formula (22), p is the number of hidden layer nodes (p > 4), min (V) k ) Is the minimum of all desired outputs in the output neurons in the training set, max (V k ) Is the maximum of all desired outputs in the output neurons in the training set.
In formula (23), min (x i ) Is all input signals in the ith input feature of the training setMinimum value of information, max (x i ) Is the maximum value of all input information in the ith input feature of the training set.
In the formula (24), d f The width adjustment coefficient is generally smaller than 1, so that each hidden layer neuron can more easily realize the sensing capability of local information, and the RBF neural network is helped to promote the local response capability.
Then, the central parameter, the width vector and the adjusting weight are adaptively adjusted by using a gradient descent method, and the iterative calculation formulas are respectively as follows:
wherein, E is RBF network evaluation function as learning factor:
and finally, judging whether the condition of exiting the iterative operation is satisfied, namely, the root mean square error RMS is less than or equal to epsilon, if so, finishing training, and if not, continuing to perform the weight iterative calculation.
After training, the values of the related parameters can be obtained, namely, the establishment of a space-time domain flow velocity distribution model of the three-dimensional cylindrical pipeline of the reactor is realized.
The fourth step is specifically division of subareas: adjusting the topological structure of the ultrasonic transducer 1: base groupAdjusting the array topology structure of the ultrasonic transducer 1 according to the error distribution of the reconstruction field and the model field; dynamic adjustment based on feedback mechanism: taking reconstruction error as a target variable, and the radius R of the inner circle of the subarea ir And the radial angle theta is used as a control variable to determine the optimal inner ring radius R ir And a radial angle θ.
The array topology structure of the ultrasonic transducer 1 is broadcast, and particularly the ultrasonic transducer 1 is uniformly arranged on the section of a pipeline to be tested.
Further, the fitting of the step 3 is a machine learning method adopting a combination of individual learner learning and individual learner strategies of more than two ultrasonic transmission characteristics. The invention divides experimental data for ultrasonic transmission time calibration into a training set, a verification set and a test set, wherein the training set is used for model training, and parameters such as weight, bias and the like are optimized; the verification set is used for optimizing super parameters, learning rate, regularization coefficient and the like; the test set is used to evaluate the generalization ability of the model. And determining an optimal data classification method, an individual learner model and an integration strategy through experiments, and finally obtaining a mapping relation description model.
The integration is carried out by adopting an average method, including a simple average method, a weighted average method and the like; when the data volume is large, the 'learning method' can be adopted for integration, namely, a new learner is used for combining a plurality of individual learners, so that the learning accuracy is improved. And finally, testing the obtained model with the minimum root mean square error by using a test set to obtain the optimal model.
According to the method for calculating the temperature flow of the nuclear reactor main pipeline coolant based on the ultrasonic array, disclosed by the invention, the measuring accuracy of the transducer group is improved by eliminating the influence factors of ultrasonic waves in a nonaqueous medium; and calculating the temperature and the flow velocity under the input parameters through the CFD and the neural network model, performing variance calculation with the temperature and the flow velocity measured by the transducer group, and further improving the measurement accuracy through adjusting the topological structure of the transducer group. The remote server is connected with the transducer group, so that real-time monitoring of measurement is realized.
Of course, a person skilled in the art shall make various corresponding changes and modifications according to the present invention without departing from the spirit and the essence of the invention, but these corresponding changes and modifications shall fall within the protection scope of the appended claims.
Claims (8)
1. The utility model provides a nuclear reactor trunk line coolant temperature flow's calculation method based on ultrasonic array, the pipeline outside is provided with ultrasonic transducer group, transducer group passes through network module and connects remote server, its characterized in that:
and (3) establishing a model field: establishing a complete continuous main pipeline coolant space-time domain temperature layered diffusion model and a flow velocity distribution model under input parameters through a CFD (computational fluid dynamics) construction model;
secondly, establishing a relation description model of three elements of ultrasonic wave propagation speed and coolant;
triplet field measurement and calculation: measuring the flying time of each effective acoustic wave propagation path; the network module transmits signals of the transducer group to a remote server through the network module; reconstructing the real-time temperature distribution and flow velocity distribution of the coolant in the pipeline; calculating average temperature and flow;
and the four-way reconstruction field and the model field are used for carrying out feedback setting on the topological structure of the ultrasonic transducer group.
