CN111722306A - Typhoon landing intensity prediction method and system based on TCN network model - Google Patents
Typhoon landing intensity prediction method and system based on TCN network model Download PDFInfo
- Publication number
- CN111722306A CN111722306A CN202010732785.7A CN202010732785A CN111722306A CN 111722306 A CN111722306 A CN 111722306A CN 202010732785 A CN202010732785 A CN 202010732785A CN 111722306 A CN111722306 A CN 111722306A
- Authority
- CN
- China
- Prior art keywords
- typhoon
- network model
- landing
- running
- tcn network
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 6
- 230000001364 causal effect Effects 0.000 claims description 17
- 230000010339 dilation Effects 0.000 claims description 14
- 230000033001 locomotion Effects 0.000 claims description 12
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 description 7
- 239000000284 extract Substances 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000013277 forecasting method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- 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
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Environmental & Geological Engineering (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Atmospheric Sciences (AREA)
- Ecology (AREA)
- Biodiversity & Conservation Biology (AREA)
- Environmental Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for predicting typhoon landing intensity based on a TCN network model, wherein the prediction method comprises the following steps: acquiring historical data of typhoon occurrence, wherein the historical data comprises typhoon factors when a set distance is reserved between a typhoon center and a landing place and typhoon landing strength; establishing a TCN network model, and training the established TCN network model by adopting the historical data to obtain a trained TCN network model; and acquiring a typhoon running path, acquiring a typhoon factor when a set distance exists between a typhoon center and a landing place, and combining the trained TCN network model to obtain the landing strength. The technical scheme provided by the invention can solve the problem of inaccurate prediction of the typhoon landing intensity in the prior art.
Description
Technical Field
The invention belongs to the technical field of typhoon landing intensity prediction, and particularly relates to a method and a system for predicting typhoon landing intensity based on a TCN (traffic control network) model.
Background
Typhoon, as an extreme weather event, not only can affect activities on the sea, but also can cause significant loss to the lives of people and urban economy in coastal areas. Thus, typhoon research and prediction has been a major concern in various coastal countries.
The Chinese patent application publication No. CN109902885A discloses a typhoon prediction method based on deep learning hybrid CNN-LSTM, which uses a historical tropical cyclone optimal path data set and global atmospheric marine variable data, extracts spatial features of the atmospheric variable data by using 3DCNN (three-dimensional CNN), extracts spatial features of the marine variable data by using 2DCNN (two-dimensional CNN), and extracts time series information in the occurrence and development process of typhoon by using LSTM, thereby constructing a typhoon prediction model of hybrid CNN-LSTM, and predicting whether the typhoon is formed and the path and intensity after the typhoon is formed by using the model.
The technical scheme disclosed in the patent application document is to predict the possibility of typhoon formation and the path and strength after formation according to the environmental factors of typhoon occurrence, but the prediction method is complex and the landing strength of typhoon cannot be accurately predicted.
The Chinese patent application publication No. CN104932035A discloses a typhoon intensity forecasting method and system, which is used for forecasting the typhoon intensity through a multiple linear regression model, and the forecasting method is poor in accuracy and cannot accurately forecast the landing intensity of the typhoon.
Disclosure of Invention
The invention aims to provide a typhoon landing intensity prediction method and a system based on a TCN network model, which aim to solve the problem that the typhoon landing intensity cannot be accurately predicted in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a typhoon landing intensity prediction method based on a TCN network model comprises the following steps:
(1) acquiring historical data of typhoon occurrence, wherein the historical data comprises typhoon factors when a set distance is reserved between a typhoon center and a landing place and typhoon landing strength;
(2) establishing a TCN network model, and training the established TCN network model by adopting the historical data to obtain a trained TCN network model;
(3) and acquiring a typhoon running path, acquiring a typhoon factor when a set distance exists between a typhoon center and a landing place, and combining the trained TCN network model to obtain the landing strength.
Further, the method for acquiring the typhoon running path in the step (3) comprises the following steps: acquiring the running track of each typhoon in historical data; matching the running track of the current typhoon with the running tracks of the typhoons in the historical data to obtain the running track of the typhoon with the highest similarity, and taking the running track of the typhoon as the running track of the current typhoon.
