CN117909666A - Intelligent sea wave correction method and system integrating numerical mode and deep learning - Google Patents
Intelligent sea wave correction method and system integrating numerical mode and deep learning Download PDFInfo
- Publication number
- CN117909666A CN117909666A CN202410308826.8A CN202410308826A CN117909666A CN 117909666 A CN117909666 A CN 117909666A CN 202410308826 A CN202410308826 A CN 202410308826A CN 117909666 A CN117909666 A CN 117909666A
- Authority
- CN
- China
- Prior art keywords
- wave
- sea
- intelligent
- numerical
- deep learning
- 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
- 238000012937 correction Methods 0.000 title claims abstract description 177
- 238000013135 deep learning Methods 0.000 title claims abstract description 96
- 238000000034 method Methods 0.000 title claims abstract description 90
- 238000004364 calculation method Methods 0.000 claims abstract description 74
- 238000004088 simulation Methods 0.000 claims abstract description 40
- 238000005259 measurement Methods 0.000 claims abstract description 29
- 230000000694 effects Effects 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 30
- 230000007613 environmental effect Effects 0.000 claims description 23
- 230000003993 interaction Effects 0.000 claims description 14
- 210000002569 neuron Anatomy 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000012876 topography Methods 0.000 claims description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 8
- 238000013473 artificial intelligence Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 4
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 4
- 230000007812 deficiency Effects 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 239000013535 sea water Substances 0.000 claims description 2
- 230000004927 fusion Effects 0.000 description 11
- 238000004590 computer program Methods 0.000 description 10
- 238000001228 spectrum Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000013277 forecasting method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of marine environment prediction and correction, and discloses an intelligent ocean wave correction method and system integrating numerical mode and deep learning. According to the method, task sea area environment data are obtained, a sea wave numerical mode calculation model is constructed, a sea wave intelligent correction data set is constructed, an artificial intelligent sea wave correction model is constructed, numerical mode and deep learning sea wave intelligent correction model generalization analysis is integrated, sea wave numerical simulation results at a certain position in a sea area are corrected by using the model, and the model is verified by adopting buoy actual measurement results so as to test the intelligent correction effect of the model on a sea area wave field. Compared with the traditional wave numerical mode, the intelligent correction model for the wave can consider the influence caused by different scheme settings of the mode, and the forecasting precision and efficiency of the wave are obviously improved. The invention provides a novel and effective method and strategy for sea wave forecasting in real sea area.
Description
Technical Field
The invention belongs to the technical field of marine environment prediction and correction, and particularly relates to an intelligent ocean wave correction method and system integrating numerical mode and deep learning.
Background
The high-precision sea wave condition is a necessary factor for guaranteeing the navigation safety of the ship. The bad sea conditions often cause dangerous phenomena such as ship instability, sway and the like. At the same time, navigation of ships, planning of routes, etc. also depend on accurate marine environment information. The method avoids severe sea conditions, and can save oil consumption to a certain extent by adjusting the navigation state of the ship in time, so that navigation economy is improved. In addition, sea waves play an important role in climate systems, and sea waves have a non-negligible effect on the interactions of the sea with the atmosphere, etc. Therefore, the high-precision sea wave simulation has important theoretical significance and practical value.
At present, a numerical solution and a deep learning method are mainly adopted for the calculation of sea waves. The correction of sea wave data mainly adopts measured data such as buoys and the like to carry out fusion assimilation.
The wave numerical mode is based on an energy balance equation, different sources and sinks in the dynamic spectrum equation are utilized to represent a complex physical process in the wave propagation process, and the process of wave energy input, propagation and dissipation and energy transfer among different simplified harmonics can be described. The deep learning can solve the problem of dependence of the traditional sea wave mode on high-performance computing equipment by virtue of the strong feature learning capability and the complex data feature extraction capability, and realizes quick forecasting of sea waves. The data assimilation can improve the simulation precision of the regional sea wave environment by carrying out fusion correction on the actual sea observation data and the prediction result of the numerical simulation and deep learning method.
Through the above analysis, the problems and defects existing in the prior art are as follows: the wave mode is based on a wave propagation equation when solving the wave, so that the dependence on calculation resources is extremely high, meanwhile, the boundary field in the mode is set, the source parameter scheme is selected, the scheme coefficient is set, the accuracy of forced field data and the like directly determine the accuracy of mode calculation, the mode calculation accuracy is limited by multiple factors, and the parameter tuning process takes a long time.
The training data of the intelligent model depends on the calculation result of the traditional numerical model, so that the forecasting accuracy of the intelligent model is limited relative to that of the traditional model.
Sea wave data assimilation needs to be based on a large amount of measured data, and the assimilation effect greatly depends on the quantity and timeliness of the measured data. The geographic position of the data and the coverage range of the acquired data can influence the assimilation precision and the global property; the timeliness of the measured data directly determines the accuracy of the assimilation model to the current environment. Therefore, the data assimilation process based on the measured information is limited by the data location and acquisition time.
Disclosure of Invention
In order to overcome the problems in the related art, the embodiment of the invention provides an intelligent ocean wave correction method and system integrating a numerical mode and deep learning.
