CN109557126B - Measuring device for soil thermophysical property parameters and seepage parameters - Google Patents
Measuring device for soil thermophysical property parameters and seepage parameters Download PDFInfo
- Publication number
- CN109557126B CN109557126B CN201910097451.4A CN201910097451A CN109557126B CN 109557126 B CN109557126 B CN 109557126B CN 201910097451 A CN201910097451 A CN 201910097451A CN 109557126 B CN109557126 B CN 109557126B
- Authority
- CN
- China
- Prior art keywords
- parameters
- soil
- seepage
- neural network
- training
- 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
- 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 invention relates to the technical field of ground source heat pump design, in particular to a device for measuring soil thermophysical parameters and seepage parameters. The measuring device can directly identify soil thermophysical property parameters and seepage parameters and display results in a field thermal response experiment, the process is quick and accurate, the practicability is higher, the adopted identification method establishes the mapping relation between the thermal response measurement parameters and the soil thermophysical property, seepage direction and speed through the field thermal response experiment and the neural network, the identification error caused by the use of a traditional identification model is avoided, and the accuracy of soil thermophysical property identification is improved.
Description
Technical Field
The invention relates to the technical field of ground source heat pump design, in particular to a device for measuring soil thermophysical parameters and seepage parameters.
Background
The ground source heat pump system has high operation efficiency, good energy saving performance and environmental friendliness, the buried pipe heat exchanger is used as an important component of the ground source heat pump, and the heat exchange performance of the buried pipe heat exchanger directly influences the energy efficiency, the operation performance and the economic benefit of the ground source heat pump system. The soil thermophysical property parameter is an important parameter when the ground heat exchanger is designed, and the numerical value of the soil thermophysical property parameter influences the design quantity and depth of a drilling well and further influences the initial investment and the operation performance of a heat pump system, so that how to accurately obtain the soil thermophysical property parameter becomes very important. The existing soil thermophysical property identification is an inverse problem of solving soil thermophysical property parameters by applying a heat transfer model on the basis of a thermal response test, and the process has the following defects: firstly, when the existing identification models (a line heat source model and a column heat source model) are used for identifying thermophysical properties, because the models have ideal construction in many aspects, more or less errors are brought to identification results, so that the identification precision is influenced, although the precision of the existing improved identification models is improved, the complexity is greatly increased, and the existing improved identification models are difficult to apply to actual engineering; secondly, the required measurement parameters of the existing thermal response experiment are only limited to the water temperature at the inlet and the outlet of the buried pipe, and the measurement form is single; meanwhile, because the stratum has a water-bearing layer, the energy accumulated by soil can be taken away by groundwater seepage, the heat exchange performance between the buried pipe and the soil is improved, and the seepage factor is often equivalent to the increment of the soil heat conductivity coefficient when the influence of the seepage factor is considered in engineering, but the assumption is inaccurate, and the design error in the buried pipe group which runs for a long time is particularly obvious, so that the method for identifying the soil thermophysical property parameter, the seepage direction and the speed and the measuring device based on the method have very important significance.
Disclosure of Invention
Technical problem to be solved
In order to solve the problems in the prior art, the invention provides a device for measuring soil thermophysical parameters and seepage parameters.
(II) technical scheme
In order to achieve the purpose, the invention adopts the main technical scheme that:
a measuring device for soil thermophysical parameters and seepage parameters comprises a data acquisition module, an identification module and an output module;
the data acquisition module comprises a temperature sensor and an information acquisition instrument and is used for acquiring configuration parameters in a thermal response experiment;
after receiving the parameters from the data acquisition module, the identification module calculates according to a neural network parameter identification method;
and the output module is used for outputting and displaying the result of the thermophysical property parameter and the seepage parameter of the soil to be identified, which are obtained by the calculation of the identification module.
