CN109557126A - A kind of measuring device of soil thermal property parameter and seepage parameters - Google Patents
A kind of measuring device of soil thermal property parameter and seepage parameters Download PDFInfo
- Publication number
- CN109557126A CN109557126A CN201910097451.4A CN201910097451A CN109557126A CN 109557126 A CN109557126 A CN 109557126A CN 201910097451 A CN201910097451 A CN 201910097451A CN 109557126 A CN109557126 A CN 109557126A
- Authority
- CN
- China
- Prior art keywords
- soil
- parameter
- neural network
- property parameter
- seepage
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N25/00—Investigating or analyzing materials by the use of thermal means
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/08—Investigating permeability, pore-volume, or surface area of porous materials
Abstract
The present invention relates to earth source heat pump design fields, more particularly to a kind of measuring device of soil thermal property parameter and seepage parameters, neural network parameter discrimination method is used in the measuring device to recognize soil thermal property parameter and seepage parameters, soil progress thermal response experiment test to be identified in identification process, measurement parameter after obtaining the thermal response experiment test of soil to be identified, it is input in the neural network that training obtains in advance again, can obtain the thermal physical property parameter and seepage parameters of the soil to be identified.The measuring device can at the scene thermal response test when directly soil thermal property parameter and seepage parameters are recognized and show result, process is quick and precisely, practicability is stronger, used discrimination method passes through the mapping relations of live thermal response experiment and neural network thermal response measurement parameter and Soil Thermal physical property, seepage direction and speed, avoid the precision that the identification of Soil Thermal physical property is improved using traditional identification model bring Identification Errors.
Description
Technical field
The present invention relates to earth source heat pump design field more particularly to a kind of soil thermal property parameters and seepage parameters
Measuring device.
Background technique
Ground-source heat pump system operational efficiency is high, has good energy saving and environment friendly, ground heat exchanger conduct
Earth source heat pump important component, the superiority and inferiority of heat exchange property directly affect the efficiency of ground-source heat pump system, runnability with
And economic benefit.Soil thermal property parameter be ground heat exchanger design when important parameter, value effect Drilling Design
Quantity and depth, and then therefore how the initial cost and runnability that influence heat pump system accurately obtain soil thermal property parameter
Become particularly significant.Current Soil Thermal physical property identification is to solve soil using heat transfer model on the basis of geo-thermal response test
The indirect problem of thermal physical property parameter, this process have the following deficiencies: firstly, using existing identification model (line source model and
Cylindrical heat source model) when carrying out the identification of hot physical property, since many aspects of model are there are Utopian construction, brought to identification result
More or less error, so that identification precision is affected, although improved identification model precision increases at present, complexity
Also it greatly increases, and is difficult to apply in Practical Project;Secondly, the required measurement parameter of existing thermal response experiment is only limited to
Water temperature is imported and exported in underground pipe, measurement form is relatively simple;Simultaneously as having water-bearing layer in stratum, seepage action of ground water can band
The energy of soil accumulation is walked, improves the heat exchange property between underground pipe and soil, in engineering when considering the influence of seepage flow factor,
Often by its equivalent increment at soil thermal conductivity, but this hypothesis be it is inaccurate, in long-term running underground pipe pipe
Design error in group is particularly evident, therefore proposes soil thermal property parameter, seepage flow side both with a kind of accuracy and practicability
It is had a very important significance to the discrimination method with speed and the measuring device based on this method.
Summary of the invention
(1) technical problems to be solved
In order to solve the above problem of the prior art, the present invention provides a kind of soil thermal property parameter and seepage parameters
Measuring device, the neural network parameter discrimination method used in the device combine thermal response simulation test platform and intelligent algorithm
Get up, establishes the Nonlinear Mapping relationship of different thermal response experiment parameter and soil thermal property parameter, seepage parameters.
