20 results about "Maritime safety" patented technology

The invention discloses a maritime accident prediction method and device based on interpretable integrated machine learning, and relates to the technical field of maritime affair safety risk analysis, and the method comprises the steps: obtaining accident investigation data, carrying out the preprocessing, balancing the data through a ten-fold layered oversampling method, and carrying out the cross verification training, and determining a performance optimal model by using the test set and carrying out interpretable analysis to explain the influence of the characteristics on the accident prediction result. By constructing a closed-loop'data processing-model optimization-explanation output 'process and adopting SMOTE oversampling and ten-fold layered cross validation training and a heterogeneous base model ensemble learning strategy, the processing capacity of the data imbalance problem of accident categories is improved, the data leakage problem of oversampling is avoided, and the possible bias of a single model is overcome. The interpretability analysis of the model prediction result can quantitatively display the contribution degree of each feature to prediction globally and locally, reveal the nonlinear relationship and interaction effect between the features, and provide transparent interpretation of model decision.

The invention discloses an offshore small target radar detection method and system, and relates to the technical field of radio detection. The method comprises the following steps: acquiring radar echo data, a visible light image, an infrared image, an AIS signal and navigation attitude data in real time; performing sea clutter suppression and pre-filtering on the radar echo data, and correcting target plots in combination with a historical radar plot set to obtain a radar single-source target track set; performing target detection on the visible light image and the infrared image to obtain an optical target detection set and an infrared target detection set; the radar single-source target track set, the optical target detection set, the infrared target detection set, the AIS signals and the navigation attitude data are associated and fused, and finally deception feature recognition and consistency verification are performed to obtain an updated fused target track set; according to the method, man-made confrontation and cheating behaviors on the sea can be accurately identified, and reliable support is provided for safety monitoring on the sea, maritime affair supervision and emergency response.

The invention provides a ship port entering and leaving yaw early warning identification method and system which are applied to the technical field of shipping informatization, ship traffic management and maritime safety monitoring, and the method comprises the steps: obtaining a channel center line and a yaw threshold value, receiving the data of a ship automatic identification system in real time, determining a track direction, determining a reference channel section, and determining a yaw threshold value; the included angle between the track direction and the reference channel section is calculated, and an early warning signal is generated when the included angle is larger than the threshold value. According to the scheme, the hysteresis quality of manual monitoring can be effectively avoided, and the problems that early warning is not timely and the accuracy is low in the prior art are solved.

The invention relates to the technical field of ship trajectory prediction, in particular to a cross-scene ship trajectory prediction fusion method based on a Transform algorithm. The method comprises the steps that a multi-source data input layer integrates basic navigation, port, sea transportation, inland river exclusive and environment and interaction data; the data preprocessing and link filtering layer cleans data noise and screens high-probability OD pairs; the multi-dimensional feature engineering layer constructs 15-dimensional feature vectors including sea transportation, inland river and interaction features; the core model is based on a Transform encoder-decoder architecture, integrates multi-head attention, time-space diagram attention and multi-mode bias, and adopts double-branch output and mixed loss function optimization; and the result post-processing and application output layer adopts a differentiation strategy for short-term and long-term prediction, and supports multi-ship collision avoidance path generation and derivative application at the same time. According to the invention, multi-scene coverage of sea transportation and inland waterway is realized, short-term and long-term prediction precision is well improved, and the method can be applied to tasks such as maritime affair safety scheduling and inland river intelligent shipping.

The invention discloses an NAVDAT, meteorological fax and NAVTEX collaborative maritime information broadcasting system, which comprises a shore-based intelligent fusion gateway, a cloud server, a cloud server, a cloud server and a cloud server, and is characterized in that the shore-based intelligent fusion gateway is used for analyzing received maritime safety information and mapping the received maritime safety information into a unified maritime event information model; based on the maritime event information model, broadcast signals conforming to NAVDAT, meteorological fax and NAVTEX system technical standards are generated in parallel, and a unified association identifier is embedded into the broadcast signals originating from the same information model; and the shipborne fusion terminal receives and analyzes the broadcast signal, automatically associates and recombines information from different systems according to the unified association identifier, and performs fusion display by taking an event as a unit. According to the invention, deep collaboration of the NAVDAT system, the weather fax system and the NAVTEX system in a protocol layer, a data layer and an application layer can be realized, and the integrity, the cognitive efficiency and the decision support capability of maritime affair safety information are remarkably improved.