According to the present invention, the neural network identification method includes the steps of:
s1, carrying out on-site thermal response experiment test on the soil to be identified, and obtaining the thermal physical property parameters of the soil to be identified and the measurement parameters required by seepage parameter identification;
s2, establishing a soil thermophysical property parameter and seepage parameter identification neural network, inputting measurement parameters required by soil thermophysical property parameter and seepage parameter identification into the neural network obtained by pre-training, and obtaining the thermophysical property parameter and the seepage parameter of the soil to be identified;
according to the present invention, the neural network obtained by the pre-training in step S2 is obtained by:
s21, establishing a high-precision three-dimensional full-scale simulation experiment platform;
s22, verifying and correcting the simulation experiment platform through a contrast experiment with controllable experiment conditions;
s23, carrying out a thermal response experiment by using the established high-precision three-dimensional full-scale simulation experiment platform to obtain a training database, and training the established neural network by using the training database to achieve good identification precision.
According to the invention, the neural network obtained by pre-training is obtained by training the training data obtained by the simulation experiment under controllable experiment conditions, wherein the controllable experiment conditions mean that the soil adopted in the experiment is various types of soil common in engineering, and the heating power and the circulating water flow rate are controllable, so that the universality and the accuracy of the model are ensured.
According to the invention, said step S22 comprises the following steps:
s21: measuring real soil thermophysical property parameters and seepage parameters by using a high-precision measurement method;
s22: inputting the real soil thermophysical property parameters and seepage parameters into thermal response experiments of a comparison experiment and a high-precision three-dimensional full-scale simulation experiment platform respectively, and obtaining measurement parameters required by soil thermophysical property parameter and seepage parameter identification of the comparison experiment and the simulation experiment respectively;
s23: and comparing the real soil thermophysical property parameters of the comparison experiment and the simulation experiment with the measurement parameters required by seepage parameter identification, and verifying and correcting the high-precision three-dimensional full-scale simulation experiment platform.
According to the invention, the high-precision three-dimensional full-scale simulation experiment platform comprises: the system comprises a three-dimensional full-scale numerical heat transfer model module, a numerical algorithm solving module and a visual post-processing module.
According to the invention, the neural network obtained by pre-training comprises an input unit, a hiding unit and an output unit, wherein the layer where the input unit is located is an input layer, the layer where the hiding unit is located is a hidden layer, and the layer where the output unit is located is an output layer.
According to the invention, a training database is adopted for training the neural network for identifying the soil thermophysical property parameters and the seepage parameters as training signals and error signals of the neural network, the soil thermophysical property parameters and the measurement parameters required by identifying the seepage parameters, which are obtained by a thermal response experiment in a simulation experiment platform in the training database, are used as input ends of the neural network, the real soil thermophysical property parameters and the seepage parameters in the training database are used as output ends, and the weight and the threshold of the neural network are determined;
calculating corresponding output of each hidden unit of the hidden layer by using each input unit in the input layer through a transfer function;
calculating the input of each output unit of the output layer by utilizing the output of each hidden unit of the hidden layer, and calculating the corresponding output of each output unit of the output layer through a transfer function;
according to the soil thermophysical property parameters and seepage parameters in the training database, identifying the required measurement parameters and the actual output of the neural network, calculating the training errors of each unit of the output layer, and correcting the weight and the threshold of the neural network by using the training errors;
and when the fitting error of the neural network is smaller than the set threshold value, obtaining the trained neural network.
According to the invention, in the training process of the neural network, a 'distance' function is adopted as a transfer function, a 'leanngdm' function is adopted as a learning function, a 'trailing' function is adopted as a training function, and a 'mse' function is adopted to evaluate the performance of the neural network.
According to the invention, the thermophysical parameters of the soil to be identified comprise: soil heat conductivity coefficient and volumetric heat capacity parameter;
the seepage parameters comprise: seepage direction and seepage velocity;
the measurement parameters required when soil thermophysical property parameters and seepage parameters of the soil to be identified are identified comprise: the temperature of the water at the inlet of the ground heat exchanger and/or the temperature of the water at the outlet of the ground heat exchanger and/or the temperature of the wall of the drill hole and/or the temperature of backfill materials in the drill hole.
