CN111612274B - Tidal water level forecasting method based on space-time correlation - Google Patents
Tidal water level forecasting method based on space-time correlation Download PDFInfo
- Publication number
- CN111612274B CN111612274B CN202010469480.1A CN202010469480A CN111612274B CN 111612274 B CN111612274 B CN 111612274B CN 202010469480 A CN202010469480 A CN 202010469480A CN 111612274 B CN111612274 B CN 111612274B
- Authority
- CN
- China
- Prior art keywords
- tidal
- water level
- time
- tidal water
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A tidal water level forecasting method based on space-time correlation utilizes tidal water level data of a relevant port which has correlation with tidal water level historical data of a target port to forecast tidal water level data of the target port, forms a tidal water level data set of the target port, utilizes a convolutional neural network to extract spatial characteristics of the tidal water level data, utilizes a gating circulation unit neural network to extract time characteristics of the tidal water level data, utilizes a two-way gating circulation unit neural network to train and verify the tidal water level data with the outdated air characteristics extracted, generates a short-time tidal water level forecasting model, and achieves short-time forecasting of the tidal water level of the target port. The method solves the problems that the tidal water level of the port is not accurately predicted or the tidal water level cannot be predicted due to less historical data of the tidal water level of the port, is convenient for the port and pilot institutions to reasonably arrange and schedule, improves the production service efficiency, and ensures the navigation safety.
Description
Technical Field
The invention relates to a tidal water level forecasting method based on space-time correlation.
Background
The accurate tidal forecast information has important roles for coastal engineers, port authorities, maritime authorities, shipping practitioners, marine accident rescue institutions, and the like. At present, a plurality of methods for forecasting the tide exist, but the current port is mostly subjected to short-time tide forecasting based on a large amount of tide historical data of the current port, and little or no historical data of small ports are not considered, and the short-time tide forecasting is also needed to provide references for activities such as production, life and the like.
Disclosure of Invention
The tidal water level forecasting method based on space-time correlation solves the problem that tidal water level forecasting or inaccurate forecasting cannot be carried out due to the fact that historical data of tidal water levels of ports are few, is convenient for ports and pilot institutions to reasonably arrange and schedule, improves production service efficiency, and guarantees navigation safety.
In order to achieve the above object, the present invention provides a tidal water level prediction method based on space-time correlation, comprising the steps of:
predicting tidal level data of the target port using tidal level data of a related port having a correlation with tidal level history data of the target port, forming a tidal level dataset of the target port;
the space features of the tidal water level data are extracted by using a convolutional neural network, the time features of the tidal water level data are extracted by using a gating circulation unit neural network, the tidal water level data of which the time-out features are extracted by using a two-way gating circulation unit neural network are trained and verified, a short-time tidal water level forecasting model is generated, and the short-time forecasting of the tidal water level of a target port is realized.
The method for forming the tidal water level dataset of the target port comprises the following steps:
s1, obtaining tidal water level data;
s2, supplementing missing tidal water level data;
s3, calculating Kendell correlation coefficients of tidal water level data of the whole points of the target port and the surrounding ports;
s4, formulating data;
the tidal level data of the target port is taken as a predicted point, the tidal level data of the port with the Kendell correlation coefficient of more than or equal to 0.8 with the tidal level data of the target port is taken as an independent variable, the tidal level data which is taken as the independent variable and the dependent variable is mapped onto a one-dimensional vector, and the one-dimensional space information vector at the same moment is expressed as:
X s ＝(X 1 ，X 2 ，…，X s )
the one-dimensional space information vectors at different moments are combined into a matrix as follows:
wherein s is different ports, and t is time;
s5, normalizing the data;
normalizing the data set X formulated in step S4 using the pandas library and the numpy library in python:
wherein the tidal water level time sequence is represented by X, X' represents a number within (-1, 1) obtained by normalizing X, min (X) represents a minimum value in the tidal water level data, and max (X) represents a maximum value in the tidal water level data.
