CN111639746B - GNSS-R sea surface wind speed inversion method and system based on CNN neural network - Google Patents

Abstract

The invention discloses a GNSS-R sea surface wind speed inversion method and system based on a CNN neural network, wherein the method comprises the following steps: inputting a DDM diagram to be tested into a pre-trained sea surface wind speed inversion model, and outputting a corresponding inversion wind speed; the sea surface wind speed inversion model is a CNN neural network. According to the invention, the CNN neural network is utilized to invert the GNSS-R sea surface wind speed, the model is simple, the modeling time and inversion time are shortened, and the inversion precision is further improved; the CNN neural network fully utilizes the physical quantity related to the wind speed in the DDM to perform characteristic learning, reduces the calculated quantity and shortens the time consumption under the condition of guaranteeing the inversion precision, and has the characteristics of simple and quick model, high result precision and the like.

Description

A GNSS-R sea surface wind speed inversion method and system based on CNN neural network

Technical field

The invention relates to the field of atmospheric scientific research, and specifically relates to a method and system for GNSS-R sea surface wind speed retrieval based on CNN neural network.

Background technique

Sea surface wind speed is a vital physical parameter in ocean state information and can currently be detected through GNSS-R satellite remote sensing technology. Because GNSS-R technology has the characteristics of high global coverage and high spatial and temporal resolution, it can obtain high-quality sea surface wind speed detection data. Currently, there are two main GNSS-R wind speed inversion methods:

Waveform matching method: First, the system status information needs to be extracted based on the measured data, then the theoretical model simulation waveform is generated, and finally the theoretical waveform diagram is obtained through normalization processing, and a simulation waveform database is established based on a large number of theoretical waveform diagrams; during inversion, it is generated from the measured data The waveform to be measured is processed with noise reduction and normalization. Match the waveform to be measured with the theoretical waveform in the database. The wind speed corresponding to the successfully matched theoretical waveform is the sea surface wind speed of the data to be measured. However, the disadvantage of this method is that the amount of calculation is large, and the establishment of a detailed database is extremely time-consuming.

Empirical function method: Through the experience summary of a large amount of measured data, one or two physical parameters with high correlation with sea surface wind speed are selected from DDM, and regression linear fitting is performed to establish a functional mapping between them and sea surface wind speed to obtain wind speed. . However, sea surface wind speed is often not determined by just one or two parameters, so the accuracy of this method will be affected by neglecting other physical parameters.

Contents of the invention

The purpose of the present invention is to overcome the shortcomings of the above two methods, which mainly include the waveform matching method, which requires a large amount of calculation and long inversion time; and the empirical function method, which has poor inversion accuracy. The present invention proposes a GNSS-R sea surface wind speed inversion method based on CNN neural network. Compared with the waveform matching method, this method does not need to establish a huge simulation database. Compared with the empirical function method, this method can establish multiple observations. The relationship between the quantity and the sea surface wind speed can make full use of the physical quantities related to the wind speed inversion. Therefore, this method can further shorten the inversion time and improve the inversion accuracy.

In order to achieve the above objectives, Embodiment 1 of the present invention proposes a GNSS-R sea surface wind speed inversion method based on CNN neural network. The method includes:

The DDM map to be measured is input into the pre-trained sea surface wind speed inversion model, and the corresponding inverted wind speed is output; the sea surface wind speed inversion model is a CNN neural network.

As an improvement of the above method, the CNN neural network includes in sequence: an input layer, a first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer and an output layer, where the input layer The input is a DDM graph, and the number of nodes is 2560; the first convolutional layer C1 outputs a feature map number of 1, and the number of neurons is 1024; the second convolutional layer outputs a feature map number of 32, 64, and 128 in sequence. , 256, the number of neurons is 512, the convolution kernel size set by the two convolutional layers is 3*3; the domain size of the first pooling layer and the second pooling layer is both 2*2; convolution The number of neurons in the fully connected layer between the layer and the output layer is 16; the number of nodes in the output layer is 1, and its output is the sea surface wind speed; the activation function of the CNN neural network is the ReLU function; the loss function is the MSE function, The evaluation index is root mean square error RMSE.