(III) advantageous effects
Compared with the prior art, the invention has the beneficial effects that:
the invention builds the measuring device which can directly identify the soil thermophysical property parameter and the seepage parameter and display the result in the field thermal response experiment, the process is rapid and accurate, the practicability is strong, the identification method adopted in the measuring device establishes the mapping relation between the thermal response measuring parameter and the soil thermophysical property, the seepage direction and the speed through the field thermal response experiment and the neural network, the identification error caused by using the traditional identification model is avoided, the identification precision of the soil thermophysical property is improved, the identification error caused by using the traditional identification model is avoided, and the neural network has certain fault tolerance.
Drawings
FIG. 1 is a schematic view of a measuring device according to the present invention;
FIG. 2 is a schematic flow chart of the present invention.
[ description of reference ]
1. Soil to be detected; 2. backfilling materials; 3. a U-shaped buried pipe; 4. a temperature measuring device; 5. a flow meter; 6. a water circulating pump; 7. an electrically heated water tank; 8. and a pressure measuring device.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
The invention provides a measuring device for soil thermophysical parameters and seepage parameters based on a neural network parameter identification method.
As shown in figure 1, in the field thermal response experiment, the device comprises soil to be measured 1, backfill materials 2, a U-shaped buried pipe 3, a temperature measuring device 4, a flowmeter 5, a circulating water pump 6, an electric heating water tank 7 and a pressure measuring device 8,
when a thermal response experiment of the buried pipe heat exchanger is carried out, a first temperature sensor for testing the water temperature at an inlet is required to be arranged at the inlet of a U-shaped buried pipe 3, a second temperature sensor for testing the water temperature at an outlet is required to be arranged at the outlet of the U-shaped buried pipe 3, a flowmeter 5 for detecting the volume flow of liquid is arranged between the U-shaped buried pipe 3 and a circulating water pump 6, an electric heating water tank 7 is arranged between a water inlet pipeline of the U-shaped buried pipe 3 and the circulating water pump 6, an electric heater is arranged in the electric heating water tank 7 for measuring the heating power of the electric heater, the whole loop is filled with water through the electric heating water tank 7 and is connected with a measuring device, the circulating loop is heated at constant power after an underground temperature field is basically recovered, water is enabled to circularly flow in the loop, the measuring device measures and collects the temperatures in the loop, the data acquisition module comprises a temperature sensor and an information acquisition instrument, and is mainly used for acquiring the temperature of each measurement parameter in a thermal response experiment, the data acquisition module is respectively connected with a temperature measurement device 4, soil to be measured 1, backfill material 2 and a U-shaped buried pipe 3, thermocouples for measuring the inlet water temperature, the outlet water temperature, the wall temperature of the buried pipe, the wall temperature of a drilled hole and the temperature of the backfill material inside the drilled hole of the buried pipe heat exchanger in the field thermal response experiment are respectively connected into the data acquisition module in the measurement device, and the data acquisition module extracts the measurement parameters of the thermal response experiment.
The identification module identifies each parameter in the data after receiving the data transmitted in the data acquisition module, the specific parameter identification method is compiled according to a neural network, and by the identification method, the identification result of the soil thermophysical property parameter and the seepage parameter of the soil to be identified can be calculated on the basis of the data acquisition module, and finally the soil pond thermophysical property parameter and the seepage parameter are output and displayed through the output module.
The parameter identification method combines a thermal response simulation experiment with a neural network system, establishes a nonlinear mapping relation of different thermal response experiment parameters, soil thermophysical parameters and seepage parameters, and aims to improve the accuracy and the practicability of the soil thermophysical parameters and the seepage parameters in the identification process, and comprises the following specific steps:
s1, carrying out on-site thermal response experiment test on the soil to be identified, and acquiring measurement parameters required by identification of thermophysical properties and seepage parameters of the soil to be identified;
s2, establishing a soil thermophysical property parameter and seepage parameter identification neural network, inputting measurement parameters required by soil thermophysical property parameter identification into the neural network obtained by pre-training, and obtaining the thermophysical property parameter and the seepage parameter of the soil to be identified.
The neural network obtained by pre-training is obtained by training the training data obtained by the simulation experiment with controllable experiment conditions.
The controllable experimental conditions mean that the soil adopted in the experiment is various types of soil common in engineering, and the heating power, the circulating water flow rate and the seepage rate are controllable, so that the universality and the accuracy of the model are ensured.