(2) technical solution
In order to achieve the above object, the main technical schemes that the present invention uses are as follows:
A kind of measuring device of soil thermal property parameter and seepage parameters, including data acquisition module, identification module and defeated
Module out;
The data acquisition module includes temperature sensor, information acquiring instrument, for acquiring the configuration in thermal response experiment
Parameter;
After the identification module receives the parameter from data acquisition module, according to neural network parameter discrimination method into
Row calculates;
The output module is used to export and show the thermal physical property parameter for the soil to be identified that recognized module is calculated
With the result of seepage parameters.
According to the present invention, the neural network identification method, comprising the following steps:
S1, live thermal response experiment test is carried out to soil to be identified, obtains soil thermal property parameter and seepage flow to be identified
Measurement parameter needed for parameter identification;
S2, soil thermal property parameter and seepage parameters identification neural network are established, soil thermal property parameter and seepage flow is joined
Measurement parameter needed for number identification is input in the neural network that training obtains in advance, obtains the hot physical property of the soil to be identified
Parameter and seepage parameters;
According to the present invention, the neural network that the preparatory training in the step S2 obtains is to obtain as follows:
S21, high-precision three-dimensional full size Simulation Experimental Platform is established;
S22, Simulation Experimental Platform is verified and is corrected by experiment condition controllable comparative experiments;
S23, thermal response experiment is carried out using the high-precision three-dimensional full size Simulation Experimental Platform of foundation, obtains training data
Library, and the neural network established using tranining database training, reach good identification precision.
According to the present invention, the neural network that the preparatory training obtains is obtained using the controllable emulation experiment of experiment condition
Training data be trained, the experiment condition controllably refers to the soil used in experiment to be common each in engineering
Seed type soil, and heating power, recirculated water flow velocity is controllable, to ensure the versatility and accuracy of model.
According to the present invention, the step S22 the following steps are included:
S21: true soil thermal property parameter and seepage parameters are measured using high-precision measuring method;
S22: it is complete with high-precision three-dimensional that the true soil thermal property parameter and seepage parameters are separately input to comparative experiments
Scale Simulation Experimental Platform thermal response experiment in, and respectively obtain comparative experiments and emulation experiment soil thermal property parameter and
Measurement parameter needed for seepage parameters identification;
S23: survey needed for comparing true soil thermal property parameter and the seepage parameters identification of comparative experiments and emulation experiment
Parameter is measured, high-precision three-dimensional full size Simulation Experimental Platform is verified and corrected.
According to the present invention, the high-precision three-dimensional full size Simulation Experimental Platform includes: three-dimensional full size Numerical Heat Transfer mould
Pattern block, numerical algorithm solve module, Visualized Post Processing module.
According to the present invention, the neural network that the preparatory training obtains includes input unit, hidden unit and output unit,
Wherein the layer where input unit is input layer, and the layer where hidden unit is hidden layer, and the layer where output unit is output
Layer.
According to the present invention, the training of soil thermal property parameter and seepage parameters identification neural network is made using tranining database
For the training signal and error signal of neural network, thermal response experiment in the Simulation Experimental Platform in tranining database is obtained
Input terminal of soil thermal property parameter and the seepage parameters measurement parameter required when recognizing as neural network, in tranining database
True soil thermal property parameter and seepage parameters as output end, and determine the weight and threshold value of neural network;
It calculates each hidden unit of hidden layer by transmission function using input unit each in input layer and exports accordingly;
Using the output of each hidden unit of hidden layer, the input of each output unit of output layer is calculated, by transmission function,
Calculate the corresponding output of each output unit of output layer;
According in tranining database soil thermal property parameter and seepage parameters recognize taken measurement parameter and nerve
The reality output of network calculates the training error of each unit of output layer, and the power of neural network is corrected using the training error
Value and threshold value;
When the error of fitting of neural network is less than set threshold value, trained neural network is obtained.
According to the present invention, it uses " tansing " function as transmission function in the training process of the neural network, uses
" learngdm " function is as learning function, using " traingd " function as training function, evaluates mind using " mse " function
Performance through network.