2. The method for calculating coolant temperature flow for a nuclear reactor main pipe based on ultrasonic arrays according to claim 1, wherein: the third step comprises ultrasonic transmission time calibration, including simulating an experimental platform and acquiring experimental data; the simulation experiment platform comprises a pipeline capable of controlling temperature and an ultrasonic transducer arranged on the outer side of the pipeline.
3. The method for calculating coolant temperature flow for a nuclear reactor main pipe based on ultrasonic arrays according to claim 2, wherein: the ultrasonic transmission time calibration specifically comprises the following steps:
1 in the empty pipe condition and temperature T 0 Under this condition, τ at this temperature is obtained by changing the ultrasonic propagation path 0 Results of (2);
2 changing the temperature, repeating the experiment to calculate tau at different temperatures 0 Results of (2);
fitting the experimental calculation results to obtain a description model tau between the time delay and the temperature of the ultrasonic wave in the nonaqueous medium 0 ＝F(T)。
4. A method of calculating coolant temperature flow for a nuclear reactor main pipe based on an ultrasonic array according to claim 3, wherein: the second step is specifically as follows: (1) Establishing a relation between ultrasonic sound velocity and coolant; (2) simulating an experimental platform and acquiring experimental data; (3) Establishing a relation description model between ultrasonic sound velocity and coolant temperature, pressure and concentration by adopting a machine learning method; the simulation experiment platform also comprises a constant temperature tank with a water inlet, a water outlet and a concentration control device and a metal pipeline for controlling pressure; the ultrasonic transducer is connected with the micro-control device.
5. A method of calculating coolant temperature flow for a nuclear reactor main pipe based on an ultrasonic array according to claim 3, wherein: the fourth step is specifically division of subareas:
adjusting the topological structure of the ultrasonic transducer: adjusting the topology of the ultrasonic transducer array based on the error distribution of the reconstructed field and the model field;
dynamic adjustment based on feedback mechanism: taking reconstruction error as a target variable, and the radius R of the inner circle of the subarea ir And the radial angle theta is used as a control variable to determine the optimal inner ring radius R ir And a radial angle θ.
6. The method of calculating coolant temperature flow for a nuclear reactor main pipe based on ultrasonic arrays of claim 5, wherein: the ultrasonic transducer array topological structure is broadcast, and particularly the ultrasonic transducers are uniformly arranged on the section of the pipeline to be tested.
7. The method of calculating coolant temperature flow for a nuclear reactor main pipe based on an ultrasonic array of claim 6, wherein: the first specific step is as follows:
calculating the temperature and flow rate data of the coolant of the typical section of the pipeline under the set parameters by calculating the fluid mechanics CFD;
b training CFD simulation results by using the neural network model: training the simulation result in the step a through a neural network model;
c prediction result: and c, predicting and obtaining continuous and complete temperature and flow velocity distribution of the coolant in the pipeline by using the neural network model in the step b.
8. A method of calculating coolant temperature flow for a nuclear reactor main pipe based on an ultrasonic array according to claim 3, wherein: the fitting of the step 3 is a machine learning method combining individual learner learning and individual learner strategies by adopting more than two ultrasonic transmission characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111251821.9A CN114068051B (en) | 2021-10-26 | 2021-10-26 | Method for calculating temperature flow of nuclear reactor main pipeline coolant based on ultrasonic array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111251821.9A CN114068051B (en) | 2021-10-26 | 2021-10-26 | Method for calculating temperature flow of nuclear reactor main pipeline coolant based on ultrasonic array |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114068051A CN114068051A (en) | 2022-02-18 |
