CN113051817A - Sea wave height prediction method based on deep learning and application thereof - Google Patents
Sea wave height prediction method based on deep learning and application thereof Download PDFInfo
- Publication number
- CN113051817A CN113051817A CN202110295881.4A CN202110295881A CN113051817A CN 113051817 A CN113051817 A CN 113051817A CN 202110295881 A CN202110295881 A CN 202110295881A CN 113051817 A CN113051817 A CN 113051817A
- Authority
- CN
- China
- Prior art keywords
- wave height
- sea wave
- model
- deep learning
- prediction
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000013135 deep learning Methods 0.000 title claims abstract description 27
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims abstract 5
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims abstract 5
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 230000015654 memory Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 abstract description 11
- 230000007246 mechanism Effects 0.000 abstract description 6
- 230000007774 longterm Effects 0.000 abstract description 4
- 238000012549 training Methods 0.000 abstract description 4
- 238000013528 artificial neural network Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000013179 statistical model Methods 0.000 description 3
- 241001123248 Arma Species 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013277 forecasting method Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241000143306 Idaea sardoniata Species 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 241000509228 Thalasseus maximus Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000009364 mariculture Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 230000009897 systematic 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
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a sea wave height prediction method based on deep learning and application thereof, wherein the method comprises the following steps: sea wave data information is respectively input into an AM-LSTM model and a Catboost model to obtain outputs P1 and P2, and then P1 and P2 are reconstructed according to the following formula to obtain a prediction sequence P; p q 2P 1+ q 1P 2, wherein w1 is the mean of MAE, RMSE and MAPE output by AM-LSTM model, w2 is the mean of MAE, RMSE and MAPE output by Catboost model,Mean of RMSE and MAPE. The sea wave height prediction method provided by the invention has the advantages of the LSTM in deep learning in the aspect of processing long-term data prediction, the characteristics of the attention mechanism and the characteristics of few parameters, fast training and difficulty in overfitting of the Catboost, reconstructs the predicted data, has high prediction precision and strong generalization performance, is particularly suitable for sea wave height prediction, and has a great application prospect.
Description
Technical Field
The invention belongs to the technical field of time series prediction, relates to a sea wave height prediction method based on deep learning and application thereof, and particularly relates to a multivariable sea wave height prediction method based on deep learning and Catboost and application thereof.
Background
In recent years, the exploration center of gravity of human beings gradually turns to the ocean from the land, and the development and utilization of ocean resources such as mariculture, ocean transportation, coastal tourism and the like have great promotion effect on economic output value. The strategic importance of developing marine economy is well recognized in all countries of the world, mainly in coastal countries. Sea height is one of the important parameters in oceanographic research, and changes of the sea height can have great influence on offshore operation, port and wharf construction and coastal resident life, and even can cause extreme situations of coastal activity suspension and offshore related work reduction. Therefore, the accurate prediction of the sea wave height plays an important role in the human ocean activity, and the research on the prediction of the sea wave height has very important practical significance.
Many researchers at home and abroad deeply explore the sea wave height prediction method. Methods of conventional numerical models, single prediction models and combined prediction models have been proposed for the prediction of the height of ocean waves.
In the early days, the traditional numerical model was the most important method for studying wave height prediction, and WAM, JSONSWAP, SWAN, WIS, WAVEWATCH III have been used to predict wave height. Comparison of wave simulation effects of a third generation numerical model on different sea ice source items is verified in a document I (comparison of wave simulation effects of Miao Qi, Xufumin, Shu Luo Ling. WAVEWATCH III on different sea ice source items [ J ]. oceanographic newspaper 2020,42(09): 22-29.). Document two (von justice, horse boat, dong sea. wave height nonlinear probability distribution high-order spectral numerical model study [ J ] oceanographic, 2019,41(03):44-51.) numerically simulates the wave height under different initial conditions by using a high-order spectral model. However, when computing large models, due to the large amount of data and computational complexity, high performance computers and significant time costs are required, and numerical computation is unreliable in emergency situations where fast results are required.