The technical scheme is as follows: the intelligent sea wave correction method integrating the numerical mode and the deep learning comprises the following steps:
s1, acquiring environmental data in a task sea area, acquiring typical sea area weather and topography data by using open source environmental information, and collecting actual measurement environmental information of buoys in the sea area; the buoy actually-measured environmental information comprises sea wave height, period and wind speed data;
s2, constructing an ocean wave numerical mode calculation model based on the acquired environmental data, and setting a plurality of schemes to carry out numerical calculation on typical ocean wave conditions by combining the mode calculation complexity;
S3, constructing an intelligent ocean wave correction data set, extracting and preprocessing a solving result of an ocean wave numerical mode calculation model aiming at an observed position of the buoy, and constructing an intelligent ocean wave correction data set with one-to-one matching observed results and numerical simulation results at the observed position of the buoy;
S4, constructing an artificial intelligent wave correction model, and analyzing an optimal input strategy of the artificial intelligent wave correction model by combining with sea area characteristics, wherein the artificial intelligent wave correction model is an intelligent wave correction model integrating a numerical mode and deep learning;
S5, performing generalization analysis of the intelligent sea wave correction model by combining the numerical mode and the deep learning, correcting the sea wave numerical simulation result at a certain position in the sea area by using the intelligent sea wave correction model by combining the numerical mode and the deep learning, verifying the intelligent sea wave correction model by combining the numerical mode and the deep learning by using the buoy actual measurement result, and testing the intelligent sea wave correction effect of the intelligent sea wave correction model by combining the numerical mode and the deep learning.
In step S1, the environmental data includes sea surface wind speed and sea water depth information, and the buoy measured environmental information includes sea wave height, period and wind speed data.
In step S2, the wave numerical mode calculation model is:
；
In the method, in the process of the invention, Is a linear input item,/>For wind energy input item,/>For nonlinear wave-wave interactions,/>Dissipation term leading to wave breaking,/>For wave-bottom interaction,/>For wave breaking caused by reduced water depth,/>Is the interaction of three waves,/>Is dispersion term,/>For wave ice interactions,/>Is a reflective item,/>Is a custom source item.
In step S2, the calculating the complexity of the combination mode, and setting a plurality of schemes to perform numerical calculation on the typical sea wave situation includes: and setting a plurality of schemes in the solving process of the wave numerical mode calculation model by using the root mean square error RMSE and the average relative error MAPE as evaluation standards.
In step S3, the extracting and preprocessing the solution result of the wave numerical mode calculation model for the buoy observation position includes: the buoy observation position is matched with the buoy actual measurement result through a data interpolation and deficiency processing mode.
In step S4, the building an artificial intelligence sea wave correction model includes: based on an ANN neural network, on the basis of the wave numerical mode calculation model, an intelligent wave correction model integrating the numerical mode and deep learning is constructed, and the actual wave data is utilized to correct the wave numerical mode forecasting result.
Further, the intelligent correction model of sea wave comprises: the sea wave numerical mode module and the artificial intelligent correction module;
The wave numerical mode module carries out numerical calculation on sea wave information in the sea area through a wave control equation based on the water depth topography data and the wind speed forced field data of the task sea area, and finally outputs wave information at the sea area wave field and the buoy station position;
The artificial intelligence correction module is used for constructing deviation between a simulation result and a real value of the sea wave by utilizing a buoy actual measurement sea wave result based on the site position sea wave information output by the numerical mode.
Further, the hidden layer of the ocean wave intelligent correction model is 3 layers, and 200, 200 and 100 neurons are respectively arranged on each layer.
Further, the intelligent sea wave correction model takes a numerical mode simulation wave height Hs as basic input, introduces wind speeds SSW and wave age WA at the sea surface 10m, analyzes the influence of the wind speeds and the wave input on the forecasting precision of the intelligent sea wave correction model, and screens and constructs an input data set.
Another object of the present invention is to provide an intelligent correction system for sea waves, which is used for implementing the intelligent correction method for sea waves by combining numerical mode and deep learning, and the system comprises:
The system comprises a task sea area environment data acquisition module, a data processing module and a data processing module, wherein the task sea area environment data acquisition module is used for acquiring environment data in a task sea area, acquiring typical sea area weather and topography data by using open source environment information, and collecting actual measurement environment information of a buoy in the sea area;
The method comprises the steps of constructing an ocean wave numerical mode calculation model module, constructing an ocean wave numerical mode calculation model based on acquired environmental data, setting a plurality of schemes to carry out numerical calculation on typical ocean wave conditions by combining the mode calculation complexity;
the method comprises the steps of constructing an intelligent ocean wave correction data set module, extracting and preprocessing a solving result of an ocean wave numerical mode calculation model aiming at a buoy observation position, and constructing an intelligent ocean wave correction data set with one-to-one matching observation result and numerical simulation result at the buoy actual measurement position;
An artificial intelligent sea wave correction model module is constructed and used for constructing an artificial intelligent sea wave correction model, analyzing the optimal input strategy of the artificial intelligent sea wave correction model by combining sea area characteristics, wherein the artificial intelligent sea wave correction model is an intelligent sea wave correction model integrating a numerical mode and deep learning;
The generalization analysis module is used for generalizing and analyzing the intelligent sea wave correction model integrating the numerical mode and the deep learning, correcting the sea wave numerical simulation result at a certain position in the sea area by utilizing the intelligent sea wave correction model integrating the numerical mode and the deep learning, verifying the intelligent sea wave correction model integrating the numerical mode and the deep learning by adopting the buoy actual measurement result, and testing the intelligent sea wave correction effect of the intelligent sea wave correction model integrating the numerical mode and the deep learning on the sea area wave field.