The invention discloses a marine operation unmanned aerial vehicle installation RIS auxiliary safety sensing method, and belongs to the field of marine safety sensing. The method comprises the following steps: installing an intelligent metasurface on the bottom surface of an unmanned aerial vehicle, controlling a reflection echo signal of a shipborne radar by using the intelligent metasurface, and establishing a cascade channel model between the shipborne radar and the unmanned aerial vehicle; calculating the received signal power of the shipborne radar based on the cascade channel; and an optimization problem is constructed by taking the maximum receiving signal power of the authorized shipborne radar as a target, and the optimization problem is solved by jointly optimizing the phase of the intelligent metasurface and the trajectory of the unmanned aerial vehicle. According to the invention, for a safety perception scene of a marine operation unmanned aerial vehicle, a safety perception scheme of installing RIS assistance at the bottom of the unmanned aerial vehicle is provided, and a reflection echo signal of a target can be directly controlled so as to respectively enhance or suppress a reflection signal towards an authorized or unauthorized radar; and meanwhile, the signal power received by the unauthorized radar is effectively suppressed, so that the safety perception performance of the unmanned aerial vehicle target is improved.

The utility model discloses an offshore safety early warning device for a communication terminal, which comprises a damping assembly, a support plate, a driving part, a damping part, a base, a damper, a monitor, a protective shell and a dismounting part, and the driving part is arranged at the bottom of the support plate. The fixing seat drives the connecting plate to rotate, the connecting plate drives the supporting seat and the sliding block to move, the sliding block slides along the outer side of the supporting rod, the sliding block drives the spring to contract, and a damper is matched for damping, so that the damping effect on the monitor is achieved; and the clamping plate rotates along the inner wall of the fixing plate, so that the effect of disassembling the protective shell is achieved, the adjusting plate drives the connecting rod to rotate along the directional seat, and the adjusting plate drives the limiting block and the limiting plate to move.

The invention discloses a safety management system for offshore wind power project operating personnel, and relates to the technical field of offshore safety management. The method comprises the following steps: a risk point module for establishing a risk point account for ship navigation of a target operator and confirming a sea approval list; the regional quota module is used for dividing the operation region into a plurality of sub-regions, setting a role cooperation quota for each sub-region, and performing role matching management and control; the rhythm monitoring module is used for constructing an operation rhythm curve, evaluating the fatigue state of the target operator and triggering a finishing instruction; the exception handling module is used for calling an exception scene template when an emergency event is monitored, and allocating an emergency task according to a preset emergency response process and a set role division requirement; and the closed-loop safety module is used for calculating a closed-loop safety index of the target operation and feeding back the closed-loop safety index to the risk point account so as to form a dynamically updated safety backtracking mechanism. According to the invention, the operation safety and the emergency disposal efficiency are obviously improved.

1. Problem 2 related to failure and malfunction during navigation of propulsion device of water transportation, Problem 3 related to defensive action on the sea, Problem 4 related to security action on the sea, and Problem related to disaster prevention action on the sea SOLUTION: (1) A device having a conical outer frame, a hollow inside, and a member rotating at a high speed by electric power or the like is used. When this device is operated, a large amount of fluid is sucked from the large-diameter side of the conical outer frame, and high-speed fluid is discharged from the small-diameter side. This device is used for propulsion. 2. Use of the device according to 1. not for propulsion purposes, but for generating a vortex flow on the water surface. 3. Use of the device according to 1. not in a propulsion application but in an application for discharging a high-speed water flow onto the water. The method of use described in (4) and (2) is used not for defense action but for disaster prevention action (disaster reduction action against natural disasters).SELECTED DRAWING: None

To provide an intelligent navigation decision support to ship staff by fully extracting and effectively fusing multimodal marine safety information.SOLUTION: In order to propose a multimodal fusion network capable of training various input information including navigation AIS latitude and longitude information, NAVTEX text information, electronic nautical chart vector information, raster information, video image RGB information, and HOG information by using various networks, a Contact fusion structure based on a multiple attention mechanism MHA is introduced into a parallel structure of the multimodal network so that features of the multimodal information are sufficiently extracted and effectively fused. A multimodal decision support method is proposed for a result display end by suggesting a Transformer model based on generative preliminary training and understanding a context of the multimodal information of fusion features.SELECTED DRAWING: Figure 1

The application discloses a ship roll risk assessment method based on AIS and ERA5 data fusion, and belongs to the field of maritime safety and ship seakeeping analysis; in view of the problem that wave parameters and ship motion states are not accurately matched in existing roll risk analysis, the method comprises the following steps: fusing AIS trajectory and ERA5 wave reanalysis data, adopting space-time nearest neighbor pairing to construct a ship-wave sample and extracting input parameters; a simplified frequency domain roll response model is established based on an encounter angle frequency and a natural roll angle frequency, a roll response standard deviation and a short-time overrun probability are calculated; a frequency proximity function, a sea state intensity function and a wave direction function are defined, and a parameter roll amplification factor is constructed; the overrun probability and the amplification factor are fused to obtain a normalized roll risk index, and are mapped to a latitude and longitude grid to generate a global or regional roll risk distribution map. The application makes full use of ship dynamic and wave information, realizes efficient and robust roll risk assessment and visualization, and supports route planning and maritime early warning.