With the continuous development of information technology and the arrival of the big data era, many researchers begin to adopt a machine learning method to predict the height of sea waves. The single prediction model is a model or algorithm, such as those in document three (span P T d. ARMA algorithms for ocean wave modeling [ J ].1983.) that uses the ARMA model to simulate a single wave height time series in a short stationary period. Artificial neural network models (ANN) have also been used for wave height prediction, and four documents (Deo M C, Naidu C. real time wave for estimating using the sea network [ J ]. Ocean engineering,1998,26(3):191-203.) propose a feed-forward network to predict wave height in real time, and their methods show more general, more flexible and more adaptive capabilities than autoregressive models. The fifth document (Mandal S, Prabaharan N. Ocean wave estimating using a recurrent neural network [ J ]. Ocean engineering,2006,33(10):1401 and 1410.) uses a Recurrent Neural Network (RNN) to perform wave height prediction, and finds that the correlation coefficient of the RNN is higher than that of the feedforward network. Document six (Fan S, Xiao N, Dong S.A novel model to predict symmetric wave length-term memory network [ J ]. Ocean Engineering,2020,205:107298.) proposes a long-short term memory (LSTM) network for fast prediction of effective wave height with higher accuracy than that of a conventional neural network. The seventh document (Mahjoobi J, mossabeb E a. prediction of systematic wave height using a generalized vector machines [ J ]. Ocean Engineering,2009,36(5): 339-.
With the progress of research, more and more technologies are applied. Document eight (Abed-Elmdous A, Kerachian R.wave height prediction using the rough set of the theory J. Ocean Engineering,2012,54: 244-. Document nine (Savitha R, Al multiple A. regional area wave height prediction using sequential neural networks [ J ]. Ocean Engineering,2017,129: 605-. The result shows that the minimum resource allocation network has a better structure and stronger adaptability. The document ten (Akbarifard S, radial f.predictingsea wave height using Symbian Organic Search (SOS) algorithm. Ocean Engineering,2018,167: 348-. The performance of the SOS algorithm is superior to that of a support vector regression, an artificial neural network and a simulated wave near-shore dynamic model.
The single prediction model has specific applicable targets and working environments, and the hybrid prediction model is generated at the same time. According to the near-shore wave forecasting method [ J ] of oceanographic (Chinese edition), 2014,36(09):18-29.) through coupling the numerical mode with the statistical model, the limitation of the near-shore wave forecasting method is realized through the logarithmic numerical mode and the statistical model, and a near-shore wave forecasting frame with the coupling of the numerical mode and the statistical model is constructed. The twelve literature (Yangjun steel, Zhangjie, Wangzui, Arctic sea area sea surface wind field and wave remote sensing observation capability analysis [ J ] oceanographic report, 2018,40(11): 105-plus 115.) mainly uses on-orbit multi-source satellite remote sensing data to develop research on distribution characteristics and change rules of the sea surface wind field and the wave in the Arctic sea area. The thirteenth document (Alexandre E, Cuadra L, Nieto-Borge J C, et al. A hybrid genetic algorithm-extreme learning machine approach for access to salt aware wave height recovery [ J ]. Ocean modeling, 2015,92:115 + 123.) applies genetic algorithms and extreme learning machine methods to reconstruct the wave height, which have proven effective in recovering lost wave data from the buoy. Fourteen documents (Duan WY, Han Y, Huang L M, et al. A hybrid EMD-SVR model for the short-term prediction of design wave height [ J ]. Ocean Engineering,2016,124:54-73.) propose a hybrid model based on Empirical Mode Decomposition (EMD) and Support Vector Regression (SVR) and are used for short-term prediction of nonlinear non-stationary wave heights. Fifteen (Kumar N K, Savitha R, Al Mamun A. ocean wave height prediction using ensemble of extreme learning machine [ J ]. neuro-prediction, 2018,277:12-20.) utilizes the randomness initialized in an Extreme Learning Machine (ELM) to obtain better generalization performance, and an ELM set is constructed to complete the prediction of the sea wave height. The sixteen (Ali M, Prasad R. signalling wave height for estimating a wave a an empirical learning machine model integrated with an improved adaptive noise integrated empirical mode decomposition [ J ]. Renewable and stationary Energy Reviews,2019,104:281-295 ]) document combines the extreme learning model with an improved adaptive noise integrated empirical mode decomposition method and is used to predict the wave height in the east coast region of Australia. Seventeen (Lu P, Liang S, Zou G, et al. M-LSTM, A HYBRID PREDICTION MODEL FOR WAVE HEIGHTS [ J ]. JOURNAL OF NONLINEAR AND CONVEX ANALYSIS,2019,20(5):775-786.) proposes to combine a Long Short Term Memory (LSTM) MODEL with a Multiple Linear Regression (MLR) MODEL, which is superior to the LSTM MODEL AND the multiple linear regression MODEL in predicting wave height.