By combining all the technical schemes, the invention has the following beneficial effects: aiming at the problem of sea wave forecasting in the real sea area, the invention provides an intelligent sea wave correction model integrating a sea wave numerical mode and a deep learning method. And constructing deviation between a numerical simulation result and a real value of the sea wave by using measured data of the buoy, and realizing intelligent correction of the sea wave simulation result. The model can solve the problem that the existing sea wave correction method such as data assimilation depends on the position and time of measured data. Compared with the traditional wave numerical mode, the intelligent correction model of the wave can consider the influence caused by different scheme settings of the mode, so that the forecasting precision and efficiency of the wave are remarkably improved, and a novel and effective method and strategy are provided for forecasting the wave in the real sea area.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure;
FIG. 1 is a flow chart of an intelligent correction method for sea waves integrating numerical modes and deep learning, which is provided by the embodiment of the invention;
FIG. 2 is a schematic diagram showing an ANN model information transfer process according to an embodiment of the present invention;
fig. 3 (a) is a schematic diagram of numerical calculation in the intelligent correction model of sea wave with fusion of numerical mode and deep learning according to the embodiment of the present invention;
FIG. 3 (b) is a schematic diagram of a neural network of an intelligent correction model for sea waves, which is provided by the embodiment of the invention and is integrated with a numerical mode and deep learning;
FIG. 4 is a generalized analysis error statistical diagram of an intelligent sea wave correction model with integrated numerical mode and deep learning provided by the embodiment of the invention;
FIG. 5 is a graph showing the intelligent correction effect (comparing measured by a No. 4 buoy) of the intelligent correction model of sea wave on the sea wave field, which is provided by the embodiment of the invention and is fused with a numerical mode and a deep learning;
FIG. 6 is a diagram of an intelligent ocean wave correction system integrating numerical modes and deep learning, which is provided by the embodiment of the invention;
In the figure: 1. the task sea area environment data acquisition module; 2. constructing a wave numerical mode calculation model module; 3. constructing an intelligent sea wave correction data set module; 4. constructing an artificial intelligent sea wave correction model module; 5. and a generalization analysis module.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit or scope of the invention, which is therefore not limited to the specific embodiments disclosed below.
The intelligent correction method and system for sea waves integrating the numerical mode and the deep learning provided by the embodiment of the invention have the following innovation points: the invention provides a sea wave correction model combining a sea wave mode and artificial intelligence, which improves the simulation precision and the calculation efficiency of sea waves. The traditional wave numerical solution method, represented by WW3, faces challenges in balancing calculation efficiency and simulation accuracy. In order to overcome the limit of dependence on a logarithmic simulation result in the training process of the existing deep learning method, the invention provides the intelligent ocean wave correction model integrating an ocean wave numerical mode and deep learning. By utilizing the buoy actual measurement data, the deviation between the numerical simulation result and the actual sea wave value is constructed, and the intelligent correction of the sea wave simulation result is realized, so that the problem that the existing sea wave correction method is dependent on the position and time of the actual measurement data and the time consumption of the parameter tuning process is long is solved. Research results show that compared with the traditional wave numerical mode, the intelligent correction model for the waves can consider the influence caused by the mode calculation problems such as wave transmission in the open sea area, mode source item simplification and the like, and the forecasting precision and efficiency of the waves are remarkably improved. The invention provides a novel and effective method and strategy for actual sea area sea wave forecast, and has important significance for improving navigation safety and coping with marine climate change.
Embodiment 1, as shown in fig. 1, the method for intelligently correcting sea waves by integrating numerical modes and deep learning provided by the embodiment of the invention comprises the following steps:
s1, acquiring environmental data in a task sea area, acquiring typical sea area weather and topography data by using open source environmental information, and collecting actual measurement environmental information of buoys in the sea area; the buoy actually-measured environmental information comprises sea wave height, period and wind speed data;
the environment data comprise sea surface wind speed and water depth information in the sea area, and the buoy actual measurement environment information comprises sea wave height, period and wind speed data.
S2, constructing an ocean wave numerical mode calculation model based on the acquired environmental data, and setting a plurality of schemes to carry out numerical calculation on typical ocean wave conditions by combining the mode calculation complexity;
the solution formula includes:
；
；
In the method, in the process of the invention, For wave action quantity,/>Is wave energy,/>Is relative frequency,/>For wave number,/>For the direction/>Is wave number direction spectrum,/>For wave action volume density spectrum,/>Representing wave action intensity spectrum related to wave number and direction,/>A wave number direction spectrum representing wave numbers and directions;
Wave propagation is expressed as:
；
In the method, in the process of the invention, For the spectrum/>Net effect of source and sink,/>Time is;
in the numerical model, the control equation is written in a conservation form, then the spectrum The equilibrium equation of (2) is expressed as:
；
In the method, in the process of the invention, Representing the change of wave action quantity with time,/>Representing the sign of the divergence calculation,/>Representation/>Divergence of/>Representing spatial displacement,/>Representing the vector sum of wave velocity and flow velocity,/>Representing wave action quantity,/>Representing wave number,/>Representing the wave number over time,/>Indicate direction,/>Representing the change of direction over time,/>Representing the net effect of spectrum sources and sinks,/>Representing the relative frequency.
The wave numerical mode calculation model is as follows:
；
In the method, in the process of the invention, Is a linear input item,/>For wind energy input item,/>For nonlinear wave-wave interactions,/>Dissipation term leading to wave breaking,/>For wave-bottom interaction,/>For wave breaking caused by reduced water depth,/>Is the interaction of three waves,/>Is dispersion term,/>For wave ice interactions,/>Is a reflective item,/>Is a custom source item.
In the embodiment of the invention, a plurality of schemes are set in the solving process of the wave numerical mode calculation model, and the wave simulation precision difference under different mode settings is analyzed.
Among them, root Mean Square Error (RMSE) and average relative error (MAPE) are used as accuracy evaluation criteria of the present invention:
；
；
In the method, in the process of the invention, For the number of samples,/>Respectively/>And calculating the wave numerical value and the buoy actual measurement result at the moment.
Aiming at the constructed wave numerical solution model, three sets of calculation schemes are set, and model calculation errors under different schemes are compared, and the method is specifically shown as follows.
Table 1 three sets of scheme error result statistics:
。
The current results can be seen: in the wave numerical mode calculation process, the mode setting scheme has a large improvement space for the simulation precision of the wave. Aiming at the current sea area, the scheme three-numerical simulation precision is the best.