According to the present invention, the thermal physical property parameter of the soil to be identified includes: soil thermal conductivity, heat capacity of volume parameter;
The seepage parameters include: seepage direction and percolation flow velocity;
It is described to carry out measurement parameter required when soil thermal property parameter and seepage parameters recognize to soil to be identified and include:
Ground heat exchanger inlet water temperature, and/or, exit water temperature, and/or, underground pipe tube wall temperature, and/or, drill wall temperature, and/
Or, bore inner backfilling material temperature.
(3) beneficial effect
Compared with prior art, the invention has the benefit that
The present invention built can at the scene thermal response test when directly soil thermal property parameter and seepage parameters are distinguished
Know and show the measuring device of result, process is quick and precisely, practical, and discrimination method employed in measuring device passes through existing
Field thermal response experiment and the mapping of neural network thermal response measurement parameter and Soil Thermal physical property, seepage direction and speed are closed
System avoids the precision that the identification of Soil Thermal physical property is improved using traditional identification model bring Identification Errors, and avoiding makes
With traditional identification model bring Identification Errors, and neural network itself also has certain fault-tolerance, while in the invention
The required measurement parameter of thermal response experiment is not limited solely to underground pipe inlet and outlet water temperature, can also be real using different thermal responses
Required measurement parameter is tested to recognize soil thermal property parameter and seepage parameters.
Detailed description of the invention
Fig. 1 is measuring device schematic diagram of the invention;
Fig. 2 is flow diagram of the invention.
[description of symbols]
1, soil to be measured;2, backfilling material;3, U-shaped underground pipe;4, temperature measuring device;5, flowmeter;6, water circulating pump;
7, electric heating water tank;8, pressure measuring unit.
Specific embodiment
In order to preferably explain the present invention, in order to understand, with reference to the accompanying drawing, by specific embodiment, to this hair
It is bright to be described in detail.
The invention proposes a kind of soil thermal property parameter based on neural network parameter discrimination method and seepage parameters
Measuring device, the measuring device include data acquisition module, identification module and output module.
As shown in Figure 1, in live thermal response experiment, including soil to be measured 1, backfilling material 2, U-shaped underground pipe 3, temperature are surveyed
Device 4, flowmeter 5, water circulating pump 6, electric heating water tank 7 and pressure measuring unit 8 are measured,
Carry out ground heat exchanger thermal response experiment when, need U-shaped underground pipe 3 entrance setting for test into
The first temperature sensor of water temperature, is arranged the second temperature for testing exit water temperature in the exit of U-shaped underground pipe 3 at mouthful
The flowmeter 5 for detecting liquid volume flow is arranged in sensor between U-shaped underground pipe 3 and water circulating pump 6, and U-shapedly
Electric heating water tank 7 is set between the inlet pipeline and water circulating pump 6 of pipe laying 3, electric heater is set in the electric heating water tank 7, it is right
The heating power of electric heater measures, and is enabled by electric heating water tank 7 full of water in entire circuit, and connects with measuring device
It connects, circulation loop is heated with constant power after underground temperature field restores substantially, water is allowed to circulate in the loop,
Measuring device is measured and is acquired to temperature each in circuit, wherein data acquisition module includes temperature sensor, information collection
Instrument, be mainly used for acquire thermal response experiment in each measurement parameter temperature, data acquisition module respectively with temperature measuring device 4,
Soil 1 to be measured, backfilling material 2, U-shaped underground pipe 3 connect, by live thermal response test measurement ground heat exchanger inlet water temperature,
Exit water temperature, underground pipe tube wall temperature, drilling wall temperature, bore inner backfilling material temperature thermo-coupler are respectively connected to measurement dress
In the data acquisition module set, the measurement parameter that data acquisition module tests thermal response is extracted.
After identification module is then the data conveyed in receiving data acquisition module, parameter each in data is carried out
Identification, specific parameter identification method, by this discrimination method, can be adopted in data made of being compiled according to neural network
Collect and the soil thermal property parameter of soil to be identified and the identification result of seepage parameters is calculated in the data basis of module, and most
The soil pool thermal physical property parameter and seepage parameters are exported and shown by output module eventually.