Further, the method for judging the track similarity of the two segments of typhoon motions comprises the following steps: converting longitude and latitude coordinates of the typhoon running track into a universal transverse-axis mercator projection coordinate system; calculating the space distance between each point in the two sections of running tracks; calculating the predicted distance between each point in the two sections of running tracks; calculating the distance between each section in the two sections of running tracks; calculating the accumulated distance of the two sections of running tracks; and normalizing the obtained accumulated distance, wherein the greater the accumulated distance is, the higher the similarity of the two running tracks is.
Further, the convolutional neural network model in the step (2) is
The output of the TCN network model is
o=RELU(z+F(z))
Wherein d is an expansion factor, the value of d is {1,2, …,2n }, and n is the number of hidden layers in the TCN network model; k is the number of filters, f (i) is the ith filter, F (z) is the result of the dilation causal convolution, ziThe ith typhoon factor z is the input to the dilation causal convolution, and F (z) is the result of the dilation causal convolution.
Further, the typhoon factor comprises the radial wind intensity, the latitudinal wind intensity, the potential height and the relative humidity of the typhoon.
A system for typhoon-landing intensity prediction based on a TCN network model, comprising a processor and a memory, the memory having stored thereon a computer program for execution on the processor; when the processor executes the calculation program, the typhoon intensity prediction method is realized as follows:
(1) acquiring historical data of typhoon occurrence, wherein the historical data comprises typhoon factors when a set distance is reserved between a typhoon center and a landing place and typhoon landing strength;
(2) establishing a TCN network model, and training the established TCN network model by adopting the historical data to obtain a trained TCN network model;
(3) and acquiring a typhoon running path, acquiring a typhoon factor when a set distance exists between a typhoon center and a landing place, and combining the trained TCN network model to obtain the landing strength.
Further, the method for acquiring the typhoon running path in the step (3) comprises the following steps: acquiring the running track of each typhoon in historical data; matching the running track of the current typhoon with the running tracks of the typhoons in the historical data to obtain the running track of the typhoon with the highest similarity, and taking the running track of the typhoon as the running track of the current typhoon.
Further, the method for judging the track similarity of the two segments of typhoon motions comprises the following steps: converting longitude and latitude coordinates of the typhoon running track into a universal transverse-axis mercator projection coordinate system; calculating the space distance between each point in the two sections of running tracks; calculating the predicted distance between each point in the two sections of running tracks; calculating the distance between each section in the two sections of running tracks; calculating the accumulated distance of the two sections of running tracks; and normalizing the obtained accumulated distance, wherein the greater the accumulated distance is, the higher the similarity of the two running tracks is.
Further, the convolutional neural network model in the step (2) is
The output of the TCN network model is
o=RELU(z+F(z))
Wherein d is an expansion factor, the value of d is {1,2, …,2n }, and n is the number of hidden layers in the TCN network model; k is the number of filters, f (i) is the ith filter, F (z) is the result of the dilation causal convolution, ziThe ith typhoon factor z is the input to the dilation causal convolution, and F (z) is the result of the dilation causal convolution.
Further, the typhoon factor comprises the radial wind intensity, the latitudinal wind intensity, the potential height and the relative humidity of the typhoon.
According to the technical scheme provided by the invention, a TCN network model for predicting typhoon landing intensity is established, a typhoon running path is obtained, a typhoon factor is obtained when a set distance is reserved between a typhoon center and a landing place, and the trained TCN network model is combined to obtain the landing intensity. According to the technical scheme provided by the invention, the typhoon landing intensity can be predicted according to the typhoon factor when the distance between the typhoon center and the landing place is set, and the problem of inaccurate prediction of the typhoon landing intensity in the prior art is solved.
Drawings
FIG. 1 is a flow chart of a method for predicting typhoon landing intensity based on a TCN network model in an embodiment of the method of the invention;
fig. 2 is a schematic diagram of a TCN network model in an embodiment of the method of the present invention.
Detailed Description
The invention provides a method and a system for predicting typhoon landing intensity based on a TCN (train communication network) model, which are used for predicting reservoir disaster and solving the problem of inaccurate prediction of reservoir disaster in the prior art.
The method comprises the following steps:
the flow of the method for predicting typhoon landing intensity based on the TCN network model provided by this embodiment is shown in fig. 1, and the method includes the following steps:
the method comprises the following steps: historical data of typhoon occurrences is obtained.