As an improvement of the above method, the method also includes: training steps of CNN neural network, specifically including:

Select multiple sets of GNSS-R data and ECMWF analysis field data, and perform spatio-temporal matching to obtain the original sample set. Each set of samples consists of a DDM map and the corresponding wind speed;

Preprocess the original sample set and split it into a training set and a test set;

The CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM diagram and establish a mapping relationship with the wind speed;

Use the test set data to test the trained CNN neural network.

As an improvement of the above method, the original sample set is preprocessed and divided into a training set and a test set; specifically including:

Filter the original sample set data based on latitude and longitude, wind speed and signal-to-noise ratio;

Preprocess the filtered original sample set data based on the sampling algorithm and normalization algorithm;

The preprocessed original sample set is divided into a training set and a test set in a ratio of 7:3.

As an improvement of the above method, the CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM diagram and establish a mapping relationship with the wind speed, specifically including:

After the weights between neurons in the CNN neural network and the threshold in each neuron are initialized, the training set data is passed from the input layer to the first convolution layer for convolution calculation, and the feature map output after convolution Through nonlinear activation function processing, the data space dimension is reduced through the first pooling layer to obtain the feature matrix; then the second convolution layer and the second pooling layer are input to perform convolution and pooling operations, and the convolution layer is processed through the fully connected layer. All the data features learned by the product are combined, normalized and transferred to the output layer. The error between the output wind speed and the real wind speed is calculated through the loss function. The result is transferred from the output layer in reverse layer by layer and each convolution layer is adjusted. and the weight of the pooling layer;

Repeat the forward propagation and back propagation processes, and the result of the loss function gradually decreases until the result is within the expected error range or reaches the set number of training times. The CNN neural network training is completed, and the nonlinear mapping of the DDM map to sea surface wind speed is realized. .

Embodiment 2 of the present invention proposes a GNSS-R sea surface wind speed inversion system based on CNN neural network. The system includes: a trained sea surface wind speed inversion model and a wind speed inversion module; the sea surface wind speed inversion model is a CNN neural network;

The wind speed inversion module is used to input the DDM map to be measured into the trained sea surface wind speed inversion model and output the corresponding inverted wind speed.

As an improvement of the above system, the CNN neural network includes in sequence: an input layer, a first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer and an output layer, where the input layer The input is a DDM graph, and the number of nodes is 2560; the first convolutional layer C1 outputs a feature map number of 1, and the number of neurons is 1024; the second convolutional layer outputs a feature map number of 32, 64, and 128 in sequence. , 256, the number of neurons is 512, the convolution kernel size set by the two convolutional layers is 3*3; the domain size of the first pooling layer and the second pooling layer is both 2*2; convolution The number of neurons in the fully connected layer between the layer and the output layer is 16; the number of nodes in the output layer is 1, and its output is the sea surface wind speed; the activation function of the CNN neural network is the ReLU function; the loss function is the MSE function, The evaluation index is root mean square error RMSE.

As an improvement to the above system, the training steps of the CNN neural network specifically include:

Select multiple sets of GNSS-R data and ECMWF analysis field data, and perform spatio-temporal matching to obtain the original sample set. Each set of samples consists of a DDM map and the corresponding wind speed;

Preprocess the original sample set and split it into a training set and a test set;

The CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM diagram and establish a mapping relationship with the wind speed;

Use the test set data to test the trained CNN neural network.

As an improvement of the above system, the original sample set is preprocessed and divided into a training set and a test set; specifically including:

Filter the original sample set data based on latitude and longitude, wind speed and signal-to-noise ratio;

Preprocess the filtered original sample set data based on the sampling algorithm and normalization algorithm;

The preprocessed original sample set is divided into a training set and a test set in a ratio of 7:3.