Although the above model or algorithm can accomplish the prediction of the wave height, it has the disadvantages: traditional numerical model prediction requires a large amount of data and computational complexity; the single prediction model has specific applicable targets and working environments; existing hybrid predictive models may not adequately mine the relationship between wave height data.
Therefore, the method for predicting the height of the sea wave, which has the advantages of small data requirement and processing amount, good adaptability and capability of fully mining the relation between the height data of the sea wave, has practical significance.
Disclosure of Invention
The invention aims to overcome the defects that the existing wave height prediction method is large in data demand and processing capacity and poor in adaptability and cannot fully mine the relation between the wave height data, and provides the wave height prediction method which is small in data demand and processing capacity and good in adaptability and can fully mine the relation between the wave height data.
In order to achieve the purpose, the invention provides the following technical scheme:
a wave height prediction method based on deep learning is applied to electronic equipment, wave data information is respectively input into an AM-LSTM model (fully called a long-short term memory neural network added with an attention mechanism) and a Catboost model to obtain outputs P1 and P2, then P1 and P2 are reconstructed according to the following formula, the relation between features is further excavated, and a prediction sequence P is obtained;
P=q2*P1+q1*P2,
wherein w1 is the mean value of MAE (mean absolute error), RMSE (root mean square error) and MAPE (mean absolute percentage error) output by the AM-LSTM model, w2 is the mean value of MAE, RMSE and MAPE output by the Catboost model, q1 and q2 are the weight coefficients of P1 and P2 respectively, and the invention distributes a larger weight coefficient to the model with smaller evaluation index value (w value), thereby reducing the error of the combined model and improving the overall prediction precision.
The method combines the LSTM and the Catboost method in the deep learning, realizes the organic combination of the advantages of the LSTM in the deep learning in the aspect of processing long-term data prediction, the characteristics of attention mechanism and the characteristics of little parameters, fast training and difficult overfitting of the Catboost, reconstructs the predicted data, greatly improves the prediction precision, has stronger generalization performance and has great application prospect.
In the reconstruction process, three indexes of MAE (mean absolute error), RMSE (root mean square error) and MAPE (mean absolute percentage error) are selected as reconstruction parameters. The MAE can avoid the problem of mutual offset of errors, so that the actual prediction error can be accurately reflected; RMSE measures the deviation between observed and true values; MAPE takes into account the error and proportion of predicted to true values. The distribution of the weight coefficients can be more accurately realized by taking the average value of the three reconstruction parameters as the final parameter.
As a preferred technical scheme:
according to the sea wave height prediction method based on deep learning, the sea wave data information is subjected to data preprocessing before being input into an AM-LSTM model and a Catboost model.
As described above, in the method for predicting sea wave height based on deep learning, the data preprocessing refers to analyzing sea wave data information to confirm abnormal values (the abnormal values include missing values and data abnormal values), calculating an average value of left and right values of the abnormal values, replacing the abnormal values with the average value, and finally performing normalization processing on the data. The model can be conveniently subjected to data processing after the data preprocessing is carried out.
The sea wave height prediction method based on deep learning is described above, and the sea wave data information includes variables such as wind speed and wind direction and a wave height value.
The invention also provides electronic equipment applying the deep learning-based sea wave height prediction method, which comprises one or more processors, one or more memories, one or more programs and parameter acquisition equipment for acquiring the sea wave data information;
the one or more programs are stored in the memory and, when executed by the processor, cause the electronic device to perform a deep learning based wave height prediction method as described above.
