CN109905190B - Modeling method for low-frequency ground wave propagation time delay variation characteristic - Google Patents
Modeling method for low-frequency ground wave propagation time delay variation characteristic Download PDFInfo
- Publication number
- CN109905190B CN109905190B CN201910072164.8A CN201910072164A CN109905190B CN 109905190 B CN109905190 B CN 109905190B CN 201910072164 A CN201910072164 A CN 201910072164A CN 109905190 B CN109905190 B CN 109905190B
- Authority
- CN
- China
- Prior art keywords
- data
- propagation delay
- low
- frequency ground
- formula
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
本发明公开了一种低频地波传播时延时变特性的建模方法，具体按照以下步骤实施：步骤1：搭建低频地波传播时延的长期实验检测系统，获取传播时延数据并对数据进行预处理；步骤2：获取接收点所在区域的温度、湿度、风速三种气象因子的气象数据；步骤3：根据步骤1获取的传播时延数据、步骤2获取的温度、湿度、风速气象数据分析低频地波传播时延的时变特性；步骤4：利用BP神经网络方法建立传播时延的预测模型。本发明提高了低频地波定位导航和授时系统的精度。
The invention discloses a modeling method for the time-delay variation characteristics of low-frequency ground wave propagation. Perform preprocessing; Step 2: Obtain the meteorological data of three meteorological factors of temperature, humidity and wind speed in the area where the receiving point is located; Step 3: According to the propagation delay data obtained in Step 1, and the temperature, humidity, and wind speed meteorological data obtained in Step 2 Analyze the time-varying characteristics of the low-frequency ground wave propagation delay; Step 4: Use the BP neural network method to establish a prediction model of the propagation delay. The invention improves the precision of the low-frequency ground wave positioning, navigation and timing system.
Description
Technical Field
The invention belongs to the technical field of land-based navigation/time service, and particularly relates to a modeling method for low-frequency ground wave propagation delay variation characteristics.
Background
Due to the characteristics of large transmitting power, strong anti-interference capability, long acting distance and wide coverage range, low-frequency ground wave signals are widely applied to positioning, navigation and time service systems. In the process of low-frequency ground wave signal propagation, propagation delay can be generated, and the precision of system positioning, navigation and time service can be seriously influenced by the propagation delay. The propagation delay is related to the propagation distance of the transmitting point and the receiving point, and shows complex spatial variation and time variation under the influence of factors such as earth conductivity, atmospheric refractive index, terrain and ground features, seasons, weather and the like on a propagation path.
In the prior art, the precision of the low-frequency ground wave propagation delay in space is obviously improved, but the systematic research on the time-varying characteristic of the propagation delay is relatively few, and the influence of long-term seasons, climate variations, short-term temperature, humidity, precipitation and other weather variations on a propagation path on the propagation delay cannot be ignored. Based on the analysis, if the time-varying characteristic of the propagation delay can be systematically researched and a propagation delay time-varying model is constructed, the method has important significance for further improving the precision of the low-frequency ground wave positioning navigation and time service system.
Disclosure of Invention
The invention aims to provide a modeling method for low-frequency ground wave propagation delay variation characteristics, which improves the precision of a low-frequency ground wave positioning navigation and time service system.
The technical scheme adopted by the invention is that the modeling method of the low-frequency ground wave propagation time delay variation characteristic is implemented according to the following steps:
step 1: constructing a long-term experiment detection system of low-frequency ground wave propagation delay, acquiring propagation delay data and preprocessing the data;
step 2: acquiring meteorological data of three meteorological factors including temperature, humidity and wind speed of an area where a receiving point is located;
and step 3: analyzing the time-varying characteristic of the low-frequency ground wave propagation delay according to the propagation delay data acquired in the step 1 and the temperature, humidity and wind speed meteorological data acquired in the step 2;
and 4, step 4: and establishing a prediction model of propagation delay by using a BP neural network method.