When the neural network is pre-trained in the step S2, firstly, a high-precision three-dimensional full-scale simulation experiment platform is established, wherein the simulation experiment is realized by simulating the reality effect on a computer through simulation software, and the simulation experiment platform is realized through a computer program and specifically comprises a three-dimensional full-scale numerical heat transfer model module, a numerical algorithm solver module and a visualization post-processor module; the thermal response experiment principle of the experiment platform is consistent with the field thermal response experiment principle, different soil thermophysical property parameters and seepage parameters are input into the simulation experiment platform, and after simulation calculation is carried out by using a computer, the soil thermophysical property parameters and the seepage parameters which correspond to the input soil thermophysical property parameters and the seepage parameters can be obtained, and the measurement parameters required by the identification of the seepage parameters are obtained.
In order to ensure the accuracy of the simulation experiment platform, the simulation experiment platform needs to be verified and corrected through a comparison experiment, wherein the basic parameters and the heat exchange conditions of the comparison experiment and the simulation experiment platform are consistent, and in order to ensure the universality and the accuracy of the model, the soil adopted in the comparison experiment is ensured to be various types of soil common in engineering.
Firstly, measuring thermal physical property parameters of real soil in advance by using a high-precision measuring method, then applying the measured parameters to a comparison experiment under the condition of controllable parameters to carry out a thermal response experiment to obtain measuring parameters required by a verification experiment for identifying the thermal physical property parameters and seepage parameters of the soil, then inputting the same thermal physical property parameters and seepage parameters of the real soil into a simulation experiment platform to carry out analog calculation to obtain measuring parameters required by the simulation experiment platform for identifying the thermal physical property parameters and the seepage parameters, and comparing the measuring parameters obtained by the simulation experiment platform with the measuring parameters obtained by the comparison experiment to ensure the accuracy of the simulation experiment platform.
The soil thermophysical parameters comprise soil heat conductivity coefficient and volumetric heat capacity;
seepage parameters include: seepage direction and seepage velocity;
the measured parameters of the thermal response experiment comprise the inlet water temperature and/or the outlet water temperature of the ground heat exchanger and/or the wall temperature of the drill hole and/or the temperature of backfill materials inside the drill hole and/or combinations of the above measured parameters.
According to a related theory, the heat exchange process between soils is complex, and the influence factors are many, so when the soil thermophysical property parameters and seepage parameters are inverted by utilizing a thermal response experiment, the relation among the measured parameters, the thermophysical property parameters and the seepage parameters is non-analytic, an accurate result is difficult to obtain by a general polynomial fitting or other fitting difference methods, and compared with the prior art, the BP neural network has strong nonlinear mapping capability, reasonably designs a network hidden layer and a transfer function of the BP neural network, and can meet the requirements of soil thermophysical property and seepage parameter modeling and data fitting.
The invention utilizes the nonlinear fitting and predicting capability of the BP neural network, trains the network by adopting a large amount of training databases, and then inputs the measurement parameters of the thermal response experiment into the trained BP neural network, thus predicting and obtaining the soil thermophysical property parameters and seepage parameters of the corresponding soil to be detected.
In step S2, a neural network training database is further established, where the database includes real soil thermophysical property parameters and percolation parameters, and soil thermophysical property parameters and percolation parameters required for identification, which are obtained by performing a thermal response experiment on a high-precision three-dimensional full-scale simulation experiment platform corresponding to the real soil thermophysical property parameters and percolation parameters, and the establishment process specifically includes: by utilizing a high-precision three-dimensional full-scale simulation experiment platform, different real soil thermophysical property parameters and seepage parameters are respectively input into the platform, a thermal response experiment is carried out, measurement parameters corresponding to the different soil thermophysical property parameters and seepage parameters and required in the identification process of the soil thermophysical property parameters and the seepage parameters can be obtained through the simulation calculation of the platform, and a neural network training database is established according to the measurement parameters. The input range of the real soil thermophysical parameters and the seepage parameters needs to cover various common geological conditions, and the times of simulation experiments can ensure the accuracy of the identification method, so the more the times of simulation thermal response experiments are, the more the accuracy of a training database can be ensured.