The parameter identification method combines thermal response emulation experiment with nerve network system, establishes different heat and rings
Answer the Nonlinear Mapping relationship of experiment parameter and soil thermal property parameter and seepage parameters, it is therefore an objective to improve soil thermal property parameter
With accuracy and practicability of the seepage parameters in identification process, the specific steps are as follows:
S1, live thermal response experiment test is carried out to soil to be identified, and obtains soil thermal property parameter and infiltration to be identified
Measurement parameter needed for flowing parameter identification;
S2, soil thermal property parameter and seepage parameters identification neural network are established, needed for soil thermal property parameter is recognized
Measurement parameter be input in the obtained neural network of training in advance, obtain the thermal physical property parameter and seepage flow of the soil to be identified
Parameter.
The neural network that the preparatory training obtains is the training data obtained using the controllable emulation experiment of experiment condition
It is trained.
The experiment condition controllably refers to that the soil that uses is for various types soil common in engineering in experiment, and heating
Power, recirculated water flow velocity, percolation flow velocity is controllable, to ensure the versatility and accuracy of model.
When train in advance to neural network in step S2, it is flat high-precision three-dimensional full size emulation experiment has been initially set up
Platform, emulation experiment are effects real with simulation software simulation on computers, which passes through computer program
It realizes, specifically includes three-dimensional full size Numerical Heat Transfer model module, numerical algorithm solver module, Visualized Post Processing device mould
Block;Thermal response experimental principle is consistent with live thermal response experimental principle in experiment porch, inputs in Simulation Experimental Platform different
Soil thermal property parameter and seepage parameters, after appliance computer carries out simulation calculating, can be obtained with input Soil Thermal object
Property parameter and the corresponding soil thermal property parameter of seepage parameters and seepage parameters identification needed for measurement parameter.
In order to ensure the accuracy of Simulation Experimental Platform, Simulation Experimental Platform need to be verified and be repaired by comparative experiments
Just, wherein the underlying parameter and heat transfer boundary condition of comparative experiments and Simulation Experimental Platform are consistent, in order to ensure leading to for model
With property and accuracy, guarantee the soil used in comparative experiments for various types soil common in engineering.
The thermal physical property parameter of true soil is measured with high-precision measuring method in advance first, then the item controllable in parameter
It is applied to progress thermal response experiment in comparative experiments under part, show that confirmatory experiment carries out soil thermal property parameter and seepage parameters are distinguished
Identical true soil thermal property parameter and seepage parameters are then input to emulation experiment again by required measurement parameter when knowledge
Simulation calculating is carried out in platform, obtains measurement ginseng required when Simulation Experimental Platform carries out thermal physical property parameter and seepage parameters identification
Number, the measurement parameter that the measurement parameter and comparative experiments that contrast simulation experiment porch obtains obtain, it is ensured that Simulation Experimental Platform
Accuracy.
Soil thermal property parameter includes soil thermal conductivity, heat capacity of volume;
Seepage parameters include: seepage direction and percolation flow velocity;
The measurement parameter of thermal response experiment includes ground heat exchanger inlet water temperature and/or exit water temperature and/or underground pipe
Tube wall temperature and/or drilling wall temperature and/or bore inner backfilling material temperature and/or be the above measurement parameter combination.
According to correlation theory, heat transfer process is complicated between soil, and influence factor is more, therefore is tested using thermal response to Soil Thermal
When physical parameter and seepage parameters carry out inverting, the relationship between measurement parameter and thermal physical property parameter and seepage parameters is non-parsing
, accurate result is hardly resulted in by general fitting of a polynomial or other fitting difference approach, under relatively, BP neural network
With very strong non-linear mapping capability, the network hidden layer and transmission function of BP neural network are reasonably designed, it can be with
Meet the needs of Soil Thermal physical property and seepage parameters modeling and data fitting.