S3, constructing an intelligent ocean wave correction data set, extracting and preprocessing a solving result of an ocean wave numerical mode calculation model aiming at an observed position of the buoy, and constructing an intelligent ocean wave correction data set with one-to-one matching observed results and numerical simulation results at the observed position of the buoy;
S4, constructing an artificial intelligent wave correction model, and analyzing an optimal input strategy of the artificial intelligent wave correction model by combining with sea area characteristics, wherein the artificial intelligent wave correction model is an intelligent wave correction model integrating a numerical mode and deep learning;
specifically, the invention constructs the intelligent correction model of the sea wave integrating the numerical mode and the deep learning on the basis of the sea wave numerical mode calculation model based on the ANN neural network, and corrects the forecasting result of the sea wave numerical mode by utilizing the measured sea wave data. By constructing the deviation between the actual value of the sea wave and the numerical simulation result, the problem that a certain deviation exists between the calculation result and the actual sea wave due to the influence of the sea wave numerical mode on multiple aspects such as mode setting in the process of resolving the sea wave in the real sea area is solved, and the high-precision forecast of the sea wave elements in the real sea area is further realized.
As shown in FIG. 2 of the ANN structure chart, in the signal transmission process, when information is transmitted to the next layer, a group of weights and biases are randomly given, and the output of each neuron is as follows:
；
In the method, in the process of the invention, For the output of the current neuron,/>For the current layer number,/>For each neuron under the current layer,/>For each neuron in the upper layer,/>Is the number of neurons in the upper layer,/>For each input element, the corresponding weight in the current neuron,/>For biasing neurons,/>Fitting a formula to a neural network,/>And outputting to an upper network.
The specific structure of the intelligent ocean wave correction model integrating the numerical mode and the deep learning is shown in fig. 3 (a) and 3 (b).
The invention constructs the intelligent sea wave correction model integrating the numerical mode and the deep learning. The system specifically comprises a sea wave numerical mode module and an artificial intelligence correction module. In the sea wave correcting process, firstly, sea wave numerical mode calculation is carried out. And obtaining a terrain water depth and wind field forcing file under the target sea area and the time period by setting the task sea area, further setting a wave numerical mode calculation scheme, and then calculating to obtain wave information at the target sea area and the target position. And then, integrating the mode calculation result and the actual measurement sea wave data such as the buoy and the like, and constructing an artificial intelligent correction module input data set by means of matching, deficiency processing and the like. The intelligent correction model of sea waves integrating the numerical mode and the deep learning is constructed through the deviation between the ANN training model output and the buoy actual measurement value, and finally the intelligent correction model of sea waves integrating the numerical mode and the deep learning can realize the intelligent correction of sea wave fields.
The wave numerical mode module in the intelligent sea wave correction model with the numerical mode and the deep learning built by the invention is used for carrying out numerical calculation on sea wave information in a sea area through a wave control equation based on topographic data such as the water depth of a task sea area and forced field data such as the wind speed, and finally outputting wave information at the sea area wave field and the position of a buoy station. The artificial intelligent correction module utilizes the buoy to actually measure the sea wave result based on the site position sea wave information output by the numerical mode, and builds the deviation between the simulation result and the real sea wave value. The intelligent sea wave correction model integrating the numerical mode and the deep learning takes the output sea wave information of the wind speed and the numerical mode as input and takes the measured wave height of the buoy as output. In the correction process, the input and the output of the sea wave intelligent correction model which are integrated with the numerical mode and the deep learning are in one-to-one correspondence in time. The hidden layer of the sea wave intelligent correction model integrating the numerical mode and the deep learning is 3 layers, and 200, 200 and 100 neurons are respectively arranged on each layer. Training and integrating the numerical mode and the deep learning ocean wave intelligent correction model parameters by adopting an Adam optimizer, wherein each layer adopts Relu as an activation function; the loss function uses a mean square error function.
And taking the numerical mode simulation wave height (Hs) as the basic input of the intelligent sea wave correction model integrating the numerical mode and the deep learning. Further, wind speeds (SSW) and wave age (WA) at the sea surface of 10m are introduced into the sea wave intelligent correction model integrating the numerical mode and the deep learning so as to test the forecasting precision of the sea wave intelligent correction model integrating the numerical mode and the deep learning.
；
In the method, in the process of the invention,Is wave phase velocity,/>Is wave age,/>For wind speed,/>Gravitational acceleration,/>Is the wave period; final wave age (WA) is shown as a function of wave period T and sea surface wind speed SSW.
The prediction effect of the intelligent sea wave correction model fused with the numerical mode and the deep learning under different inputs is shown in the following table 2.
Table 2 model correction error comparison table under multiple inputs (MAPE):
。
table 2 shows the tuning process of model input strategy, and finally filters and constructs the optimal input data set of the model by exploring the influence of various inputs such as wind speed, wave and the like on model prediction accuracy.
By comparison, the intelligent sea wave correction model integrating the numerical mode and the deep learning can be found, and the model prediction accuracy is highest when (Hs+WA) is taken as input.
S5, performing generalization analysis of the intelligent sea wave correction model by combining the numerical mode with the deep learning, correcting the sea wave numerical simulation result at a certain position in the sea area by using the intelligent sea wave correction model by combining the numerical mode with the deep learning, verifying the intelligent sea wave correction model by using the buoy actual measurement result, and testing the intelligent sea wave correction effect of the intelligent sea wave correction model by combining the numerical mode with the deep learning;
The invention selects (Hs+WA) as input and trains the model by using the measured data of No. 5 buoy. And correcting the output result of the sea wave mode under the scheme 1. Finally, after the intelligent correction model of the sea wave through the fusion of the numerical mode and the deep learning is corrected, the sea wave forecasting errors at the positions of the No. 1-No. 4 buoys are shown in the following table 3.