Generally, a neural network is composed of an input unit, a hidden unit and an output unit, wherein a layer where the input unit is located is an input layer, a layer where the hidden unit is located is a hidden layer, and a layer where the output unit is located is an output layer. All the units are connected with the units in the two side layers, input information can firstly enter the input layer at the left side, then the hidden layer in the middle is excited, and finally the result is output from the output layer at the right side, and the neural network can gradually adjust the connecting force of every two units through learning.
When the BP neural network is trained, the measurement parameters in the training database are used as input, and the corresponding soil thermophysical property parameters and seepage parameters are used as output.
In step S2, the neural network training database is used as a training signal and an error signal of the neural network to train the neural network, the measurement parameters of the simulated thermal response experiment are used as the input end of the neural network, the thermal physical property parameters and the seepage parameters of the real soil are used as the output end in the training process, and the weight and the threshold of the neural network are determined;
processing each layer through a transfer function, calculating the input of each unit in the hidden layer by using an input signal, a weight and a threshold, calculating the corresponding output of each unit in the hidden layer by using the input of each unit through the transfer function, calculating the input of each unit in the output layer by using the output, the weight and the threshold of the hidden layer, and calculating the corresponding output of each unit in the output layer through the transfer function;
the difference between the theoretical output and the actual output of the network, namely the difference between the measurement parameter of the simulation thermal response experiment and the actual output of the neural network, wherein the error is the training error of each unit of the output layer;
calculating the training error of the hidden layer by using the connection weight, the training error of each unit of the output layer and the output of the hidden layer;
correcting the weight and the threshold of the neural network by using the training errors of all units of the output layer and the output of the hidden layer;
when the fitting error of the neural network is smaller than the set threshold, the neural network learning is finished, and the trained neural network can be obtained, wherein the threshold can be set according to the actual situation.
In the whole neural network training process, the neural network adopts a 'ranging' function (hyperbolic tangent type transfer function) as a transfer function, a 'leangdm' function (gradient descent momentum learning function) as a learning function, a 'trailing' function (gradient descent algorithm) as a training function to train the BP neural network, and a 'mse' function to evaluate the performance of the neural network. Wherein, the training function comprises a learning function, and the learning function is a part of the training function; the training function globally adjusts the weight and the threshold, and the learning function locally adjusts the weight and the threshold; in terms of errors, the training function has the smallest error for the whole, and the learning function has the smallest error for the single neural unit.
And after the neural network training is finished, carrying out thermal response experiment test on the soil to be identified, and inputting the obtained soil thermophysical property parameters and the measurement parameters required in the seepage parameter identification process into the trained neural network to obtain the soil thermophysical property parameters and the seepage parameters of the soil to be identified.
The specific implementation mode of the whole soil thermophysical property parameter and seepage parameter identification process is as follows:
thermocouples for measuring the inlet water temperature, the outlet water temperature, the wall temperature of the buried pipe, the wall temperature of a drilled hole and the temperature of backfill materials in the drilled hole of the buried pipe heat exchanger in a field thermal response experiment are respectively connected into corresponding jacks of a data acquisition module in a measuring device, and the device starts to extract measurement parameters of the thermal response experiment; and then the extracted data is transmitted to an identification module, the identification module compiled based on the identification method can input the measurement parameters into the trained BP neural network, the calculation is carried out according to the corresponding neural network mapping relation, the trained BP neural network outputs the final soil thermophysical property parameters and seepage parameters, and the soil thermophysical property parameters and the seepage parameters are displayed through an output module.
The mapping relation between the thermal response test parameters and the soil thermophysical parameters and the seepage parameters established through the field thermal response experiment and the neural network avoids the identification error brought by the traditional identification model, the neural network has certain fault tolerance, and meanwhile, the measuring device based on the method can directly identify the soil thermophysical parameters and the seepage parameters in the field thermal response experiment process, so that the engineering operation is more concise and convenient, and the practicability is strong.
It should be understood that the above description of specific embodiments of the present invention is only for the purpose of illustrating the technical lines and features of the present invention, and is intended to enable those skilled in the art to understand the contents of the present invention and to implement the present invention, but the present invention is not limited to the above specific embodiments. It is intended that all such changes and modifications as fall within the scope of the appended claims be embraced therein.