Interpolation is used to supplement the missing tidal level data.
Dividing data in tidal water level data sets of a target port, wherein the first 80% is used as a training set, the second 20% is used as a verification set, and the third 10% is used as a test set.
The method for generating the short-time tide water level forecast model comprises the following steps of:
s1, spatial feature extraction is carried out on tidal water level data by using a convolutional neural network CNN;
s2, performing time feature extraction on tidal water level data with the extracted spatial features by using a gate control circulation unit GRU neural network;
and S3, training and verifying the tidal water level data with the extracted spatial features and time features by using a Bi-gate control circulation unit Bi-GRU neural network, and establishing a short-time tidal forecast model to realize end-to-end forecast output.
The spatial feature extraction method comprises the following steps:
filling operation;
a convolution operation;
G(i)＝F(Aw+B)
wherein w is the filtering weight of the node, B is the deviation, A is the value of the input node, G (i) is the convolved spatial feature data, F is the activation function, and a ReLu activation function is used;
and (5) pooling operation.
Using a mean square error MSE as a loss function during training of the short time tide forecast model;
wherein y is t Is the observed value of the tide water level,Is a predicted tidal level.
Using an Adam algorithm to perform deep optimization on the neural network during training of the short-time tide prediction model;
m＝β 1 ·m+(1-β 1 )·d x
wherein beta is 1 And beta 2 Is two constants, x is the updated parameter, d x Is the derivative vector of x, m is used to store the first order matrix and v is used to store the second order matrix.
After each training, the short-time tide forecasting model is used for verifying the verification set and calculating the precision of the short-time tide forecasting model, and when the precision is not lower than 97 percent or the training times reach 2000 times, the training is stopped and the short-time tide forecasting model is stored;
the accuracy calculation formula is as follows:
wherein y is t Is the observed value of the tide water level,Is a predicted tidal level.
Testing the short time tide forecasting model by using a test set, and evaluating the short time tide forecasting model by using an average absolute error MAE, an average absolute percentage error MAPE, a root mean square error RMSE and a correlation coefficient CC;
wherein y is t Is the observed value of the tide water level,Is a tidal level predictor,/->Is the average value of tidal water level observation,/>Is the predicted average of tidal levels.
The present invention also provides a processing apparatus comprising: a processor adapted to execute each piece of program code; and a data storage device adapted to store a plurality of program codes; the program code is adapted to be loaded and executed by a processor to implement the method of tidal level prediction based on spatiotemporal correlation.
According to the method, the tidal water level data of the port with higher correlation with the tidal water level historical data of the target port can be utilized to predict the tidal water level data of the target port in a space-time mode, the problem that the tidal water level historical data of the port are few and the tidal water level prediction cannot be carried out or the tidal water level prediction is inaccurate is solved, the reasonable arrangement and the scheduling of ports and navigation institutions are facilitated, the production service efficiency is improved, and the navigation safety is guaranteed.
Drawings
FIG. 1 is a schematic diagram of a tidal level prediction method based on spatio-temporal correlation according to the present invention.
Fig. 2 is a flow chart of forming a tidal level dataset for a target port.
FIG. 3 is a schematic diagram of the generation of a short time tidal level forecast model in an embodiment of the invention.
Fig. 4 is a schematic diagram of the location of a target port and its associated ports in an embodiment of the present invention.
Detailed Description
The following describes a preferred embodiment of the present invention with reference to fig. 1 to 4.
As shown in fig. 1, the present invention provides a tidal water level prediction method based on space-time correlation, comprising the steps of:
the tide level data preprocessing module predicts tide level data of the target port by using tide level data of a relevant port with higher relativity with tide level historical data of the target port, and forms a tide level data set of the target port;
the tide level forecasting module extracts spatial features from tide level data by using a convolutional neural network, extracts time features from the tide level data with the spatial features by using a gating circulation unit neural network, generates tide level data with obvious space-time features, learns and trains the tide level data with the temporal-spatial features by using a two-way gating circulation unit neural network, generates a short-time tide level forecasting model, and realizes short-time forecasting of tide level of a target port.
