Marine disaster monitoring and early warning system

By integrating various monitoring methods and modules, multi-dimensional and all-round real-time monitoring of the marine environment is achieved, solving the problem of untimely early warning of marine disasters and improving monitoring and early warning capabilities and emergency response efficiency.

CN119049227AInactive Publication Date: 2024-11-29NATIONAL MARINE ENVIRONMENTAL MONITORING CENTRE

Patent Information

Application Number
CN202411535454.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-31
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure CN119049227A_ABST
    Figure CN119049227A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of disaster early warning, and particularly discloses a marine disaster monitoring and early warning system which comprises a data acquisition module, a data processing module, a data analysis module, an early warning issuing module, an information sharing module and a system management module. According to the invention, by integrating the buoy with the sensor network, the remote sensing monitoring network and the video monitoring network, multi-dimensional and omnibearing real-time monitoring of the marine environment is realized, basic sea condition data such as temperature, salinity, flow velocity, wind speed, wind direction, wave height and the like are covered, and large-range sea area information is obtained through satellite and unmanned aerial vehicle technologies. And the video monitoring equipment is used for capturing sea surface conditions in real time, so that comprehensive and reliable data support is provided for accurate early warning and timely response of marine disasters, and the monitoring efficiency and the disaster response capability are remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of disaster early warning, and specifically discloses a marine disaster monitoring and early warning system. Background Art

[0002] Marine disasters such as tsunamis, storm surges, and wave erosion pose a huge threat to human society and the marine ecosystem. In order to reduce the losses caused by these disasters, it is necessary to establish an effective monitoring and early warning system. Traditional marine environmental monitoring methods often have problems such as limited monitoring range and untimely data acquisition, which makes it difficult to meet the needs of accurate early warning and timely response to marine disasters.

[0003] With the continuous development of sensor technology, remote sensing technology and video monitoring technology, these technologies have been able to achieve multi-dimensional, all-round real-time monitoring of the marine environment. Buoys and sensor networks can measure key parameters such as temperature, salinity, and flow rate; remote sensing monitoring technology obtains large-scale, high-resolution marine disaster monitoring data through satellites, aircraft and other platforms; video monitoring equipment can capture sea surface conditions in real time and provide intuitive information for marine disaster warning. However, there is currently no marine disaster monitoring and early warning system that can integrate the above-mentioned multiple monitoring methods to achieve comprehensive and real-time monitoring of the marine environment. Summary of the invention

[0004] In view of the above deficiencies in the prior art, the present invention provides a marine disaster monitoring and early warning system that can integrate multiple monitoring methods to achieve comprehensive and real-time monitoring of the marine environment.

[0005] The technical solution adopted by the present invention to solve its technical problem is: The marine disaster monitoring and early warning system includes: The data acquisition module is used to collect sea condition data in real time, wherein the collected sea condition data includes sea surface temperature, seabed temperature, seabed salinity, seabed current velocity, wind speed, wind direction, wave height and sea ice distribution; The data processing module is used to pre-process the raw data collected by the data acquisition module to improve the data quality. The data pre-processing methods used include data cleaning, data verification and data formatting; The data analysis module is used to analyze the pre-processed data using data analysis methods and algorithms, identify disaster signs, and predict marine disasters based on historical data and real-time data; The warning release module is used to automatically generate disaster warning information based on data analysis results and prediction results, and release the disaster warning information through the set information release channels; The information sharing module is used to share marine disaster monitoring data with other relevant institutions, promote the exchange of information, and establish cooperation mechanisms with other emergency management departments and scientific research institutions to jointly respond to marine disasters.