The present invention utilizes the nonlinear fitting and predictive ability of BP neural network, by using a large amount of tranining databases to net
Network is trained, then the measurement parameter that thermal response is tested is input to trained BP neural network is i.e. predictable to be obtained accordingly
The soil thermal property parameter and seepage parameters for the treatment of survey soil.
It in step s 2, further include establishing neural metwork training database, database includes true soil thermal property parameter
And seepage parameters, and it is corresponding with the two parameter high-precision three-dimensional full size Simulation Experimental Platform carry out thermal response experiment obtain
Measurement parameter needed for soil thermal property parameter and the seepage parameters identification obtained, establishment process specifically: utilize high-precision three-dimensional
Full size Simulation Experimental Platform inputs different true soil thermal property parameter and seepage parameters respectively in platform, carries out heat
Response experiment, by platform simulation calculate, can be obtained it is corresponding with different soils thermal physical property parameter and seepage parameters
Measurement parameter needed for soil thermal property parameter and seepage parameters identification process, and neural metwork training data are created accordingly
Library.Wherein, the true soil thermal property parameter of input and seepage parameters range need to cover various common geological conditions, due to imitative
The number of true experiment can guarantee the accuracy of the discrimination method, therefore emulation thermal response experiment number is more, can more guarantee to instruct
Practice the accuracy of database.
Under normal conditions, neural network is by input unit, hidden unit and output unit composition, wherein where input unit
Layer be input layer, layer where hidden unit is hidden layer, and the layer where output unit is output layer.All units are all with two
Unit connection in boundary layer, input information can initially enter the input layer in left side, intermediate hidden layer then be excited, finally from the right side
Output layer output in side is as a result, neural network can be by learning the dynamics that adjustment every two unit connects gradually.
When being trained to BP neural network, using the measurement parameter in tranining database as input, with corresponding soil
Thermal physical property parameter and seepage parameters are as output.
In step s 2, neural metwork training database is as the training signal and error signal of neural network to nerve net
Network is trained, using emulation thermal response experiment measurement parameter as the input terminal of neural network, true Soil Thermal in training process
Physical parameter and seepage parameters determine the weight and threshold value of neural network as output end;
It is handled between each layer by transmission function, utilizes unit a in input signal, weight and threshold calculations hidden layer
Then the input of each unit is used in input, by transmission function, calculate hidden layer each unit and export accordingly, utilize hidden layer
Output, the input of weight and threshold calculations output layer each unit calculate the corresponding defeated of output layer each unit by transmission function
Out;
The difference of the reality output of theory output and network, the i.e. reality of the measurement parameter and neural network of emulation thermal response experiment
The difference of border output, the error are the training error of output layer each unit;
Using connection weight, the output of the training error of output layer each unit and hidden layer, the training error of hidden layer is calculated;
The weight and threshold value of neural network are corrected using the output of the training error and hidden layer of output layer each unit;
When the error of fitting of neural network is less than set threshold value, neural network learning terminates, and training can be obtained
Neural network afterwards, the threshold value can be set according to the actual situation.
In entire neural network training process, using " tansing " function, (tanh type transmits letter to neural network
Number) it is used as transmission function, learning function is used as using " learngdm " function (gradient declines momentum learning function), is used
" traingd " function (gradient descent algorithm) is trained BP neural network as training function, is evaluated using " mse " function
The performance of neural network.Wherein, training function includes learning function, and learning function is a part for belonging to trained function;Training
Function carries out local directed complete set to weight and threshold value to weight and threshold value global adaptation, learning function;From error, training function
It is minimum to generally error, learning function error for single neural unit is minimum.
After completing neural metwork training, thermal response experiment test is carried out on soil to be identified, by the Soil Thermal of acquisition
Measurement parameter needed for physical parameter and seepage parameters identification process is input in the neural network after training to get to wait distinguish
Know the soil thermal property parameter and seepage parameters of soil.