The invention is also characterized in that:
in step 1, the detection system comprises a long wave receiver antenna; the long wave receiver antenna is in communication connection with the long wave receiver; the GPS receiver antenna is in communication connection with the GPS receiver; the long wave receiver and the GPS receiver are connected with the counter through serial interfaces; the long wave receiver, the GPS receiver and the counter are connected with the frosting box through serial interfaces; the frosting box is electrically connected with the industrial PC.
The model of the frosting box is USB-RS 232.
In the step 1, the pretreatment process is as follows: removing potentially invalid data; denoising by a wavelet filtering method so as to extract effective propagation delay measurement data; the data which is possibly invalid comprises data when the GPS receiver is unlocked, data when the long wave receiver is unlocked and decoding fails, data of sudden jump in the measuring process, sudden change data caused by starting and stopping of an instrument, partial data when measurement starts and ends, and data measured when signal-to-noise ratio is low.
The specific process of step 3 is as follows:
step 3.1: acquiring the time resolution T of the propagation delay data actually measured in the step 11；
Step 3.2: acquiring the time resolution of the temperature, humidity and wind speed meteorological data in the step 2; taking the maximum time resolution T of the three time resolutions2；
Step 3.3: calculating the correlation values between the propagation delay data acquired in the step 1 and the temperature, humidity and wind speed meteorological data acquired in the step 2
The concrete formula is as follows:
in the formula, Xi(i ═ 1,2, …, n) represents meteorological data; y represents the propagation delay; d (X)i) Representing a variance of the meteorological data; d (y) represents the variance of the propagation delay.
The specific process of the step 4 is as follows:
step 4.1: data set collation
The propagation delay data obtained in the step 1 and the temperature, humidity and wind speed meteorological data obtained in the step 2 are sorted, and effective data are selected; wherein the temporal resolution is taken as T2；
Step 4.2: BP neural network construction
The input layer of the BP neural network structure is n nodes; the output layer is m nodes; determining the range of the number of the hidden layer nodes according to an empirical formula; continuously reducing the range of the number of hidden layer nodes by utilizing a dichotomy so as to determine an optimal hidden layer node formula; wherein n is consistent with the number of meteorological factors in the step 2; m is 1; the empirical formula is as follows:
in the formula, n represents the number of nodes of an input layer; l is the number of hidden layer nodes; m is the number of nodes of the output layer; p is a constant between 0 and 10;
step 4.3: BP neural network training
Step 4.3.1: carrying out normalization processing on the data set to enable the data to be distributed between [0 and 1] so as to provide high-quality data for BP neural network training;
step 4.3.2: initializing a network; includes initializing connection weight w between neurons in input layer, hidden layer and output layerijAnd wjkHidden layer threshold ajOutput layer threshold bkInitializing a learning rate eta and a neuron excitation function f (x);
(1)wij、wjkthe calculation formula of (a) is as follows:
wjk＝wjk+ηHjek j＝1，2，…，l；k＝1，2，…，m (4)
wherein x (i) represents input temperature, humidity and wind speed meteorological data; eta represents learning rate, is a positive number, and is between 0.01 and 0.1;
(2)aj、bkthe calculation formula of (a) is as follows:
bk＝bk+ek k＝1，2，…，m (6)
(3) the calculation formula of f (x) is as follows:
step 4.3.3: computing the output H of the hidden layerj：
Step 4.3.4: computing output O of output layerk：
Step 4.3.5: calculation of error ek：
ek＝Yk-Ok k＝1，2，…，m (10)
Step 4.3.6: BP neural network prediction and verification
Randomly selecting half of data to train the network, and testing the network performance by half of the data; comparing the measured value with the predicted value, and carrying out quantitative performance comparison by using a Root Mean Square Error (RMSE) value and an average absolute error (MAE) value; the calculation formula is as follows:
in the formula, yiAnd y'iRespectively, an actual measurement value and a predicted value of the propagation delay.