Table 3 model generalization analysis error statistics table:
。
tables 3 and 4 focus on the generalization of constructing the intelligent ocean wave correction model integrating the numerical mode and the deep learning. The actual measurement and simulation results at the buoy 5 are used as the input of the intelligent correction model of the sea wave with the fusion numerical mode and the deep learning, and the intelligent correction model of the sea wave with the fusion numerical mode and the deep learning is tested by using the buoys 1-4, so that the intelligent correction model of the sea wave with the fusion numerical mode and the deep learning still shows higher prediction precision to other untrained positions, and the intelligent correction model of the sea wave with the fusion numerical mode and the deep learning is proved to have better generalization performance.
By comparison, the current intelligent correction model of the sea wave integrating the numerical mode and the deep learning can be found, and the effective correction of the numerical simulation result of the wave field in the sea area can be realized. After correction, the sea wave forecasting result of the sea wave mode in the scheme 1 is equivalent to the numerical simulation effect of the numerical mode in the scheme 3 (namely the current precision optimal scheme). The invention selects the position of the No. 4 buoy, and displays the wave correction effect of the intelligent sea wave correction model integrating the numerical mode and the deep learning, as shown in fig. 5; fig. 5 includes measured wave height data of a buoy, a plurality of sets of wave numerical mode calculation results (including two sets of results with poor precision and highest precision), and wave prediction results after the intelligent correction model of the wave with the established fusion numerical mode and deep learning. The picture shows that the calculated result under the scheme of poor accuracy of the numerical mode is used as the input of the intelligent correction model of the sea wave integrating the numerical mode and the deep learning, and after correction, the sea wave forecasting result is superior to the calculated result with highest accuracy of the numerical mode. Further proves the superiority of the intelligent sea wave correction model integrating the numerical mode and the deep learning.
Therefore, the intelligent correction model for sea waves integrating the numerical mode and the deep learning can effectively correct the sea wave numerical simulation result. Meanwhile, the model can utilize actual measurement information of a single buoy position in the sea area to construct deviation between a wave simulation value and a true value, so that intelligent correction of the whole sea area wave field is realized.
According to the embodiment, the intelligent correction model of the sea wave is provided by combining the sea wave numerical mode and the deep learning method. And constructing the deviation between the numerical simulation result and the real value of the sea wave by using the measured data. And the existing numerical forecasting means is effectively corrected through a deep learning method. The calculation resources are greatly saved, and the forecasting precision of sea waves is improved. Meanwhile, the influence caused by the calculation complexity of the wave mode can be considered, and the wave numerical simulation result can be effectively optimized by designing different inputs. The intelligent sea wave correction model constructed by the invention provides a novel and effective method and strategy for sea wave forecast in real sea areas. The method can ensure the navigation safety of the ship, provides an environment foundation for marine climate analysis, and has important research significance for the application in the related fields.
Meanwhile, according to the embodiment, expected benefits and commercial values after the technical scheme is converted are as follows: according to the intelligent correction method and system for the sea wave, through fusion of the numerical mode and deep learning, the problems of optimization, time consumption in calculation and the like of the traditional sea wave numerical mode are effectively solved. The technical scheme remarkably improves the forecasting precision and efficiency in real sea area sea wave forecasting, and provides a novel and effective environmental data forecasting method for the ocean related industry. Expected benefits and commercial value are reflected in improving reliability of marine environment predictions, providing more reliable data support for decisions and planning in marine, fishery and other related fields.
The technical scheme of the invention fills the technical blank in the domestic and foreign industries: compared with the traditional wave numerical mode, the technical scheme can optimize the numerical mode calculation result and has good generalization, so that the forecasting precision and the practical applicability are improved. The innovation fills the technical blank that the traditional sea wave forecasting method cannot balance the calculation accuracy efficiency, and introduces more advanced and comprehensive technical means for the field of ocean environment forecasting.
The technical scheme of the invention solves the technical problems which are always desired to be solved but are not successful all the time: the technical scheme of the invention effectively solves a technical problem in the field of marine environment prediction for a long time. The traditional sea wave numerical mode is a problem which is long and desired to be solved by people because of higher calculation cost and resource requirement of calculation, and how to realize high-precision and faster prediction of real sea environment. The technical scheme omits time consumption of adjusting and optimizing a large number of numerical mode parameters, has higher prediction precision compared with the numerical mode, and provides a breakthrough solution for improving the accuracy and the applicability of marine environment prediction. The achievement fills the technical blank and provides an advanced technical means meeting the actual demands for the industry, and marks the successful response to the long-term challenges of the field.
The technical scheme of the invention overcomes the technical bias: through fusion of the numerical mode and deep learning, the scheme can realize effective correction of the wave numerical calculation result, can greatly save the numerical mode tuning time, and breaks through the problem that the numerical mode has difference between the calculation result and the true value due to the problems of simplified calculation process and the like. The innovative method not only improves the precision of the prediction model, but also plays an important step in solving the technical bottleneck which cannot be overcome in the past. The invention challenges the technical prejudice in the field, and brings new prospects and possibilities for the field of marine environment prediction.