The acquired historical data of typhoon occurrence comprises typhoon factors influencing typhoon landing intensity and intensity levels of typhoon landing when a set distance is reserved between a typhoon center and a landing place.
In this embodiment, the typhoon factor affecting the typhoon landing intensity includes the radial wind intensity, the latitudinal wind intensity, the potential height, and the relative humidity of the typhoon.
Step two: and establishing a TCN network model, and training the established TCN network model by adopting historical data of typhoon occurrence to obtain a trained TCN neural network model.
Step three: the method comprises the steps of predicting the running track of typhoon to obtain the running track, obtaining typhoon factors when a set distance is reserved between the center of the typhoon and a landing place, and finally predicting the landing intensity of the typhoon by combining a trained TCN network model.
As shown in fig. 2, the TCN network model established in step two of this embodiment calculates the target sequence of each group by using dilation causal convolution, where the formula used in the calculation formula is:
let the input quantity of the established TCN network model be z1、z2、z3And z4The sequence formed is Z ═ (Z)1,z2,z3,z4) Wherein z is1Is radial strength of wind, z2Is the intensity of the weft wind, z3Is the potential height, z4Is the relative humidity. Filter F ═ F1,f2,…,fk) Calculating the target sequence corresponding to each group by using the expansion causal convolution, wherein the calculation formula is as follows:
wherein d is an expansion factor, the value of d is {1,2, …,2n }, and n is the number of hidden layers in the TCN network model; k is the number of filters, f (i) is the ith filter, F (z) is the result of the dilation causal convolution, ziIs the ith typhoon factor.
The output of the TCN network model is
o=RELU(z+F(z))
z is the input to the dilated causal convolution, and F (z) is the result of the dilated causal convolution.
In the third step of this embodiment, the method for predicting the typhoon running track includes:
and acquiring the running track of each typhoon from historical data of typhoon occurrence.
Acquiring a running path of the current typhoon, and taking the running path as a running track of the current typhoon;
matching the running track of the current typhoon with the running tracks of the typhoons in the historical library to obtain the running track of the typhoon with the highest similarity, and taking the running track of the typhoon as the running track of the current typhoon.
The specific steps for judging the track similarity of the two segments of typhoon motions are as follows:
converting longitude and latitude coordinates of the typhoon running track into a universal transverse-axis mercator projection coordinate system;
calculating the space distance between each point in the two sections of running tracks;
calculating the predicted distance between each point in the two sections of running tracks;
calculating the distance between each section in the two sections of running tracks;
calculating the accumulated distance of the two sections of running tracks;
and normalizing the obtained accumulated distance to be between [0 and 1], wherein 0 represents that the two running tracks are completely dissimilar, and 1 represents that the two running tracks are completely similar, so that the running track with the highest similarity to the current typhoon running track in the historical data is obtained and is taken as the running track of the current typhoon.
The specific steps of calculating the space distance between each point in the two typhoon running tracks are as follows:
setting the running track of one typhoon as a running track L1, setting the running track of the other typhoon as a running track L2, selecting the ith track point in the running track L1 and the jth track point in the running track L2, judging whether the distance between the point segment from the ith track point to the jth track point is equal to the distance between the jth track point and the ith track point, and correcting if the distances are not equal to each other.
The specific steps of calculating the predicted distance between each point in the travel locus L1 and the travel locus L2 are as follows:
an ith track point P in the running track L1 is takeni(xi,yi) J-th track point SP in the travel track L2j(xj,yj) (ii) a Comparing point trajectories PiWith locus point SPjThe time sequence relation of (1) is that the trace point with larger time is A, the trace point with smaller time is B, and the time difference is delta t-tA-tB;
Calculating a predicted position B' of the track point B; supposing that the track point B is a motion point, traversing the time of each track point on the motion track where the track point B is located, and searching for tBAnd when the track point B is positioned between the two track points at + delta t, the predicted position of the track point B is obtained. Suppose at tBWhen point B is between the i-1 th and i-th track points at + Δ t, the spatial coordinates (x) of the predicted position B' for track point BB′,yB′) The calculation formula is as follows:
the motion between any two track points on the track is assumed to be uniform linear motion, so that the motion speed between the two track points can be obtained, and the solving formula is as follows:
if time tB+ Δ t is not present on the corresponding track, then B' is calculated as follows:
wherein N is the total number of points of the corresponding running track of the track point B;
when the predicted distance between the track point A and the track point B is calculated, the calculation formula is as follows:
distt(A,B)=dist(A,B′)
where dist (A, B ') is the Euclidean distance of A and B' in space coordinates.