[0006] The data analysis module uses data analysis methods and algorithms to analyze the pre-processed data and identify disaster signs, including the following steps: S21: Data loading and integration: loading preprocessed data from the data storage medium, integrating data from different sources and in different formats to form a unified data set, and dividing the data set to obtain a training data set and a validation data set; S22: Feature extraction and selection: extract key features that can reflect changes in the marine environment from the original data, and select features that have a significant impact on the prediction results according to the needs of disaster prediction; S23: Model construction and training: Based on historical data and real-time data, select appropriate machine learning algorithms and statistical models to build feature extraction models, train the models using training data sets, and adjust model parameters to improve prediction accuracy; S24: Model evaluation and optimization: Use the validation data set to evaluate the trained model, calculate the prediction accuracy and recall rate, and optimize the model based on the accuracy and recall rate indicators; S25: Use the trained model to process and analyze real-time data to identify disaster signs.

[0007] When the data analysis module selects appropriate machine learning algorithms and statistical models to build a prediction model, the machine learning algorithms and statistical models used include: A21: Linear regression model is used to predict the linear relationship between one or more independent variables and dependent variables. Its principle formula is: ; Among them, Y is the dependent variable, X1, X2 and X n are the independent variables, β0, β1, β2, and β n is the regression coefficient, ∈ is the error term; A22: Support vector machine, used for classification and regression problems, separates data of different categories by finding the optimal hyperplane. Its principle formula is: ; Where w is the normal vector of the hyperplane, C is the penalty parameter, ξ i is the slack variable; A23: The neural network model simulates the connection and transmission mode of human brain neurons to perform nonlinear transformation and classification on data. Its principle formula is: ; Among them, f is the activation function, w i is the weight, x i is the input data and b is the bias term.

[0008] Preferably, the data acquisition module includes: A11: Buoy and sensor network, including buoys and seabed sensors, used to collect real-time sea condition data including temperature, salinity, current velocity, wind speed, wind direction and wave height; A12: Remote sensing monitoring network, which uses satellite and drone technology to obtain large-scale marine information, including sea surface temperature, sea ice distribution and wave conditions; A13: Video surveillance network is used to utilize video surveillance equipment set up in key sea areas to capture sea surface conditions in real time and assist in determining disaster situations.

[0009] Preferably, the data processing module pre-processes the raw data collected by the data collection module to improve the data quality, including the following steps: S11: Data reception: the data processing module first receives the raw data from the data acquisition module; S12: Data cleaning, including removing invalid data, removing duplicate data and processing abnormal data; S13: Data verification: Verify the accuracy and rationality of the data, and verify whether the format and unit of the data are consistent; S14: Data formatting, converting the data into a unified format to facilitate subsequent data analysis and processing, including data standardization and data normalization. The principle formula for data standardization is: z = (x - μ) / σ; Among them, z is the data after standardization, x is the original data, μ is the mean, and σ is the standard deviation; The principle formula for normalizing data is: x'=(x-min) / (max-min); Among them, x' is the data after normalization, x is the original data, min and max are the minimum and maximum values ​​in the data set respectively; S15: Data storage: storing the preprocessed data in a suitable data storage medium, wherein the data storage medium includes a database, a file system, and a cloud storage.

[0010] Preferably, the data analysis module includes the following steps when predicting marine disasters based on historical data and real-time data: S31: Feature dataset construction, integrating the extracted disaster sign features into a feature dataset; S32: Select a prediction model, select a suitable prediction model from a pre-trained model library according to the disaster type and prediction requirements, load the selected prediction model, and initialize it; S33: Data preprocessing and feature engineering: normalize or standardize the feature data set to eliminate the dimensional differences between different features and improve the prediction accuracy of the model. Use correlation analysis and principal component analysis to perform feature selection and dimensionality reduction on the extracted feature data. S34: Use historical data to train the prediction model and tune the model parameters through cross-validation and grid search methods; S35: Disaster prediction and result analysis. First, the disaster sign features collected in real time are input into the trained prediction model. The model outputs the probability or level of disaster occurrence based on the input feature data. Finally, the prediction results are analyzed and interpreted.