As an improvement of the above system, the CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM diagram and establish a mapping relationship with the wind speed, specifically including:

After the weights between neurons in the CNN neural network and the threshold in each neuron are initialized, the training set data is passed from the input layer to the first convolution layer for convolution calculation, and the feature map output after convolution Through nonlinear activation function processing, the data space dimension is reduced through the first pooling layer to obtain the feature matrix; then the second convolution layer and the second pooling layer are input to perform convolution and pooling operations, and the convolution layer is processed through the fully connected layer. All the data features learned by the product are combined, normalized and transferred to the output layer. The error between the output wind speed and the real wind speed is calculated through the loss function. The result is transferred from the output layer in reverse layer by layer and each convolution layer is adjusted. and the weight of the pooling layer;

Repeat the forward propagation and back propagation processes, and the result of the loss function gradually decreases until the result is within the expected error range or reaches the set number of training times. The CNN neural network training is completed, and the nonlinear mapping of the DDM map to sea surface wind speed is realized. .

The advantages of the present invention are:

1. The present invention proposes a GNSS-R sea surface wind speed inversion method based on CNN (Convolutional Neural Network) convolutional neural network. Compared with the waveform matching method, this method does not require the establishment of a huge simulation database, and is compared with the empirical function method. Compared with this method, this method can establish the relationship between multiple observations and sea surface wind speed, and can make full use of the physical quantities related to wind speed inversion. Therefore, this method can further shorten the inversion time and improve the inversion accuracy;

2. This invention uses CNN neural network to invert GNSS-R sea surface wind speed. The model is simple, shortens the modeling time and inversion time, and further improves the inversion accuracy;

3. This invention makes full use of the physical quantities related to wind speed in the DDM diagram for feature learning based on the CNN neural network. It reduces the amount of calculation and shortens the time-consuming while ensuring the accuracy of the inversion. It has the characteristics of simple and fast model and high accuracy of the results. ;

4. The method of the present invention has the advantage of relatively efficiently retrieving sea surface wind speed based on CNN neural network, and can meet the needs of using a large amount of GNSS-R satellite data to conduct atmospheric research related to sea surface wind fields.

Description of the drawings

Figure 1 is a flow chart of the GNSS-R sea surface wind speed inversion method based on CNN neural network of the present invention;

Figure 2 is a schematic diagram of the CNN neural network of the sea surface wind speed inversion model.

Detailed ways

The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

Example 1

As shown in Figure 1, Embodiment 1 of the present invention provides a GNSS-R sea surface wind speed inversion method based on CNN neural network, which mainly includes the following steps:

Step one: Construction of original data sample set. Match the GNSS-R data and ECMWF data in space and time to form the original sample set;

A large amount of GNSS-R data and ECMWF analysis field data are selected for spatio-temporal matching to obtain the original sample set. Each set of samples is composed of a DDM map and corresponding wind speed information.

Step 2: Generate training set and test set. Preprocess the original sample set and split it into a training set and a test set;

In order to avoid data anomalies and noise interference, it is necessary to screen the longitude and latitude, wind speed, and signal-to-noise ratio (SNR) of the data; after screening, the problems of uneven distribution and inconsistent dimensions of the data set are solved based on the sampling algorithm and normalization algorithm. Divide the preprocessed sample set into a training set and a test set according to 7:3.

Step 3: Build CNN neural network. Build a CNN neural network model with DDM map as input and wind speed as output;

First determine the activation function of the network, then determine the number of network layers, the number of network nodes in each layer, and the optimal number of iterations of the model based on continuous experiments, then select the evaluation indicators of the model, and finally build a model with the DDM diagram as input and wind speed as output. CNN neural network model.

Step 4: CNN neural network training. The CNN model is continuously trained through the training set data, so that the model can continuously capture the data characteristics in the DDM diagram and establish a mapping relationship with the wind speed;

After the weights between neurons in the network and the threshold in each neuron are initialized, the training set data is passed from the input layer to the first convolution layer for convolution calculation, and the feature map output after convolution is activated through nonlinear Function processing, and then reducing the data space dimension through the pooling layer to obtain the feature matrix. By analogy, the convolution and pooling operations are repeated. After completing all the convolution operations, all the data features learned by convolution are combined through the fully connected layer in the network, and finally the output of the feature map is translated through normalization. When the sensitivity of the transformation decreases, the result is transferred to the output layer, and the error between the output wind speed and the real wind speed is calculated through the loss function. The result is transferred back from the output layer layer by layer and the weight of each hidden layer is adjusted. Repeat the above forward propagation and back propagation processes, and the result of the loss function gradually decreases until the result is within the expected error range of the model or reaches the number of training times set by the model. The predicted wind speed is close to the true wind speed, and the model training is deemed to be completed. , to achieve nonlinear mapping from DDM map to sea surface wind speed.