The specific steps of calculating the distance between each segment in the running locus L1 and the running locus 2 are as follows:
let SmFor the m-th section of the travel path L1, SSnIs the o-th segment of the travel locus L2; first calculate SmAnd SSnAngle theta, S ofmTwo end points of (A) are Pm(xm,ym) And Pm+1(xm+1,ym+1),SSoAre SPo(xo,yo) And SPn+1(xo+1,yo+1) The calculation formula of the angle theta is as follows:
θ=|arctan2(ym+1-ym,xm+1-xm)-arctan2(yo+1-yo,xo+1-xo)|
the spatiotemporal distance between two segments is the sum of the spatiotemporal distances of the two end points of the segment, hence SmAnd SSoInter-segment distance dist ofs(Sm,SSo) The calculation formula of (2) is as follows:
dists(Sm,SSo)=f(θ)(distst(Pm,SPo)+distst(Pm+1,SPo+1))
solving f (theta) by the calculation formula:
where ω is the negative factor of the shape of the adjustable parameter, the larger ω is the insensitive to the shape factor, and ω is 1, dist in this embodimentsmid(Sm,SSo) Is the midpoint SmAnd SSoSpace-time distance of midpoints, distmax(R, S) is the maximum space-time distance between any two points of the movement locus L1 and the movement locus L2.
The embodiment of the system is as follows:
the embodiment provides a typhoon login strength prediction system based on a TCN network model, which includes a processor and a memory, where the memory stores a computer program for execution on the processor, and when the processor executes the computer program, the prediction method based on the TCN network model typhoon login strength provided in the above method embodiments is implemented.
The embodiments of the present invention disclosed above are intended merely to help clarify the technical solutions of the present invention, and it is not intended to describe all the details of the invention nor to limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (10)
1. A typhoon landing intensity prediction method based on a TCN network model is characterized by comprising the following steps:
(1) acquiring historical data of typhoon occurrence, wherein the historical data comprises typhoon factors when a set distance is reserved between a typhoon center and a landing place and typhoon landing strength;
(2) establishing a TCN network model, and training the established TCN network model by adopting the historical data to obtain a trained TCN network model;
(3) and acquiring a typhoon running path, acquiring a typhoon factor when a set distance exists between a typhoon center and a landing place, and combining the trained TCN network model to obtain the landing strength.
2. The method for predicting typhoon landing intensity based on the TCN network model as claimed in claim 1, wherein the method for acquiring the typhoon running path in the step (3) is as follows: acquiring the running track of each typhoon in historical data; matching the running track of the current typhoon with the running tracks of the typhoons in the historical data to obtain the running track of the typhoon with the highest similarity, and taking the running track of the typhoon as the running track of the current typhoon.
3. The method for predicting typhoon landing intensity based on the TCN network model as claimed in claim 2, wherein the method for judging the track similarity of the two segments of typhoon motions comprises the following steps: converting longitude and latitude coordinates of the typhoon running track into a universal transverse-axis mercator projection coordinate system; calculating the space distance between each point in the two sections of running tracks; calculating the predicted distance between each point in the two sections of running tracks; calculating the distance between each section in the two sections of running tracks; calculating the accumulated distance of the two sections of running tracks; and normalizing the obtained accumulated distance, wherein the greater the accumulated distance is, the higher the similarity of the two running tracks is.
4. The method for predicting typhoon landing intensity based on TCN network model according to claim 1, wherein the convolutional neural network model in step (2) is
The output of the TCN network model is
o=RELU(z+F(z))
Wherein d is an expansion factor, the value of d is {1,2, …,2n }, and n is the number of hidden layers in the TCN network model; k is the number of filters, f (i) is the ith filter, F (z) is the result of the dilation causal convolution, ziThe ith typhoon factor z is the input to the dilation causal convolution, and F (z) is the result of the dilation causal convolution.
5. The TCN network model-based typhoon landing intensity prediction method according to claim 1, wherein the typhoon factors comprise radial wind intensity, latitudinal wind intensity, potential altitude and relative humidity of typhoon.