Embodiment 2, as shown in fig. 6, the intelligent ocean wave correction system integrating numerical mode and deep learning provided by the embodiment of the invention includes:
The system comprises a task sea area environment data acquisition module 1, a buoy acquisition module and a data processing module, wherein the task sea area environment data acquisition module 1 is used for acquiring environment data in a task sea area, acquiring typical sea area weather and topography data by using open source environment information, and collecting actual measurement environment information of a buoy in the sea area;
the wave numerical mode calculation model module 2 is used for constructing a wave numerical mode calculation model, constructing a wave numerical mode calculation model based on the acquired environmental data, and setting a plurality of schemes to carry out numerical calculation on the wave conditions of the typical sea area by combining the mode calculation complexity;
The intelligent ocean wave correction data set constructing module 3 is used for constructing an intelligent ocean wave correction data set, extracting and preprocessing the solving result of the ocean wave numerical mode calculation model aiming at the buoy observation position, and constructing an intelligent ocean wave correction data set with one-to-one matching observation result and numerical simulation result at the buoy actual measurement position;
An artificial intelligent sea wave correction model module 4 is constructed and used for constructing an artificial intelligent sea wave correction model, analyzing the optimal input strategy of the artificial intelligent sea wave correction model by combining sea area characteristics, wherein the artificial intelligent sea wave correction model is an intelligent sea wave correction model integrating a numerical mode and deep learning;
The generalization analysis module 5 is used for carrying out generalization analysis on the intelligent sea wave correction model by combining the numerical mode and the deep learning, correcting the sea wave numerical simulation result at a certain position in the sea area by utilizing the intelligent sea wave correction model by combining the numerical mode and the deep learning, verifying the intelligent sea wave correction model by combining the numerical mode and the deep learning by adopting the buoy actual measurement result, and testing the intelligent sea wave correction effect of the intelligent sea wave correction model by combining the numerical mode and the deep learning.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
The content of the information interaction and the execution process between the devices/units and the like is based on the same conception as the method embodiment of the present invention, and specific functions and technical effects brought by the content can be referred to in the method embodiment section, and will not be described herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present invention. For specific working processes of the units and modules in the system, reference may be made to corresponding processes in the foregoing method embodiments.
The embodiment of the invention also provides a computer device, which comprises: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, which when executed by the processor performs the steps of any of the various method embodiments described above.
Embodiments of the present invention also provide a computer readable storage medium storing a computer program which, when executed by a processor, performs the steps of the respective method embodiments described above.
The embodiment of the invention also provides an information data processing terminal, which is used for providing a user input interface to implement the steps in the method embodiments when being implemented on an electronic device, and the information data processing terminal is not limited to a mobile phone, a computer and a switch.
The embodiment of the invention also provides a server, which is used for realizing the steps in the method embodiments when being executed on the electronic device and providing a user input interface.
Embodiments of the present invention provide a computer program product which, when run on an electronic device, causes the electronic device to perform the steps of the method embodiments described above.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application may implement all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing device/terminal apparatus, recording medium, computer memory, read-only memory (ROM), random access memory (Random Access Memory, RAM), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc.
While the invention has been described with respect to what is presently considered to be the most practical and preferred embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (10)
1. The intelligent sea wave correction method integrating the numerical mode and the deep learning is characterized by comprising the following steps of:
s1, acquiring environmental data in a task sea area, acquiring typical sea area weather and topography data by using open source environmental information, and collecting actual measurement environmental information of buoys in the sea area; the buoy actually-measured environmental information comprises sea wave height, period and wind speed data;
s2, constructing an ocean wave numerical mode calculation model based on the acquired environmental data, and setting a plurality of schemes to carry out numerical calculation on typical ocean wave conditions by combining the mode calculation complexity;
S3, constructing an intelligent ocean wave correction data set, extracting and preprocessing a solving result of an ocean wave numerical mode calculation model aiming at an observed position of the buoy, and constructing an intelligent ocean wave correction data set with one-to-one matching observed results and numerical simulation results at the observed position of the buoy;
S4, constructing an artificial intelligent wave correction model, and analyzing an optimal input strategy of the artificial intelligent wave correction model by combining with sea area characteristics, wherein the artificial intelligent wave correction model is an intelligent wave correction model integrating a numerical mode and deep learning;
S5, performing generalization analysis of the intelligent sea wave correction model by combining the numerical mode and the deep learning, correcting the sea wave numerical simulation result at a certain position in the sea area by using the intelligent sea wave correction model by combining the numerical mode and the deep learning, verifying the intelligent sea wave correction model by combining the numerical mode and the deep learning by using the buoy actual measurement result, and testing the intelligent sea wave correction effect of the intelligent sea wave correction model by combining the numerical mode and the deep learning.
2. The intelligent correction method for sea waves by combining numerical modes and deep learning according to claim 1, wherein in the step S1, the environmental data comprises sea surface wind speed and sea water depth information, and the buoy actual measurement environmental information comprises sea wave height, period and wind speed data.
3. The intelligent correction method for sea waves integrating numerical modes and deep learning according to claim 1, wherein in step S2, the calculation model of the numerical mode of sea waves is as follows:
；
In the method, in the process of the invention, Is a linear input item,/>For wind energy input item,/>For nonlinear wave-wave interactions,/>Dissipation term leading to wave breaking,/>For wave-bottom interaction,/>For wave breaking caused by reduced water depth,/>Is the interaction of three waves,/>Is dispersion term,/>For wave ice interactions,/>Is a reflective item,/>Is a custom source item.
4. The intelligent correction method for sea waves by combining numerical modes and deep learning according to claim 1, wherein in step S2, the calculating complexity of the combination modes, and setting a plurality of schemes to perform numerical calculation on typical sea wave conditions comprises: and setting a plurality of schemes in the solving process of the wave numerical mode calculation model by using the root mean square error RMSE and the average relative error MAPE as evaluation standards.
5. The intelligent correction method for sea waves integrating numerical modes and deep learning according to claim 1, wherein in step S3, the extracting and preprocessing the solution result of the calculation model of the numerical mode of sea waves for the buoy observation position comprises: the buoy observation position is matched with the buoy actual measurement result through a data interpolation and deficiency processing mode.
6. The method for intelligent correction of sea waves with integrated numerical mode and deep learning according to claim 1, wherein in step S4, the constructing an artificial intelligent sea wave correction model comprises: based on an ANN neural network, on the basis of the wave numerical mode calculation model, an intelligent wave correction model integrating the numerical mode and deep learning is constructed, and the actual wave data is utilized to correct the wave numerical mode forecasting result.