Step 5: Model testing. Verify the accuracy and reliability of the trained model through the test set;

Use the test set data to test the trained model. If the error results of several test sets are similar, the model is robust. Then judge the accuracy of the model inversion results based on whether the RMSE obtained by the model is within 2m/s. and reliability. If the above conditions are met, the CNN neural network sea surface wind speed inversion model is obtained.

Step 6: Data inversion. Input the DDM map to be measured into the CNN neural network sea surface wind speed inversion model to obtain the corresponding inverted wind speed.

Using the observation data of the SGR-ReSI GNSS-R receiver on the TDS-1 satellite, the CNN neural network based on the present invention is used to invert the GNSS-R sea surface wind speed. The TDS-1 satellite was launched in 2014. It operates in an orbit with an altitude of 635km and an inclination angle of 98.4°. It uses the SSTL-150 platform. The eight payloads on the satellite perform periodic turns. During one working cycle, the SGR -ReSI's working time is 1-2 days, during which it receives and processes reflected signals from GNSS satellites on the surface. The TDS-1 satellite incoherently accumulates the attribute data attached to the reflected signal to generate a Doppler (Delay-DopplerMap, DDM) map. The size of the DDM map is 128 delay pixels multiplied by 20 Doppler pixels, and the Doppler resolution is 500Hz and the latency resolution is 0.25chips.

As shown in Figure 1, it includes the following six steps:

The first step is to construct the original data sample: space-time matching of GNSS-R data and ECMWF data. This example uses TDS-1 satellite data and ECMWF analysis field data from February to October 2018, and matches them according to time, longitude, and latitude to obtain the original sample set. The data volume of the entire sample set reaches more than two million samples, basically Covering all ocean areas, the wind speed range is between 0 and 20m/s.

The second step is to generate a training set and a test set: filter the original sample set to eliminate sample points in the sea ice area with a latitude higher than 55° in the northern and southern hemispheres. At the same time, in order to avoid noise affecting the wind speed inversion accuracy, only leaving 3 to Data within the 18m/s range with a signal-to-noise ratio (SNR) greater than 3. There are approximately 370,000 sample data after screening, and most of the wind speeds are between 3-10m/s. A hybrid sampling algorithm is used to make the distribution even, and a normalization algorithm is used to map the wind speed of the sample set to the range of 0 to 1 to ensure consistent dimensions. , and finally the processed sample set is divided into a training set and a test set according to 7:3, where the number of training sets is 73,500 and the number of test sets is 31,500.

The third step is to build a CNN neural network: first determine that the activation function of the network is the ReLU function, and then determine based on continuous experiments that the network consists of input layer i, convolution layer C1, pooling layer S1, convolution layer C2, pooling layer S2, and all It consists of connection layer and output layer. The number of input layer nodes is 2560, the number of output feature maps of convolution layer C1 is 1, and the number of neurons is 1024. The number of output feature maps of convolution layer C2 is 32, 64, 128, 256, and the number of neurons is 512, the convolution kernel size set in the convolution layer is 3*3; the domain size of the pooling layers S1 and S2 are both 2*2, and the number of neurons in the fully connected layer between the convolution layer and the output layer There are 16, the number of output layer nodes is 1, the loss function is the MSE function, and then the evaluation index of the model is selected to be the root mean square error RMSE. In this way, a CNN neural network model is built with the DDM map as the input and the wind speed as the output.