The specific embodiment of entire soil thermal property parameter and seepage parameters identification process are as follows:
Measurement ground heat exchanger inlet water temperature, exit water temperature, underground pipe tube wall temperature, drilling are tested into live thermal response
Wall temperature, bore inner backfilling material temperature thermo-coupler are respectively connected to the corresponding of the data acquisition module in measuring device and insert
Kong Zhong, the device start to extract the measurement parameter that thermal response is tested;The data of extraction are then transmitted to identification module,
Measurement parameter can be input to BP neural network after training by the identification module based on the compiling of this discrimination method, according to corresponding
Neural network mapping relations are calculated, BP neural network output final output soil thermal property parameter and seepage flow ginseng after training
Number, and soil thermal property parameter and seepage parameters are shown by output module.
The geo-thermal response test parameter and soil thermal property parameter got up by live thermal response experiment and neural network,
The mapping relations of seepage parameters are avoided using tradition identification model bring Identification Errors, and neural network itself also has
Certain fault-tolerance, meanwhile, the measuring device based on this method can be at the scene in thermal response experimentation directly to Soil Thermal
Physical parameter and seepage parameters are recognized, and more succinct convenience is operated in engineering, practical.
It is to be appreciated that describing the skill simply to illustrate that of the invention to what specific embodiments of the present invention carried out above
Art route and feature, its object is to allow those skilled in the art to can understand the content of the present invention and implement it accordingly, but
The present invention is not limited to above-mentioned particular implementations.All various changes made within the scope of the claims are repaired
Decorations, should be covered by the scope of protection of the present invention.
Claims (10)
1. the measuring device of a kind of soil thermal property parameter and seepage parameters, which is characterized in that the device includes data acquisition module
Block, identification module and output module;
The data acquisition module includes temperature sensor, information acquiring instrument, for acquiring the configuration parameter in thermal response experiment;
After the identification module receives the parameter from data acquisition module, counted according to neural network parameter discrimination method
It calculates;
The output module is used to export and show the soil thermal property parameter for the soil to be identified that recognized module is calculated
With the result of seepage parameters.
2. measuring device according to claim 1, which is characterized in that the neural network identification method, including following step
It is rapid:
S1, live thermal response experiment test is carried out to soil to be identified, obtains soil thermal property parameter and seepage parameters to be identified
Measurement parameter needed for identification;
S2, soil thermal property parameter and seepage parameters identification neural network are established, soil thermal property parameter and seepage parameters is distinguished
Measurement parameter needed for knowing is input in the neural network that training obtains in advance, obtains the thermal physical property parameter of the soil to be identified
And seepage parameters.
3. measuring device according to claim 2, which is characterized in that
The neural network that preparatory training in the step S2 obtains is to obtain as follows:
S21, high-precision three-dimensional full size Simulation Experimental Platform is established;
S22, Simulation Experimental Platform is verified and is corrected by experiment condition controllable comparative experiments;
S23, thermal response experiment is carried out using the high-precision three-dimensional full size Simulation Experimental Platform of foundation, obtains tranining database,
And the neural network established using tranining database training, reach good identification precision.
4. measuring device according to claim 3, which is characterized in that
The neural network that the preparatory training obtains is that the training data obtained using the controllable emulation experiment of experiment condition is carried out
Training obtains, the experiment condition controllably refer to the soil used in experiment for various types soil common in engineering, and
Heating power, recirculated water flow velocity is controllable, to ensure the versatility and accuracy of model.
5. experimental provision according to claim 3, which is characterized in that
The step S22 the following steps are included:
S21: true soil thermal property parameter and seepage parameters are measured using high-precision measuring method;
S22: the true soil thermal property parameter and seepage parameters are separately input to comparative experiments and high-precision three-dimensional full size
In the thermal response experiment of Simulation Experimental Platform, and respectively obtain the soil thermal property parameter and seepage flow of comparative experiments and emulation experiment
Measurement parameter needed for parameter identification;
S23: measurement needed for comparing true soil thermal property parameter and the seepage parameters identification of comparative experiments and emulation experiment is joined
Number, is verified and is corrected to high-precision three-dimensional full size Simulation Experimental Platform.