Has the advantages that:
(1) the sea wave height prediction method based on deep learning disclosed by the invention has the advantages of the LSTM in the deep learning in the aspect of processing long-term data prediction, the characteristics of an attention mechanism and the characteristics of less parameters, fast training and difficulty in overfitting of the Catboost, and the predicted data is reconstructed;
(2) the sea wave height prediction method based on deep learning has high prediction precision and strong generalization performance, is particularly suitable for sea wave height prediction, and has great application prospect.
Drawings
Fig. 1 is a processing flow chart of a sea wave height prediction method based on deep learning according to the present invention.
Detailed Description
The present invention will be described in more detail with reference to the accompanying drawings, in which embodiments of the invention are shown and described, and it is to be understood that the embodiments described are merely illustrative of some, but not all embodiments of the invention.
The wave data information used in the following embodiments is specifically wave data information of the gulf of mexico station 42002 buoy, the measured data is within an acceptable error range specified by the national buoy center, and meets the national world weather organization (WMO) of relevant regulations, the original wave data information includes wind speed, wind direction and other variables and wave height values, and the summary of the gulf of mexico station 42002 is shown in the following table:
the wave height data is recorded once every hour from 1 day of 2015 to 13 days of 2015 and 10 months, and 6884 data are recorded.
Example 1
A sea wave height prediction method based on deep learning is applied to electronic equipment and comprises the following steps (specifically shown in figure 1):
(1) the method comprises the following steps of carrying out data preprocessing on sea wave data information including variables such as wind speed and wind direction and wave height values, specifically: after the wave data information is analyzed to confirm an abnormal value, calculating the average value of the left and right values of the abnormal value, replacing the abnormal value with the average value, and finally carrying out normalization processing on the data;
(2) sea wave data information after data preprocessing is respectively input into an AM-LSTM model and a Catboost model to obtain outputs P1 and P2;
(3) reconstructing P1 and P2 according to the following formula to obtain a predicted sequence P;
P=q2*P1+q1*P2,
wherein w1 is the mean value of MAE, RMSE and MAPE output by AM-LSTM model, and w2 is the mean value of MAE, RMSE and MAPE output by Catboost model.
The method also independently selects five algorithm models of SVR, ANN, LSTM, AM-LSTM and Catboost (compared with the embodiment 1, the difference is that no step (3) exists, the step (2) is changed into the step of inputting sea wave data information after data preprocessing into SVR, ANN, LSTM, AM-LSTM or Catboost algorithm models to obtain output) to predict the same data so as to verify the prediction precision of the method, and the method inspects the prediction performance of the method and each model through three indexes of MAE, RMSE and MAPE, and is specifically shown in the following table;
algorithm | MAE | RMSE | MAPE(%) |
SVR | 0.2652 | 0.3131 | 37.1725 |
ANN | 0.0882 | 01129 | 13.8608 |
LSTM | 0.0695 | 0.0925 | 10.0203 |
AM-LSTM | 0.0612 | 0.0815 | 8.7985 |
CatBoost | 0.0636 | 0.0856 | 11.2437 |
Method of the invention | 0.0521 | 0.0703 | 7.4512 |
As can be seen from the results in the above table, after the LSTM is further improved by using the attention mechanism, the predicted performance of the AM-LSTM is generally higher than that of the LSTM neural network. In addition, compared with an AM-LSTM model, the method can predict the sea wave height more accurately, and is far better than the prediction precision of a Catboost model, which shows that the prediction performance of the method is higher than that of a model without a reconstruction method by adopting the reconstruction method, and the reconstruction mode is proved to be meaningful.
In the method provided by the invention, an AM-LSTM model and a Catboost model are independently trained; then independently predicting by using the trained AM-LSTM model and the trained Catboost model; and finally, reconstructing the prediction results of the prediction results to be used as final prediction.
By verification, the method disclosed by the invention is found to be more effective in predicting and fitting the sea wave height data, and generate the performance incomparable with a single model. Compared with other models, the method has obvious superiority, more deeply excavates the relation between the characteristics, better fits the prediction target, and is particularly suitable for predicting the height of the sea wave.