The application relates to the technical field of detection and recognition, and discloses a mobile ship dynamic tracking and locking method and system based on target detection and recognition, which comprises the following steps: constructing a manifold embedding network to map ship multi-modal features to a Riemann manifold space; constructing a continuous evolution model of a neural differential equation to predict the continuous evolution trajectory of the features under time, perspective and scale changes; constructing a multi-scale bridging network to realize bidirectional conversion between features representing long and short distances; constructing a differential geometric pose encoder to represent the ship pose in an SO(3) Lie group space; realizing a feature memory and reconstruction mechanism to solve the problem of short-time disappearing target recovery; and integrating the above components to construct a unified tracking and locking system. The application solves the problems of discretization processing limitations, feature representation fragmentation and cross-condition consistency loss in the prior art, and remarkable technical effects are achieved in terms of cross-condition recognition capability, continuous feature evolution capability and short-time disappearing target recovery capability, and the application is suitable for scenes such as Yangtze River shipping monitoring and maritime safety supervision.

The invention relates to the technical field of maritime affair safety monitoring, in particular to an intelligent ship identification management method and system, which ensures the comprehensiveness of data sources and provides a rich basis for subsequent analysis by collecting original data from a plurality of heterogeneous sensors. Original data are fused to generate a fusion feature vector, effective integration of multi-source data is achieved, ship feature representation is more comprehensive and accurate, and a reliable basis is provided for judgment of unfamiliar ships; whether the target ship is a strange ship or not is judged based on the fusion feature vector, manual experience judgment is replaced with data-driven judgment logic, and subjectivity and negligence are avoided; risk assessment is carried out on the unfamiliar ship, and corresponding early warning is triggered, so that seamless connection of identification and early warning is realized, and rapid risk response to the unfamiliar ship is ensured; and according to the risk level and the real-time state information, a disposal suggestion is generated and linkage response is started, so that a complete process from identification to disposal is formed, and automation and high efficiency of ship management are improved.

The present application belongs to the technical field of flow field intelligent prediction based on numerical simulation, and particularly relates to a ship surface fluid pressure field intelligent prediction method, program, device and storage medium based on a combination of a reduced-order model and a neural network. The present application maps the unstructured grid data obtained by numerical simulation into structured grid data by using a surrogate model, ensuring the uniformity of the subsequent data processing format. On this basis, the main modal of the surface pressure field and the corresponding grid node normal vector is extracted by the reduced-order model. Then, the full-link neural network is used to realize the rapid prediction of the new ship type basis coefficient, so that for the new ship type, only the design parameters need to be input without inputting the ship grid, the rapid prediction of the ship surface load distribution, the external normal vector distribution and the pressure resistance can be realized, and reliable technical support can be provided for the ship structure strength design and safe navigation at sea.

An image-based floating marine object detection and distance measurement system for a small vessel of less than 10 tons is disclosed. The system comprises: an infrared stereo camera and an IR projector provided on a vessel; and an image display system which is connected to the infrared stereo camera and the IR projector through a control API, includes an image correction filter for correcting blurriness due to light reflection, shadow, and sea fog, detects the floating marine object (a discarded fishing net, a rope, a buoy, and marine litter) by a deep learning-based object detection model on the basis of pre-training data of the floating marine object, measures a distance between a camera of a vessel of less than 10 tons and the floating marine object, and displays an image to prevent vessel collision, damage to a propulsion shafting system, and an accident of entanglement between a vessel and floating marine debris through a forward hazard situation alert and a corresponding response. The system uses an infrared stereo camera and an IR projector provided on a small vessel of less than 10 tons in which hull motion, sea fog, and light reflection due to wind and wave heights exist on the sea, so as to detect a floating marine object posing a hazard during vessel navigation, by a vision recognition technology using a floating marine object recognition distance measurement algorithm and a deep learning-based floating marine object recognition model using an AI technology, prevents vessel collision, damage to a propulsion shafting system, and an accident of entanglement between a vessel and floating marine debris through a hazard situation alert and a corresponding response, monitors marine litter, and is used as a marine safe navigation assistance system for a small vessel that is vulnerable to the floating marine object.