The invention has the beneficial effects that:
(1) further, the change of the low-frequency ground wave propagation delay is not only space-variant, but also time-variant, and for a high-precision navigation time service system, the time-variant characteristic of the propagation delay cannot be ignored;
(2) the analysis of the propagation time delay variation characteristics is simplified; the main causes of the propagation delay variation with time are: firstly, the refractive index of the atmosphere on a propagation path changes along with time; the atmospheric refractive index shows the characteristic of time variation due to the change of factors such as atmospheric temperature, humidity and pressure in the troposphere along with time, so that the propagation characteristic of electric waves is influenced, and the influence is more obvious when the propagation distance is longer; secondly, the earth electrical parameter (mainly earth electrical conductivity) on the propagation path changes along with time; the ground conductivity is related to the type of ground medium (such as soil, desert, lake, seawater, etc.), and also related to the temperature, water content, salinity, minerals, etc. of the medium, which are closely related to weather, climate, seasonal variation, etc.; the time-varying characteristic of the propagation delay is indirectly reflected by meteorological factors such as temperature, humidity and the like, so that the analysis of the time-varying characteristic of the propagation delay is simplified;
(3) establishing a propagation delay prediction model; based on the analysis of the propagation delay time-varying characteristics, a prediction model for predicting the propagation delay by using three meteorological data of temperature, humidity and wind speed is established by adopting a BP neural network method; through the prediction model, the propagation delay can be predicted more simply, conveniently and accurately, so that beneficial guidance is provided for improving the precision of the low-frequency ground wave positioning navigation and time service system.
Drawings
FIG. 1 is a flow chart of a modeling method of low-frequency ground wave propagation delay variation characteristics according to the invention;
FIG. 2 is a schematic diagram of the daily variation of propagation delay of Xian 2016.11.2-2016.11.5 days actually measured in the modeling method of low-frequency ground wave propagation delay variation characteristics of the present invention;
FIG. 3 is a schematic diagram of daily variation of weather factors of temperature of the Saian at 2016.11.2-2016.11.5 days actually acquired in the modeling method of the low-frequency ground wave propagation delay time variation characteristic of the invention;
FIG. 4 is a schematic diagram of daily variation of humidity meteorological factors of Xian 2016.11.2-2016.11.5 days actually acquired in the modeling method of low-frequency ground wave propagation delay variation characteristics of the present invention;
FIG. 5 is a schematic diagram of daily variation of wind speed meteorological factors of 2016.11.2-2016.11.5 days of Xian actually acquired in the modeling method of low-frequency ground wave propagation delay time variation characteristics of the invention;
FIG. 6 is a comparison graph of the predicted propagation delay values of Weian 2016.11.2-2016.11.5 days and the actual measurement values of the propagation delay of Weian 2016.11.2-2016.11.5 days by using the modeling method of the low-frequency ground wave propagation delay time-varying characteristic of the invention;
FIG. 7 is a schematic connection diagram of a detection system built in the modeling method for low-frequency ground wave propagation delay variation characteristics of the invention;
FIG. 8 is a workflow diagram of software used by an industrial PC in a detection system built in the modeling method of low-frequency ground wave propagation delay variation characteristics of the invention.
In the figure, 1 is a long wave receiver antenna, 2 is a long wave receiver, 3 is a GPS receiver antenna, 4 is a GPS receiver, 5 is a counter, 6 is a sanding box, and 7 is an industrial PC.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, the modeling method for the low-frequency ground wave propagation delay time-varying characteristic of the present invention is specifically implemented according to the following steps:
step 1: constructing a long-term experiment detection system of low-frequency ground wave propagation delay, acquiring propagation delay data and preprocessing the data; (as shown in FIG. 2)
The pretreatment process comprises the following steps: removing potentially invalid data; denoising by a wavelet filtering method so as to extract effective propagation delay measurement data; the data which is possibly invalid comprises data when the GPS receiver is unlocked, data when the long wave receiver is unlocked and decoding fails, data of sudden jump in the measuring process, sudden change data caused by starting and stopping of an instrument, partial data when measurement starts and ends, and data measured when signal-to-noise ratio is low.