[0011] Preferably, the principle algorithms used by the data analysis module when performing disaster prediction include: A31: Storm surge water increase formula, the principle formula is: ; Among them, Δh represents the storm surge, α represents the first weight, p represents the storm intensity, β represents the second weight, V represents the wind speed, γ represents the third weight, S represents the moving speed, θ represents the path angle, and d represents the geographical feature; A32: Tsunami wave propagation formula, its principle formula is: ; Among them, v is the propagation speed of the tsunami wave, g is the acceleration of gravity, h is the depth of the ocean, and λ is the wavelength of the tsunami wave.

[0012] As a preferred option, the warning release module also has the following functions: A41: According to the early warning information, the corresponding emergency plan is activated, which includes the personnel evacuation plan and material deployment plan; A42: Provide real-time disaster information and disaster prediction information to rescue teams to ensure the efficiency and safety of rescue operations; A43: After a disaster occurs, the disaster losses are assessed.

[0013] Preferably, when the warning release module starts the corresponding emergency plan according to the warning information, the following steps are included: S41: The warning release module first receives the disaster warning information from the data analysis module, including the disaster type, expected occurrence time and impact range; S42: emergency plan matching, matching the corresponding emergency plan from the plan library according to the specific content of the warning information; S43: Emergency resource allocation. Before launching the emergency plan, the required emergency resources need to be allocated, including rescue teams, material reserves and transportation tools; S44: Emergency plan initiation and response, through the set emergency plan initiation process, the emergency plan initiation phase is carried out step by step.

[0014] Preferably, a system management module is further included, and the system management module is used to remotely monitor and manage the hardware devices. When the system management module remotely monitors and manages the hardware devices, the following steps are included: S51: The hardware device periodically sends a verification data packet to the hardware device according to a set time interval; S52: After receiving the verification data packet, the system management module receives and processes the verification data packet sent by the hardware device; S53: the system management module records the time when the verification data packet is received, and uses the time difference between the time when the verification data packet is received and the time when it is sent to determine the connection status; S54: When the connection status is poor or the connection is disconnected, the system management module records it and sends it to the staff to notify the staff to perform inspection and maintenance.

[0015] The beneficial effects of the present invention are: 1. By integrating modules such as data collection, preprocessing, analysis, warning release and information sharing, it realizes comprehensive real-time collection and efficient processing of sea condition data, can accurately identify disaster signs and predict marine disasters, release warning information in a timely manner, and promote information sharing and cooperation, effectively improving the monitoring and warning capabilities of marine disasters and the efficiency of emergency response; 2. By integrating buoys and sensor networks, remote sensing monitoring networks and video surveillance networks, multi-dimensional and all-round real-time monitoring of the marine environment is achieved. It not only covers basic sea condition data such as temperature, salinity, current velocity, wind speed, wind direction, and wave height, but also obtains a wide range of sea area information through satellite and drone technology, and uses video surveillance equipment to capture sea surface conditions in real time, providing comprehensive and reliable data support for accurate early warning and timely response to marine disasters, significantly improving monitoring efficiency and disaster response capabilities; 3. By loading and integrating pre-processed data, extracting key features and building an optimized machine learning model, we can accurately capture the characteristics of marine environmental changes and effectively identify disaster signs. By dividing the training and validation data sets, feature selection and model training, evaluation and optimization, we can not only improve the prediction accuracy, but also ensure the stability and reliability of the model, providing a scientific basis and technical support for the timely warning and accurate prediction of marine disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the framework of the present invention; Figure 2 It is a schematic diagram of the architecture of the data acquisition module in the present invention; Figure 3 Schematic diagram of part of the workflow of the data analysis module in the present invention. DETAILED DESCRIPTION

[0017] The present invention will be described and illustrated in detail below in conjunction with the accompanying drawings.