The invention belongs to the technical field of maritime safety and intelligent navigation, and particularly relates to a maritime high-rise protection distance prediction analysis method based on a radar technology, which comprises the following steps of: establishing a space grid by acquiring a detection efficiency attenuation field around a maritime high-rise, and extracting a radar signal attenuation value, a space change gradient and a distance-height ratio of each grid unit; after normalization and nonlinear transformation, the radar interference risk is determined according to the range of three transformation results, the real-time ship density and the ship encounter threat value of each grid unit are calculated based on the real-time ship AIS data, and the navigation conflict risk is determined; weighting and fusing the two corresponding risks into a comprehensive risk, and dynamically delimiting the boundary of the grading protection area according to the comprehensive risk; and monitoring the ships entering the area, and generating and sending a navigation instruction according to the position risk levels of the ships. According to the method, through grid analysis, multi-feature fusion and a dynamic decision-making mechanism, the complex radar interference and navigation conflict mixed risk caused by the towering object is effectively dealt with.

The application discloses a water ship traffic flow complexity calculation method, device, equipment and medium, and the method comprises the steps of obtaining ship AIS data of a target water area; determining ship clustering clusters in the target water area according to the ship AIS data by using an improved DBSCAN algorithm; performing ship state feature clustering on the ship clustering clusters to obtain state feature probability distribution of the ship clustering clusters; and determining ship traffic flow complexity of the target water area based on information entropy according to the state feature probability distribution. The improved DBSCAN algorithm is innovatively obtained based on a ship field model, so that the algorithm is more suitable for navigation practice; the uncertainty of the water traffic flow state is described by information entropy according to the state feature probability, the method for measuring the complexity of the water ship traffic flow in the field of water traffic engineering is expanded, and the application has important theoretical and practical significance for promoting the informatization of maritime safety supervision.

The application discloses a maritime accident prediction method and device based on an interpretable integrated machine learning, relates to the technical field of maritime safety risk analysis, and comprises the following steps: acquiring accident investigation data and preprocessing, balancing the data through a ten-fold stratified oversampling method and training through cross-validation, determining a performance-optimal model by using a test set and performing interpretable analysis, and explaining the influence of features on accident prediction results. Through the construction of a closed-loop "data processing-model optimization-explanation output" process, the SMOTE oversampling, the ten-fold stratified cross-validation training and the heterogeneous base model integrated learning strategy are adopted, the processing capacity for the data imbalance problem of the accident category is improved, the data leakage problem of oversampling is avoided, and the possible bias of a single model is overcome. The interpretable analysis of the model prediction result can quantitatively show the contribution of each feature to the prediction from the global and local aspects, reveals the nonlinear relationship and interaction effect among the features, and provides a transparent explanation for the model decision.

The invention discloses an intelligent gateway for a heterogeneous maritime affair broadcast protocol, which comprises a data access layer, an intelligent processing layer and a protocol output layer, and is characterized in that the intelligent processing layer comprises a natural language processing module and is configured to perform semantic analysis on input maritime affair safety information and extract key elements including time, places and event types; the content association engine is configured to establish an association relationship among multi-source information for the same maritime event based on the key elements; the multi-protocol converter is configured to generate broadcast signals conforming to the NAVDAT protocol, the meteorological fax protocol and the NAVTEX protocol according to the incidence relation; and a guidance instruction generator configured to generate a standardized guidance instruction associated across the protocol. According to the intelligent gateway for the heterogeneous maritime affair broadcast protocol, intelligent understanding and automatic association of maritime affair information can be achieved, the problem that multi-system fusion depends on manual operation is solved, and unified, efficient and enhanced information service is provided.

This invention relates to the interdisciplinary field of maritime safety supervision in transportation and artificial intelligence, and discloses an intelligent auxiliary method and system for Port State Control (PSC) inspections of liquefied natural gas (LNG) vessels. The method includes structured processing and relational mapping of multi-source basic data, semantic storage of maritime regulatory texts in parent-child segmentation, and generation of a specialized knowledge base data with a five-dimensional relational mapping network. Responding to the identification information of the vessel to be inspected, the method calls upon the specialized knowledge base data and performs calculations using a risk analysis model combining a rule engine and machine learning classification to generate personalized inspection checklist data labeled with risk level, high-risk inspection areas, and inspection priorities. Through the synergy of the method and system, the method addresses pain points such as inefficient regulatory retrieval for LNG vessel PSC inspections, reliance on experience for defect determination, and cumbersome report generation, significantly improving inspection efficiency and accuracy, standardizing the inspection process, and enabling personalized training for professionals and continuous system optimization.