Step 2: acquiring meteorological data of three meteorological factors including temperature, humidity and wind speed of an area where a receiving point is located; (as shown in FIGS. 3-5)
The meteorological data of the three meteorological factors of temperature, humidity and wind speed can be acquired through a Chinese weather website.
And step 3: analyzing the time-varying characteristic of the low-frequency ground wave propagation delay according to the propagation delay data acquired in the step 1 and the temperature, humidity and wind speed meteorological data acquired in the step 2; the specific process is as follows:
step 3.1: acquiring the time resolution T of the propagation delay data actually measured in the step 11；
Step 3.2: acquiring the time resolution of the temperature, humidity and wind speed meteorological data in the step 2; taking the maximum time resolution T of the three time resolutions2；
Step 3.3: calculating the correlation values between the propagation delay data acquired in the step 1 and the temperature, humidity and wind speed meteorological data acquired in the step 2
The concrete formula is as follows:
in the formula, Xi(i ═ 1,2, …, n) represents meteorological data; y represents the propagation delay; d (X)i) Representing a variance of the meteorological data; d (y) represents the variance of the propagation delay.
And 4, step 4: establishing a prediction model of propagation delay by using a BP neural network method; the specific process is as follows:
step 4.1: data set collation
The propagation delay data obtained in the step 1 and the temperature, humidity and wind speed meteorological data obtained in the step 2 are sorted, and effective data are selected; wherein the temporal resolution is taken as T2；
Step 4.2: BP neural network construction
The input layer of the BP neural network structure is n nodes; the output layer is m nodes; determining the range of the number of the hidden layer nodes according to an empirical formula; continuously reducing the range of the number of hidden layer nodes by utilizing a dichotomy so as to determine an optimal hidden layer node formula; wherein n is consistent with the number of meteorological factors in the step 2; m is 1; the empirical formula is as follows:
in the formula, n represents the number of nodes of an input layer; l is the number of hidden layer nodes; m is the number of nodes of the output layer; p is a constant between 0 and 10;
step 4.3: BP neural network training
Step 4.3.1: carrying out normalization processing on the data set to enable the data to be distributed between [0 and 1] so as to provide high-quality data for BP neural network training;
step 4.3.2: initializing a network; includes initializing connection weight w between neurons in input layer, hidden layer and output layerijAnd wjkHidden layer threshold ajOutput layer threshold bkInitializing a learning rate eta and a neuron excitation function f (x);
(1)wij、wjkthe calculation formula of (a) is as follows:
wjk＝wjk+ηHjek j＝1，2，…l；k＝1，2，…，m (4)
wherein x (i) represents input temperature, humidity and wind speed meteorological data; eta represents learning rate, is a positive number, and is between 0.01 and 0.1;
(2)aj、bkthe calculation formula of (a) is as follows:
bk＝bk+ek k＝1，2，…，m (6)
(3) the calculation formula of f (x) is as follows:
step 4.3.3: computing the output H of the hidden layerj：
Step 4.3.4: computing output O of output layerk：
Step 4.3.5: calculation of error ek：
ek＝Yk-Ok k＝12…，m (1O)
Step 4.3.6: BP neural network prediction and verification
Randomly selecting half of data to train the network, and testing the network performance by half of the data; comparing the measured value with the predicted value (as shown in fig. 6), and performing quantitative performance comparison by using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values; the calculation formula is as follows:
in the formula, yiAnd y'iRespectively, an actual measurement value and a predicted value of the propagation delay.
As shown in fig. 7, in step 1, the detection system includes a long wave receiver antenna 1; the long wave receiver antenna 1 is in communication connection with the long wave receiver 2; the GPS receiver antenna 3 is in communication connection with the GPS receiver 4; the long wave receiver 2 and the GPS receiver 4 are both connected with the counter 5 through serial interfaces; the long wave receiver 2, the GPS receiver 4 and the counter 5 are connected with the sanding box 6 through serial interfaces; the sanding box 6 is electrically connected with the industrial PC 7; wherein the model of the sanding box 6 is USB-RS 232.