7. The intelligent correction method for sea waves integrating numerical modes and deep learning according to claim 6, wherein the intelligent correction model for sea waves comprises: the sea wave numerical mode module and the artificial intelligent correction module;
The wave numerical mode module carries out numerical calculation on sea wave information in the sea area through a wave control equation based on the water depth topography data and the wind speed forced field data of the task sea area, and finally outputs wave information at the sea area wave field and the buoy station position;
The artificial intelligence correction module is used for constructing deviation between a simulation result and a real value of the sea wave by utilizing a buoy actual measurement sea wave result based on the site position sea wave information output by the numerical mode.
8. The intelligent correction method for sea waves integrating numerical modes and deep learning according to claim 7, wherein the hidden layers of the intelligent correction model for sea waves are 3 layers, and 200, 200 and 100 neurons are respectively arranged in each layer.
9. The intelligent correction method for sea waves integrating numerical modes and deep learning according to claim 6, wherein the intelligent correction model for sea waves takes a numerical mode simulation wave height Hs as basic input, introduces wind speeds SSW and wave age WA at the sea surface of 10m, analyzes the influence of wind speeds and wave inputs on the forecasting precision of the intelligent correction model for sea waves, and screens and constructs an input dataset.
10. An intelligent correction system for sea waves integrating numerical modes and deep learning, which is characterized in that the system implements the intelligent correction method for sea waves integrating numerical modes and deep learning according to any one of claims 1-9, and the system comprises:
The system comprises a task sea area environment data acquisition module (1) for acquiring environment data in a task sea area, acquiring typical sea area weather and topography data by using open source environment information, and collecting actual measurement environment information of a buoy in the sea area;
The wave numerical mode calculation model module (2) is used for constructing a wave numerical mode calculation model, constructing a wave numerical mode calculation model based on acquired environmental data, and setting a plurality of schemes to carry out numerical calculation on the wave conditions of a typical sea area by combining the mode calculation complexity;
The intelligent ocean wave correction data set constructing module (3) is used for constructing an intelligent ocean wave correction data set, extracting and preprocessing the solving result of the ocean wave numerical mode calculation model aiming at the buoy observation position, and constructing an intelligent ocean wave correction data set with one-to-one matching observation result and numerical simulation result at the buoy actual measurement position;
an artificial intelligent wave correction model module (4) is constructed and used for constructing an artificial intelligent wave correction model, analyzing the optimal input strategy of the artificial intelligent wave correction model by combining with sea area characteristics, wherein the artificial intelligent wave correction model is an intelligent wave correction model integrating a numerical mode and deep learning;
The generalization analysis module (5) is used for carrying out generalization analysis on the intelligent sea wave correction model by combining the numerical mode and the deep learning, correcting the sea wave numerical simulation result at a certain position in the sea area by utilizing the intelligent sea wave correction model by combining the numerical mode and the deep learning, verifying the intelligent sea wave correction model by combining the numerical mode and the deep learning by adopting the buoy actual measurement result, and testing the intelligent sea wave correction effect of the intelligent sea wave correction model by combining the numerical mode and the deep learning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410308826.8A CN117909666A (en) | 2024-03-19 | 2024-03-19 | Intelligent sea wave correction method and system integrating numerical mode and deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410308826.8A CN117909666A (en) | 2024-03-19 | 2024-03-19 | Intelligent sea wave correction method and system integrating numerical mode and deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117909666A true CN117909666A (en) | 2024-04-19 |
Family
ID=90689228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410308826.8A Pending CN117909666A (en) | 2024-03-19 | 2024-03-19 | Intelligent sea wave correction method and system integrating numerical mode and deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117909666A (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102831313A (en) * | 2012-08-22 | 2012-12-19 | 河海大学 | Radial sand ridge group typhoon-generated wave numerical simulation method |
| CN104331599A (en) * | 2014-09-30 | 2015-02-04 | 江苏省交通规划设计院股份有限公司 | Unstructured grid nesting wave numerical simulation method |
| CN108920877A (en) * | 2018-08-02 | 2018-11-30 | 中交第四航务工程勘察设计院有限公司 | A kind of global wave method for numerical simulation based on MIKE21-SW model |
| CN112199840A (en) * | 2020-09-30 | 2021-01-08 | 国家海洋环境预报中心 | Numerical-mode sea-surface wind speed correction method and device, electronic equipment and storage medium |
| CN113468773A (en) * | 2021-04-27 | 2021-10-01 | 交通运输部天津水运工程科学研究所 | Global table wind wave numerical simulation method and system |
| KR102386532B1 (en) * | 2022-01-12 | 2022-04-15 | 주식회사 아라종합기술 | Methods for learning and testing a multi-layer perceptron based marine environment information correction model for correcting marine environment information using wave prediction information and devices using the same |
| CN114491386A (en) * | 2022-02-21 | 2022-05-13 | 国家海洋环境预报中心 | Sea wave effective wave height correction method based on sea gas temperature difference |
| CN114611608A (en) * | 2022-03-14 | 2022-06-10 | 中国人民解放军国防科技大学 | Sea surface height numerical value prediction deviation correction method based on deep learning model |
| CN114912077A (en) * | 2022-05-27 | 2022-08-16 | 中国海洋大学 | Sea wave forecasting algorithm integrating random search and mixed decomposition error correction |
| CN114936620A (en) * | 2022-03-04 | 2022-08-23 | 中国人民解放军国防科技大学 | Sea surface temperature numerical value forecast deviation correction method based on attention mechanism |
| CN116029378A (en) * | 2022-12-19 | 2023-04-28 | 中国石油大学(华东) | Ocean numerical mode single-point deviation correction method and system based on random forest |
| CN116644608A (en) * | 2023-06-14 | 2023-08-25 | 青岛哈尔滨工程大学创新发展中心 | Real sea area ship motion forecasting method and system based on marine environment data |
| CN117113828A (en) * | 2023-08-21 | 2023-11-24 | 中船航海科技有限责任公司 | Numerical forecast correction method based on ship-based navigation observation |
| KR20230164518A (en) * | 2022-05-25 | 2023-12-04 | 에이치디현대중공업 주식회사 | Ocean information prediction system and vessel comprising the same |
| CN117540625A (en) * | 2023-10-27 | 2024-02-09 | 中科星图维天信科技股份有限公司 | Training method and device for correction model of sea wave numerical forecast data |
-
2024
- 2024-03-19 CN CN202410308826.8A patent/CN117909666A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102831313A (en) * | 2012-08-22 | 2012-12-19 | 河海大学 | Radial sand ridge group typhoon-generated wave numerical simulation method |
| CN104331599A (en) * | 2014-09-30 | 2015-02-04 | 江苏省交通规划设计院股份有限公司 | Unstructured grid nesting wave numerical simulation method |
| CN108920877A (en) * | 2018-08-02 | 2018-11-30 | 中交第四航务工程勘察设计院有限公司 | A kind of global wave method for numerical simulation based on MIKE21-SW model |
| CN112199840A (en) * | 2020-09-30 | 2021-01-08 | 国家海洋环境预报中心 | Numerical-mode sea-surface wind speed correction method and device, electronic equipment and storage medium |
| CN113468773A (en) * | 2021-04-27 | 2021-10-01 | 交通运输部天津水运工程科学研究所 | Global table wind wave numerical simulation method and system |
| KR102386532B1 (en) * | 2022-01-12 | 2022-04-15 | 주식회사 아라종합기술 | Methods for learning and testing a multi-layer perceptron based marine environment information correction model for correcting marine environment information using wave prediction information and devices using the same |
| CN114491386A (en) * | 2022-02-21 | 2022-05-13 | 国家海洋环境预报中心 | Sea wave effective wave height correction method based on sea gas temperature difference |
| CN114936620A (en) * | 2022-03-04 | 2022-08-23 | 中国人民解放军国防科技大学 | Sea surface temperature numerical value forecast deviation correction method based on attention mechanism |
| CN114611608A (en) * | 2022-03-14 | 2022-06-10 | 中国人民解放军国防科技大学 | Sea surface height numerical value prediction deviation correction method based on deep learning model |
| KR20230164518A (en) * | 2022-05-25 | 2023-12-04 | 에이치디현대중공업 주식회사 | Ocean information prediction system and vessel comprising the same |
| CN114912077A (en) * | 2022-05-27 | 2022-08-16 | 中国海洋大学 | Sea wave forecasting algorithm integrating random search and mixed decomposition error correction |
| CN116029378A (en) * | 2022-12-19 | 2023-04-28 | 中国石油大学(华东) | Ocean numerical mode single-point deviation correction method and system based on random forest |
| CN116644608A (en) * | 2023-06-14 | 2023-08-25 | 青岛哈尔滨工程大学创新发展中心 | Real sea area ship motion forecasting method and system based on marine environment data |
| CN117113828A (en) * | 2023-08-21 | 2023-11-24 | 中船航海科技有限责任公司 | Numerical forecast correction method based on ship-based navigation observation |
| CN117540625A (en) * | 2023-10-27 | 2024-02-09 | 中科星图维天信科技股份有限公司 | Training method and device for correction model of sea wave numerical forecast data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sutherland et al. | Evaluation of coastal area modelling systems at an estuary mouth | |
| CN116644608B (en) | Real sea area ship motion forecasting method and system based on marine environment data | |
| CN107193060B (en) | A kind of multipath Typhoon Storm Surge Over method for quick predicting and system | |
| CN108985521A (en) | Output power predicting method, device, equipment and the medium of photovoltaic generating system | |
| CN113344291A (en) | Urban inland inundation range forecasting method, device, medium and equipment | |
| CN107480781A (en) | The nuclear accident Source Term Inversion method of neutral net adaptive Kalman filter | |
| Sandhya et al. | An operational wave forecasting system for the east coast of India | |
| CN112100922A (en) | Wind resource prediction method based on WRF and CNN convolutional neural network | |
| CN115018022A (en) | Quality control method, device, equipment and medium for gridding environment monitoring data | |
| CN113515798B (en) | Urban three-dimensional space expansion simulation method and device | |
| CN113609783B (en) | Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information | |
| CN114511061A (en) | Shoreside region sea fog visibility forecasting method based on depth neural network | |
| CN116976227B (en) | Storm water increasing forecasting method and system based on LSTM machine learning | |
| CN117909666A (en) | Intelligent sea wave correction method and system integrating numerical mode and deep learning | |
| CN111811465A (en) | Method for predicting sea wave effective wave height based on multi-sine function decomposition neural network | |
| Hu et al. | Adaptive environmental sampling for underwater vehicles based on ant colony optimization algorithm | |
| CN111126827A (en) | Input-output accounting model construction method based on BP artificial neural network | |
| Xu et al. | Monthly NDVI prediction using spatial autocorrelation and nonlocal attention networks | |
| CN113343924B (en) | Modulation signal identification method based on cyclic spectrum characteristics and generation countermeasure network | |
| CN114325877B (en) | Assessment method and device for weather forecast data | |
| Wang et al. | Xihe: A data-driven model for global ocean eddy-resolving forecasting | |
| Gong et al. | Development of a coupled genetic algorithm and empirical typhoon wind model and its application | |
| CN117935023B (en) | Method and system for identifying mesoscale vortex and analyzing features based on attention mechanism | |
| Limareva et al. | The climate change of the Caucasus as a result of the global warming | |
| CN117471575B (en) | Typhoon wave height forecasting method based on BO-LSTM neural network model |
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 |