Airline avoiding method and system based on iceberg

Abstract

The invention relates to an iceberg-based route avoidance method and system, and the method comprises the steps: obtaining original route information, and recognizing a target route region; acquiring ice region environment data and meteorological environment data and inputting the data into the ocean numerical model; acquiring ship performance parameters, converting the ship performance parameters into a dynamic parameter sequence, and inputting the dynamic parameter sequence and original route information into a navigation numerical model; inputting corresponding features output by the model into a risk prediction model; and dynamically adjusting the original route information based on the risk characteristics. According to the method, the ice region environment data and the meteorological environment data are obtained, dynamic changes of ship performance parameters are combined, multiple risk factors are comprehensively considered, the dynamic nature and uncertainty of the ice region environment can be adapted in real time, the accuracy of route planning is improved, and meanwhile, the risk prediction accuracy is improved according to the risk characteristics output by the risk prediction model. The original ship route information is dynamically adjusted to ensure that the ship can be adjusted and avoided in time when encountering potential risks, and the effect of improving the sailing safety and efficiency is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of route planning, and in particular to a route avoidance method and system based on icebergs. Background Art

[0002] In the field of marine transportation, route planning is an important part of ensuring safe and efficient navigation of ships. Traditional route planning methods mainly rely on nautical charts, historical navigation data and empirical rules, which usually achieve good results in open waters. However, with global climate change and the opening of polar routes, the demand for ships to sail in ice areas is increasing, which poses a higher challenge to route planning.

[0003] The ice navigation environment is complex and changeable. Icebergs, dense sea ice areas, extreme weather and other factors may pose a serious threat to the safety of ship navigation. Traditional route planning methods often find it difficult to combine the performance of the ship and the impact of the environment on the performance of the ship, and to consider the dynamic changes and uncertainties of the ice environment in many aspects for planning, which may lead to unexpected risks in actual navigation. Summary of the invention

[0004] In order to evaluate the risk of ice navigation in real time and accurately, and provide an effective route avoidance strategy, thereby significantly improving the safety and efficiency of ice navigation, the present application provides an iceberg-based route avoidance method and system.

[0005] The above-mentioned invention objective of the present application is achieved through the following technical solutions:

[0006] A route avoidance method based on icebergs comprises the following steps:

[0007] Acquire original route information of the target ship, and identify the target route area associated with the original route information;

[0008] Acquiring ice area environment data and meteorological environment data of the target route area;

[0009] Obtaining the ship performance parameters of the target ship at intervals and converting them into a dynamic parameter sequence;

[0010] Inputting the ice area environment data and the meteorological environment data into a pre-trained ocean numerical model to simulate and predict the characteristics of waves and icebergs;

[0011] Inputting the original route information and the dynamic parameter sequence into a pre-trained navigation numerical model to simulate and predict navigation characteristics;

[0012] Inputting the time-series wave characteristics and iceberg characteristics output by the ocean numerical model, the time-series navigation characteristics output by the navigation numerical model, and the original route information into the risk prediction model;

[0013] When the risk characteristics output by the risk prediction model are received, the original route information is dynamically adjusted based on the risk characteristics.

[0014] By adopting the above technical solution, the original route information of the target ship is obtained, and the target route area corresponding to the original route information is identified, that is, the route area included in the original route of the target ship is identified, and the ice environment data and meteorological environment data of the target route area are obtained; during the navigation of the ship, the ship performance parameters of the target ship are obtained at intervals, and converted into a dynamic parameter sequence, so as to reflect the real-time state of the ship during the navigation; the ice environment data and the meteorological environment data are input into a pre-trained ocean numerical model, so that it can simulate and predict the characteristics of the waves and the characteristics of the icebergs, so as to evaluate the dynamic changes of the ocean and ice environment; the original route information and the dynamic parameter sequence are input into the pre-trained navigation numerical model, so that it can simulate and predict the navigation characteristics, so as to evaluate the changes in the navigation state of the ship during the navigation; the ocean numerical model is input into the navigation numerical model, and the navigation characteristics are simulated and predicted. The time-series wave characteristics and iceberg characteristics output by the navigation numerical model, the time-series navigation characteristics output by the navigation numerical model, and the original route information are input into the risk prediction model to make a comprehensive risk prediction and assessment. Based on the risk characteristics output by the risk prediction model, the original route information is dynamically adjusted to enable the ship to avoid icebergs during navigation in ice areas, thereby ensuring the safety of its navigation. The present application obtains ice area environment data and meteorological environment data, and combines the dynamic changes of ship performance parameters, comprehensively considers icebergs, sea ice, waves, meteorological conditions and other risk factors, can adapt to the dynamics and uncertainty of the ice area environment in real time, improve the accuracy of route planning, and dynamically adjusts the original route information according to the risk characteristics output by the risk prediction model to ensure that the ship can make timely adjustments and avoid risks when encountering potential risks, which has the effect of improving the safety and efficiency of navigation.