As shown in fig. 8, the working process of the software employed by the industrial PC7 is: starting a serial interface; receiving data output by the long wave receiver 2; receiving data output by the GPS receiver 4; receiving data output by the counter 5; and meanwhile, displaying the data in real time and storing the data.
The invention has the advantages that:
(1) further, the change of the low-frequency ground wave propagation delay is not only space-variant, but also time-variant, and for a high-precision navigation time service system, the time-variant characteristic of the propagation delay cannot be ignored;
(2) the analysis of the propagation time delay variation characteristics is simplified; the main causes of the propagation delay variation with time are: firstly, the refractive index of the atmosphere on a propagation path changes along with time; the atmospheric refractive index shows the characteristic of time variation due to the change of factors such as atmospheric temperature, humidity and pressure in the troposphere along with time, so that the propagation characteristic of electric waves is influenced, and the influence is more obvious when the propagation distance is longer; secondly, the earth electrical parameter (mainly earth electrical conductivity) on the propagation path changes along with time; the ground conductivity is related to the type of ground medium (such as soil, desert, lake, seawater, etc.), and also related to the temperature, water content, salinity, minerals, etc. of the medium, which are closely related to weather, climate, seasonal variation, etc.; the time-varying characteristic of the propagation delay is indirectly reflected by meteorological factors such as temperature, humidity and the like, so that the analysis of the time-varying characteristic of the propagation delay is simplified;
(3) establishing a propagation delay prediction model; based on the analysis of the propagation delay time-varying characteristics, a prediction model for predicting the propagation delay by using three meteorological data of temperature, humidity and wind speed is established by adopting a BP neural network method; through the prediction model, the propagation delay can be predicted more simply, conveniently and accurately, so that beneficial guidance is provided for improving the precision of the low-frequency ground wave positioning navigation and time service system.
Claims (5)
1. A modeling method for low-frequency ground wave propagation delay variation characteristics is characterized by comprising the following steps:
step 1: constructing a long-term experiment detection system of low-frequency ground wave propagation delay, acquiring propagation delay data and preprocessing the data;
step 2: acquiring meteorological data of three meteorological factors including temperature, humidity and wind speed of an area where a receiving point is located;
and step 3: analyzing the time-varying characteristic of the low-frequency ground wave propagation delay according to the propagation delay data acquired in the step 1 and the temperature, humidity and wind speed meteorological data acquired in the step 2;
and 4, step 4: establishing a prediction model of propagation delay by using a BP neural network method;
the specific process of the step 4 is as follows:
step 4.1: data set collation
The propagation delay data obtained in the step 1 and the temperature, humidity and wind speed meteorological data obtained in the step 2 are sorted, and effective data are selected; wherein the temporal resolution is taken as T2；
Step 4.2: BP neural network construction
The input layer of the BP neural network structure is n nodes; the output layer is m nodes; determining the range of the number of the hidden layer nodes according to an empirical formula; continuously reducing the range of the number of hidden layer nodes by utilizing a dichotomy so as to determine an optimal hidden layer node formula; wherein n is consistent with the number of meteorological factors in the step 2; m is 1; the empirical formula is as follows:
in the formula, n represents the number of nodes of an input layer; l is the number of hidden layer nodes; m is the number of nodes of the output layer; p is a constant between 0 and 10;
step 4.3: BP neural network training
Step 4.3.1: carrying out normalization processing on the data set to enable the data to be distributed between [0 and 1] so as to provide high-quality data for BP neural network training;
step 4.3.2: initializing a network; includes initializing connection weight w between neurons in input layer, hidden layer and output layerijAnd wjkHidden layer threshold ajAnd an outputLayer threshold bkInitializing a learning rate eta and a neuron excitation function f (x);
(1)wij、wjkthe calculation formula of (a) is as follows:
wjk＝wjk+ηHjek j＝1,2,…,l；k＝1,2,…,m (4)
in the formula, xiRepresenting input temperature, humidity and wind speed meteorological data; eta represents learning rate, is a positive number, and is between 0.01 and 0.1;
(2)aj、bkthe calculation formula of (a) is as follows:
bk＝bk+ek k＝1,2,…,m (6)
(3) the calculation formula of f (x) is as follows:
step 4.3.3: computing the output H of the hidden layerj：
Step 4.3.4: computing output O of output layerk：
Step 4.3.5: calculation of error ek：
ek＝Yk-Ok k＝1,2,…,m (10)
Step 4.3.6: BP neural network prediction and verification
Randomly selecting half of data to train the network, and testing the network performance by half of the data; comparing the measured value with the predicted value, and carrying out quantitative performance comparison by using a Root Mean Square Error (RMSE) value and an average absolute error (MAE) value; the calculation formula is as follows:
in the formula, yiAnd y'iRespectively, an actual measurement value and a predicted value of the propagation delay.