6. experimental provision according to claim 3, which is characterized in that
The high-precision three-dimensional full size Simulation Experimental Platform includes: three-dimensional full size Numerical Heat Transfer model module, numerical algorithm
Solve module, Visualized Post Processing module.
7. experimental provision according to claim 3, which is characterized in that
The neural network that the preparatory training obtains includes input unit, hidden unit and output unit, wherein input unit institute
Layer be input layer, layer where hidden unit is hidden layer, and the layer where output unit is output layer.
8. experimental provision according to claim 7, which is characterized in that
The training of soil thermal property parameter and seepage parameters identification neural network uses instruction of the tranining database as neural network
Practice signal and error signal, by the soil thermal property parameter that thermal response experiment obtains in the Simulation Experimental Platform in tranining database
Input terminal with required measurement parameter when seepage parameters identification as neural network, the true Soil Thermal object in tranining database
Property parameter and seepage parameters as output end, and determine the weight and threshold value of neural network;
It calculates each hidden unit of hidden layer by transmission function using input unit each in input layer and exports accordingly;
Using the output of each hidden unit of hidden layer, the input of each output unit of output layer is calculated, by transmission function, is calculated
The corresponding output of each output unit of output layer;
According in tranining database soil thermal property parameter and seepage parameters recognize taken measurement parameter and neural network
Reality output, calculate the training error of each unit of output layer, corrected using the training error neural network weight and
Threshold value;
When the error of fitting of neural network is less than set threshold value, trained neural network is obtained.
9. experimental provision according to claim 8, which is characterized in that
Using " tansing " function as transmission function in the training process of the neural network, using " learngdm " function
As learning function, using " traingd " function as training function, using the performance of " mse " function evaluation neural network.
10. experimental provision according to claim 1, which is characterized in that
The thermal physical property parameter of the soil to be identified includes: soil thermal conductivity, heat capacity of volume parameter;
The seepage parameters include: seepage direction and percolation flow velocity;
It is described that measurement parameter required when soil thermal property parameter and seepage parameters recognize is carried out to soil to be identified includes: buried
Heat exchange of heat pipe inlet water temperature, and/or, exit water temperature, and/or, underground pipe tube wall temperature, and/or, drill wall temperature, and/or,
Bore inner backfilling material temperature.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811495562 | 2018-12-07 | ||
CN2018114955622 | 2018-12-07 |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109557126A true CN109557126A (en) | 2019-04-02 |
CN109557126B CN109557126B (en) | 2021-03-26 |
Family
ID=65874002
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910097451.4A Active CN109557126B (en) | 2018-12-07 | 2019-01-31 | Measuring device for soil thermophysical property parameters and seepage parameters |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109557126B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110006803A (en) * | 2019-04-04 | 2019-07-12 | 杨国华 | A kind of device and monitoring method of long-range monitoring seepage action of ground water speed |
CN110515930A (en) * | 2019-09-03 | 2019-11-29 | 清华大学 | Critical, carbon dioxide and hydrogen mixture thermal physical property data library and construction method |
WO2021135719A1 (en) * | 2020-01-03 | 2021-07-08 | 山东天岳先进科技股份有限公司 | Method and apparatus for deducing physical property parameter |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101430317A (en) * | 2008-12-09 | 2009-05-13 | 山东建筑大学 | Rock thermophysics on-site test method under non-constant power condition |
