CN114139437A - Method and system for inverting submarine topography by using satellite height measurement data - Google Patents
Method and system for inverting submarine topography by using satellite height measurement data Download PDFInfo
- Publication number
- CN114139437A CN114139437A CN202111216713.8A CN202111216713A CN114139437A CN 114139437 A CN114139437 A CN 114139437A CN 202111216713 A CN202111216713 A CN 202111216713A CN 114139437 A CN114139437 A CN 114139437A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- training
- gravity anomaly
- acquiring
- 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.)
- Pending
Links
- 238000012876 topography Methods 0.000 title claims abstract description 74
- 238000005259 measurement Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000005484 gravity Effects 0.000 claims abstract description 90
- 238000012549 training Methods 0.000 claims abstract description 64
- 238000001514 detection method Methods 0.000 claims abstract description 16
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 45
- 238000003062 neural network model Methods 0.000 claims description 37
- 230000002159 abnormal effect Effects 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 15
- 238000012937 correction Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 7
- 239000013535 sea water Substances 0.000 claims description 7
- 239000000523 sample Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 12
- 238000013527 convolutional neural network Methods 0.000 description 9
- 230000004913 activation Effects 0.000 description 7
- 238000004590 computer program Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000011176 pooling Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000013049 sediment Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V7/00—Measuring gravitational fields or waves; Gravimetric prospecting or detecting
- G01V7/02—Details
- G01V7/06—Analysis or interpretation of gravimetric records
-
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a method and a system for inverting a submarine topography by using satellite height measurement data, and belongs to the field of submarine topography inversion. The method comprises the following steps: establishing a marine gravity anomaly model, and outputting marine gravity anomaly information of an observation point; acquiring seabed detection data of an observation force point, inputting the training data into a network model for training by taking the seabed detection data and ocean gravity anomaly data of the observation point as training data, and acquiring a landform inversion model; acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a landform and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point through the submarine topography data. The invention can mine more and complex relations between satellite height measurement data and submarine topography data.
Description
Technical Field
The present invention relates to the field of inversion of seafloor topography, and more particularly, to a method and system for inverting seafloor topography using satellite altimetry data.
Background
The submarine topography is an important component of global topography, reflects fluctuation change of the seabed, and plays an important role in marine scientific research, marine military, marine engineering and the like, such as submarine plate motion, sediment migration change, safe navigation of water surface/underwater carriers, underwater matched navigation, underwater pipe joint placement, sunken ship salvaging, oil gas exploration, environment monitoring and the like. The submarine topography reflects potential geological action to a great extent, high-precision and high-quality submarine topography information is of great importance to important activities such as offshore operation and navy safe navigation, and research on tsunami waves, ocean circulation and the like is limited by the precision of a sea depth model, so that the research on a high-precision submarine topography inversion technology has great significance.
Disclosure of Invention
The invention aims to obtain more comprehensive and effective satellite height measurement data and ocean gravity anomaly measurement data, and provides a method for inverting submarine topography by using the satellite height measurement data, which comprises the following steps:
establishing a marine gravity anomaly model, and outputting marine gravity anomaly information of an observation point according to the marine gravity anomaly model;
the establishing of the marine gravity anomaly model specifically comprises the following steps:
acquiring satellite height measurement data of a satellite on an observation point, and preprocessing the satellite height measurement data to acquire ground level height information and plumb line deviation information;
taking ocean gravity anomaly measurement data, geodetic level height information and plumb line deviation information of the observation points as training data;
inputting training data into a neural network model for training, optimizing the neural network model, acquiring performance weight parameters in the optimization process, and acquiring an optimal neural network model, namely a marine gravity anomaly model, according to the performance weight parameters;
acquiring seabed detection data of an observation force point, inputting the training data into a network model for training by taking the seabed detection data and ocean gravity anomaly data of the observation point as training data, and acquiring a landform inversion model;
the obtaining of the landform inversion model specifically includes:
feeding training data into a neural network model for training, and extracting submarine topography characteristics of an observation point;
acquiring water depth data and ship-borne water depth data according to seabed ground movement characteristics, performing loss calculation on the water depth data and the ship-borne water depth data, and performing parameter learning through back propagation to acquire optimal performance weight parameters of the model;
taking the neural network model with the optimal performance weight parameters as an optimal neural network model, namely a submarine topography inversion model;
acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a landform and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point through the submarine topography data.