| CN114068051B true CN114068051B (en) | 2023-06-16 |
Family
ID=80235653
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111251821.9A Active CN114068051B (en) | 2021-10-26 | 2021-10-26 | Method for calculating temperature flow of nuclear reactor main pipeline coolant based on ultrasonic array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114068051B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113947045A (en) * | 2021-10-26 | 2022-01-18 | 四川大学 | Method for calculating pipeline flow based on setting of topological structure of neural network and energy converter |
| CN113990539B (en) * | 2021-10-26 | 2023-05-23 | 四川大学 | Ultrasonic measurement reconstruction method for temperature distribution of coolant in hot section of main pipeline of nuclear reactor |
| CN114877954B (en) * | 2022-07-12 | 2022-09-23 | 杭州春来科技有限公司 | Fixed pollution source measuring method and system |
| CN117057212B (en) * | 2023-06-05 | 2024-05-17 | 四川大学 | Acoustic reconstruction method for dynamic temperature field of nuclear power device |
| CN117473873B (en) * | 2023-11-13 | 2024-04-26 | 上海交通大学 | Nuclear thermal coupling realization method based on DeepM & Mnet neural network |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4318776A (en) * | 1978-07-21 | 1982-03-09 | Kraftwerk Union Aktiengesellschaft | Measuring lance for boiling-water nuclear reactors |
| GB8529446D0 (en) * | 1985-11-29 | 1986-01-08 | Univ Aberdeen | Divergent ultrasound arrays |
| EP2657658A1 (en) * | 2012-04-25 | 2013-10-30 | General Electric Company | Ultrasonic flow measurement system |
| WO2014166847A1 (en) * | 2013-04-10 | 2014-10-16 | Areva Np | Methods for simulating the flow of a fluid in a vessel of a nuclear reactor and for calculating the mechanical deformation of assemblies of a nuclear reactor core, and associated computer program products |
| KR101776567B1 (en) * | 2017-01-11 | 2017-09-11 | 대한민국 | Method for improving temperature stratification in order to efficient heating and cooling and energy saving in buildings and heating control system using thereof |
| CN207132929U (en) * | 2017-08-11 | 2018-03-23 | 北京清环智慧水务科技有限公司 | A kind of flow probe |
| CN108154672A (en) * | 2018-02-07 | 2018-06-12 | 天津中德应用技术大学 | System and method is configured in a kind of ultrasonic liquid flow meter Automatic parameter |
| CN208061413U (en) * | 2018-02-07 | 2018-11-06 | 天津中德应用技术大学 | A kind of ultrasonic liquid flow meter Automatic parameter configuration system |
| CN110268478A (en) * | 2016-12-15 | 2019-09-20 | 斯特凡Tto有限公司 | The method and process of subject's specificity computation model of decision support and diagnosis for cardiovascular disease are provided |
| CN110618197A (en) * | 2019-02-01 | 2019-12-27 | 中国石油化工股份有限公司 | Long-distance oil and gas pipeline full-defect combined ultrasonic array nondestructive scanning method |
| CN113001070A (en) * | 2015-03-26 | 2021-06-22 | 克里凯文斯管线国际有限公司 | Pipeline cooling system |
| CN113375828A (en) * | 2021-05-10 | 2021-09-10 | 电子科技大学 | Temperature field reconstruction method based on ultrasonic waves |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8249825B2 (en) * | 2009-05-08 | 2012-08-21 | American Power Conversion Corporation | System and method for predicting cooling performance of arrangements of equipment in a data center |
| US11011280B2 (en) * | 2016-03-10 | 2021-05-18 | Westinghouse Electric Company Llc | Reactor coolant system piping temperature distribution measurement system |
-
2021
- 2021-10-26 CN CN202111251821.9A patent/CN114068051B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4318776A (en) * | 1978-07-21 | 1982-03-09 | Kraftwerk Union Aktiengesellschaft | Measuring lance for boiling-water nuclear reactors |
| GB8529446D0 (en) * | 1985-11-29 | 1986-01-08 | Univ Aberdeen | Divergent ultrasound arrays |
| EP2657658A1 (en) * | 2012-04-25 | 2013-10-30 | General Electric Company | Ultrasonic flow measurement system |
| WO2014166847A1 (en) * | 2013-04-10 | 2014-10-16 | Areva Np | Methods for simulating the flow of a fluid in a vessel of a nuclear reactor and for calculating the mechanical deformation of assemblies of a nuclear reactor core, and associated computer program products |
| CN113001070A (en) * | 2015-03-26 | 2021-06-22 | 克里凯文斯管线国际有限公司 | Pipeline cooling system |
| CN110268478A (en) * | 2016-12-15 | 2019-09-20 | 斯特凡Tto有限公司 | The method and process of subject's specificity computation model of decision support and diagnosis for cardiovascular disease are provided |