Through verification, the sea wave height prediction method based on deep learning disclosed by the invention has the advantages of the LSTM in the deep learning in the aspect of processing long-term data prediction, the characteristics of an attention mechanism and the characteristics of less parameters, fast training and difficulty in overfitting of the Catboost, reconstructs the predicted data, has high prediction precision and strong generalization performance, is particularly suitable for sea wave height prediction, and has a great application prospect.
Example 2
An electronic device comprises one or more processors, one or more memories, one or more programs and a parameter acquisition device for acquiring sea wave data information;
one or more programs are stored in the memory that, when executed by the processor, cause the electronic device to perform the wave height prediction method as described in embodiment 1.
Although specific embodiments of the present invention have been described above, it will be appreciated by those skilled in the art that these embodiments are merely illustrative and various changes or modifications may be made without departing from the principles and spirit of the invention.
Claims (5)
1. A sea wave height prediction method based on deep learning is applied to electronic equipment and is characterized in that sea wave data information is respectively input into an AM-LSTM model and a Catboost model to obtain outputs P1 and P2, and then P1 and P2 are reconstructed according to the following formula to obtain a prediction sequence P;
P=q2*P1+q1*P2,
wherein w1 is the mean value of MAE, RMSE and MAPE output by AM-LSTM model, and w2 is the mean value of MAE, RMSE and MAPE output by Catboost model.
2. A sea wave height prediction method based on deep learning according to claim 1, characterized in that the sea wave data information is subjected to data preprocessing before being input into an AM-LSTM model and a Catboost model.
3. A sea wave height prediction method based on deep learning as claimed in claim 2, wherein the data preprocessing refers to analyzing sea wave data information to confirm an abnormal value, calculating an average value of the left and right values of the abnormal value, replacing the abnormal value with the average value, and finally normalizing the data.
4. A deep learning based sea wave height prediction method according to claim 1, wherein the sea wave data information comprises wind speed, wind direction and wave height values.
5. An electronic device applying the deep learning based sea wave height prediction method according to any one of claims 1 to 4, wherein the electronic device comprises one or more processors, one or more memories, one or more programs and a parameter acquisition device for acquiring the sea wave data information;
the one or more programs are stored in the memory and, when executed by the processor, cause the electronic device to perform a deep learning based wave height prediction method as described above.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110295881.4A CN113051817B (en) | 2021-03-19 | 2021-03-19 | Deep learning-based sea wave height prediction method and application thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110295881.4A CN113051817B (en) | 2021-03-19 | 2021-03-19 | Deep learning-based sea wave height prediction method and application thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113051817A true CN113051817A (en) | 2021-06-29 |
CN113051817B CN113051817B (en) | 2024-04-12 |
Family
ID=76514077
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110295881.4A Active CN113051817B (en) | 2021-03-19 | 2021-03-19 | Deep learning-based sea wave height prediction method and application thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113051817B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114265087A (en) * | 2021-12-24 | 2022-04-01 | 青岛镭测创芯科技有限公司 | Sea wave parameter inversion method and related assembly |
CN114445634A (en) * | 2022-02-28 | 2022-05-06 | 南京信息工程大学 | Sea wave height prediction method and system based on deep learning model |
CN114580509A (en) * | 2022-02-21 | 2022-06-03 | 国家海洋环境预报中心 | Sea wave macroscopic characteristic quantity prediction system based on convolution length memory network |
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 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106769051A (en) * | 2017-03-10 | 2017-05-31 | 哈尔滨理工大学 | A kind of rolling bearing remaining life Forecasting Methodology based on MCEA KPCA and combination S VR |
CN109213967A (en) * | 2018-10-22 | 2019-01-15 | 中国人民解放军战略支援部队航天工程大学 | A kind of carrier rocket data predication method and device, storage medium and electronic equipment |
CN110415013A (en) * | 2019-06-12 | 2019-11-05 | 河海大学 | A kind of combination forecasting method of net about vehicle trip requirements in short-term |