6. A system for typhoon-landing intensity prediction based on a TCN network model, comprising a processor and a memory, the memory having stored thereon a computer program for execution on the processor; wherein the processor implements the method for predicting typhoon intensity as follows when executing the calculation program:
(1) acquiring historical data of typhoon occurrence, wherein the historical data comprises typhoon factors when a set distance is reserved between a typhoon center and a landing place and typhoon landing strength;
(2) establishing a TCN network model, and training the established TCN network model by adopting the historical data to obtain a trained TCN network model;
(3) and acquiring a typhoon running path, acquiring a typhoon factor when a set distance exists between a typhoon center and a landing place, and combining the trained TCN network model to obtain the landing strength.
7. The system for predicting typhoon landing intensity based on the TCN network model as claimed in claim 6, wherein the method for obtaining the typhoon running path in the step (3) is as follows: acquiring the running track of each typhoon in historical data; matching the running track of the current typhoon with the running tracks of the typhoons in the historical data to obtain the running track of the typhoon with the highest similarity, and taking the running track of the typhoon as the running track of the current typhoon.
8. The system for predicting typhoon landing intensity based on the TCN network model of claim 7, wherein the method for judging the track similarity of the two segments of typhoon motions comprises the following steps: converting longitude and latitude coordinates of the typhoon running track into a universal transverse-axis mercator projection coordinate system; calculating the space distance between each point in the two sections of running tracks; calculating the predicted distance between each point in the two sections of running tracks; calculating the distance between each section in the two sections of running tracks; calculating the accumulated distance of the two sections of running tracks; and normalizing the obtained accumulated distance, wherein the greater the accumulated distance is, the higher the similarity of the two running tracks is.
9. The TCN-network-model-based typhoon-landing intensity prediction system according to claim 6, wherein the convolutional neural network model in the step (2) is
The output of the TCN network model is
o=RELU(z+F(z))
Wherein d is an expansion factor, the value of d is {1,2, …,2n }, and n is the number of hidden layers in the TCN network model; k is the number of filters, f (i) is the ith filter, F (z) is the result of the dilation causal convolution, ziThe ith typhoon factor z is the input to the dilation causal convolution, and F (z) is the result of the dilation causal convolution.
10. The TCN-network-model-based typhoon landing intensity prediction system of claim 6, wherein the typhoon factors comprise radial wind intensity, latitudinal wind intensity, potential altitude and relative humidity of typhoons.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010732785.7A CN111722306A (en) | 2020-07-27 | 2020-07-27 | Typhoon landing intensity prediction method and system based on TCN network model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010732785.7A CN111722306A (en) | 2020-07-27 | 2020-07-27 | Typhoon landing intensity prediction method and system based on TCN network model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111722306A true CN111722306A (en) | 2020-09-29 |
Family
ID=72573626
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010732785.7A Pending CN111722306A (en) | 2020-07-27 | 2020-07-27 | Typhoon landing intensity prediction method and system based on TCN network model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111722306A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114548606A (en) * | 2022-04-25 | 2022-05-27 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Construction method of cyclone strength prediction model and cyclone strength prediction method |
CN115061223A (en) * | 2022-07-14 | 2022-09-16 | 郑州航空工业管理学院 | Intelligent forecasting system about west boundary flow |
CN116822140A (en) * | 2023-04-25 | 2023-09-29 | 中国气象局上海台风研究所(上海市气象科学研究所) | Space-time integrated similar typhoon influence inversion method and system combined with weather mechanism |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104200081A (en) * | 2014-08-22 | 2014-12-10 | 清华大学 | Method and system for forecasting landed typhoon characterization factors based on historical data |
CN104570161A (en) * | 2015-01-21 | 2015-04-29 | 中国南方电网有限责任公司 | Typhoon automated forecasting method based on EC/JMA global lattice point forecast data |