CN102628817A (en) * | 2012-03-30 | 2012-08-08 | 长安大学 | Unattended type soil source heat pump rock soil thermal property tester |
CN103149234A (en) * | 2013-03-20 | 2013-06-12 | 上海理工大学 | Movable ground source heat pump thermal response test instrument and testing method |
CN107153077A (en) * | 2017-05-10 | 2017-09-12 | 安徽新富地能源科技有限公司 | A kind of portable ground temperature energy geo-thermal response test equipment |
-
2019
- 2019-01-31 CN CN201910097451.4A patent/CN109557126B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101430317A (en) * | 2008-12-09 | 2009-05-13 | 山东建筑大学 | Rock thermophysics on-site test method under non-constant power condition |
CN102628817A (en) * | 2012-03-30 | 2012-08-08 | 长安大学 | Unattended type soil source heat pump rock soil thermal property tester |
CN103149234A (en) * | 2013-03-20 | 2013-06-12 | 上海理工大学 | Movable ground source heat pump thermal response test instrument and testing method |
CN107153077A (en) * | 2017-05-10 | 2017-09-12 | 安徽新富地能源科技有限公司 | A kind of portable ground temperature energy geo-thermal response test equipment |
Non-Patent Citations (2)
Title |
---|
何发祥 等: "用BP网络求解土体的导热系数", 《岩土力学》 * |
陈尚沅 等: "基于神经网络的垂直地埋管换热量预测", 《太阳能学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110006803A (en) * | 2019-04-04 | 2019-07-12 | 杨国华 | A kind of device and monitoring method of long-range monitoring seepage action of ground water speed |
CN110515930A (en) * | 2019-09-03 | 2019-11-29 | 清华大学 | Critical, carbon dioxide and hydrogen mixture thermal physical property data library and construction method |
WO2021135719A1 (en) * | 2020-01-03 | 2021-07-08 | 山东天岳先进科技股份有限公司 | Method and apparatus for deducing physical property parameter |
Also Published As
Publication number | Publication date |
---|---|
CN109557126B (en) | 2021-03-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102721722B (en) | In-situ thermal response testing method of stratified thermal properties of underground rock and soil | |
CN109557126A (en) | A kind of measuring device of soil thermal property parameter and seepage parameters | |
CN106017965B (en) | A kind of U-shaped ground heat exchanger heat-moisture transmission performance simulating test device and test method | |
Bauer et al. | Transient 3D analysis of borehole heat exchanger modeling | |
CN104237301B (en) | In-situ thermal response testing method for layered rock and soil thermophysical properties | |
CN206974932U (en) | Ground geo-thermal response test system | |
CN204203709U (en) | A kind of ground thermal property tester based on PLC | |
CN109297077A (en) | A kind of hot heating system cascade utilization of mid-deep strata interference-free rock and monitoring system and method | |
CN203561596U (en) | Buried pipe rock thermal response tester | |
CN206235584U (en) | A kind of experimental system for verifying energy stake Calculation of Heat Transfer model under the conditions of seepage action of ground water | |
CN205879561U (en) | Wet transmission performance simulating measurement setup of U type ground pipe laying heat exchanger heat | |
JP2013238419A (en) | Analysis method and analysis program for thermal response test and pumping test | |
CN204694654U (en) | The hot physical property of combined type ground source heat pump rock-soil and underground pipe heat exchange tester | |
CN103792255B (en) | Rock soil cold and hot response testing system | |
CN102539471A (en) | Well group rechecking testing device and testing method thereof | |
CN109521051A (en) | A kind of soil thermal property parameter discrimination method | |
CN102288637B (en) | Single-hole heat exchange power processing method in ground source heat pump | |
CN104458130B (en) | Method for detecting multiple concentrated leakage positions of rock and earth mass based on ground temperature data | |
CN202599888U (en) | Thermal response testing equipment with double-U heat exchange tube | |
CN202189026U (en) | Thermal response test device | |
CN208206849U (en) | Ground heat exchanger scene thermophysical property measurement instrument calibration equipment Special temperature measurement stick | |
CN202599885U (en) | Thermal response test device | |
CN202599887U (en) | Thermal response test vehicle for double-U heat exchange conduit | |
CN214953183U (en) | Buried hole rock-soil thermal physical property tester | |
CN202101957U (en) | Group Hole Thermal Response Test System |
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 |