Optionally, the seafloor probe data comprises: average water depth, maximum water depth, crust-to-crust and sea water density differences.
Optionally, the preprocessing is performed on the satellite altimetry data, and includes: data editing, error correction and analysis refinement, collinear processing and cross point adjustment.
The invention also provides a system for inverting the submarine topography by using the satellite altimetry data, which comprises the following steps:
the abnormal information extraction unit is used for establishing a marine gravity abnormal model and outputting marine gravity abnormal information of the observation point according to the marine gravity abnormal model;
the establishing of the marine gravity anomaly model specifically comprises the following steps:
acquiring satellite height measurement data of a satellite on an observation point, and preprocessing the satellite height measurement data to acquire ground level height information and plumb line deviation information;
taking ocean gravity anomaly measurement data, geodetic level height information and plumb line deviation information of the observation points as training data;
inputting training data into a neural network model for training, optimizing the neural network model, acquiring performance weight parameters in the optimization process, and acquiring an optimal neural network model, namely a marine gravity anomaly model, according to the performance weight parameters;
the training unit is used for acquiring seabed detection data of the observation force point, inputting the training data into the network model for training by taking the seabed detection data and ocean gravity abnormal data of the observation point as training data, and acquiring a landform inversion model;
the obtaining of the landform inversion model specifically includes:
feeding training data into a neural network model for training, and extracting submarine topography characteristics of an observation point;
acquiring water depth data and ship-borne water depth data according to seabed ground movement characteristics, performing loss calculation on the water depth data and the ship-borne water depth data, and performing parameter learning through back propagation to acquire optimal performance weight parameters of the model;
taking the neural network model with the optimal performance weight parameters as an optimal neural network model, namely a submarine topography inversion model;
determining a submarine topography unit, acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a topography and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point through the submarine topography data.
Optionally, the seafloor probe data comprises: average water depth, maximum water depth, crust-to-crust and sea water density differences.
Optionally, the preprocessing is performed on the satellite altimetry data, and includes: data editing, error correction and analysis refinement, collinear processing and cross point adjustment.
The method is not based on the traditional physical model, does not need to manually set model parameters, can better fit the nonlinear relation between the satellite height measurement data and the submarine topography data, and excavates more and complex relations between the satellite height measurement data and the submarine topography data.
Drawings
FIG. 1 is a flow chart of a method for inverting seafloor topography using satellite altimetry data of the present invention:
FIG. 2 is a schematic diagram of the establishment of a gravity anomaly inversion model based on a convolutional neural network;
FIG. 3 is a schematic diagram of a network structure of a gravity anomaly model;
FIG. 4 is a schematic diagram of the construction of a submarine topography inversion model based on a convolutional neural network;
FIG. 5 is a block diagram of a system for inverting seafloor topography using satellite altimetry data in accordance with the present invention;
Detailed Description
The exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, however, the present invention may be embodied in many different forms and is not limited to the embodiments described herein, which are provided for complete and complete disclosure of the present invention and to fully convey the scope of the present invention to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, the same units/elements are denoted by the same reference numerals.
Unless otherwise defined, terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Further, it will be understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense.
The invention provides a method for inverting submarine topography by using satellite altimetry data, which comprises the following steps of:
establishing a marine gravity anomaly model, and outputting marine gravity anomaly information of an observation point according to the marine gravity anomaly model;
acquiring seabed detection data of an observation force point, inputting the training data into a network model for training by taking the seabed detection data and ocean gravity anomaly data of the observation point as training data, and acquiring a landform inversion model;
acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a landform and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point through the submarine topography data.
The method for establishing the marine gravity anomaly model specifically comprises the following steps:
acquiring satellite height measurement data of a satellite on an observation point, and preprocessing the satellite height measurement data to acquire ground level height information and plumb line deviation information;
taking ocean gravity anomaly measurement data, geodetic level height information and plumb line deviation information of the observation points as training data;
inputting training data into the neural network model for training, optimizing the neural network model, obtaining performance weight parameters in the optimization process, and obtaining an optimal neural network model, namely a marine gravity anomaly model, according to the performance weight parameters.