[0015] In a preferred example, the present application may be further configured as follows: the step of obtaining the original route information of the target ship and identifying the target route area associated with the original route information comprises the steps of:

[0016] The route area set is divided based on a preset division rule to obtain a plurality of route sub-areas;

[0017] Building an R-tree index with the obtained plurality of route sub-areas as nodes;

[0018] Convert the original route information into continuous coordinate points, and quickly query each coordinate point based on the R-tree index;

[0019] The target route sub-areas obtained from the quick query results are integrated to obtain the target route area, wherein the target route sub-areas include: route sub-areas intersecting with at least one of the coordinate points and route sub-areas including at least one of the coordinate points.

[0020] By adopting the above technical solution, the entire route area set is divided into several smaller route sub-areas based on the preset division rules, and an index is constructed with the route sub-areas as nodes through the R-tree; the original route information is converted into a series of continuous coordinate points, and each coordinate point is quickly queried through the R-tree index to identify the route sub-areas that contain or intersect at least one coordinate point, and the identified route sub-areas are integrated to form the final target route area; by accurately identifying all route areas related to the original route of the ship, accurate spatial location information is provided for subsequent analysis and decision-making.

[0021] In a preferred example, the present application can be further configured as follows: the step of obtaining the ship performance parameters of the target ship at intervals and converting them into a dynamic parameter sequence includes the steps of:

[0022] Acquire the ship performance parameters of the target ship based on the frequency corresponding to the preset dynamic acquisition rule;

[0023] Assign a timestamp of the accuracy corresponding to the preset dynamic acquisition rule to each group of acquired ship performance parameters;

[0024] The ship performance parameters are converted into dynamic parameter sequences in time sequence.

[0025] By adopting the above technical solution, the ship performance parameters of the target ship are collected based on the frequency corresponding to the preset dynamic collection rules, and each group of collected ship performance parameters is assigned an accurate timestamp corresponding to the current dynamic collection rule preset, and all ship performance parameters with timestamps are sorted and converted into a dynamic parameter sequence based on time sequence; by adopting dynamic collection rules and accurate timestamps, the collected ship performance parameters have a high degree of accuracy and reliability.

[0026] In a preferred example, the present application may be further configured as follows: before the step of acquiring the ship performance parameters of the target ship based on the frequency corresponding to the preset dynamic acquisition rule, the following steps are performed:

[0027] Inputting ice area environmental data and meteorological environmental data of the target route area into a pre-trained risk assessment model, so that the risk assessment model outputs the risk level of each route sub-area in the target route area;

[0028] The current location information of the target ship is obtained, and a corresponding preset dynamic collection rule is matched based on the risk level associated with the current location information.

[0029] By adopting the above technical scheme, the ice area environmental data and meteorological environmental data of the target route area are input into the pre-trained risk assessment model, so that the model can analyze and calculate the risk level of each route sub-area in the target route area according to the input environmental data, identify the current position information of the target ship and determine the route sub-area and its corresponding risk level based on the current position information, and match the preset corresponding dynamic collection rules based on the risk level to determine the collection frequency of the ship performance parameters; the application dynamically adjusts the collection frequency of the ship performance parameters by performing a risk assessment on the current position of the ship, so that data can be collected more frequently in high-risk areas, while the collection frequency can be appropriately reduced in low-risk areas, thereby improving the pertinence and effectiveness of data collection, while avoiding excessive data collection when it is not necessary, saving storage space and transmission resources, and optimizing resource utilization.

[0030] In a preferred example, the present application can be further configured as follows: the ocean numerical model includes an extraction layer, a first simulation layer, and a second simulation layer, and the step of inputting ice area environment data and meteorological environment data into the pre-trained ocean numerical model to simulate and predict wave characteristics and iceberg characteristics includes the steps of:

[0031] The extraction layer extracts ice zone sequence features, meteorological sequence features and associated sequence features from the ice zone environment data and meteorological environment data respectively based on time series;

[0032] The first simulation layer simulates and predicts iceberg characteristics based on ice zone sequence characteristics and associated sequence characteristics, and outputs iceberg characteristics;

[0033] The second simulation layer simulates and predicts the wave characteristics based on the meteorological sequence characteristics and the associated sequence characteristics, and outputs the wave characteristics.

[0034] By adopting the above technical scheme, the ocean numerical model includes an extraction layer, a first simulation layer and a second simulation layer. The extraction layer extracts corresponding ice zone sequence characteristics, meteorological sequence characteristics and associated sequence characteristics from the input ice zone environment data and meteorological environment data based on time series analysis. The first simulation layer simulates and predicts iceberg characteristics based on the ice zone sequence characteristics and associated sequence characteristics output by the extraction layer and outputs the iceberg characteristics. The second simulation layer simulates and predicts ocean wave characteristics based on the meteorological sequence characteristics and associated sequence characteristics output by the extraction layer and outputs the ocean wave characteristics.

[0035] In a preferred example, the present application can be further configured as follows: the first simulation layer simulates and predicts iceberg characteristics based on ice zone sequence characteristics and associated sequence characteristics, and outputs the steps of iceberg characteristics, including the steps of:

[0036] Extracting a first key vector from the associated sequence features and the ice zone sequence features, wherein the first key vector includes a flow field velocity vector, a wind field velocity vector, and a velocity vector of the interaction between the iceberg and the sea ice within a preset range;

[0037] Calculate the velocity vector and position vector of the iceberg based on the extracted first key vector: in, is the position vector of the iceberg, is the flow field velocity vector, is the wind velocity vector, is the velocity vector of the iceberg interacting with the sea ice within its preset range;

[0038] The motion characteristics of the iceberg are calculated based on the velocity vector and the position vector of the iceberg, and the calculation results are output as iceberg characteristics.