CN104932035A (en) * | 2015-05-26 | 2015-09-23 | 中国科学院深圳先进技术研究院 | Typhoon intensity prediction method and system |
CN110837137A (en) * | 2019-11-07 | 2020-02-25 | 刘健华 | Typhoon prediction alarm method |
CN111126680A (en) * | 2019-12-11 | 2020-05-08 | 浙江大学 | Road section traffic flow prediction method based on time convolution neural network |
CN111126718A (en) * | 2019-11-07 | 2020-05-08 | 刘健华 | Typhoon path prediction method |
-
2020
- 2020-07-27 CN CN202010732785.7A patent/CN111722306A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104200081A (en) * | 2014-08-22 | 2014-12-10 | 清华大学 | Method and system for forecasting landed typhoon characterization factors based on historical data |
CN104570161A (en) * | 2015-01-21 | 2015-04-29 | 中国南方电网有限责任公司 | Typhoon automated forecasting method based on EC/JMA global lattice point forecast data |
CN104932035A (en) * | 2015-05-26 | 2015-09-23 | 中国科学院深圳先进技术研究院 | Typhoon intensity prediction method and system |
CN110837137A (en) * | 2019-11-07 | 2020-02-25 | 刘健华 | Typhoon prediction alarm method |
CN111126718A (en) * | 2019-11-07 | 2020-05-08 | 刘健华 | Typhoon path prediction method |
CN111126680A (en) * | 2019-12-11 | 2020-05-08 | 浙江大学 | Road section traffic flow prediction method based on time convolution neural network |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114548606A (en) * | 2022-04-25 | 2022-05-27 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Construction method of cyclone strength prediction model and cyclone strength prediction method |
CN115061223A (en) * | 2022-07-14 | 2022-09-16 | 郑州航空工业管理学院 | Intelligent forecasting system about west boundary flow |
CN115061223B (en) * | 2022-07-14 | 2023-09-29 | 郑州航空工业管理学院 | Intelligent forecasting system for west boundary flow |
CN116822140A (en) * | 2023-04-25 | 2023-09-29 | 中国气象局上海台风研究所(上海市气象科学研究所) | Space-time integrated similar typhoon influence inversion method and system combined with weather mechanism |
CN116822140B (en) * | 2023-04-25 | 2024-02-27 | 中国气象局上海台风研究所(上海市气象科学研究所) | Space-time integrated similar typhoon influence inversion method and system combined with weather mechanism |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11657708B2 (en) | Large-scale real-time traffic flow prediction method based on fuzzy logic and deep LSTM | |
JP6759474B2 (en) | Vessel automatic tracking methods and systems based on deep learning networks and average shifts | |
Zürn et al. | Self-supervised visual terrain classification from unsupervised acoustic feature learning | |
CN111722306A (en) | Typhoon landing intensity prediction method and system based on TCN network model | |
CN105206041B (en) | Smart-phone track chain-cluster identification method considering sequential DBSCAN | |
CN110796168A (en) | Improved YOLOv 3-based vehicle detection method | |
Liu et al. | A survey on deep-learning approaches for vehicle trajectory prediction in autonomous driving | |
CN113159403B (en) | Intersection pedestrian track prediction method and device | |
CN113762338B (en) | Traffic flow prediction method, equipment and medium based on multiple graph attention mechanism | |
CN112651374B (en) | Future trajectory prediction method based on social information and automatic driving system | |
CN111739066B (en) | Visual positioning method, system and storage medium based on Gaussian process | |
CN111462180B (en) | Object tracking method based on AND-OR graph AOG | |
CN110348492A (en) | A kind of correlation filtering method for tracking target based on contextual information and multiple features fusion | |
CN111126718A (en) | Typhoon path prediction method | |
CN113903173B (en) | Vehicle track feature extraction method based on directed graph structure and LSTM | |
CN105184096A (en) | Virtual human movement pose calculating method | |
Xu et al. | Trajectory prediction for autonomous driving with topometric map | |
Meihong | Vehicle Detection Method of Automatic Driving based on Deep Learning | |
Zhang et al. | Multi-task learning for location prediction with deep multi-model ensembles | |
CN110081890A (en) | A kind of dynamic K arest neighbors map-matching method of combination depth network | |
CN111353509B (en) | Key point extractor generation method of visual SLAM system | |
CN115062708A (en) | Abnormal node detection method based on track deviation point embedding and depth clustering | |
CN115187614A (en) | Real-time simultaneous positioning and mapping method based on STDC semantic segmentation network | |
Chen et al. | Q‐EANet: Implicit social modeling for trajectory prediction via experience‐anchored queries | |
Chen et al. | Towards bio-inspired place recognition over multiple spatial scales |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200929 |
|
RJ01 | Rejection of invention patent application after publication |