Wherein the seafloor probe data comprises: average water depth, maximum water depth, crust-to-crust and sea water density differences.
Wherein, carry out the preliminary treatment to satellite altimetry data, include: data editing, error correction and analysis refinement, collinear processing and cross point adjustment.
The method for obtaining the landform inversion model specifically comprises the following steps:
feeding training data into a preset network model for training, and extracting submarine topography characteristics of an observation point;
acquiring water depth data and ship-borne water depth data according to seabed ground movement characteristics, performing loss calculation on the water depth data and the ship-borne water depth data, and performing parameter learning through back propagation to acquire optimal performance weight parameters of the model;
and taking the neural network model with the optimal performance weight parameters as an optimal neural network model, namely a submarine topography inversion model.
The invention is explained in detail below with reference to the drawings, in which:
the specific implementation flow is shown in fig. 2, and includes:
1. the method comprises the steps of preprocessing satellite height measurement data, wherein the purpose of processing is to control the quality of original height measurement data and improve the precision of the data, and the processing mainly comprises data editing, error correction and analysis refinement, collinear processing and intersection adjustment.
Data editing and removing, wherein the height measurement data records polluted under the conditions of land, sea ice, high sea conditions and the like are removed according to surface type marks, echo type marks, rain marks in the height measurement data and threshold range detection of each data item in geophysical records, so that the quality of the used data is ensured; the error correction and analysis refinement are to analyze different models or algorithms of ionospheric delay correction, atmospheric inverse pressure correction, ocean tide and the like and select a proper method to correct the accuracy aiming at correction mechanisms of dry-wet tropospheric delay, ionospheric delay, atmospheric inverse pressure, sea state deviation, high-frequency oscillation correction items, ocean tide, extreme tide, solid earth tide and the like in satellite height measurement data.
In order to reduce the sea surface time-varying effect of the height measurement data of the repeated orbit period, namely reduce the satellite orbit error, collinear processing is carried out, mainly a reference orbit and a normal point are determined, the sea surface height after collinear processing on the reference orbit is obtained by calculation through a distance weighted average method, and in order to reduce the radial orbit error influence weakening the height measurement data, a delay compensation method is used for carrying out cross point averaging on the satellite height measurement data.
2. And calculating the deviation of the vertical line at the intersection point of the satellite orbit by using the satellite height measurement data, namely calculating the numerical derivative of the height measurement section by using the primary difference of the height measurement observation values by using the position and time information of the lateral height point. At the intersection, the measured sea height profile gives two linearly independent estimates of the horizon horizontal gradient, one along each of the satellite ascent and descent trajectories. These two estimates may determine the direction of the ground level vertical. Since the estimates of the height of the sea at the two sides of the intersection are referenced to the ellipsoid, the direction of the geodetic surface perpendicular determines the local perpendicular deviation.
The derivatives of the ground level surface along the rising arc and the falling arc to the time t are respectively shown as a formula (1) and a formula (2).
Wherein λ and
respectively the longitude and the latitude of the earth, respectively,
and
when the orbit of the satellite approximates a circular orbit, there are equations (3) and (4), as follows:
the derivative of the geohorizon in the longitudinal direction is shown below.
The derivative of the geohorizon in the longitudinal direction is shown below.
Wherein the content of the first and second substances,
can be obtained from the time and position information of the height measuring section measuring point.
Wherein the content of the first and second substances,
and
the component of (a) can be calculated by formula (5) and formula (6), so that the meridian direction component xi and the prime unit direction component eta of the vertical deviation can be determined:
because the vertical deviation contains abundant gravity field information, the east-west component and the south-north component of the obtained vertical deviation are used for determining gravity anomaly.
3. Designing a deep learning network structure, constructing a marine gravity model, and using satellite height measurement data to invert marine gravity abnormal information. When designing the gravity model, the satellite height measurement observation data and the vertical line deviation information calculated in the step 2 are used as input, the gravity abnormal data such as the disturbance of the ocean gravity field and the like are used as output, and the purpose of network training is to obtain a neural network system which can quickly output the ocean gravity information by inputting the observation information. And after the training is finished, storing the trained parameters, and only calling the stored network parameters when the gravity anomaly needs to be inverted.
The detailed steps are as follows:
firstly, designing a convolutional neural network structure, as shown in fig. 3, the first layer of the network is a convolution with a convolution kernel size of (3 × 3) and a channel number of 16, activating by using a ReLU activation function, and outputting to the next layer, keeping the size of each intermediate layer unchanged by adopting a padding (filling) mode for edges, and ensuring that the sizes of the intermediate layers are consistent by adopting the same filling operation in the next convolutional layer; next, performing convolution with convolution kernel size (3 × 3) and channel number 32, and using a ReLU activation function; then carrying out convolution with convolution kernel size of (3 × 3) and channel number of 64, and using a ReLU activation function; next, performing convolution with a convolution kernel size of (3 × 3) and a channel number of 128, and using a ReLU activation function; finally, using convolution with convolution kernel size (3 x 3) and number of channels 256, tanh is used as the activation function. A mean square error log loss function (MSLE) is employed and an Adam optimizer is used as the optimizer for the network.
And training the network. And inputting the satellite height measurement observation data and the vertical line deviation information as training data, and inputting the ship measurement gravity anomaly data as label data to the network to train the network model. Loss calculation is carried out on a loss function in the forward propagation process, parameter learning is carried out through backward propagation, weight parameters enabling a neural network model to obtain the best performance are obtained through repeated iteration and continuous model optimization, and the network parameters are stored after training is finished, namely the marine gravity model.
And inverting the marine gravity data through the trained network model. And (3) inputting the satellite height measurement observation data and the plumb line deviation information as input data into a trained network model, namely a marine gravity model, and carrying out forward calculation by using a neural network, and outputting the predicted marine gravity anomaly information.
After the intelligent inversion of the marine gravity anomaly based on the satellite altimetry data is realized, the invention constructs a submarine topography inversion model by using a deep learning method based on the marine gravity anomaly data obtained by inversion, and inverts the submarine topography data, as shown in fig. 4, the process of the intelligent inversion of the submarine topography using the marine gravity anomaly is as follows:
firstly, acquiring ocean gravity abnormal data by an ocean gravity model, and taking the data and sea bottom depth detection data, average water depth, maximum water depth, density difference between the crust and the sea water and other ship-borne water depth data of a corresponding sea area as training data.
Then, a convolutional neural network is designed. Performing submarine topography inversion using convolutional neural networks requires transforming convolutional neural networks that solve the classification problem into convolutional neural networks that solve the regression problem. The last layer in the convolutional neural network is usually a Softmax layer to solve the classification problem, and the output dimension is the same as the classified class. The invention aims to solve the inversion problem, namely the regression problem, of the submarine topography, so that the output of the last layer is a submarine water depth data value. The structure in front of the network output layer adopts a classic LeNet network mechanism, and because the study of seabed terrain inversion is not carried out by using a convolutional neural network, a model is initialized to a classic structure, and then parameters are continuously adjusted. The network structure is composed of two convolutional layers, two pooling layers and two fully-connected layers, and the output of each convolutional layer passes through a ReLU activation function. Unlike sigmoid used by classical LeNet, the ReLU activation function is used in the present network. The convolutional layer of the first layer adopts Valid convolution with the step of 1, and 20 convolution kernels are selected, wherein the size of each convolution kernel is 5 multiplied by 5. The second convolutional layer was similar to the first convolutional layer, and 50 convolutional kernels were selected. The step in the pooling layer is 2, the convolution kernel is 2, and the efficiency of parameter reduction in the network structure is improved through the pooling layer. After passing through the convolutional layer and the pooling layer, the data features are flattened into multidimensional vectors, which can also be called a plurality of neurons or nodes, and then enter the fully-connected layer.
And (4) training the neural network model constructed in the step (2). And (2) feeding the training data prepared in the step (1) into a network model for training, extracting relevant features of the submarine topography, performing loss calculation on the generated water depth data and the ship-borne water depth, performing parameter learning through back propagation, continuously optimizing the model through repeated iteration to obtain weight parameters when the model obtains the best performance, and storing the network parameters after the training is finished to serve as a trained neural network model, namely a submarine topography inversion model.
And (3) performing submarine topography inversion through the trained submarine topography inversion model based on the ocean gravity anomaly in the step (3), and outputting corresponding submarine topography data when inputting data of the ocean gravity anomaly, the average water depth, the maximum water depth and the like in the sea area.
The present invention also provides a system 200 for inverting seafloor topography using satellite altimetry data, as shown in fig. 5, comprising:
an abnormal information extraction unit 205 that establishes a marine gravity abnormal model and outputs marine gravity abnormal information of an observation point according to the marine gravity abnormal model;
the training unit 204 is used for acquiring seabed detection data of the observation force point, inputting the training data into the network model for training by taking the seabed detection data and ocean gravity abnormal data of the observation point as training data, and acquiring a landform inversion model;
determining a submarine topography unit 203, acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a topography and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point according to the submarine topography data.
The method for establishing the marine gravity anomaly model specifically comprises the following steps:
acquiring satellite height measurement data of a satellite on an observation point, and preprocessing the satellite height measurement data to acquire ground level height information and plumb line deviation information;
taking ocean gravity anomaly measurement data, geodetic level height information and plumb line deviation information of the observation points as training data;
inputting training data into the neural network model for training, optimizing the neural network model, obtaining performance weight parameters in the optimization process, and obtaining an optimal neural network model, namely a marine gravity anomaly model, according to the performance weight parameters.
Wherein the seafloor probe data comprises: average water depth, maximum water depth, crust-to-crust and sea water density differences.
Wherein, carry out the preliminary treatment to satellite altimetry data, include: data editing, error correction and analysis refinement, collinear processing and cross point adjustment.
The method for obtaining the landform inversion model specifically comprises the following steps:
feeding training data into a preset network model for training, and extracting submarine topography characteristics of an observation point;
acquiring water depth data and ship-borne water depth data according to seabed ground movement characteristics, performing loss calculation on the water depth data and the ship-borne water depth data, and performing parameter learning through back propagation to acquire optimal performance weight parameters of the model;
and taking the neural network model with the optimal performance weight parameters as an optimal neural network model, namely a submarine topography inversion model.
The method is not based on the traditional physical model, does not need to manually set model parameters, can better fit the nonlinear relation between the satellite height measurement data and the submarine topography data, and excavates more and complex relations between the satellite height measurement data and the submarine topography data.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The scheme in the embodiment of the application can be implemented by adopting various computer languages, such as object-oriented programming language Java and transliterated scripting language JavaScript.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (6)
1. A method of inverting seafloor topography using satellite altimetry data, the method comprising:
establishing a marine gravity anomaly model, and outputting marine gravity anomaly information of an observation point according to the marine gravity anomaly model;
the establishing of the marine gravity anomaly model specifically comprises the following steps:
acquiring satellite height measurement data of a satellite on an observation point, and preprocessing the satellite height measurement data to acquire ground level height information and plumb line deviation information;
taking ocean gravity anomaly measurement data, geodetic level height information and plumb line deviation information of the observation points as training data;
inputting training data into a neural network model for training, optimizing the neural network model, acquiring performance weight parameters in the optimization process, and acquiring an optimal neural network model, namely a marine gravity anomaly model, according to the performance weight parameters;
acquiring seabed detection data of an observation force point, inputting the training data into a network model for training by taking the seabed detection data and ocean gravity anomaly data of the observation point as training data, and acquiring a landform inversion model;
the obtaining of the landform inversion model specifically includes:
feeding training data into a neural network model for training, and extracting submarine topography characteristics of an observation point;
acquiring water depth data and ship-borne water depth data according to seabed ground movement characteristics, performing loss calculation on the water depth data and the ship-borne water depth data, and performing parameter learning through back propagation to acquire optimal performance weight parameters of the model;
taking the neural network model with the optimal performance weight parameters as an optimal neural network model, namely a submarine topography inversion model;
acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a landform and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point through the submarine topography data.
2. The method of claim 1, the seafloor probe data comprising: average water depth, maximum water depth, crust-to-crust and sea water density differences.
3. The method of claim 1, the pre-processing satellite altimetry data comprising: data editing, error correction and analysis refinement, collinear processing and cross point adjustment.
4. A system for inverting seafloor topography using satellite altimetry data, the system comprising:
the abnormal information extraction unit is used for establishing a marine gravity abnormal model and outputting marine gravity abnormal information of the observation point according to the marine gravity abnormal model;
the establishing of the marine gravity anomaly model specifically comprises the following steps:
acquiring satellite height measurement data of a satellite on an observation point, and preprocessing the satellite height measurement data to acquire ground level height information and plumb line deviation information;
taking ocean gravity anomaly measurement data, geodetic level height information and plumb line deviation information of the observation points as training data;
inputting training data into a neural network model for training, optimizing the neural network model, acquiring performance weight parameters in the optimization process, and acquiring an optimal neural network model, namely a marine gravity anomaly model, according to the performance weight parameters;
the training unit is used for acquiring seabed detection data of the observation force point, inputting the training data into the network model for training by taking the seabed detection data and ocean gravity abnormal data of the observation point as training data, and acquiring a landform inversion model;
the obtaining of the landform inversion model specifically includes:
feeding training data into a neural network model for training, and extracting submarine topography characteristics of an observation point;
acquiring water depth data and ship-borne water depth data according to seabed ground movement characteristics, performing loss calculation on the water depth data and the ship-borne water depth data, and performing parameter learning through back propagation to acquire optimal performance weight parameters of the model;
taking the neural network model with the optimal performance weight parameters as an optimal neural network model, namely a submarine topography inversion model;
determining a submarine topography unit, acquiring satellite height measurement data of an observation point to be detected, inputting the satellite height measurement data into a marine gravity anomaly model to acquire marine gravity anomaly data, inputting the acquired marine gravity anomaly data into a topography and landform inversion model to calculate output data, namely submarine topography data, and determining the submarine topography of the target observation point through the submarine topography data.
5. The system of claim 4, the seafloor probe data comprising: average water depth, maximum water depth, crust-to-crust and sea water density differences.
6. The system of claim 4, the pre-processing satellite altimetry data comprising: data editing, error correction and analysis refinement, collinear processing and cross point adjustment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111216713.8A CN114139437A (en) | 2021-10-19 | 2021-10-19 | Method and system for inverting submarine topography by using satellite height measurement data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111216713.8A CN114139437A (en) | 2021-10-19 | 2021-10-19 | Method and system for inverting submarine topography by using satellite height measurement data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114139437A true CN114139437A (en) | 2022-03-04 |
Family
ID=80394418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111216713.8A Pending CN114139437A (en) | 2021-10-19 | 2021-10-19 | Method and system for inverting submarine topography by using satellite height measurement data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114139437A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114993268A (en) * | 2022-04-13 | 2022-09-02 | 南京信息工程大学 | Water depth inversion method and device combined with Catboost and storage medium |
| CN117908108A (en) * | 2024-03-20 | 2024-04-19 | 山东省地质矿产勘查开发局第二水文地质工程地质大队(山东省鲁北地质工程勘察院) | Real-time marine seismic monitoring system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851790A (en) * | 2019-10-29 | 2020-02-28 | 王金虎 | Ocean current dynamic optimization forecasting model based on deep learning algorithm |
| CN111045099A (en) * | 2019-12-27 | 2020-04-21 | 武汉大学 | Method for inverting ocean gravity field by imaging type altimeter data |
| CN112556660A (en) * | 2021-02-20 | 2021-03-26 | 中国测绘科学研究院 | Sea area gravity anomaly inversion method and system based on satellite height measurement data |
-
2021
- 2021-10-19 CN CN202111216713.8A patent/CN114139437A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851790A (en) * | 2019-10-29 | 2020-02-28 | 王金虎 | Ocean current dynamic optimization forecasting model based on deep learning algorithm |
| CN111045099A (en) * | 2019-12-27 | 2020-04-21 | 武汉大学 | Method for inverting ocean gravity field by imaging type altimeter data |
| CN112556660A (en) * | 2021-02-20 | 2021-03-26 | 中国测绘科学研究院 | Sea area gravity anomaly inversion method and system based on satellite height measurement data |
Non-Patent Citations (4)
| Title |
|---|
| 周武;林明森;李延民;王振占;黄磊;: "海洋二号扫描微波辐射计冷空定标和地球物理参数反演研究", 中国工程科学, no. 07, 15 July 2013 (2013-07-15) * |
| 杨磊: ""基于BP神经网络的重力异常分离"", 《工程地球物理学报》, vol. 18, no. 1, 31 January 2021 (2021-01-31), pages 90 - 97 * |
| 陈梅森: ""基于卫星测高数据的海底地形反演研究"", 《经纬天地》, 28 February 2021 (2021-02-28), pages 59 - 63 * |
| 陈浩等: ""基于神经网络的杭州湾北岸水下地形遥感反演研究"", 《世界科技研究与发辰》, vol. 33, no. 3, 30 June 2011 (2011-06-30), pages 390 - 392 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114993268A (en) * | 2022-04-13 | 2022-09-02 | 南京信息工程大学 | Water depth inversion method and device combined with Catboost and storage medium |
| CN117908108A (en) * | 2024-03-20 | 2024-04-19 | 山东省地质矿产勘查开发局第二水文地质工程地质大队(山东省鲁北地质工程勘察院) | Real-time marine seismic monitoring system |
| CN117908108B (en) * | 2024-03-20 | 2024-05-28 | 山东省地质矿产勘查开发局第二水文地质工程地质大队(山东省鲁北地质工程勘察院) | Real-time marine seismic monitoring system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109709603B (en) | Seismic horizon identification and tracking method and system | |
| RU2321064C2 (en) | Method for building a reversible three-dimensional hydrodynamic model of earth, calibrated in real time during drilling | |
| CN114139437A (en) | Method and system for inverting submarine topography by using satellite height measurement data | |
| US11828894B2 (en) | Multi-scale unsupervised seismic velocity inversion method based on autoencoder for observation data | |
| CN110031895B (en) | Multipoint geostatistical stochastic inversion method and device based on image stitching | |
| CN108957502B (en) | GNSS multi-system multi-path error real-time weakening method based on deep learning LSTM | |
| CN104180873B (en) | Single-wave-beam depth finder water depth gross error detection and correction method and system | |
| CN113341476B (en) | Method for improving ocean gravity spatial resolution based on submarine topography-gravity combination | |
| CN114137624B (en) | Method and system for inverting submarine topography based on satellite altimeter | |
| CN111561313B (en) | Compact sandstone reservoir parameter prediction method based on physical model and machine learning | |
| CN112904419B (en) | Microseism imaging method and terminal equipment | |
| CN103018778A (en) | Method and device for conducting reservoir prediction through correction acoustic well logging curves | |
| CN116187168A (en) | Method for improving water depth inversion accuracy based on neural network-gravity information wavelet decomposition | |
| Behzadi et al. | Introducing a novel digital elevation model using artificial neural network algorithm | |
| CN104316961A (en) | Method for obtaining geological parameters of weathered layer | |
| Fokker et al. | Disentangling and parameterizing shallow sources of subsidence: application to a reclaimed coastal area, Flevoland, the Netherlands | |
| CN116861799A (en) | Submarine topography inversion model generation and submarine topography inversion method based on residual errors | |
| CN110907990B (en) | Quantitative prediction method and system for post-stack seismic cracks | |
| Le Ravalec-Dupin | Pilot block method methodology to calibrate stochastic permeability fields to dynamic data | |
| CN115542393A (en) | Tunneling-following roadway full-waveform inversion method based on multi-parameter constraint and structural correction | |
| CN107255832A (en) | A kind of inversion method of subsurface structure | |
| Jo et al. | Conditioning stratigraphic, rule-Based models with generative adversarial networks: A deepwater lobe, deep learning example | |
| Yang et al. | Gippsland Basin 3D forward modelling in Badlands | |
| Lloyd | Controls upon the location and size of glacial overdeepenings | |
| Hess et al. | Analysis of coastal-plain fluvial architecture and high-frequency stacking patterns in the Upper Cretaceous Masuk Formation, Utah, USA: Climate-driven cyclicity? |
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 |