As shown in fig. 2, the method for forming the tidal water level dataset of the target port specifically includes:
and S1, acquiring tidal water level data.
And obtaining the whole-point tidal water level data of the port around the target port by using the tidal water level data recorder sensor.
And S2, supplementing missing tidal water level data.
Interpolation is used for interpolation and filling of the missing whole-point tidal water level data.
And S3, calculating a correlation coefficient.
And calculating Kendell (Kendall) correlation coefficients of the whole-point tidal level data of the target port and the port around the target port.
And S4, formulating the data.
The tidal level data of the target port is taken as a predicted point, the tidal level data of the port with higher correlation with the tidal level data of the target port (the correlation coefficient is more than or equal to 0.8) is determined as an independent variable, the tidal level data which is taken as the independent variable and the dependent variable is mapped onto a one-dimensional vector, and the one-dimensional space information vector at the same moment can be expressed as:
X s ＝(X 1 ，X 2 ，…，X s )
the one-dimensional space information vectors at different moments are combined into a matrix as follows:
wherein s is different ports, and t is time.
And S5, normalizing the data.
The dataset X formulated in step S4 was normalized using the pandas library and the numpy library in python, as follows:
wherein the tidal water level time sequence is represented by X, X' represents a number within (-1, 1) obtained by normalizing X, min (X) represents a minimum value in the tidal water level data, and max (X) represents a maximum value in the tidal water level data.
And S6, dividing the data set.
Taking the first 80% of the data set X' obtained in the step S5 as a training set, taking the last 20% as a verification set and taking the last 10% as a test set.
As shown in fig. 1, the method for generating the short-time tidal water level forecast model specifically includes the following steps:
and S1, extracting spatial features.
Spatial feature extraction is performed on tidal level data using convolutional neural networks (CNN, convolutional Neural Network).
And S2, extracting time characteristics.
And (3) performing time feature extraction on the tidal water level data with the extracted spatial features by using a gating and circulating unit (GRU, gate Recurrent Unit) neural network.
And S3, model training.
And (3) learning and training the tidal water level data with the spatial characteristics and the time characteristics extracted by using a Bi-gating circulation unit (Bi-GRU, bi-directional Gate Recurrent Unit) neural network, and establishing a short-time tidal forecast model to realize end-to-end forecast output.
In one embodiment of the invention, the tidal level forecast is performed for the upper sea margin (31 DEG 25.5'N,122 DEG 14.1' E), with the following steps:
and step 1, obtaining tidal water level data.
As shown in fig. 4, full-point tidal water level data at 1 month 0 to 29 days 23 of the year 2020 of the Shanghai and Shanghai (31 ° 25.5'n,122 ° 14.1' e), and surrounding reed harbors (30 ° 50.0'n,121 ° 50.0' e), mid-dredging (31 ° 6.8'n,121 ° 54.2' e), wu Song (31 ° 23.5'n,121 ° 30.5' e), jinshan (30 ° 45.0'n,121 ° 22.0' e), huangpu (31 ° 14.0'n,121 ° 29.0' e), high bridges (31 ° 19.8'n,121 ° 33.5' e), chongming (31 ° 32.0'n,121 ° 38.0' e) were obtained by using tidal water level data recorder sensors, respectively.
And 2, supplementing missing tide water level data.
And interpolating and filling the missing whole-point tidal water level data by using a cubic spline interpolation method.
And step 3, calculating a correlation coefficient.
The kendel (Kendall) correlation coefficients of tidal level data of the mid-and peripheral reed switch ports, mid-dredging, wu Song, golden mountain, huangpu, high bridge and Chongming are calculated, and the calculated correlation coefficients are respectively 0.826, 0.814, 0.278, 0.488, 0.089, 0.38 and 0.341, so that the correlation of the tidal level data of the mid-and reed switch ports and the mid-and peripheral reed switch ports is the highest.