The fourth step of CNN neural network training: After the weights between neurons in the network and the threshold in each neuron are initialized, the input layer receives the DDM graph as input data and passes it into the first convolution layer, using this layer Each convolution kernel acts on all feature spaces of the DDM map. The pixels in each sensing area and their corresponding connection weights are calculated as a weighted sum. The result plus the offset value is the calculation result of the convolution kernel. Convolution The finally output feature map is processed by the ReLU nonlinear activation function, and then the dimension of the data space is reduced by max pooling to obtain the feature matrix. By analogy, the convolution and pooling operations are repeated. After completing all the convolution operations, all the data features learned by convolution are combined through the fully connected layer in the network, and finally the output of the feature map is translated through normalization. When the sensitivity of the transformation decreases, the result is transferred to the output layer. The error between the output wind speed and the real wind speed is calculated through the loss function MSE. The result is transferred from the output layer in reverse layer by layer and the weight of each hidden layer is adjusted. By repeating the above forward propagation and back propagation processes, the calculation error of the neural network model continues to decrease, making the predicted value gradually approach the true wind speed. When the calculation error no longer decreases as the number of training times increases, the optimal model can be obtained, and the model training is deemed to be completed, and the nonlinear mapping of the DDM map to sea surface wind speed is achieved.

Step 5: Model testing: Use the test set data to test the trained model. If the error results of several test sets are similar, the model is robust. The model is then judged based on whether the RMSE obtained by the model is within 2m/s. Regarding the accuracy and reliability of the inversion results, the error range in this example is 1.5-1.8m/s, which meets the accuracy requirements. From this, the CNN neural network sea surface wind speed inversion model is obtained.

Step 6: Data inversion: Input the DDM map to be measured into the CNN neural network sea surface wind speed inversion model, and output the corresponding inverted wind speed. The obtained RMSE is 1.52m/s.

Example 2

Embodiment 2 of the present invention proposes a GNSS-R sea surface wind speed inversion system based on CNN neural network. The system includes: a trained sea surface wind speed inversion model and a wind speed inversion module; the sea surface wind speed inversion model is a CNN neural network;

The wind speed inversion module is used to input the DDM map to be measured into the trained sea surface wind speed inversion model and output the corresponding inverted wind speed.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the embodiments, those of ordinary skill in the art will understand that modifications or equivalent substitutions may be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and they shall all be covered by the scope of the present invention. within the scope of the claims.

Claims (4)

1. A GNSS-R sea surface wind speed inversion method based on a CNN neural network, the method comprising:

inputting a DDM diagram to be tested into a pre-trained sea surface wind speed inversion model, and outputting a corresponding inversion wind speed; the sea surface wind speed inversion model is a CNN neural network;

the CNN neural network sequentially comprises: the input layer is a DDM graph, and the node number of the input layer is 2560; the number of the output characteristic graphs of the first convolution layer C1 is 1, and the number of the neurons is 1024; the number of the output characteristic graphs of the second convolution layer is 32, 64, 128 and 256 in sequence, the number of the neurons is 512, and the convolution kernels arranged on the two convolution layers are all of the same sizeThe method comprises the steps of carrying out a first treatment on the surface of the First, theThe domain sizes of the first pooling layer and the second pooling layer are bothThe method comprises the steps of carrying out a first treatment on the surface of the The number of neurons of the full connection layer between the convolution layer and the output layer is 16; the number of output layer nodes is 1, and the output of the node is sea surface wind speed; the activation function of the CNN neural network is a ReLU function; the loss function is an MSE function, and the evaluation index is Root Mean Square Error (RMSE);

the method further comprises the steps of: the training step of the CNN neural network specifically comprises the following steps:

selecting a plurality of groups of GNSS-R data and ECMWF analysis field data, and performing space-time matching to obtain an original sample set, wherein each group of samples consists of a DDM graph and corresponding wind speed;

preprocessing an original sample set and cutting the original sample set into a training set and a testing set;

the CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM graph and establish a mapping relation with wind speed;

testing the trained CNN neural network by using test set data;

the CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM graph and establish the mapping relation with the wind speed, and the CNN neural network specifically comprises the following steps:

after weights among neurons in the CNN neural network and threshold values in each neuron are initialized, training set data are transmitted into a first convolution layer from an input layer to carry out convolution calculation, a feature map output after convolution is processed through a nonlinear activation function, and then the data space dimension is reduced through a first pooling layer to obtain a feature matrix; then, a second convolution layer and a second pooling layer are input to carry out convolution and pooling operations, all data features obtained by convolution are combined through a full connection layer, the normalized data features are transmitted to an output layer, an error between the output wind speed and the real wind speed is calculated through a loss function, the error is reversely transmitted from the output layer by layer, and weights of the convolution layers and the pooling layers are adjusted;

and repeating the forward propagation and the backward propagation processes, gradually reducing the result of the loss function until the result is in a desired error range or reaches the set training times, and completing the training of the CNN neural network to realize the nonlinear mapping from the DDM graph to the sea surface wind speed.