CN112488392A (en) * | 2020-12-01 | 2021-03-12 | 重庆邮电大学 | Intelligent water affair daily water consumption prediction method based on machine learning |
-
2021
- 2021-03-19 CN CN202110295881.4A patent/CN113051817B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106769051A (en) * | 2017-03-10 | 2017-05-31 | 哈尔滨理工大学 | A kind of rolling bearing remaining life Forecasting Methodology based on MCEA KPCA and combination S VR |
CN109213967A (en) * | 2018-10-22 | 2019-01-15 | 中国人民解放军战略支援部队航天工程大学 | A kind of carrier rocket data predication method and device, storage medium and electronic equipment |
CN110415013A (en) * | 2019-06-12 | 2019-11-05 | 河海大学 | A kind of combination forecasting method of net about vehicle trip requirements in short-term |
CN112488392A (en) * | 2020-12-01 | 2021-03-12 | 重庆邮电大学 | Intelligent water affair daily water consumption prediction method based on machine learning |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114265087A (en) * | 2021-12-24 | 2022-04-01 | 青岛镭测创芯科技有限公司 | Sea wave parameter inversion method and related assembly |
CN114580509A (en) * | 2022-02-21 | 2022-06-03 | 国家海洋环境预报中心 | Sea wave macroscopic characteristic quantity prediction system based on convolution length memory network |
CN114580509B (en) * | 2022-02-21 | 2022-10-28 | 国家海洋环境预报中心 | Sea wave macroscopic characteristic quantity prediction system based on convolution length memory network |
CN114445634A (en) * | 2022-02-28 | 2022-05-06 | 南京信息工程大学 | Sea wave height prediction method and system based on deep learning model |
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 |
Also Published As
Publication number | Publication date |
---|---|
CN113051817B (en) | 2024-04-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113051817A (en) | Sea wave height prediction method based on deep learning and application thereof | |
Wang et al. | A compound framework for wind speed forecasting based on comprehensive feature selection, quantile regression incorporated into convolutional simplified long short-term memory network and residual error correction | |
Ni et al. | An integrated long-short term memory algorithm for predicting polar westerlies wave height | |
Lu et al. | Short-term wind power forecasting based on meteorological feature extraction and optimization strategy | |
Li et al. | Multi-step ahead wind speed forecasting approach coupling maximal overlap discrete wavelet transform, improved grey wolf optimization algorithm and long short-term memory | |
CN111144644B (en) | Short-term wind speed prediction method based on variation variance Gaussian process regression | |
CN114549925A (en) | Sea wave effective wave height time sequence prediction method based on deep learning | |
CN111242351A (en) | Tropical cyclone track prediction method based on self-encoder and GRU neural network | |
Zhou et al. | A hybrid grey prediction model for small oscillation sequence based on information decomposition | |
Wang et al. | Quantile deep learning model and multi-objective opposition elite marine predator optimization algorithm for wind speed prediction | |
Li et al. | An innovative combined model based on multi-objective optimization approach for forecasting short-term wind speed: A case study in China | |
Hossain et al. | Optimized forecasting model to improve the accuracy of very short-term wind power prediction | |
Wang et al. | Wind speed forecasting based on hybrid model with model selection and wind energy conversion | |
Song et al. | Application of artificial intelligence based on synchrosqueezed wavelet transform and improved deep extreme learning machine in water quality prediction | |
Wang et al. | Hybrid deep learning model for short-term wind speed forecasting based on time series decomposition and gated recurrent unit | |
Wang et al. | A hybrid intelligent framework for forecasting short-term hourly wind speed based on machine learning | |
Huang et al. | Short-Term PV Power Forecasting Based on CEEMDAN and Ensemble DeepTCN | |
Sareen et al. | An integrated decomposition algorithm based bidirectional lstm neural network approach for predicting ocean wave height and ocean wave energy | |
Alharbi et al. | Short-term wind speed and temperature forecasting model based on gated recurrent unit neural networks | |
Xuan et al. | A comprehensive evaluation of statistical, machine learning and deep learning models for time series prediction | |
Deng et al. | Abnormal data detection for structural health monitoring: State-of-the-art review | |
Li et al. | A multi-factor combination prediction model of carbon emissions based on improved CEEMDAN | |
Hazra et al. | Generating Synthetic Data to Reduce Prediction Error of Energy Consumption. | |
Pang et al. | A novel multivariable hybrid model to improve short and long-term significant wave height prediction | |
Wang et al. | Filling gaps in significant wave height time series records using bidirectional gated recurrent unit and cressman analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |