CN117168545A - Ocean phenomenon observation method and system based on buoy end local identification - Google Patents
Ocean phenomenon observation method and system based on buoy end local identification Download PDFInfo
- Publication number
- CN117168545A CN117168545A CN202311411493.3A CN202311411493A CN117168545A CN 117168545 A CN117168545 A CN 117168545A CN 202311411493 A CN202311411493 A CN 202311411493A CN 117168545 A CN117168545 A CN 117168545A
- Authority
- CN
- China
- Prior art keywords
- history information
- sampling
- neural network
- observation
- rainfall
- 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 40
- 238000005070 sampling Methods 0.000 claims abstract description 96
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 33
- 238000005516 engineering process Methods 0.000 claims abstract description 10
- 238000013473 artificial intelligence Methods 0.000 claims description 24
- 238000013527 convolutional neural network Methods 0.000 claims description 24
- 238000013528 artificial neural network Methods 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- 230000015654 memory Effects 0.000 claims description 6
- 230000000306 recurrent effect Effects 0.000 claims description 4
- 230000006403 short-term memory Effects 0.000 claims description 4
- 238000002054 transplantation Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 3
- 230000007787 long-term memory Effects 0.000 claims description 2
- 238000005265 energy consumption Methods 0.000 abstract description 9
- 238000000605 extraction Methods 0.000 description 10
- 230000000875 corresponding effect Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 3
- 238000004146 energy storage Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- 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
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The application belongs to the technical field of ocean phenomenon observation, and provides an ocean phenomenon observation method and system based on buoy end local identification, wherein an artificial intelligent algorithm is used for predicting a key ocean phenomenon, and the artificial intelligent algorithm is integrated to a buoy end through an edge computing technology, so that the method for autonomously predicting the key ocean phenomenon by the buoy end is realized; meanwhile, before one or two signals of internal waves and typhoons do not appear, a small number of sensors are adopted to sample at a first sampling frequency with a low frequency, so that energy consumption is saved in a non-encryption observation mode, after one or two signals of internal waves and typhoons appear, a large number of sensors are adopted to sample at a second sampling frequency with a high frequency, accurate grading of sampling information is achieved in an encryption observation mode, and the energy consumption is reduced on the basis of guaranteeing observation quality in a non-encryption observation and encryption observation matched mode.
Description
Technical Field
The application belongs to the technical field of marine phenomenon observation, and particularly relates to a marine phenomenon observation method and system based on buoy end local identification.
Background
At present, the ocean phenomenon observation methods based on the buoy end are mainly divided into two types, one type is to transmit sensor data to a local base station by using satellites and observe the sensor data, the other type is a buoy end local identification method, and a more common method is a threshold method.
The inventor finds that the time required by the local base station prediction method is longer and the time delay is higher because buoy data is required to be sent to the local through a satellite; moreover, when observing internal waves and typhoons in the ocean phenomenon based on the buoy end, the capacity of storing and acquiring electric energy of the buoy end in the ocean is limited, the capacity of acquiring electric energy is greatly influenced by the ocean environment, equipment on the buoy end needs to be subjected to continuous high-frequency sampling in order to ensure the real-time performance, accuracy and the like of the observation, in the long-time high-frequency sampling process, the equipment is faster in loss of electric energy, the capacity of storing electric energy and acquiring electric energy of the buoy end in the ocean cannot meet the requirement of electric energy under long-time high-frequency sampling, if the electric energy cannot be timely supplemented, the problem of insufficient electric energy can be caused, normal operation and service life of equipment on the buoy end are influenced, and normal observation is influenced.
Disclosure of Invention
In order to solve the problems, the application provides a marine phenomenon observation method and a marine phenomenon observation system based on the local identification of a buoy end, wherein the method for predicting the key marine phenomenon by using an artificial intelligent algorithm and integrating the artificial intelligent algorithm into the buoy end through an edge computing technology is realized; meanwhile, before no internal wave or typhoon signal appears, the corresponding sensors with fewer numbers are adopted to sample at the first sampling frequency with lower frequency, so that the energy consumption is saved in a non-encryption observation mode, after the internal wave or typhoon signal appears, the corresponding sensors with more numbers are adopted to sample at the second sampling frequency with higher frequency, the accurate classification of sampling information is realized in an encryption observation mode, and the energy consumption is reduced on the basis of ensuring the observation quality in a mode of non-encryption observation and encryption observation matched operation, so that the problem that the capacity of storing electric energy and acquiring electric energy at the buoy end in the ocean cannot meet the requirement of electric energy under long-time high-frequency sampling is solved.
In order to achieve the above object, the present application is realized by the following technical scheme:
in a first aspect, the present application provides a marine phenomenon observation method based on local identification of a buoy end, including:
acquiring ocean atmospheric wind speed history information, rainfall history information, air pressure history information and underwater temperature profile data history information;
obtaining a prediction result of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature profile data by using the obtained atmospheric wind speed history information, the rainfall history information, the air pressure history information and the underwater temperature profile data history information and a preset prediction model based on an artificial intelligence algorithm; transplanting a prediction model based on an artificial intelligence algorithm to a buoy end by using an edge computing technology;
and obtaining a marine phenomenon observation result according to the prediction results of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature and a preset neural network recognition model.
Further, the prediction model based on the artificial intelligence algorithm is a long-term and short-term memory recurrent neural network; transplanting the artificial intelligence algorithm-based predictive model to the float end using edge computing techniques includes: firstly, expanding a port for receiving sensor data, transmitting prediction and transmitting identification results on the basis of a calculation unit according to the data transmission requirement; secondly, converting the prediction model into an open neural network switching architecture which is convenient for edge end transplantation; and finally, transplanting the prediction model of the open neural network switching architecture into the expanded computing unit.
Further, the buoy end is controlled to sample at a first sampling frequency by a first number of sensors to obtain sampling information;
judging whether the sampling information has internal waves or typhoons, if so, controlling the buoy end to sample at a second sampling frequency by using a second number of sensors, otherwise, continuing to sample at a first sampling frequency by using a first number of sensors; wherein the second number is greater than the first number and the second sampling frequency is higher than the first sampling frequency;
and grading the sampling information obtained by sampling at the second sampling frequency.
Further, after the time of sampling at the second sampling frequency by the second number of sensors reaches the preset time, the buoy end is controlled to sample at the first sampling frequency by the first number of sensors.
Further, judging whether the sampling information has internal waves or typhoons or not, and grading the sampling information obtained by sampling at the second sampling frequency, wherein the sampling information is realized by adopting a preset neural network identification model;
training the convolutional neural network by using the atmospheric wind speed history information, the rainfall history information and the air pressure history information; predicting whether typhoons exist according to the atmospheric wind speed, the rainfall and the air pressure obtained by the sensor and the trained convolutional neural network;
meanwhile, predicting the convolutional neural network by using historical information of the underwater temperature profile data; and predicting whether an internal wave exists according to the underwater temperature profile data acquired by the sensor and the trained convolutional neural network.
Further, a plurality of sensors are arranged at different depth positions of the observation area; a plurality of sensors of different depths are controlled to alternately sample at a second sampling frequency.
Further, a plurality of temperature sensors are provided at different depth positions of the observation area.
In a second aspect, the present application also provides a marine phenomenon observation system based on local identification of a buoy end, including:
a data acquisition module configured to: acquiring ocean atmospheric wind speed history information, rainfall history information, air pressure history information and underwater temperature profile data history information;
an artificial intelligence algorithm prediction module configured to: obtaining a prediction result of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature profile data by using the obtained atmospheric wind speed history information, the rainfall history information, the air pressure history information and the underwater temperature profile data history information and a preset prediction model based on an artificial intelligence algorithm; transplanting a prediction model based on an artificial intelligence algorithm to a buoy end by using an edge computing technology;
an identification module configured to: and obtaining a marine phenomenon observation result according to the prediction results of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature and a preset neural network recognition model.
In a third aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of the marine phenomenon observation method according to the first aspect based on local identification of the float end.
In a fourth aspect, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the marine phenomenon observation method based on local identification at the buoy end according to the first aspect when the processor executes the program.
Compared with the prior art, the application has the beneficial effects that:
1. the method for predicting the key ocean phenomenon by using the artificial intelligent algorithm and integrating the artificial intelligent algorithm to the buoy end through the edge computing technology, so that the method for autonomously predicting the key ocean phenomenon by the buoy end is realized, and the problems of long time and high delay caused by the need of sending buoy data to the local through a satellite are solved;
2. the method comprises the steps that a buoy end is controlled to sample at a first sampling frequency by a first number of sensors, so that sampling information is obtained; judging whether the sampling information has internal waves or typhoons, if so, controlling the buoy end to sample at a second sampling frequency by using a second number of sensors, otherwise, continuing to sample at a first sampling frequency by using a first number of sensors; before no internal wave or typhoon signal appears, the sensors with fewer numbers are adopted to sample at a first sampling frequency with lower frequency, so that energy consumption is saved in a non-encryption observation mode, after the internal wave or typhoon signal appears, the sensors with more numbers are adopted to sample at a second sampling frequency with higher frequency, the accurate grading of sampling information is realized in an encryption observation mode, and the energy consumption is reduced on the basis of ensuring the observation quality through the non-encryption observation and encryption observation matched operation mode, so that the problem that the electric energy requirement under long-time high-frequency sampling cannot be met due to the electric energy storage and electric energy acquisition capacity of a buoy end in the ocean is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of this specification, illustrate and explain the embodiments and together with the description serve to explain the embodiments.
FIG. 1 is a flow chart of embodiment 1 of the present application;
fig. 2 is a schematic diagram of a sampling flow chart in embodiment 1 of the present application.
Detailed Description
The application will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
Example 1:
in view of the above problems, as shown in fig. 1, this embodiment provides a marine phenomenon observation method based on local identification of a buoy end, including:
acquiring ocean atmospheric wind speed history information, rainfall history information, air pressure history information and underwater temperature profile data history information;
obtaining a prediction result of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature profile data by using the obtained atmospheric wind speed history information, the rainfall history information, the air pressure history information and the underwater temperature profile data history information and a preset prediction model based on an artificial intelligence algorithm; the prediction model based on the artificial intelligence algorithm is a Long Short-Term Memory recurrent neural network (LSTM); transplanting a prediction model based on an artificial intelligence algorithm to a buoy end by using an edge computing technology;
obtaining a marine phenomenon observation result according to the prediction results of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature and a preset neural network recognition model; the neural network recognition model may be trained using convolutional neural networks (Convolutional Neural Networks, CNN).
Specifically, the key ocean phenomenon is predicted by using an artificial intelligent algorithm, and the artificial intelligent algorithm is integrated to the buoy end through an edge computing technology, so that the method for autonomously predicting the key ocean phenomenon by the buoy end is realized, and the problems of long time and high delay caused by the need of sending buoy data to the local through a satellite are solved.
In the convolutional neural network in the embodiment, the original data is a 14-layer temperature sequence with the length of 30, and contains 420 feature quantities in total, the data becomes a 5-layer feature sequence with the length of 8 after passing through a feature extraction network formed by one-dimensional convolution, and the feature extraction efficiency can reach 90.48% in total, and 40 feature quantities are contained in total; the feature extraction mode is realized by two modes of expanding the sampling step of convolution operation and reducing the number of convolution kernels.
Collected seaThe correlation between adjacent moments of the ocean temperature profile data is relatively strong, so that the algorithm effect is not influenced by properly improving the sampling step of convolution operation; first, theStrip original data and the corresponding +.>The convolution calculation formula of the strip convolution kernel is as follows:
（1）
wherein:arepresenting the raw data;ConvOutputrepresenting a convolution output;wrepresenting a convolution kernel;nrepresenting the number of bits of the convolved output;la sampling stride representing a convolution operation;mis the length;lengthrepresenting a length calculation formula.
Inputting feature numberskAnd output characteristic numbernSampling stride for convolution operationslThe corresponding relation is as follows:
（2）
identifying internal waves through ocean profile temperature data of multiple depth layers, wherein the temperature change trend of adjacent depth layers is similar when the internal waves arrive; for this problem, in the present embodiment, the effectiveness of the spatial feature is improved by designing an appropriate number of convolution kernels. From equation (1), a convolution output can be obtainedConvOutputAnd original dataaThe corresponding relation, and the characteristic extraction network output is obtained by averaging the convolution output, as shown in a formula (3):
（3）
wherein,output FE extracting network output of representative features;Nthe number of layers representing the original data;Mrepresenting the number of groups of convolution kernels.
The formula (3) can show that the spatial dimension of the feature extraction output is irrelevant to the spatial dimension of the original data relative to the number of the convolution kernels, so that the embodiment improves the effectiveness of the spatial features by testing the number of the convolution kernels with different numbers.
Optionally, using edge computing techniques to migrate the artificial intelligence algorithm-based predictive model to the float side includes: first, ports for receiving sensor data, transmitting predictions, and transmitting recognition results are extended on the basis of a core computing unit, which can be understood as a computing board, according to the data transmission requirements. And secondly, converting a prediction model under the PyTorch architecture into an open neural network exchange (Open Neural Network Exchange, ONNX) architecture which is convenient for edge end transplantation, and finally transplanting the prediction model of the open neural network exchange architecture into an expanded computing board to obtain an edge computing board, and testing the edge computing board which completes the prediction model transplantation, wherein the testing comprises prediction, recognition result testing, data transmission testing, energy consumption testing and the like.
In this embodiment, the artificial intelligence algorithm is optionally a Long Short-Term Memory recurrent neural network (LSTM). Specifically, firstly, the input of the artificial intelligence algorithm comprises meteorological parameters such as rainfall, air pressure, wind speed and the like, and ocean parameters such as ocean vertical profile and the like. Second, the artificial intelligence algorithm uses a two-layer LSTM, the first of which receives the input parameter sequence and learns the correlation between the meteorological parameters and the hydrological parameters and the time dependence of the input sequence parameters. The second layer LSTM can further process the output of the first layer LSTM and capture the information in the sequence in a deeper manner; by using two layers of LSTM, correlated phenomena based on meteorological parameters and marine parameters can be better modeled and predicted. The addition of a layer of Dropout then helps reduce the excessive dependence of the model on the input data, thereby improving the generalization ability of the model. And finally, connecting an LSTM layer, integrating the Dropout output data, further processing the upper output layer, capturing the information in the sequence in a deeper mode, and finally connecting the prediction module and the output for one layer of full connection, wherein the output is weather parameters such as rainfall, air pressure, wind speed and the like at the next moment, and ocean parameters such as ocean vertical profile and the like, such as underwater temperature profile data. In this embodiment, the neural network identification model is a convolutional neural network, and optionally, the convolutional neural network is input with time series data such as rainfall, air pressure, wind speed, etc. for identifying typhoons, or the convolutional neural network is input with time series data of underwater temperature profile data for identifying internal waves; the convolutional neural network comprises a feature extraction module and a feature recognition module, firstly, the input is subjected to feature extraction through the feature extraction module, the feature extraction module comprises a layer of convolutional neural network, and the optimal feature extraction effect is obtained by adjusting the convolutional step length and the number of convolutional kernels of the convolutional operation. And the characteristic recognition module comprises two layers of fully-connected neural networks, and the function of the last two layers of fully-connected neural networks is to linearly combine the characteristics of the previous hidden layers so as to generate a final output result. And outputting a recognition result of whether typhoons or internal waves exist.
The buoy end in the ocean is limited in electric energy storage and acquisition capacity, the electric energy acquisition capacity is greatly influenced by the ocean environment, equipment on the buoy end needs to be subjected to continuous high-frequency sampling in order to ensure the real-time performance, accuracy and the like of observation, the equipment is fast in electric energy loss in the long-time high-frequency sampling process, the electric energy storage and acquisition capacity of the buoy end in the ocean cannot meet the electric energy requirement under the long-time high-frequency sampling, and if the electric energy can not be timely supplemented, the problem of insufficient electric energy can be caused, and normal observation is influenced. In view of the above problems, as shown in fig. 2, in this embodiment, the buoy end is controlled to sample at a first sampling frequency by a first number of sensors to obtain sampling information;
judging whether the sampling information has internal waves or typhoons, if so, controlling the buoy end to sample at a second sampling frequency by using a second number of sensors, otherwise, continuing to sample at a first sampling frequency by using a first number of sensors; wherein the second number is greater than the first number and the second sampling frequency is higher than the first sampling frequency;
and grading the sampling information obtained by sampling at the second sampling frequency.
It can be understood that the data used in predicting typhoons are the atmospheric wind speed, the rainfall and the air pressure, that is to say, when judging that the sampling information appears in typhoons, the sensor for collecting the atmospheric wind speed, the rainfall and the air pressure data is controlled to sample at the second sampling frequency. The data used when the wind is predicted is underwater temperature profile data, namely when the internal wave appears in the sampling information, the sensor for collecting the underwater temperature profile data is controlled to sample at the second sampling frequency. For different sensors, the specific values of the corresponding second sampling frequency and the first sampling frequency are different, and the second number is required to be larger than the first number.
Specifically, before no internal wave or typhoon signal appears, a first sampling frequency with a lower frequency is adopted for sampling, the energy consumption is saved in a non-encryption observation mode, after the internal wave or typhoon signal appears, a second sampling frequency with a higher frequency is adopted for sampling, the accuracy grade of sampling information is divided in an encryption observation mode, and the energy consumption is reduced on the basis of ensuring the observation quality by adopting a mode of combining non-encryption observation and encryption observation, so that the problem that the electric energy demand of the buoy end in the ocean under long-time high-frequency sampling cannot be met is solved.
In order to further avoid the loss of electric energy in the encrypted observation state, in this embodiment, after the time for sampling at the second sampling frequency by the second number of sensors reaches the preset time, the buoy end is controlled to sample at the first sampling frequency by the first number of sensors. It can be understood that the time of existence of the internal wave and typhoon sampling signals is different, and when aiming at different sampling signals, the preset time is adjusted, so that the data volume of the sampling signals in the encrypted observation state is ensured, and the observation effect is ensured.
The method for transmitting sensor data to a local base station and then observing by using a satellite is characterized in that the satellite transmission is needed, the acquisition frequency of the sensor is low, a part of information can be lost, the satellite transmission speed is limited, and the real-time property of the observation can not be ensured; the buoy end local identification method is adopted, the key ocean phenomenon is identified through the change range of the sensor data, and the method has certain subjectivity and poor identification effect; in order to solve the above problems, in this embodiment, a buoy-end-based local recognition method is adopted, and in judging whether the sampled information has internal waves or typhoons or not, and in grading the sampled information obtained by sampling at the second sampling frequency, a trained convolutional neural network is adopted to achieve the purpose, so that the observation effect of the ocean phenomenon is improved, and the problem of poor recognition effect caused by subjectivity is solved. Alternatively, as shown in fig. 2, the convolutional neural network employs a simple one-dimensional convolutional neural network; specifically, the convolutional neural network is trained by using historical profile temperature data; predicting whether an internal wave exists according to the time point profile temperature data acquired by the sensor and the trained convolutional neural network; training the convolutional neural network by using the historical atmospheric wind speed and rainfall; and predicting whether typhoons exist according to the atmospheric wind speed and the rainfall obtained by the sensor and the trained convolutional neural network.
The internal wave and typhoon have certain difference in passing time, the internal wave passing time is generally 20-30 minutes, and the typhoon passing time is generally 2-3 days or even longer; thus, observations can be made in different ways for different marine phenomena. Because the internal wave passing time is relatively short, the unencrypted observation cannot be carried out in a mode of reducing the sampling frequency in the unencrypted stage, a plurality of temperature sensors can be arranged at different depth positions of an observation area, and the temperature sensors at different depths are controlled to alternately sample at the second sampling frequency. Specifically, a plurality of temperature sensors with relatively large influence on internal wave observation are selected through historical acquisition data, and a plurality of temperature sensors with different prevention depths are alternately sampled at high frequency, so that the problem that the internal wave passing time is relatively short is solved, and the power consumption problem of the sensor is also solved.
Example 2:
the embodiment provides a marine phenomenon observation system based on buoy end local identification, which comprises:
a data acquisition module configured to: acquiring ocean atmospheric wind speed history information, rainfall history information, air pressure history information and underwater temperature profile data history information;
an artificial intelligence algorithm prediction module configured to: obtaining a prediction result of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature profile data by using the obtained atmospheric wind speed history information, the rainfall history information, the air pressure history information and the underwater temperature profile data history information and a preset prediction model based on an artificial intelligence algorithm; transplanting a prediction model based on an artificial intelligence algorithm to a buoy end by using an edge computing technology;
an identification module configured to: and obtaining a marine phenomenon observation result according to the prediction results of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature and a preset neural network recognition model.
The working method of the system is the same as the marine phenomenon observation method based on the local identification of the buoy end in embodiment 1, and will not be described here again.
Example 3:
the present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the marine phenomenon observation method based on local identification of a float end described in embodiment 1.
Example 4:
the present embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the steps of the marine phenomenon observation method based on local identification of a buoy end described in embodiment 1 are implemented when the processor executes the program.
The above description is only a preferred embodiment of the present embodiment, and is not intended to limit the present embodiment, and various modifications and variations can be made to the present embodiment by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present embodiment should be included in the protection scope of the present embodiment.
Claims (10)
1. The marine phenomenon observation method based on the buoy end local identification is characterized by comprising the following steps of:
acquiring ocean atmospheric wind speed history information, rainfall history information, air pressure history information and underwater temperature profile data history information;
obtaining a prediction result of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature profile data by using the obtained atmospheric wind speed history information, the rainfall history information, the air pressure history information and the underwater temperature profile data history information and a preset prediction model based on an artificial intelligence algorithm; transplanting a prediction model based on an artificial intelligence algorithm to a buoy end by using an edge computing technology;
and obtaining a marine phenomenon observation result according to the prediction results of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature and a preset neural network recognition model.
2. The marine phenomenon observation method based on buoy end local recognition according to claim 1, wherein the prediction model based on artificial intelligence algorithm is a long-term and short-term memory recurrent neural network; transplanting the artificial intelligence algorithm-based predictive model to the float end using edge computing techniques includes: firstly, expanding a port for receiving sensor data, transmitting prediction and transmitting identification results on the basis of a calculation unit according to the data transmission requirement; secondly, converting the prediction model into an open neural network switching architecture which is convenient for edge end transplantation; and finally, transplanting the prediction model of the open neural network switching architecture into the expanded computing unit.
3. The marine observation method based on the local identification of the buoy end as claimed in claim 1, wherein the buoy end is controlled to sample at a first sampling frequency with a first number of sensors to obtain sampling information;
judging whether the sampling information has internal waves or typhoons, if so, controlling the buoy end to sample at a second sampling frequency by using a second number of sensors, otherwise, continuing to sample at a first sampling frequency by using a first number of sensors; wherein the second number is greater than the first number and the second sampling frequency is higher than the first sampling frequency;
and grading the sampling information obtained by sampling at the second sampling frequency.
4. A method of observing marine phenomena based on local identification of a float end as claimed in claim 3 wherein the float end is controlled to sample at the first sampling frequency with the first number of sensors after the time for sampling at the second sampling frequency with the second number of sensors reaches a predetermined time.
5. The marine phenomenon observation method based on buoy end local identification as claimed in claim 3, wherein judging whether the sampling information has internal waves or typhoons or not and grading the sampling information obtained by sampling at the second sampling frequency are realized by adopting a preset neural network identification model;
training the convolutional neural network by using the atmospheric wind speed history information, the rainfall history information and the air pressure history information; predicting whether typhoons exist according to the atmospheric wind speed, the rainfall and the air pressure obtained by the sensor and the trained convolutional neural network;
meanwhile, predicting the convolutional neural network by using historical information of the underwater temperature profile data; and predicting whether an internal wave exists according to the underwater temperature profile data acquired by the sensor and the trained convolutional neural network.
6. A method of observing marine phenomena based on local identification of the float end as claimed in claim 3 wherein a plurality of sensors are provided at different depth locations of the observation area; a plurality of sensors of different depths are controlled to alternately sample at a second sampling frequency.
7. A method of observing marine phenomena based on local identification of a float end as claimed in claim 6 wherein a plurality of temperature sensors are provided at different depth locations of the observation area.
8. A marine phenomenon observation system based on local identification of a buoy end, comprising:
a data acquisition module configured to: acquiring ocean atmospheric wind speed history information, rainfall history information, air pressure history information and underwater temperature profile data history information;
an artificial intelligence algorithm prediction module configured to: obtaining a prediction result of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature profile data by using the obtained atmospheric wind speed history information, the rainfall history information, the air pressure history information and the underwater temperature profile data history information and a preset prediction model based on an artificial intelligence algorithm; transplanting a prediction model based on an artificial intelligence algorithm to a buoy end by using an edge computing technology;
an identification module configured to: and obtaining a marine phenomenon observation result according to the prediction results of the atmospheric wind speed, the rainfall, the air pressure and the underwater temperature and a preset neural network recognition model.
9. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of a method of marine phenomenon observation based on local identification of a float end as claimed in any one of claims 1 to 7.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the buoy end local identification based marine phenomenon observation method of any one of claims 1-7 when the program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311411493.3A CN117168545A (en) | 2023-10-30 | 2023-10-30 | Ocean phenomenon observation method and system based on buoy end local identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311411493.3A CN117168545A (en) | 2023-10-30 | 2023-10-30 | Ocean phenomenon observation method and system based on buoy end local identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117168545A true CN117168545A (en) | 2023-12-05 |
Family
ID=88937753
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311411493.3A Pending CN117168545A (en) | 2023-10-30 | 2023-10-30 | Ocean phenomenon observation method and system based on buoy end local identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117168545A (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103123269A (en) * | 2012-12-19 | 2013-05-29 | 中国海洋大学 | Real-time monitoring apparatus for ocean internal waves |
| CN105867232A (en) * | 2016-04-19 | 2016-08-17 | 沈阳大学 | Soil environment monitoring system based on wireless sensor network and covering control method |
| CN206075474U (en) * | 2016-08-30 | 2017-04-05 | 云南云鹰消防设施维护管理有限公司 | A kind of Intelligent fire-fighting detection means |
| CN206618872U (en) * | 2017-04-19 | 2017-11-07 | 国家海洋局第二海洋研究所 | A kind of typhoon real-time monitoring system and real-time system for monitoring and pre-warning |
| CN109974670A (en) * | 2019-04-17 | 2019-07-05 | 清华大学深圳研究生院 | A kind of interior wave detecting method based on drifting buoy |
| CN110837137A (en) * | 2019-11-07 | 2020-02-25 | 刘健华 | Typhoon prediction alarm method |
| CN111401297A (en) * | 2020-04-03 | 2020-07-10 | 天津理工大学 | Triphibian robot target recognition system and method based on edge calculation and neural network |
| CN111913237A (en) * | 2020-08-10 | 2020-11-10 | 中国海洋大学 | Large-scale buoy oceanographic monitoring system of intermediate latitude |
| CN112099110A (en) * | 2020-09-17 | 2020-12-18 | 中国科学院海洋研究所 | Ocean internal wave forecasting method based on machine learning and remote sensing data |
| CN114282574A (en) * | 2021-12-16 | 2022-04-05 | 中国人民解放军海军潜艇学院 | Inversion method and system for ocean internal wave characteristic parameters |
| CN115107931A (en) * | 2021-06-22 | 2022-09-27 | 青岛海洋科学与技术国家实验室发展中心 | Offshore buoy system |
| CN115219280A (en) * | 2022-07-20 | 2022-10-21 | 上海铃恒环境科技有限公司 | Water quality detection device and detection method |
| CN115618988A (en) * | 2021-07-13 | 2023-01-17 | 中国科学院沈阳自动化研究所 | Three-dimensional space-time field joint prediction method for seawater temperature, salinity and flow velocity based on deep neural network |
| CN116147883A (en) * | 2021-11-15 | 2023-05-23 | 上海海洋大学 | Ocean internal wave detection method based on buoy observation data |
| CN116625327A (en) * | 2023-07-24 | 2023-08-22 | 海南智慧海事科技有限公司 | Big data driven marine environment monitoring and early warning method |
| CN116704688A (en) * | 2023-08-04 | 2023-09-05 | 自然资源部第一海洋研究所 | Ocean buoy passive defense method and system |
-
2023
- 2023-10-30 CN CN202311411493.3A patent/CN117168545A/en active Pending
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103123269A (en) * | 2012-12-19 | 2013-05-29 | 中国海洋大学 | Real-time monitoring apparatus for ocean internal waves |
| CN105867232A (en) * | 2016-04-19 | 2016-08-17 | 沈阳大学 | Soil environment monitoring system based on wireless sensor network and covering control method |
| CN206075474U (en) * | 2016-08-30 | 2017-04-05 | 云南云鹰消防设施维护管理有限公司 | A kind of Intelligent fire-fighting detection means |
| CN206618872U (en) * | 2017-04-19 | 2017-11-07 | 国家海洋局第二海洋研究所 | A kind of typhoon real-time monitoring system and real-time system for monitoring and pre-warning |
| CN109974670A (en) * | 2019-04-17 | 2019-07-05 | 清华大学深圳研究生院 | A kind of interior wave detecting method based on drifting buoy |
| CN110837137A (en) * | 2019-11-07 | 2020-02-25 | 刘健华 | Typhoon prediction alarm method |
| CN111401297A (en) * | 2020-04-03 | 2020-07-10 | 天津理工大学 | Triphibian robot target recognition system and method based on edge calculation and neural network |
| CN111913237A (en) * | 2020-08-10 | 2020-11-10 | 中国海洋大学 | Large-scale buoy oceanographic monitoring system of intermediate latitude |
| CN112099110A (en) * | 2020-09-17 | 2020-12-18 | 中国科学院海洋研究所 | Ocean internal wave forecasting method based on machine learning and remote sensing data |
| CN115107931A (en) * | 2021-06-22 | 2022-09-27 | 青岛海洋科学与技术国家实验室发展中心 | Offshore buoy system |
| CN115618988A (en) * | 2021-07-13 | 2023-01-17 | 中国科学院沈阳自动化研究所 | Three-dimensional space-time field joint prediction method for seawater temperature, salinity and flow velocity based on deep neural network |
| CN116147883A (en) * | 2021-11-15 | 2023-05-23 | 上海海洋大学 | Ocean internal wave detection method based on buoy observation data |
| CN114282574A (en) * | 2021-12-16 | 2022-04-05 | 中国人民解放军海军潜艇学院 | Inversion method and system for ocean internal wave characteristic parameters |
| CN115219280A (en) * | 2022-07-20 | 2022-10-21 | 上海铃恒环境科技有限公司 | Water quality detection device and detection method |
| CN116625327A (en) * | 2023-07-24 | 2023-08-22 | 海南智慧海事科技有限公司 | Big data driven marine environment monitoring and early warning method |
| CN116704688A (en) * | 2023-08-04 | 2023-09-05 | 自然资源部第一海洋研究所 | Ocean buoy passive defense method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Marine wireless big data: Efficient transmission, related applications, and challenges | |
| CN102034239B (en) | Local gray abrupt change-based infrared small target detection method | |
| CN112615804B (en) | Short burst underwater acoustic communication signal modulation identification method based on deep learning | |
| CN111797712A (en) | Remote sensing image cloud and cloud shadow detection method based on multi-scale feature fusion network | |
| CN112990316B (en) | Hyperspectral remote sensing image classification method and system based on multi-saliency feature fusion | |
| CN113269077A (en) | Underwater acoustic communication signal modulation mode identification method based on improved gating network and residual error network | |
| CN114239749B (en) | Modulation identification method based on residual shrinkage and two-way long-short-term memory network | |
| CN116599602B (en) | Low-energy-consumption long-distance sonar communication system for ocean monitoring | |
| CN114724019A (en) | Remote sensing image sea ice intelligent monitoring method based on wavelet transformation and separable convolution semantic segmentation | |
| CN111768372B (en) | Method and system for detecting foreign matters in cavity of GIS (gas insulated switchgear) | |
| CN115062527A (en) | Geostationary satellite sea temperature inversion method and system based on deep learning | |
| CN111222640A (en) | Signal recognition convolutional neural network convolutional kernel partition pruning method | |
| CN111507299A (en) | Method for identifying STBC (space time Block coding) signal on frequency domain by using convolutional neural network | |
| CN117168545A (en) | Ocean phenomenon observation method and system based on buoy end local identification | |
| CN114861985A (en) | Ice damage early warning method, system and medium based on open neural network exchange algorithm | |
| CN112463877B (en) | Compressed sensing observation and reconstruction method for three-dimensional marine environment monitoring data | |
| CN108133387B (en) | Multi-label K nearest neighbor algorithm based on soft information | |
| CN116667966A (en) | Intelligent interference model rewarding poisoning defense and training method and system | |
| CN111208586A (en) | Weather forecasting method and system based on mesoscale sea air coupling mode | |
| CN113034445B (en) | Multi-scale connection image defogging algorithm based on UNet3+ | |
| Li et al. | Research on modulation recognition of underwater acoustic communication signal based on deep learning | |
| CN110211124B (en) | Infrared imaging frozen lake detection method based on MobileNet V2 | |
| Yin et al. | Few-Shot Domain Adaption-Based Specific Emitter Identification Under Varying Modulation | |
| Randell et al. | Modelling covariate effects in extremes of storm severity on the Australian North West Shelf | |
| CN106249241A (en) | A kind of self-adapting clutter power statistic algorithm |
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 |