2. A method for modelling the characteristics of delay variation of the propagation of low-frequency ground waves according to claim 1, characterized in that in step 1 said detection system comprises a long-wave receiver antenna (1); the long wave receiver antenna (1) is in communication connection with the long wave receiver (2); the GPS receiver antenna (3) is in communication connection with the GPS receiver (4); the long wave receiver (2) and the GPS receiver (4) are connected with the counter (5) through serial interfaces; the long wave receiver (2), the GPS receiver (4) and the counter (5) are connected with the sanding box (6) through serial interfaces; the sanding box (6) is electrically connected with the industrial PC (7).
3. A modeling method of low frequency ground wave propagation delay time variation characteristics according to claim 2, characterized in that the model of the said sanding box (6) is USB-RS 232.
4. The modeling method for the low-frequency ground wave propagation delay time variation characteristic according to claim 1, wherein in the step 1, the preprocessing specifically comprises the following steps: removing potentially invalid data; denoising by a wavelet filtering method so as to extract effective propagation delay measurement data; the data which is possibly invalid comprises data when the GPS receiver is unlocked, data when the long wave receiver is unlocked and decoding fails, data of sudden jump in the measuring process, sudden change data caused by starting and stopping of an instrument, partial data when measurement starts and ends, and data measured when signal-to-noise ratio is low.
5. The modeling method of low-frequency ground wave propagation delay time variation characteristics according to claim 1, characterized in that the specific process of the step 3 is as follows:
step 3.1: acquiring the time resolution T of the propagation delay data actually measured in the step 11；
Step 3.2: acquiring the time resolution of the temperature, humidity and wind speed meteorological data in the step 2; taking the maximum time resolution T of the three time resolutions2；
Step 3.3: calculating a correlation value rho between the propagation delay data acquired in the step 1 and the temperature, humidity and wind speed meteorological data acquired in the step 2; the concrete formula is as follows:
in the formula, xiRepresenting meteorological data; y isiRepresenting the propagation delay; d (x)i) Representing a variance of the meteorological data; d (y)i) Representing the variance of the propagation delay.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910072164.8A CN109905190B (en) | 2019-01-25 | 2019-01-25 | Modeling method for low-frequency ground wave propagation time delay variation characteristic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910072164.8A CN109905190B (en) | 2019-01-25 | 2019-01-25 | Modeling method for low-frequency ground wave propagation time delay variation characteristic |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109905190A CN109905190A (en) | 2019-06-18 |
| CN109905190B true CN109905190B (en) | 2021-09-10 |
Family
ID=66944081
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910072164.8A Active CN109905190B (en) | 2019-01-25 | 2019-01-25 | Modeling method for low-frequency ground wave propagation time delay variation characteristic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109905190B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111985684B (en) * | 2020-07-14 | 2023-09-22 | 西安理工大学 | Long-wave ground wave propagation time delay time-varying characteristic prediction method applied to long distance |
| CN113627074B (en) * | 2021-07-13 | 2024-04-19 | 西安理工大学 | Ground wave propagation delay prediction method based on transfer learning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059543A (en) * | 2007-05-16 | 2007-10-24 | 西安理工大学 | Measuring method for time-delay correction value for long wave transmission |
| CN105740204A (en) * | 2016-03-14 | 2016-07-06 | 西安理工大学 | Low-frequency-band earth conductivity rapid inversion method under irregular terrain |