And 4, formulating the data.
Formulating the whole-point tidal water level data of the middle dredging, the reed switch and the rest mountains, namely:
wherein s is 1 Representing the dredging, s 2 Representing the tidal ports, s 3 Representing the rest, t is the time.
And 5, normalizing the data.
The dataset formulated in step 4 was normalized using the pandas and numpy libraries in python, as follows:
wherein the tidal water level time sequence is represented by X, X' represents a number within (-1, 1) obtained by normalizing X, min (X) represents a minimum value in the tidal water level data, and max (X) represents a maximum value in the tidal water level data.
And 6, dividing the data set.
Taking the first 80% of the data set in the step 5 as a training set, taking the last 20% as a verification set and taking the last 10% as a test set.
And 7, filling (packing) operation.
As shown in fig. 3, in order to put the convolutional neural network in a matrix form, a zero dimension is added, that is, padding operation in the convolutional neural network, so that the original data is changed into a small matrix of 3×3.
And 8, convolution operation.
As shown in fig. 3, each element in the matrix is processed using a one-dimensional convolution, with a convolution kernel size of 2 x 2, with a stride of 1, due to the simple time series. And (3) using a one-dimensional convolution kernel filter as a convolution layer, obtaining convolution information of a local perception domain through a sliding filter, sliding unit nodes generated by each step on vectors, and aggregating local features into global features. The convolved data becomes a 2 x 2 matrix.
G(i)＝F(Aw+B)
Wherein w is the filtering weight of the node, B is the deviation, A is the value of the input node, G (i) is the convolved spatial feature data, F is the activation function, and a ReLu activation function is used.
And 9, pooling operation.
As shown in fig. 3, the pooling filter is applied as a pooling layer to the spatial feature data G (i) after the convolution operation, so as to effectively reduce the size of the feature map, thereby reducing parameters and learning burden. Some unnecessary spatial signature information is filtered out during pooling to obtain more abstract tidal level data spatial signatures. In order not to miss important information, the average pooling is adopted, the pooling core size is 2 multiplied by 2, the stride is 1, and the characteristic sequence G (i) generated in the previous step is reduced to half of the original size through the average pooling, namely, the characteristic sequence G (i) is changed into 1 multiplied by 1. The convolved and pooled data vector is denoted as C t ＝(C 1 ，C 2 ，…，C t )。
And step 10, extracting space-time characteristics.
As shown in FIG. 3, the convolved and pooled tidal level data C with distinct spatial features t Inputting the tidal water level data CG into a GRU module for time feature extraction to generate tidal water level data CG with obvious space-time features t ＝(CG 1 ，CG 2 ，…，CG t ) It is used as input to the Bi-GRU module.
And 11, model training and verification.
As shown in fig. 3, the model is trained and validated using a training set and a validation set. Using the mean square error MSE (Mean Squared Error) as a loss function when training the short time tidal forecast model, it can accurately describe the difference between the true and predicted values.
The MSE expression is as follows:
wherein y is t Is the observed value of the tide water level,Is a predicted tidal level.
The Adam algorithm was used to deeply optimize the neural network during training. The Adam algorithm is an update of RMSProp algorithm by integrating momentum, is an optimizer based on first order gradient, has very high calculation efficiency, and almost does not need to occupy memory.
The core part of Adam algorithm is as follows:
m＝β 1 ·m+(1-β 1 )·d x
wherein beta is 1 And beta 2 Is two constants, x is the updated parameter, d x Is a derivative vector of x, m is used to store a first order matrix, vIs used to store the second order matrix.
And after each training, verifying the verification set by using the short-time tide prediction model, calculating the precision of the short-time tide prediction model, and stopping training and storing the model when the precision is not lower than 97% or the training times reach 2000 times. The accuracy calculation formula is as follows:
wherein y is t Is the observed value of the tide water level,Is a predicted tidal level.
And step 12, testing a tide forecasting method based on space-time correlation.
And (3) testing the tide forecasting method based on the space-time correlation by using the short-time tide forecasting model stored in the step (11) and using a testing set.
The tidal prediction method based on space-time correlation is evaluated by adopting an average absolute error MAE, an average absolute percentage error MAPE, a root mean square error RMSE and a correlation coefficient CC, and the specific calculation formula is as follows:
wherein y is t Is the observed value of the tide water level,Is a tidal level predictor,/->Is the average value of tidal water level observation,/>Is the predicted average of tidal levels.
Experiments are carried out by taking the above-sea Sheshan as an example by using the tidal prediction method based on space-time correlation, and the experimental results show that the average absolute error MAE is as follows: 0.5602cm, average absolute percentage error MAPE: 0.32%, root mean square error RMSE: 0.7879cm, the correlation coefficient CC is 99.99%.
According to the method, the tidal water level data of the port with higher correlation with the tidal water level historical data of the target port can be utilized to predict the tidal water level data of the target port in a space-time mode, the problem that the tidal water level historical data of the port are few and the tidal water level prediction cannot be carried out or the tidal water level prediction is inaccurate is solved, the reasonable arrangement and the scheduling of ports and navigation institutions are facilitated, the production service efficiency is improved, and the navigation safety is guaranteed.
While the present invention has been described in detail through the foregoing description of the preferred embodiment, it should be understood that the foregoing description is not to be considered as limiting the invention. Many modifications and substitutions of the present invention will become apparent to those of ordinary skill in the art upon reading the foregoing. Accordingly, the scope of the invention should be limited only by the attached claims.
Claims (8)
1. A tidal water level forecasting method based on space-time correlation is characterized by comprising the following steps:
predicting tidal level data of the target port using tidal level data of a related port having a correlation with tidal level history data of the target port, forming a tidal level dataset of the target port;
the method comprises the steps of extracting spatial features from tidal water level data by using a convolutional neural network, extracting time features from the tidal water level data by using a gating circulation unit neural network, training and verifying the tidal water level data with the extracted time-out features by using a bidirectional gating circulation unit neural network, generating a short-time tidal water level forecast model, and realizing short-time forecast of the tidal water level of a target port;
the method for forming the tidal water level dataset of the target port comprises the following steps:
s1, obtaining tidal water level data;
s2, supplementing missing tidal water level data;
s3, calculating Kendell correlation coefficients of tidal water level data of the whole points of the target port and the surrounding ports;
s4, formulating data;
the tidal level data of the target port is taken as a predicted point, the tidal level data of the port with the Kendell correlation coefficient of more than or equal to 0.8 with the tidal level data of the target port is taken as an independent variable, the tidal level data which is taken as the independent variable and the dependent variable is mapped onto a one-dimensional vector, and the one-dimensional space information vector at the same moment is expressed as:
X s ＝(X 1 ，X 2 ，…，X s )
the one-dimensional space information vectors at different moments are combined into a matrix as follows:
wherein s is different ports, and t is time;
s5, normalizing the data;
normalizing the data set X formulated in step S4 using the pandas library and the numpy library in python:
wherein the tidal water level time sequence is represented by X, X' represents a number in (-1, 1) obtained by normalizing X, min (X) represents a minimum value in tidal water level data, and max (X) represents a maximum value in tidal water level data;
and S6, dividing data in the tidal water level data set of the target port, wherein the first 80% is used as a training set, the second 20% is used as a verification set, and the third 20% is used as a test set.
2. The method for forecasting tidal water level based on space-time correlation according to claim 1, wherein the method for generating the model for forecasting tidal water level in short time comprises the following steps:
s1, spatial feature extraction is carried out on tidal water level data by using a convolutional neural network CNN;
s2, performing time feature extraction on tidal water level data with the extracted spatial features by using a gate control circulation unit GRU neural network;
and S3, training and verifying the tidal water level data with the extracted spatial features and time features by using a Bi-gate control circulation unit Bi-GRU neural network, and establishing a short-time tidal forecast model to realize end-to-end forecast output.
3. The tidal water level prediction method based on space-time correlation according to claim 2, wherein the spatial feature extraction method comprises:
filling operation;
a convolution operation;
C(i)＝F(Aw+B)
wherein w is the filtering weight of the node, B is the deviation, A is the value of the input node, G (i) is the convolved spatial feature data, F is the activation function, and a ReLu activation function is used;
and (5) pooling operation.
4. A method of tidal level prediction based on spatiotemporal correlation as claimed in claim 3 wherein the mean square error MSE is used as a loss function when training the short term tidal prediction model;
wherein y is t Is the observed value of the tide water level,Is a predicted tidal level.
5. The tidal level prediction method based on space-time correlation according to claim 4, wherein a Adam algorithm is used to perform depth optimization on the neural network during training of the short-time tidal prediction model;
m＝β 1 ·m+(1-β 1 )·d x
wherein beta is 1 And beta 2 Is two constants, x is the updated parameter, d x Is the derivative vector of x, m is used to store the first order matrix and v is used to store the second order matrix.
6. The tidal level prediction method based on time-space correlation according to claim 5, wherein the short time tidal prediction model is used to verify the verification set after each training and calculate the accuracy of the short time tidal prediction model, and when the accuracy is not lower than 97%, or the training times reach 2000 times, the training is stopped and the short time tidal prediction model is stored;
the accuracy calculation formula is as follows:
wherein y is t Is the observed value of the tide water level,Is a predicted tidal level.
7. The tidal level prediction method based on space-time correlation as claimed in claim 6, wherein the short time tidal prediction model is tested by using a test set, and is evaluated by using mean absolute error MAE, mean absolute percentage error MAPE, root mean square error RMSE and correlation coefficient CC;
wherein y is t Is the observed value of the tide water level,Is a tidal level predictor,/->Is the tide water level observationAverage value->Is the predicted average of tidal levels.
8. A processing apparatus, comprising: a processor adapted to execute each piece of program code; and a data storage device adapted to store a plurality of program codes; characterized in that the program code is adapted to be loaded and executed by a processor to implement the space-time correlation based tidal level forecasting method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010469480.1A CN111612274B (en) | 2020-05-28 | 2020-05-28 | Tidal water level forecasting method based on space-time correlation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010469480.1A CN111612274B (en) | 2020-05-28 | 2020-05-28 | Tidal water level forecasting method based on space-time correlation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111612274A CN111612274A (en) | 2020-09-01 |
| CN111612274B true CN111612274B (en) | 2023-09-12 |
Family
ID=72196668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010469480.1A Active CN111612274B (en) | 2020-05-28 | 2020-05-28 | Tidal water level forecasting method based on space-time correlation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111612274B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113077110A (en) * | 2021-04-21 | 2021-07-06 | 国家海洋信息中心 | GRU-based harmonic residual segmented tide level prediction method |
| CN113506000B (en) * | 2021-07-15 | 2024-04-02 | 上海海事大学 | Method for calculating storm overflow risk of urban water collecting area |