[0018] Example 1 like Figure 1-3 As shown, the marine disaster monitoring and early warning system includes: The data acquisition module is used to collect sea condition data in real time, wherein the collected sea condition data includes sea surface temperature, seabed temperature, seabed salinity, seabed current velocity, wind speed, wind direction, wave height and sea ice distribution; The data processing module is used to pre-process the raw data collected by the data acquisition module to improve the data quality. The data pre-processing methods used include data cleaning, data verification and data formatting; The data analysis module is used to analyze the pre-processed data using data analysis methods and algorithms, identify disaster signs, and predict marine disasters based on historical data and real-time data; The warning release module is used to automatically generate disaster warning information based on data analysis results and prediction results, and release the disaster warning information through the set information release channels; The information sharing module is used to share marine disaster monitoring data with other relevant institutions, promote the exchange of information, and establish cooperation mechanisms with other emergency management departments and scientific research institutions to jointly respond to marine disasters.

[0019] As described above, the present invention realizes comprehensive real-time collection and efficient processing of sea condition data by integrating modules such as data collection, preprocessing, analysis, warning release and information sharing. It can accurately identify disaster signs and predict marine disasters, release warning information in a timely manner, and promote information sharing and cooperation, effectively improving the monitoring and early warning capabilities of marine disasters and the efficiency of emergency response.

[0020] The data analysis module uses data analysis methods and algorithms to analyze the pre-processed data and identify disaster signs, including the following steps: S21: Data loading and integration: loading preprocessed data from the data storage medium, integrating data from different sources and in different formats to form a unified data set, and dividing the data set to obtain a training data set and a validation data set; S22: Feature extraction and selection: extract key features that can reflect changes in the marine environment from the original data, and select features that have a significant impact on the prediction results according to the needs of disaster prediction; S23: Model construction and training: Based on historical data and real-time data, select appropriate machine learning algorithms and statistical models to build feature extraction models, train the models using training data sets, and adjust model parameters to improve prediction accuracy; S24: Model evaluation and optimization: Use the validation data set to evaluate the trained model, calculate the prediction accuracy and recall rate, and optimize the model based on the accuracy and recall rate indicators; S25: Use the trained model to process and analyze real-time data to identify disaster signs.

[0021] As described in the above steps S21-S24, the present invention achieves accurate capture of the characteristics of marine environmental changes and effective identification of disaster signs by loading and integrating preprocessed data, extracting key features and constructing an optimized machine learning model. By dividing the training and verification data sets, feature selection and model training, evaluation and optimization, etc., not only the prediction accuracy is improved, but also the stability and reliability of the model are ensured, providing a scientific basis and technical support for timely warning and accurate prediction of marine disasters.

[0022] When the data analysis module selects appropriate machine learning algorithms and statistical models to build a prediction model, the machine learning algorithms and statistical models used include: A21: Linear regression model is used to predict the linear relationship between one or more independent variables and dependent variables. Its principle formula is: ; Among them, Y is the dependent variable, X1, X2 and X n are the independent variables, β0, β1, β2, and β n is the regression coefficient, ∈ is the error term; A22: Support vector machine, used for classification and regression problems, separates data of different categories by finding the optimal hyperplane. Its principle formula is: ; Where w is the normal vector of the hyperplane, C is the penalty parameter, ξ i is the slack variable; A23: The neural network model simulates the connection and transmission mode of human brain neurons to perform nonlinear transformation and classification on data. Its principle formula is: ; Among them, f is the activation function, Wi is the weight, Xi is the input data, and b is the bias term.

[0023] As described above, the data analysis module in the present invention integrates the linear relationship prediction ability of the linear regression model, the classification and regression accuracy of the support vector machine, and the nonlinear transformation and classification advantages of the neural network model when selecting machine learning algorithms and statistical models. The application of these algorithms and models not only improves the accuracy of marine disaster prediction, but also enhances the model's adaptability and generalization ability to different types of sea condition data, providing powerful data analysis and prediction support for marine disaster monitoring and early warning systems.

[0024] The data acquisition module includes: A11: Buoy and sensor network, including buoys and seabed sensors, used to collect real-time sea condition data including temperature, salinity, current velocity, wind speed, wind direction and wave height; A12: Remote sensing monitoring network, which uses satellite and drone technology to obtain large-scale marine information, including sea surface temperature, sea ice distribution and wave conditions; A13: Video surveillance network is used to utilize video surveillance equipment set up in key sea areas to capture sea surface conditions in real time and assist in determining disaster situations.

[0025] As described above, the present invention realizes multi-dimensional, all-round real-time monitoring of the marine environment by integrating buoys and sensor networks, remote sensing monitoring networks and video monitoring networks. It not only covers basic sea condition data such as temperature, salinity, flow velocity, wind speed, wind direction, wave height, etc., but also obtains a wide range of sea area information through satellite and drone technology, and uses video monitoring equipment to capture sea surface conditions in real time, providing comprehensive and reliable data support for accurate early warning and timely response to marine disasters, and significantly improving monitoring efficiency and disaster response capabilities.

[0026] The data processing module pre-processes the raw data collected by the data acquisition module to improve the data quality, including the following steps: S11: Data reception: the data processing module first receives the raw data from the data acquisition module; S12: Data cleaning, including removing invalid data, removing duplicate data and processing abnormal data; S13: Data verification: Verify the accuracy and rationality of the data, and verify whether the format and unit of the data are consistent; S14: Data formatting, converting the data into a unified format to facilitate subsequent data analysis and processing, including data standardization and data normalization. The principle formula for data standardization is: z = (x - μ) / σ; Among them, z is the data after standardization, x is the original data, μ is the mean, and σ is the standard deviation; The principle formula for normalizing data is: x'=(x-min) / (max-min); Among them, x' is the data after normalization, x is the original data, min and max are the minimum and maximum values ​​in the data set respectively; S15: Data storage: storing the preprocessed data in a suitable data storage medium, wherein the data storage medium includes a database, a file system, and a cloud storage.

[0027] As described in the above steps S11-S15, the present invention effectively improves data quality and ensures data accuracy, consistency and availability by receiving raw data and performing a series of pre-processing steps such as data cleaning, verification, formatting and storage. By removing invalid and duplicate data, processing outliers, verifying data accuracy and rationality, and converting data into a unified format through standardization and normalization, a high-quality data foundation is provided for subsequent data analysis and disaster prediction. At the same time, a variety of data storage media options, such as databases, file systems and cloud storage, ensure the reliability and security of data, and provide strong support for the stable operation of marine disaster monitoring and early warning systems.

[0028] The data analysis module includes the following steps when predicting marine disasters based on historical data and real-time data: S31: Feature dataset construction, integrating the extracted disaster sign features into a feature dataset; S32: Select a prediction model, select a suitable prediction model from a pre-trained model library according to the disaster type and prediction requirements, load the selected prediction model, and initialize it; S33: Data preprocessing and feature engineering: normalize or standardize the feature data set to eliminate the dimensional differences between different features and improve the prediction accuracy of the model. Use correlation analysis and principal component analysis to perform feature selection and dimensionality reduction on the extracted feature data. S34: Use historical data to train the prediction model and tune the model parameters through cross-validation and grid search methods; S35: Disaster prediction and result analysis. First, the disaster sign features collected in real time are input into the trained prediction model. The model outputs the probability or level of disaster occurrence based on the input feature data. Finally, the prediction results are analyzed and interpreted.

[0029] As described in the above steps S31-S35, the present invention integrates disaster sign features to construct a feature data set, selects a suitable prediction model and initializes it, and then normalizes or standardizes the feature data set to improve the model prediction accuracy, uses historical data to train the model and tune parameters, and finally achieves accurate prediction and result analysis of marine disasters. This technical solution significantly improves the accuracy of disaster prediction and provides scientific and effective decision support for marine disaster monitoring and early warning systems.

[0030] The principle algorithms used by the data analysis module for disaster prediction include: A31: Storm surge water increase formula, the principle formula is: ; Among them, Δh represents the storm surge, α represents the first weight, p represents the storm intensity, β represents the second weight, V represents the wind speed, γ represents the third weight, S represents the moving speed, θ represents the path angle, and d represents the geographical feature; A32: Tsunami wave propagation formula, its principle formula is: ; Among them, v is the propagation speed of the tsunami wave, g is the acceleration of gravity, h is the depth of the ocean, and λ is the wavelength of the tsunami wave.

[0031] As described above, the present invention adopts principle algorithms such as storm surge water increase formula and tsunami wave propagation formula when predicting marine disasters. These algorithms can accurately simulate the physical processes of marine disasters such as storm surges and tsunamis, improve the accuracy and reliability of disaster prediction, and provide a more scientific prediction basis for marine disaster monitoring and early warning systems.

[0032] The warning release module also has the following functions: A41: According to the early warning information, the corresponding emergency plan is activated, which includes the personnel evacuation plan and material deployment plan; A42: Provide real-time disaster information and disaster prediction information to rescue teams to ensure the efficiency and safety of rescue operations; A43: After a disaster occurs, the disaster losses are assessed.

[0033] As described above, the warning release module in the present invention not only initiates emergency plans including personnel evacuation and material deployment according to the warning information, but also provides real-time disaster and forecast information to the rescue team to ensure efficient and safe rescue, and assess the losses after the disaster. This technical solution effectively improves the efficiency of disaster response and the scientific nature of rescue operations, and provides important support for reducing the impact of disasters.

[0034] When the early warning release module starts the corresponding emergency plan according to the early warning information, it includes the following steps: S41: The warning release module first receives the disaster warning information from the data analysis module, including the disaster type, expected occurrence time and impact range; S42: emergency plan matching, matching the corresponding emergency plan from the plan library according to the specific content of the warning information; S43: Emergency resource allocation. Before launching the emergency plan, the required emergency resources need to be allocated, including rescue teams, material reserves and transportation tools; S44: Emergency plan initiation and response, through the set emergency plan initiation process, the emergency plan initiation phase is carried out step by step.

[0035] As described in the above steps S41-S44, the present invention receives disaster warning information, accurately matches emergency plans, efficiently allocates emergency resources, and strictly executes the plan startup process. This technical solution ensures the rapid response and effective implementation of emergency plans, and provides strong organizational guarantees and resource support for responding to marine disasters.

[0036] The system also includes a system management module, which is used to remotely monitor and manage the hardware devices. When the system management module remotely monitors and manages the hardware devices, the system management module includes the following steps: S51: The hardware device periodically sends a verification data packet to the hardware device according to a set time interval; S52: After receiving the verification data packet, the system management module receives and processes the verification data packet sent by the hardware device; S53: the system management module records the time when the verification data packet is received, and uses the time difference between the time when the verification data packet is received and the time when it is sent to determine the connection status; S54: When the connection status is poor or the connection is disconnected, the system management module records it and sends it to the staff to notify the staff to perform inspection and maintenance.

[0037] As described in the above steps S51-S54, the present invention periodically receives and processes verification data packets sent by hardware devices, uses time differences to determine connection status, and promptly records and notifies staff to handle connection problems, thereby achieving remote and effective monitoring and management of hardware devices and ensuring stable operation and timely maintenance of the system.

Claims

1. A marine disaster monitoring and early warning system, characterized in that: Included are: The data acquisition module is used to collect sea condition data in real time, wherein the collected sea condition data includes sea surface temperature, seabed temperature, seabed salinity, seabed current velocity, wind speed, wind direction, wave height and sea ice distribution; The data processing module is used to pre-process the raw data collected by the data acquisition module to improve the data quality. The data pre-processing methods used include data cleaning, data verification and data formatting; The data analysis module is used to analyze the pre-processed data using data analysis methods and algorithms, identify disaster signs, and predict marine disasters based on historical data and real-time data; The warning release module is used to automatically generate disaster warning information based on data analysis results and prediction results, and release the disaster warning information through the set information release channels; The information sharing module is used to share marine disaster monitoring data with other relevant institutions, promote the exchange of information, and establish a cooperation mechanism with other emergency management departments and scientific research institutions to jointly respond to marine disasters; The data analysis module uses data analysis methods and algorithms to analyze the pre-processed data and identify disaster signs, including the following steps: S21: Data loading and integration: loading preprocessed data from the data storage medium, integrating data from different sources and in different formats to form a unified data set, and dividing the data set to obtain a training data set and a validation data set; S22: Feature extraction and selection: extract key features that can reflect changes in the marine environment from the original data, and select features that have a significant impact on the prediction results according to the needs of disaster prediction; S23: Model construction and training: Based on historical data and real-time data, select appropriate machine learning algorithms and statistical models to build feature extraction models, train the models using training data sets, and adjust model parameters to improve prediction accuracy; S24: Model evaluation and optimization: Use the validation data set to evaluate the trained model, calculate the prediction accuracy and recall rate, and optimize the model based on the accuracy and recall rate indicators; S25: Use the trained model to process and analyze real-time data to identify disaster signs; When the data analysis module selects appropriate machine learning algorithms and statistical models to build a prediction model, the machine learning algorithms and statistical models used include: A21: Linear regression model is used to predict the linear relationship between one or more independent variables and dependent variables. Its principle formula is: ; Among them, Y is the dependent variable, X1, X2 and X n are independent variables, β0, β1, β2, and β n is the regression coefficient, ∈ is the error term; A22: Support vector machine, used for classification and regression problems, separates data of different categories by finding the optimal hyperplane. Its principle formula is: ; Where w is the normal vector of the hyperplane, C is the penalty parameter, ξ i is the slack variable; A23: The neural network model simulates the connection and transmission mode of human brain neurons to perform nonlinear transformation and classification on data. Its principle formula is: ; Where f is the activation function, W i is the weight, X i is the input data and b is the bias term.

2. The marine disaster monitoring and early warning system according to claim 1 is characterized in that: The data acquisition module includes: A11: Buoy and sensor network, including buoys and seabed sensors, used to collect real-time sea condition data including temperature, salinity, current velocity, wind speed, wind direction and wave height; A12: Remote sensing monitoring network, which uses satellite and drone technology to obtain large-scale marine information, including sea surface temperature, sea ice distribution and wave conditions; A13: Video surveillance network is used to utilize video surveillance equipment set up in key sea areas to capture sea surface conditions in real time and assist in determining disaster situations.

3. The marine disaster monitoring and early warning system according to claim 1 is characterized in that: The data processing module pre-processes the raw data collected by the data acquisition module to improve the data quality, including the following steps: S11: Data reception: the data processing module first receives the raw data from the data acquisition module; S12: Data cleaning, including removing invalid data, removing duplicate data and processing abnormal data; S13: Data verification: Verify the accuracy and rationality of the data, and verify whether the format and unit of the data are consistent; S14: Data formatting, converting the data into a unified format to facilitate subsequent data analysis and processing, including data standardization and data normalization. The principle formula for data standardization is: z = (x - μ) / σ; Among them, z is the data after standardization, x is the original data, μ is the mean, and σ is the standard deviation; The principle formula for normalizing data is: x'=(x-min) / (max-min); Among them, x' is the data after normalization, x is the original data, min and max are the minimum and maximum values ​​in the data set respectively; S15: Data storage: storing the preprocessed data in a suitable data storage medium, wherein the data storage medium includes a database, a file system, and a cloud storage.

4. The marine disaster monitoring and early warning system according to claim 1 is characterized in that: The data analysis module includes the following steps when predicting marine disasters based on historical data and real-time data: S31: Feature dataset construction, integrating the extracted disaster sign features into a feature dataset; S32: Select a prediction model, select a suitable prediction model from a pre-trained model library according to the disaster type and prediction requirements, load the selected prediction model, and initialize it; S33: Data preprocessing and feature engineering: normalize or standardize the feature data set to eliminate the dimensional differences between different features and improve the prediction accuracy of the model. Use correlation analysis and principal component analysis to perform feature selection and dimensionality reduction on the extracted feature data. S34: Use historical data to train the prediction model and tune the model parameters through cross-validation and grid search methods; S35: Disaster prediction and result analysis. First, the disaster sign features collected in real time are input into the trained prediction model. The model outputs the probability or level of disaster occurrence based on the input feature data. Finally, the prediction results are analyzed and interpreted.

5. The marine disaster monitoring and early warning system according to claim 4 is characterized in that: The principle algorithms used by the data analysis module for disaster prediction include: A31: Storm surge water increase formula, the principle formula is: ; Among them, Δh represents the storm surge, α represents the first weight, p represents the storm intensity, β represents the second weight, V represents the wind speed, γ represents the third weight, S represents the moving speed, θ represents the path angle, and d represents the geographical feature; A32: Tsunami wave propagation formula, its principle formula is: ; Among them, v is the propagation speed of the tsunami wave, g is the acceleration of gravity, h is the depth of the ocean, and λ is the wavelength of the tsunami wave.

6. The marine disaster monitoring and early warning system according to claim 1, characterized in that: The warning release module also has the following functions: A41: According to the early warning information, the corresponding emergency plan is activated, which includes the personnel evacuation plan and material deployment plan; A42: Provide real-time disaster information and disaster prediction information to rescue teams to ensure the efficiency and safety of rescue operations; A43: After a disaster occurs, the disaster losses are assessed.

7. The marine disaster monitoring and early warning system according to claim 6, characterized in that: When the early warning release module starts the corresponding emergency plan according to the early warning information, it includes the following steps: S41: The warning release module first receives the disaster warning information from the data analysis module, including the disaster type, expected occurrence time and impact range; S42: emergency plan matching, matching the corresponding emergency plan from the plan library according to the specific content of the warning information; S43: Emergency resource allocation. Before launching the emergency plan, the required emergency resources need to be allocated, including rescue teams, material reserves and transportation tools; S44: Emergency plan initiation and response, through the set emergency plan initiation process, the emergency plan initiation phase is carried out step by step.

8. The marine disaster monitoring and early warning system according to claim 1, characterized in that: The system also includes a system management module, which is used to remotely monitor and manage the hardware devices. When the system management module remotely monitors and manages the hardware devices, the system management module includes the following steps: S51: The hardware device periodically sends a verification data packet to the hardware device according to a set time interval; S52: After receiving the verification data packet, the system management module receives and processes the verification data packet sent by the hardware device; S53: the system management module records the time when the verification data packet is received, and uses the time difference between the time when the verification data packet is received and the time when it is sent to determine the connection status; S54: When the connection status is poor or the connection is disconnected, the system management module records it and sends it to the staff to notify the staff to perform inspection and maintenance.

Citation Information

Patent Citations

  • Marine disaster risk prevention and control early warning system

    CN116797019A

  • Volcano tsunami monitoring method based on remote sensing spectral characteristics

    CN117743828A

  • Rapid forecasting method for storm surge water increase

    CN118567004A

  • Marine disaster data processing method based on remote sensing monitoring

    CN118865600A

  • Process for Recovering Lithium from Spent Lithium Batteries

    KR1020230123881A

Cited By

  • Marine monitoring buoy network construction method and device based on satellite and storage medium

    CN119316850A

  • Satellite-based ocean monitoring buoy network construction method, device and storage medium

    CN119316850B

  • Marine disaster risk prevention and control early warning system

    CN119672912A

  • Marine three-dimensional layered right setting project monitoring and early warning method and system

    CN120278531A

  • Multi-source interference compensation tsunami monitoring and early warning method and system and storage medium

    CN120299221A