CN108764461B - Earthen site temperature prediction method based on solar term characteristics - Google Patents
Earthen site temperature prediction method based on solar term characteristics Download PDFInfo
- Publication number
- CN108764461B CN108764461B CN201810508671.7A CN201810508671A CN108764461B CN 108764461 B CN108764461 B CN 108764461B CN 201810508671 A CN201810508671 A CN 201810508671A CN 108764461 B CN108764461 B CN 108764461B
- Authority
- CN
- China
- Prior art keywords
- matrix
- earthen
- site temperature
- earthen site
- neural network
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Optimization (AREA)
- Computing Systems (AREA)
- Mathematical Analysis (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Algebra (AREA)
- Entrepreneurship & Innovation (AREA)
- Databases & Information Systems (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an earthen archaeological site temperature prediction method based on solar terms, which mainly solves the problems that the earthen archaeological site temperature prediction result obtained by the prior art has large relative error and low prediction accuracy. The method comprises the following implementation steps: 1. extracting time characteristics of earthen site temperature data; 2. extracting gas saving characteristics of the earthen site temperature data; 3. obtaining the total characteristics of the temperature data according to the time characteristics and the gas saving characteristics; 4. introducing a neural network, and performing learning training by using the total characteristics of the earthen site temperature data and the earthen site temperature data sample to obtain a neural network model for forecasting the earthen site temperature; 5. and predicting the temperature of the earthen site through the obtained neural network model to obtain a temperature prediction result matrix of the earthen site. According to the method, the temperature of the earthen archaeological site is predicted by adding the solar term characteristic to the membership function on the basis of the characteristics of year, day and hour, minute and second, so that the relative error generated in temperature prediction is reduced, and the prediction accuracy is improved.
Description
Technical Field
The invention belongs to the technical field of signal and information processing, and further relates to temperature prediction, in particular to a gas term characteristic-based earthen site temperature prediction method which can be used for predicting the temperature of the environment where an earthen site is located.
Background
The ancient site is an ancient site which takes soil as a main building material and has historical, cultural and scientific values, is not only a trace of human activities for thousands of years, but also an important carrier of historical information, and has extremely high research value. As the only one of the four civilization ancient countries to continue and continuously inherit the civilization to the present, the research and protection problems of the Chinese earthen site should be more emphasized. The study of earthen sites began earlier abroad, but most of them were conducted in open air environments. In the 80 s of the 20 th century, the study and protection work of the earthen site begins in China, but the study specially aiming at the problem of the earthen site in the indoor closed environment is still less. The burial pit site outside Hanyang tomb in Shaanxi Yanyang is the first totally-enclosed museum of the earthen site in China, and the museum separates the site protection area from the tourist visiting area through a totally-enclosed showpiece alignment mode, thereby effectively avoiding site damage caused by direct contact between the tourist and the site, and still generating stealthy influence on the earthen site by factors such as temperature, humidity and the like.
Twenty-four solar terms are 24 festivals representing season change in the Chinese lunar calendar, carefully reflect the climate characteristics of the four-season alternation in China, and have marking significance in the aspects of seasons, climate, phenological climate, crop growth and the like. Twenty-four solar terms originate in the ancient yellow river basin. In 104 years before the Gongyuan, the first Chinese calendar-Sun calendar-written in complete letters written in the first place of China, which was worked out by Deng Ping et al, defines the twenty-four solar terms in the calendar and defines the astronomical position. According to the position change of the earth on the yellow track (namely the revolution orbit of the earth), from zero degree of the yellow longitude, the earth names the date as a solar term every 15 degrees of the earth operation; the earth runs for 360 degrees in one year, namely 24 solar terms are in total in one year. The solar terms reflecting the change of four seasons include: the spring, spring equinox, summer solstice, autumn equinox, winter solstice and winter solstice are 8 solar terms; the following are reflected in the temperature change: the heat is saved by 5 solar terms of minor summer heat, major summer heat, sunstroke, minor cold and major cold; the weather phenomenon is reflected by: 7 solar terms of rain, white dew, cold dew, frost, small snow and big snow; reflecting the phenological phenomenon, there are frightening, clear, discontented and mango seeds. The factors of temperature, humidity and the like which can influence the earthen site can be reflected by the solar term.