2. The CNN neural network-based GNSS-R sea surface wind speed inversion method according to claim 1, wherein the original sample set is preprocessed and segmented into a training set and a testing set; the method specifically comprises the following steps:

screening the original sample set data based on longitude and latitude, wind speed and signal to noise ratio;

preprocessing the screened original sample set data based on a sampling algorithm and a normalization algorithm;

the preprocessed original sample set is segmented into a training set and a testing set according to the proportion of 7:3.

3. A GNSS-R sea surface wind speed inversion system based on a CNN neural network, the system comprising: a trained sea surface wind speed inversion model and a wind speed inversion module; the sea surface wind speed inversion model is a CNN neural network;

the wind speed inversion module is used for inputting the DDM diagram to be tested into a trained sea surface wind speed inversion model and outputting a corresponding inversion wind speed;

the CNN neural network sequentially comprises: the input layer is a DDM graph, and the node number of the input layer is 2560; the number of the output characteristic graphs of the first convolution layer C1 is 1, and the number of the neurons is 1024; the number of the output characteristic graphs of the second convolution layer is 32, 64, 128 and 256 in sequence, the number of the neurons is 512, and the convolution kernels arranged on the two convolution layers are all of the same sizeThe method comprises the steps of carrying out a first treatment on the surface of the The domain sizes of the first pooling layer and the second pooling layer are bothThe method comprises the steps of carrying out a first treatment on the surface of the Convolutional layerThe number of neurons of the full connection layer between the output layers is 16; the number of output layer nodes is 1, and the output of the node is sea surface wind speed; the activation function of the CNN neural network is a ReLU function; the loss function is an MSE function, and the evaluation index is Root Mean Square Error (RMSE);

the training step of the CNN neural network specifically comprises the following steps:

selecting a plurality of groups of GNSS-R data and ECMWF analysis field data, and performing space-time matching to obtain an original sample set, wherein each group of samples consists of a DDM graph and corresponding wind speed;

preprocessing an original sample set and cutting the original sample set into a training set and a testing set;

the CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM graph and establish a mapping relation with wind speed;

testing the trained CNN neural network by using test set data;

the CNN neural network is continuously trained through the training set data, so that the network can continuously capture the data characteristics in the DDM graph and establish the mapping relation with the wind speed, and the CNN neural network specifically comprises the following steps:

after weights among neurons in the CNN neural network and threshold values in each neuron are initialized, training set data are transmitted into a first convolution layer from an input layer to carry out convolution calculation, a feature map output after convolution is processed through a nonlinear activation function, and then the data space dimension is reduced through a first pooling layer to obtain a feature matrix; then, a second convolution layer and a second pooling layer are input to carry out convolution and pooling operations, all data features obtained by convolution are combined through a full connection layer, the normalized data features are transmitted to an output layer, an error between the output wind speed and the real wind speed is calculated through a loss function, the error is reversely transmitted from the output layer by layer, and weights of the convolution layers and the pooling layers are adjusted;

and repeating the forward propagation and the backward propagation processes, gradually reducing the result of the loss function until the result is in a desired error range or reaches the set training times, and completing the training of the CNN neural network to realize the nonlinear mapping from the DDM graph to the sea surface wind speed.

4. The CNN neural network-based GNSS-R sea surface wind speed inversion system according to claim 3, wherein said preprocessing and slicing the original sample set into a training set and a test set; the method specifically comprises the following steps:

screening the original sample set data based on longitude and latitude, wind speed and signal to noise ratio;

preprocessing the screened original sample set data based on a sampling algorithm and a normalization algorithm;

the preprocessed original sample set is segmented into a training set and a testing set according to the proportion of 7:3.