| KR101776567B1 (en) * | 2017-01-11 | 2017-09-11 | 대한민국 | Method for improving temperature stratification in order to efficient heating and cooling and energy saving in buildings and heating control system using thereof |
| CN207132929U (en) * | 2017-08-11 | 2018-03-23 | 北京清环智慧水务科技有限公司 | A kind of flow probe |
| CN108154672A (en) * | 2018-02-07 | 2018-06-12 | 天津中德应用技术大学 | System and method is configured in a kind of ultrasonic liquid flow meter Automatic parameter |
| CN208061413U (en) * | 2018-02-07 | 2018-11-06 | 天津中德应用技术大学 | A kind of ultrasonic liquid flow meter Automatic parameter configuration system |
| CN110618197A (en) * | 2019-02-01 | 2019-12-27 | 中国石油化工股份有限公司 | Long-distance oil and gas pipeline full-defect combined ultrasonic array nondestructive scanning method |
| CN113375828A (en) * | 2021-05-10 | 2021-09-10 | 电子科技大学 | Temperature field reconstruction method based on ultrasonic waves |
Non-Patent Citations (4)
| Title |
|---|
| Ultrasonic Thermometry Algorithm Based on Inverse Quadratic Function;Xinzhi Zhou;《IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control》;第第68卷卷(第第5期期);全文 * |
| 反应堆冷却剂系统不可隔离滞流分支管热疲劳瞬态的CFD研究;秦洁;《核动力工程》;第第32卷卷(第第S1期期);全文 * |
| 基于TDC7200的超声测温系统设计;崔岢;《仪表技术与传感器》;第第439卷卷(第第8期期);全文 * |
| 高温铸锻件内部温度场电磁超声重构方法研究;石文泽;《仪器仪表学报》;第第41卷卷(第第8期期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114068051A (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114068051B (en) | Method for calculating temperature flow of nuclear reactor main pipeline coolant based on ultrasonic array | |
| CN113990539B (en) | Ultrasonic measurement reconstruction method for temperature distribution of coolant in hot section of main pipeline of nuclear reactor | |
| CN105444923A (en) | Mechanical temperature instrument error prediction method based on genetic-algorithm optimized least square support vector machine | |
| CN113947045A (en) | Method for calculating pipeline flow based on setting of topological structure of neural network and energy converter | |
| Wang et al. | Data-driven CFD modeling of turbulent flows through complex structures | |
| CN105258839A (en) | Array type air pressure measurement compensation device and method based on quantum particle swarm wavelet neural network | |
| CN102419827B (en) | Radial basis function (RBF) neural network-based boiling heat exchanging prediction method | |
| Sridharan | Applications of artificial intelligence techniques in heat exchanger systems | |
| CN113551744A (en) | Ultrasonic flowmeter performance online monitoring method and system | |
| CN103617563B (en) | A kind of water supply network theoretical based on geo-statistic spatial analysis is without monitoring node pressure determination statement | |
| Jing et al. | Investigation of open channel flow with unsubmerged rigid vegetation by the lattice Boltzmann method | |
| Ali et al. | Prediction of discharge through a sharp-crested triangular weir using ANN model trained with Levenberg–Marquardt algorithm | |
| Kémajou et al. | Application of artificial neural network for predicting the indoor air temperature in modern building in humid region | |
| Bin et al. | Progressive, extrapolative machine learning for near-wall turbulence modeling | |
| Al-Salaymeh | Optimization of hot-wire thermal flow sensor based on a neural net model | |
| Zhou et al. | Leakage diagnosis of heating pipe-network based on BP neural network | |
| CN117272832B (en) | Post-dry out heat transfer area simulation method based on neural network | |
| CN114239397A (en) | Soft measurement modeling method based on dynamic feature extraction and local weighted deep learning | |
| Zhang et al. | Optimization of transducer array for cross-sectional velocity field reconstruction in acoustic tomography | |
| Sestito et al. | Vibration-based multiphase-flow pattern classification via machine learning techniques | |
| CN114216582A (en) | High-precision rapid temperature field reconstruction method, system, equipment and terminal | |
| Chen et al. | Effect of transducer shadowing of ultrasonic anemometers on wind velocity measurement | |
| Wang et al. | Thermal performance and evaluation of a novel stratified and mixed flexible transformation solar heat storage unit | |
| CN114462336B (en) | Method for calculating average temperature of coolant of main pipeline of nuclear reactor | |
| CN116595845A (en) | Structure optimization method of spiral fin tube type heat exchanger |
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 |