Claims (7)
1. A measuring device for soil thermophysical parameters and seepage parameters is characterized by comprising a data acquisition module, an identification module and an output module;
the data acquisition module comprises a temperature sensor and an information acquisition instrument and is used for acquiring configuration parameters in a thermal response experiment;
after receiving the parameters from the data acquisition module, the identification module calculates according to a neural network parameter identification method;
the output module is used for outputting and displaying the results of the soil thermophysical property parameters and the seepage parameters of the soil to be identified, which are obtained by the calculation of the identification module;
the neural network parameter identification method comprises the following steps:
s1, carrying out on-site thermal response experiment test on the soil to be identified, and obtaining the thermal physical property parameters of the soil to be identified and the measurement parameters required by seepage parameter identification;
s2, establishing a soil thermophysical property parameter and seepage parameter identification neural network, inputting measurement parameters required by soil thermophysical property parameter and seepage parameter identification into the neural network obtained by pre-training, and obtaining the thermophysical property parameter and the seepage parameter of the soil to be identified;
the neural network obtained by the pre-training in the step S2 is obtained by:
s21, establishing a high-precision three-dimensional full-scale simulation experiment platform;
s22, verifying and correcting the simulation experiment platform through a contrast experiment with controllable experiment conditions;
s23, carrying out a thermal response experiment by using the established high-precision three-dimensional full-scale simulation experiment platform to obtain a training database, and training the established neural network by using the training database to achieve good identification precision;
the step S22 includes the steps of:
s221: measuring real soil thermophysical property parameters and seepage parameters by using a high-precision measurement method;
s222: inputting the real soil thermophysical property parameters and seepage parameters into thermal response experiments of a comparison experiment and a high-precision three-dimensional full-scale simulation experiment platform respectively, and obtaining measurement parameters required by soil thermophysical property parameter and seepage parameter identification of the comparison experiment and the simulation experiment respectively;
s223: comparing real soil thermophysical property parameters of the comparison experiment and the simulation experiment with measurement parameters required by seepage parameter identification, and verifying and correcting the high-precision three-dimensional full-scale simulation experiment platform;
when the neural network is trained, the measurement parameters in the training database are used as the input of the neural network, and the corresponding soil thermophysical property parameters and seepage parameters are used as the output.
2. The measuring device according to claim 1, wherein the neural network obtained by pre-training is obtained by training with training data obtained by simulation experiments under controllable experimental conditions, the soil adopted in the experiments is various types of soil common in engineering, and the heating power and the circulating water flow rate are controllable to ensure the universality and the accuracy of the model.
3. The measurement device of claim 1, wherein the high-precision three-dimensional full-scale simulation experiment platform comprises: the system comprises a three-dimensional full-scale numerical heat transfer model module, a numerical algorithm solving module and a visual post-processing module.
4. The measurement device according to claim 1, wherein the neural network obtained by pre-training comprises an input unit, a hidden unit and an output unit, wherein a layer where the input unit is located is an input layer, a layer where the hidden unit is located is a hidden layer, and a layer where the output unit is located is an output layer.
5. The measuring device according to claim 4, wherein the training of the neural network for identifying the soil thermophysical property parameter and the seepage parameter adopts a training database as a training signal and an error signal of the neural network, the soil thermophysical property parameter and the measurement parameter required for identifying the seepage parameter, which are obtained by a thermal response experiment in a simulation experiment platform in the training database, are used as input ends of the neural network, the real soil thermophysical property parameter and the seepage parameter in the training database are used as output ends, and the weight and the threshold of the neural network are determined;
calculating corresponding output of each hidden unit of the hidden layer by using each input unit in the input layer through a transfer function;
calculating the input of each output unit of the output layer by utilizing the output of each hidden unit of the hidden layer, and calculating the corresponding output of each output unit of the output layer through a transfer function;
according to the soil thermophysical property parameters and seepage parameters in the training database, identifying the required measurement parameters and the actual output of the neural network, calculating the training errors of each unit of the output layer, and correcting the weight and the threshold of the neural network by using the training errors;
and when the fitting error of the neural network is smaller than the set threshold value, obtaining the trained neural network.
6. The measurement device according to claim 5, wherein the performance of the neural network is evaluated using a "distance" function as a transfer function, a "learngdm" function as a learning function, a "train" function as a training function, and a "mse" function during the training of the neural network.
7. The measuring device of claim 1,
the thermophysical property parameters of the soil to be identified comprise: soil heat conductivity coefficient and volumetric heat capacity parameter;
the seepage parameters comprise: seepage direction and seepage velocity;
the measurement parameters required when soil thermophysical property parameters and seepage parameters of the soil to be identified are identified comprise: the temperature of the water at the inlet of the ground heat exchanger and/or the temperature of the water at the outlet of the ground heat exchanger and/or the temperature of the wall of the drill hole and/or the temperature of backfill materials in the drill hole.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2018114955622 | 2018-12-07 | ||
| CN201811495562 | 2018-12-07 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109557126A CN109557126A (en) | 2019-04-02 |
| CN109557126B true 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) |
Families Citing this family (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 |
| CN111210877B (en) * | 2020-01-03 | 2023-11-24 | 山东天岳先进科技股份有限公司 | Method and device for deducing physical parameters |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430317B (en) * | 2008-12-09 | 2012-01-04 | 山东建筑大学 | Rock thermophysics on-site test method under non-constant power condition |
| CN102628817B (en) * | 2012-03-30 | 2013-09-11 | 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
Also Published As
| Publication number | Publication date |
|---|---|
| CN109557126A (en) | 2019-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109557126B (en) | Measuring device for soil thermophysical property parameters and seepage parameters | |
| CN102721722B (en) | In-situ thermal response testing method of stratified thermal properties of underground rock and soil | |
| CN2771856Y (en) | Analogue monitoring device for oily saturation field | |
| CN104863557A (en) | Experimental device and method of super heavy oil reservoir steam-nitrogen-viscosity reducer composite steam stimulation | |
| CN104895550A (en) | Tight gas fracturing horizontal well numerical value well testing model building and solving method | |
| CN104237301B (en) | In-situ thermal response testing method for layered rock and soil thermophysical properties | |
| CN103114842A (en) | Experimental facility and model system thereof for simulating thickened oil steam huff and puff oil extraction | |
| CN105823932A (en) | Grounding resistance testing method for frozen earth and snow-covered environments | |
| CN105092815A (en) | Rock acoustic and electrical parameter joint testing device capable of simulating reservoir conditions | |
| CN103675452A (en) | Rock resistivity testing system and method capable of simulating reservoir conditions | |
| CN103884738B (en) | Underground heat individual well stratum hot physical property distribution appraisal procedure | |
| CN105649603A (en) | Method for measuring accumulated liquid quantity of well bore of gas well in real time | |
| CN204694654U (en) | The hot physical property of combined type ground source heat pump rock-soil and underground pipe heat exchange tester | |
| CN203145918U (en) | Experimental apparatus for simulating heavy oil steam huff-and-puff recovery and model system thereof | |
| CN109521051B (en) | Soil thermophysical property parameter identification method | |
| CN113252532A (en) | Simulation device for breakthrough of abnormal high-temperature and high-pressure stratum sand body through seepage barrier | |
| CN202649147U (en) | Device for testing field thermal response of underground geotechnical layered thermal properties | |
| TWI734029B (en) | Production capacity estimation method for geothermal well | |
| CN201397310Y (en) | Novel ground source heat pump soil thermal property test system | |
| CN107817317A (en) | One kind is by mining influence coal seam bottom water bursting simulation experiment system | |
| CN105672996A (en) | Comprehensive recognition system for loss of drilling fluid | |
| CN105043600B (en) | Intelligent heavy caliber calorimeter detection device and method | |
| CN114413184A (en) | Intelligent pipeline, intelligent pipeline management system and leakage detection method thereof | |
| CN203643301U (en) | Device for measuring rock core permeability | |
| CN208400130U (en) | A kind of multi-dielectric Groundwater movement simulation 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 |