| CN108053061A (en) * | 2017-12-08 | 2018-05-18 | 天津大学 | A kind of solar energy irradiation level Forecasting Methodology based on improvement convolutional neural networks |
-
2019
- 2019-01-25 CN CN201910072164.8A patent/CN109905190B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059543A (en) * | 2007-05-16 | 2007-10-24 | 西安理工大学 | Measuring method for time-delay correction value for long wave transmission |
| CN105740204A (en) * | 2016-03-14 | 2016-07-06 | 西安理工大学 | Low-frequency-band earth conductivity rapid inversion method under irregular terrain |
| CN108053061A (en) * | 2017-12-08 | 2018-05-18 | 天津大学 | A kind of solar energy irradiation level Forecasting Methodology based on improvement convolutional neural networks |
Non-Patent Citations (2)
| Title |
|---|
| Analysis of Diurnal Variation on Long-wave Ground Wave Propagation Delay;Yang Hongjuan等;《2017 Sixth Asia-Pacific Conference on Antennas and Propagation (APCAP)》;20171019;全文 * |
| 基于熵权_BP神经网络模型的生态系统安全评价;杨红娟;《昆明理工大学学报》;20171031;第42卷(第5期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109905190A (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108241779B (en) | Ground PM2.5 Density feature vector space filter value modeling method based on remotely-sensed data | |
| CN111401602B (en) | Assimilation method for satellite and ground rainfall measurement values based on neural network | |
| CN113255972B (en) | Short-term precipitation prediction method based on Attention mechanism | |
| CN114935759B (en) | Method and system for filling missing values in wave field based on high-frequency ground wave radar observation | |
| CN114861537A (en) | GNSS-R sea surface wind speed inversion method and system based on CNN multi-information fusion | |
| CN111721714B (en) | A Soil Moisture Content Estimation Method Based on Multi-source Optical Remote Sensing Data | |
| CN113468804B (en) | Underground pipeline identification method based on matrix bundles and deep neural network | |
| CN115544706A (en) | An Estimation Method of Atmospheric Fine Particle Concentration Based on Wavelet and XGBoost Model Integration | |
| CN109905190B (en) | Modeling method for low-frequency ground wave propagation time delay variation characteristic | |
| CN117216480B (en) | A remote sensing estimation method for near-surface ozone based on deep coupling of geographic temporal and spatial information | |
| CN115357847A (en) | A daily-scale satellite-terrestrial precipitation fusion method based on error decomposition | |
| CN116990772A (en) | Ground penetrating radar double-parameter real-time inversion method based on multi-scale convolution network | |
| CN116821694A (en) | Soil moisture inversion method based on multi-branch neural network and segmented model | |
| CN119337725A (en) | A spaceborne GNSS-R soil moisture inversion method and system | |
| CN110806444B (en) | Identification and classification of seabed sediments based on shallow stratigraphic profiler and SVM | |
| CN116304590A (en) | Automatic global ocean mesoscale frontal surface identification method based on sliding window threshold | |
| CN114692692B (en) | Snowfall recognition method based on microwave attenuation signal fusion kernel extreme learning machine | |
| CN114239274A (en) | Method for calculating root layer soil water by multi-source remote sensing data driven index filtering model | |
| CN118886312A (en) | A GNSS-IR soil moisture inversion method considering climate and environment | |
| Wang et al. | Early warning of debris flow using optimized self-organizing feature mapping network | |
| CN117195958A (en) | Surface sea current prediction method based on attention mechanism TCN-LSTM model | |
| Choi et al. | Storm identification and tracking algorithm for modeling of rainfall fields using 1-h NEXRAD rainfall data in Texas | |
| Guo et al. | Research on tunnel lining image target recognition method based on YOLOv3 | |
| Yang et al. | Reconstruction of Arctic sea ice thickness and its impact on sea ice forecasting in the melting season | |
| CN114966902B (en) | A method and system for automatic rainfall estimation based on neural network |
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 |