| CN116542392B (en) * | 2023-05-24 | 2024-02-13 | 珠江水利委员会珠江水利科学研究院 | Real-time forecasting method and system for tidal range water level and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102214262A (en) * | 2010-04-02 | 2011-10-12 | 上海海洋大学 | Tide predicting method |
| CN104376230A (en) * | 2014-12-03 | 2015-02-25 | 大连海事大学 | Tidal prediction method |
| CN105956709A (en) * | 2016-05-12 | 2016-09-21 | 大连海事大学 | GUI based modular support vector machine tide forecasting method |
| KR101754030B1 (en) * | 2016-06-09 | 2017-07-05 | 군산대학교산학협력단 | Tide Level Data Processing Method And Device Of Thereof |
| CN108694482A (en) * | 2018-07-27 | 2018-10-23 | 西南石油大学 | Based on fractal theory and improved least square method supporting vector machine tide flow velocity prediction technique |
-
2020
- 2020-05-28 CN CN202010469480.1A patent/CN111612274B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102214262A (en) * | 2010-04-02 | 2011-10-12 | 上海海洋大学 | Tide predicting method |
| CN104376230A (en) * | 2014-12-03 | 2015-02-25 | 大连海事大学 | Tidal prediction method |
| CN105956709A (en) * | 2016-05-12 | 2016-09-21 | 大连海事大学 | GUI based modular support vector machine tide forecasting method |
| KR101754030B1 (en) * | 2016-06-09 | 2017-07-05 | 군산대학교산학협력단 | Tide Level Data Processing Method And Device Of Thereof |
| CN108694482A (en) * | 2018-07-27 | 2018-10-23 | 西南石油大学 | Based on fractal theory and improved least square method supporting vector machine tide flow velocity prediction technique |
Non-Patent Citations (1)
| Title |
|---|
| 李佳文 ; 王景升.基于深度学习的短时交通流预测.《山东交通科技》.2018,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111612274A (en) | 2020-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111612274B (en) | Tidal water level forecasting method based on space-time correlation | |
| US11921566B2 (en) | Abnormality detection system, abnormality detection method, abnormality detection program, and method for generating learned model | |
| Alencar et al. | Hybrid approach combining SARIMA and neural networks for multi-step ahead wind speed forecasting in Brazil | |
| GB2547816B (en) | Actually-measured marine environment data assimilation method based on sequence recursive filtering three-dimensional variation | |
| Jia et al. | Physics guided recurrent neural networks for modeling dynamical systems: Application to monitoring water temperature and quality in lakes | |
| US20150254554A1 (en) | Information processing device and learning method | |
| CN112712214B (en) | Method, system, device and storage medium for predicting track of maritime search and rescue object | |
| CN111027686A (en) | Landslide displacement prediction method, device and equipment | |
| Imani et al. | Caspian Sea level prediction using satellite altimetry by artificial neural networks | |
| CN111242351A (en) | Tropical cyclone track prediction method based on self-encoder and GRU neural network | |
| CN115545334B (en) | Land utilization type prediction method and device, electronic equipment and storage medium | |
| CN112668238B (en) | Rainfall processing method, rainfall processing device, rainfall processing equipment and storage medium | |
| CN114498619A (en) | Wind power prediction method and device | |
| CN114826947B (en) | Flow matrix recovery prediction method and system based on automatic encoder | |
| Jin et al. | Deep learning for seasonal precipitation prediction over China | |
| CN117575111A (en) | Agricultural remote sensing image space-time sequence prediction method based on transfer learning | |
| Wang et al. | Motionhint: Self-supervised monocular visual odometry with motion constraints | |
| CN117271979A (en) | Deep learning-based equatorial Indian ocean surface ocean current velocity prediction method | |
| Pashova et al. | Gap filling of daily sea levels by artificial neural networks | |
| CN112966815A (en) | Target detection method, system and equipment based on impulse neural network | |
| CN110648030A (en) | Method and device for predicting seawater temperature | |
| US20220327335A1 (en) | Controlling asynchronous fusion of spatio-temporal multimodal data | |
| Sabililah et al. | Time Series Forecasting of Sea Level by Using Transformer Approach, with a Case Study in Pangandaran, Indonesia | |
| CN108960406B (en) | MEMS gyroscope random error prediction method based on BFO wavelet neural network | |
| Champenois et al. | Real-time reconstruction of 3D ocean temperature fields from reanalysis data and satellite and buoy surface measurements |
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 |