Many temperature prediction methods for different applications have been proposed in the literature, for example: the method comprises the steps of establishing a rammed earth building site surface temperature forecasting model by using a regression fitting method, establishing a temperature forecasting model under certain environments by using a linear fitting method, establishing a temperature forecasting model of an agricultural greenhouse by using a support vector machine, and establishing a temperature forecasting method based on a statistical theory. In addition, a neural network based temperature prediction method is also proposed, for example: BP neural network and RBF neural network based temperature prediction methods and the like. However, these methods for predicting temperature are less applicable to research and protection of earthen sites, and have larger errors.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a method for predicting the earthen site temperature based on the solar term characteristic. The method introduces a neural network, and utilizes the membership function and the solar term characteristic to predict the temperature on the basis of the characteristics of year, day and hour, minute and second, thereby reducing the relative error generated during prediction and improving the accuracy of temperature prediction.
In order to achieve the purpose, the technical scheme adopted by the invention comprises the following steps:
(1) extracting time characteristics:
(1.1) selecting n data from the earthen site temperature data actually measured in the past year as data samples, wherein n is an integer greater than or equal to 1;
(1.2) inputting a time matrix T corresponding to the data sample:
T=[t1,t2,…,tj,…,tn]T,
wherein, tjThe unit of the time corresponding to the j earthen site temperature data is as follows: j is more than or equal to 1 and less than or equal to n, and j is an integer; [. the]TA transpose operation representing a matrix;
setting matrixThe minimum value of all elements in T is TminMaximum value of tmax;
(1.3) calculating time characteristics of the jth earthen site temperature data in the data sample, including a time minute second characteristic α1jDay characteristics α2jAnd year characteristics α3j:
α1j=tjmod1,
(2) extracting solar terms:
(2.1) selecting m continuous throttle points, and inputting a corresponding time matrix Z:
Z=[z1,z2,…,zi,…,zm]T,
wherein m is an integer greater than or equal to 1; z is a radical ofiThe unit of time corresponding to the ith throttle point is as follows: i is more than or equal to 1 and less than or equal to m, and i is an integer; [. the]TA transpose operation representing a matrix;
let the minimum of all elements in the matrix Z be ZminMaximum value of zmaxAnd z ismin≤tmin、zmax≥tmax;
(2.2) calculating the solar terms characteristic α of the jth earthen site temperature data through the membership function4j:
Wherein the content of the first and second substances,is a time tjThe time corresponding to the throttle point closest thereto before,is a time tjThe time corresponding to the throttle point closest to the throttle point;is tjTo pairThe function of the degree of membership of (c),is tjTo pairA membership function of;
(3) obtaining the overall characteristics of the data sample:
calculating the total characteristic x of the jth earthen site temperature dataj:
xj=ω1α1j+ω2α2j+ω3α3j+ω4α4j<1>
Wherein, ω is1、ω2、ω3And ω4The weights of the time, minute and second characteristic, the day characteristic, the year characteristic and the solar term characteristic of the earthen site temperature data are sequentially set;
taking j as 1 to n, and calculating by a formula <1> to obtain a total feature matrix X of the data sample:
X=[x1,x2,…,xj,…,xn]Τ;
(4) using a neural network to predict the temperature of the earthen site:
(4.1) determining an earthen site temperature data matrix for training Γ from the data samples:
Γ=[τ1,τ2,…,τj,…,τn]Τ,
wherein, taujThe j th earthen site temperature data in the data sample is obtained;
(4.2) taking the total characteristic matrix X of the data sample as an input layer input matrix of a neural network, taking the earthen site temperature data matrix gamma as an output layer expected output matrix of the neural network, and performing learning training by using the neural network to obtain a neural network model for forecasting the earthen site temperature;
(4.3) setting the total characteristic matrix of the earthen site temperature data to be predicted as XcAnd substituting the matrix as an input layer input matrix into the neural network model for the earthen site temperature prediction to obtain an output layer output matrix, namely an earthen site temperature prediction result matrix.
Compared with the prior art, the invention has the following advantages:
firstly, the method comprises the following steps: because the relationship between the solar terms and the time corresponding to each temperature data is established by utilizing the membership function, the problems that the solar terms are dispersed and the corresponding time cannot correspond to the time corresponding to each temperature data one by one are effectively solved, and the extraction of the solar terms is more accurate;
secondly, the method comprises the following steps: in the process of temperature prediction, the invention not only uses the characteristics of year, day and hour, minute and second, but also comprehensively considers the characteristics of gas saving, thereby effectively improving the accuracy of temperature prediction and reducing the relative error generated when the temperature is predicted.
Drawings
FIG. 1 is a flow chart of an implementation of the present invention;
FIG. 2 is a graph of Gaussian membership functions for two adjacent solar terms;
figure 3 is a data graph of temperature of 110116 station in 2011 as a function of days;
FIG. 4 is a data graph of temperature of 110116 station in 2011 as a function of the throttle;
FIG. 5 is a graph of the predicted results of the random snapshot after addition of the solar term feature from day 110 to day 117 of the 110116 website in 2011;
FIG. 6 is a graph comparing the relative error between the addition of a throttle feature and the absence of a throttle feature from day 110 to day 117 at station # 110116 in 2011;
FIG. 7 is a graph of the predicted results of the random snapshot after addition of the solar term feature from day 110 to day 140 of the 110116 station in 2011;
FIG. 8 is a chart comparing relative errors between the 110 th day to 140 th day of the 110116 station in 2011 with and without the addition of a throttle feature;
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the accompanying drawings and specific embodiments.
Referring to fig. 1, the specific implementation steps of the present invention are as follows:
step 1, extracting time characteristics:
(1.1) selecting n data from the earthen site temperature data actually measured and recorded in the past years as data samples, wherein n is an integer greater than or equal to 1;
(1.2) inputting a time matrix T corresponding to the data sample:
T=[t1,t2,…,tj,…,tn]T,
wherein, tjThe time corresponding to the jth earthen site temperature data is represented, a public element 0 year is taken as a positive and negative boundary, and the unit is as follows: j is more than or equal to 1 and less than or equal to n, and j is an integer; [. the]TA transpose operation representing a matrix;
let the minimum of all elements in the matrix T be TminMaximum value of tmax;
(1.3) the time/minute/second characteristic, the day characteristic and the year characteristic are collectively called as time characteristics, and the time characteristics of the jth earthen site temperature data in the data sample are calculated and comprise the time/minute/second characteristic α1jDay characteristics α2jAnd year characteristics α3j:
α1j=tjmod1,
step 2, extracting solar terms characteristics:
(2.1) selecting m continuous throttle points, wherein m is an integer greater than or equal to 1, and inputting a corresponding time matrix Z:
Z=[z1,z2,…,zi,…,zm]T,
wherein z isiThe unit of time corresponding to the ith throttle point is as follows: i is more than or equal to 1 and less than or equal to m, and i is an integer; [. the]TA transpose operation representing a matrix;
let the minimum of all elements in the matrix Z be ZminMaximum value of zmaxAnd z ismin≤tmin、zmax≥tmax;
(2.2) introducing a membership function, and calculating the solar terms characteristic α of the jth earthen site temperature data4j:
Wherein the content of the first and second substances,is a time tjThe time corresponding to the throttle point closest thereto before,is a time tjThe time corresponding to the throttle point closest to the throttle point;is tjTo pairThe function of the degree of membership of (c),is tjTo pairA membership function of;
corresponding time t to jth earthen site temperature datajComparing the sizes of the elements with all the elements in the matrix Z one by one, calculating the difference values, selecting the minimum two of the difference values, and determining the elements in the matrix Z corresponding to the minimum two values as theAnd
the common membership function includes gaussian membership function, sigmoid membership function, bell-shaped membership function, parabolic membership function, customized membership function, etc., in this embodiment, the gaussian membership function is taken, and the image is shown in fig. 2, and is specifically calculated by the following formulaAnd
wherein, the mean value mu is 0, and the standard deviation sigma is 3.7;
step 3, obtaining the total characteristics of the data sample:
calculating the total characteristic x of the jth earthen site temperature dataj:
xj=ω1α1j+ω2α2j+ω3α3j+ω4α4j<1>
Wherein, ω is1、ω2、ω3And ω4The weights of the time-minute-second characteristic, the day characteristic, the year characteristic and the solar term characteristic of the earthen site temperature data are obtained from experience and are constants, but different values can be generated according to different data. Here, the embodiment takes the time feature weightWhen adding the solar term feature, the solar term feature weight is takenOtherwise ω is4=0。
Taking j as 1 to n, and calculating by a formula <1> to obtain a total feature matrix X of the data sample:
X=[x1,x2,…,xj,…,xn]Τ;
step 4, using a neural network to predict the temperature of the earthen site:
the neural network used in this step may be any one of a plurality of neural networks such as a single hidden layer feedforward neural network, a back propagation neural network, a convolutional neural network, and the like, and this embodiment takes the single hidden layer feedforward neural network as an example, and the specific prediction steps are as follows:
(4.1) determining an earthen site temperature data matrix for training Γ from the data samples:
Γ=[τ1,τ2,…,τj,…,τn]Τ,
wherein, taujThe j th earthen site temperature data in the data sample is obtained;
(4.2) solving a neural network model for predicting the temperature of the earthen site:
(4.2.1) taking the total feature matrix X of the data samples as an input layer input matrix of the neural network, resulting in a hidden layer output matrix h (X) for the total feature matrix X:
wherein, L is the number of hidden layer neurons and is an integer greater than or equal to 1; a iskAs a connection weight between input layer and k-th neuron of the hidden layer, bkIs the threshold of the kth neuron of the hidden layer, k is more than or equal to 1 and less than or equal to L, k is an integer, akAnd bkRandomly generating through a computer; g (-) is a stimulus function, and commonly used stimulus functions are a sin function, a sigmoid function, a hardlim function and the like. Here, the present embodiment sets the excitation function format to g (a, b, x) ═ g (ax + b), and takes the sin function as the excitation function, that is, g (a, b, x) ═ sin (ax + b).
(4.2.2) taking the temperature data matrix gamma of the earthen archaeological site as an expected output matrix of an output layer of the single-hidden-layer feedforward neural network;
(4.2.3) assuming that the connection weight matrix between the hidden layer and the output layer is β and is an L × 1 matrix, then:
Η(X)β=Γ<2>
transforming equation <2> yields:
wherein the content of the first and second substances,Moore-Penrose generalized inverse matrix of H (X), which can be represented as·TTranspose operation of the representation matrix [ ·]-1Representing the inverse of the matrix.
The connection weight matrix β between the hidden layer and the output layer is calculated by the formula <3 >.
(4.3) forecasting the earthen site temperature:
(4.3.1) setting the total characteristic matrix of the earthen site temperature data to be predicted as XcThen the hidden layer output matrix is H (X)c);
(4.3.2) calculating an earthen site temperature prediction result matrix Y according to the following formula:
Y=Η(Xc)β。
the effect of the present invention can be further illustrated by the following simulation results:
1. simulation conditions
The following simulation experiment selects partial temperature data of the site 110116 in 2011 and corresponding time data as input, and the annual data of the site 110116 in 2011 are shown in fig. 3 and 4. Time feature weight settingThe experiment times are set to be 100, 99% of temperature data and corresponding time data are randomly taken out for training each time, and the rest data are used for prediction.
2. Emulated content
Simulation 1, inputting temperature data of 110 days to 117 days (total 7 days) of the 110116 station in 2011. Without introducing a throttle feature, i.e. omega 40, obtaining the relative error mean value of the predicted value and the true value of 1.5 multiplied by 10-3(ii) a Introduction of gas-saving features, i.e.Obtain the mean value of the relative errors between the predicted value and the true value 3.3473 × 10-4The prediction result of the random snapshot is shown in fig. 5. The relative error ratio of adding and not adding throttles is shown in fig. 6.
Simulation 2, inputting temperature data of 110116 station 110 th day to 140 th day (30 days in total) in 2011 and corresponding time data. Without introducing a throttle feature, i.e. omega 40, obtaining the relative error mean value of the predicted value and the true value of 1.5 multiplied by 10-3(ii) a Introduction of gas-saving features, i.e.Obtain the mean value of the relative errors between the predicted value and the true value 5.9418 × 10-4The prediction result of the random snapshot is shown in fig. 7. Joining nodeThe relative error ratio of gas to unthrottled gas is shown in fig. 8.
5-8 show that the throttle characteristic can be effectively added to improve the temperature prediction precision and reduce the relative error whether the data is short-time data or long-time data.
The invention has not been described in detail in part of the common general knowledge of those skilled in the art.
The above description is only one specific example of the present invention and does not constitute any limitation of the present invention. It will be apparent to persons skilled in the relevant art that various modifications and changes in form and detail can be made therein without departing from the principles and arrangements of the invention, but these modifications and changes are still within the scope of the invention as defined in the appended claims.
Claims (6)
1. An earthen site temperature prediction method based on solar terms features is characterized by comprising the following steps:
(1) extracting time characteristics:
(1.1) selecting n data from the earthen site temperature data actually measured in the past year as data samples, wherein n is an integer greater than or equal to 1;
(1.2) inputting a time matrix T corresponding to the data sample:
T=[t1,t2,…,tj,…,tn]T,
wherein, tjThe unit of the time corresponding to the j earthen site temperature data is as follows: j is more than or equal to 1 and less than or equal to n, and j is an integer; [. the]TA transpose operation representing a matrix;
let the minimum of all elements in the matrix T be TminMaximum value of tmax;
(1.3) calculating time characteristics of the jth earthen site temperature data in the data sample, including a time minute second characteristic α1jDay characteristics α2jAnd year characteristics α3j:
α1j=tjmod 1,
(2) extracting solar terms:
(2.1) selecting m continuous throttle points, and inputting a corresponding time matrix Z:
Z=[z1,z2,…,zi,…,zm]T,
wherein m is an integer greater than or equal to 1; z is a radical ofiThe unit of time corresponding to the ith throttle point is as follows: i is more than or equal to 1 and less than or equal to m, and i is an integer; [. the]TA transpose operation representing a matrix;
let the minimum of all elements in the matrix Z be ZminMaximum value of zmaxAnd z ismin≤tmin、zmax≥tmax;
(2.2) calculating the solar terms characteristic α of the jth earthen site temperature data through the membership function4j:
Wherein the content of the first and second substances,is a time tjThe time corresponding to the throttle point closest thereto before,is a time tjThe time corresponding to the throttle point closest to the throttle point;is tjTo pairThe function of the degree of membership of (c),is tjTo pairA membership function of;
(3) obtaining the overall characteristics of the data sample:
calculating the total characteristic x of the jth earthen site temperature dataj:
xj=ω1α1j+ω2α2j+ω3α3j+ω4α4j<1>
Wherein, ω is1、ω2、ω3And ω4The weights of the time, minute and second characteristic, the day characteristic, the year characteristic and the solar term characteristic of the earthen site temperature data are sequentially set;
taking j as 1 to n, and calculating by a formula <1> to obtain a total feature matrix X of the data sample:
X=[x1,x2,…,xj,…,xn]T;
(4) using a neural network to predict the temperature of the earthen site:
(4.1) determining an earthen site temperature data matrix for training Γ from the data samples:
Γ=[τ1,τ2,…,τj,…,τn]T,
wherein, taujThe j th earthen site temperature data in the data sample is obtained;
(4.2) taking the total characteristic matrix X of the data sample as an input layer input matrix of a neural network, taking the earthen site temperature data matrix gamma as an output layer expected output matrix of the neural network, and performing learning training by using the neural network to obtain a neural network model for forecasting the earthen site temperature;
(4.3) setting the total characteristic matrix of the earthen site temperature data to be predicted as XcAnd substituting the matrix as an input layer input matrix into the neural network model for the earthen site temperature prediction to obtain an output layer output matrix, namely an earthen site temperature prediction result matrix.
2. The method of claim 1, wherein: the membership function in the step (2.2) is a Gaussian membership function, a sigmoid membership function, a bell-shaped membership function, a parabolic membership function or a self-defined membership function.
4. The method of claim 1, wherein: the weights omega of the time, minute and second characteristic, the day characteristic, the year characteristic and the solar term characteristic of the earthen site temperature data in the step (3)1、ω2、ω3And ω4Are empirically derived and are constants.
5. The method of claim 1, wherein: the neural network in the step (4) comprises: a single hidden layer feedforward neural network, a back propagation neural network, and a convolutional neural network.
6. The method of claim 5, wherein: the method for predicting the earthen archaeological site temperature by using the single hidden layer feedforward neural network comprises the following steps:
(4.1.1) determining an earthen site temperature data matrix for training Γ from the data samples:
Γ=[τ1,τ2,…,τj,…,τn]T,
wherein, taujThe j th earthen site temperature data in the data sample is obtained;
(4.2.1) taking the total feature matrix X of the data sample as an input layer input matrix of the single hidden layer feedforward neural network, and obtaining a hidden layer output matrix h (X) about the total feature matrix X;
(4.2.2) taking the earthen site temperature data matrix gamma as an expected output matrix of an output layer of the single-hidden-layer feedforward neural network;
(4.2.3) calculating a connection weight matrix β between the hidden layer and the output layer:
wherein the content of the first and second substances,Moore-Penrose generalized inverse matrix of h (X);
(4.3.1) setting the total characteristic matrix of the earthen site temperature data to be predicted as XcThen the hidden layer output matrix is H (X)c);
(4.3.2) calculating an earthen site temperature prediction result matrix Y according to the following formula:
Y=Η(Xc)β。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810508671.7A CN108764461B (en) | 2018-05-24 | 2018-05-24 | Earthen site temperature prediction method based on solar term characteristics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810508671.7A CN108764461B (en) | 2018-05-24 | 2018-05-24 | Earthen site temperature prediction method based on solar term characteristics |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108764461A CN108764461A (en) | 2018-11-06 |
CN108764461B true CN108764461B (en) | 2020-04-14 |
Family
ID=64005416
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810508671.7A Active CN108764461B (en) | 2018-05-24 | 2018-05-24 | Earthen site temperature prediction method based on solar term characteristics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108764461B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106779067A (en) * | 2016-12-02 | 2017-05-31 | 清华大学 | Soil moisture method for reconstructing and system based on multi- source Remote Sensing Data data |
CN107907227A (en) * | 2017-11-10 | 2018-04-13 | 敦煌研究院 | A kind of earthen ruins surface and the measuring method of internal temperature change |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101295022A (en) * | 2008-06-25 | 2008-10-29 | 中国农业科学院农业资源与农业区划研究所 | Method for ground surface temperature and emissivity inversion by remote sensing data ASTER |
CN103984980B (en) * | 2014-01-28 | 2017-12-19 | 中国农业大学 | The Forecasting Methodology of temperature extremal in a kind of greenhouse |
-
2018
- 2018-05-24 CN CN201810508671.7A patent/CN108764461B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106779067A (en) * | 2016-12-02 | 2017-05-31 | 清华大学 | Soil moisture method for reconstructing and system based on multi- source Remote Sensing Data data |
CN107907227A (en) * | 2017-11-10 | 2018-04-13 | 敦煌研究院 | A kind of earthen ruins surface and the measuring method of internal temperature change |
Non-Patent Citations (3)
Title |
---|
A neural network algorithm to retrieve nearsurface air temperature from landsat ETM+ imagery over the hangjiang river basin,china;DengZhong Zhao etal.;《2007 IEEE International Geoscience and Remote Sensing Symposium》;20070728;第1705-1708页 * |
支持向量机在土遗址保护中的应用研究;汝佳;《中国优秀硕士学位论文全文数据库 信息科技辑》;20131215;第15-35页 * |
汉阳陵帝陵外藏坑遗址温度变化规律及预报模型;姚雪 等;《敦煌研究》;20141231;第24卷(第6期);第69-74页 * |
Also Published As
Publication number | Publication date |
---|---|
CN108764461A (en) | 2018-11-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Falamarzi et al. | Estimating evapotranspiration from temperature and wind speed data using artificial and wavelet neural networks (WNNs) | |
Engelbrecht et al. | Multi-scale climate modelling over Southern Africa using a variable-resolution global model | |
Bates et al. | Stochastic downscaling of numerical climate model simulations | |
Ouali et al. | Fully nonlinear statistical and machine‐learning approaches for hydrological frequency estimation at ungauged sites | |
Weiss et al. | Incorporating bias error in calculating solar irradiance: implications for crop yield simulations | |
CN112163375B (en) | Long-time sequence near-ground ozone inversion method based on neural network | |
CN111783987A (en) | Farmland reference crop evapotranspiration prediction method based on improved BP neural network | |
Liang et al. | Regional climate model simulation of US–Mexico summer precipitation using the optimal ensemble of two cumulus parameterizations | |
CN107133686A (en) | City-level PM2.5 concentration prediction methods based on Spatio-Temporal Data Model for Spatial | |
CN104598765A (en) | Building energy consumption prediction method based on elastic adaptive neural network | |
Parker | Simulation and understanding in the study of weather and climate | |
CN104992068A (en) | Method for predicting nitrogen distribution of surface soil | |
Ratnam et al. | Improvements to the WRF seasonal hindcasts over South Africa by bias correcting the driving SINTEX-F2v CGCM fields | |
Elanchezhian et al. | Evaluating different models used for predicting the indoor microclimatic parameters of a greenhouse. | |
CN108764461B (en) | Earthen site temperature prediction method based on solar term characteristics | |
Zeng et al. | Effects of land surface schemes on WRF-simulated geopotential heights over China in summer 2003 | |
CN109272144A (en) | The prediction technique of grassland in northern China area NDVI based on BPNN | |
CN106202920A (en) | A kind of application and interpretation method of single sea level pressure of standing | |
CN210895535U (en) | Air temperature prediction system based on convolution cyclic neural network | |
CN113077110A (en) | GRU-based harmonic residual segmented tide level prediction method | |
Lee et al. | Assessment of cloud retrieval for IASI 1D-Var cloudy-sky assimilation and improvement with an ANN approach | |
Kang et al. | A case study for ANN-based rainfall–runoff model considering antecedent soil moisture conditions in Imha Dam watershed, Korea | |
Dunbar et al. | Ensemble-based experimental design for targeted high-resolution simulations to inform climate models | |
CN112734073A (en) | Photovoltaic power generation short-term prediction method based on long and short-term memory network | |
Li et al. | Application and validation of AquaCrop model in simulating biomass and yield of oil flax in Northwest China |
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 |