[0039] By adopting the above technical scheme, the first key vector is extracted from the associated sequence characteristics and the ice zone sequence characteristics: the flow field velocity vector, the wind field velocity vector and the velocity vector of the interaction between the iceberg and the sea ice within a preset range, and the velocity vector of the iceberg is calculated based on the three extracted velocity vectors; at the same time, the position vector of the iceberg is calculated based on its velocity vector, that is, the position of the iceberg at a specific point in time; the motion characteristics of the iceberg are further calculated based on the velocity vector and the position vector of the iceberg, and the calculated results are output as iceberg characteristics; the present application improves the accuracy of predicting the motion characteristics of icebergs by comprehensively considering the flow field, wind field and the interaction between icebergs and sea ice.

[0040] In a preferred example, the present application can be further configured as follows: the second simulation layer simulates and predicts the wave characteristics based on the meteorological sequence characteristics and the associated sequence characteristics, and outputs the wave characteristics, including the steps of:

[0041] Extracting a second key vector from the association sequence features and the meteorological sequence features, wherein the second key vector includes wind speed, tidal height, and ocean current speed;

[0042] Calculate the wave spectral density based on the extracted second key vector:

[0043] Where S(ω) is the wave spectral density, and β are dimensionless constants, g is the acceleration of gravity, U is the wind speed, and ω is the angular frequency;

[0044] Calculate the corrected wave spectral density based on the wave spectral density and the second key vector:

[0045] S modified (ω) = S (ω)·f (T tide ,V current ,ω), where S modified (ω) is the modified wave spectral density, S(ω) is the wave spectral density, f(T tide ,V current ,ω) is the correction function of the wave spectral density, T tide is the tidal height, V current is the ocean current velocity, ω is the angular frequency;

[0046] The wave height characteristics, period characteristics and propagation characteristics are obtained based on the modified wave spectral density and output as wave characteristics.

[0047] By adopting the above technical scheme, the second key vector is extracted from the associated sequence features and the meteorological sequence features: wind speed, tidal height and current speed, and the wave spectral density is calculated based on the wind speed in the extracted second key vector. At the same time, in order to further combine the influence of tidal height and current speed on waves, the wave spectral density is corrected based on a correction function including tidal height and current speed. The corrected wave spectral density further improves the accuracy of the actual wave conditions. The wave height characteristics, period characteristics and propagation characteristics of the waves are calculated based on the corrected wave spectral density, and the calculated characteristics are output as wave characteristics. The present application has the effect of significantly improving the prediction accuracy of wave characteristics by comprehensively considering multiple key factors affecting waves (such as wind speed, tidal height, current speed, etc.).

[0048] The second object of the invention is achieved by the following technical solutions:

[0049] An iceberg-based route avoidance system, comprising:

[0050] A route area acquisition module, used to acquire original route information of a target ship and identify a target route area associated with the original route information;

[0051] An environmental data acquisition module, used to acquire ice area environmental data and meteorological environmental data of the target route area;

[0052] A dynamic parameter sequence acquisition module is used to obtain the ship performance parameters of the target ship at intervals and convert them into a dynamic parameter sequence;

[0053] A first data input module is used to input the ice area environment data and the meteorological environment data into a pre-trained ocean numerical model to simulate and predict the characteristics of waves and icebergs;

[0054] A second data input module is used to input the original route information and the dynamic parameter sequence into a pre-trained navigation numerical model to perform navigation characteristic simulation prediction;

[0055] A feature input module, used to input the time-series wave features and iceberg features output by the ocean numerical model, the time-series navigation features output by the navigation numerical model, and the original route information into the risk prediction model;

[0056] The route information adjustment module is used to dynamically adjust the original route information based on the risk characteristics output by the risk prediction model when the risk characteristics are received.

[0057] By adopting the above technical solution, a route area acquisition module is used to acquire the original route information of the target ship and identify the target route area associated with the original route information; an environmental data acquisition module is used to acquire the ice area environmental data and meteorological environment data of the target route area; a dynamic parameter sequence acquisition module is used to obtain the ship performance parameters of the target ship at intervals and convert them into a dynamic parameter sequence; a first data input module is used to input the ice area environmental data and the meteorological environment data into a pre-trained ocean numerical model for wave feature simulation prediction and iceberg feature simulation prediction; a second data input module is used to input the original route information and the dynamic parameter sequence into a pre-trained navigation numerical model for navigation feature simulation prediction; a feature input module is used to input the wave features and iceberg features with time series output by the ocean numerical model, the navigation features with time series output by the navigation numerical model, and the original route information into a risk prediction model; a route information adjustment module is used to dynamically adjust the original route information based on the risk features when receiving the risk features output by the risk prediction model.

[0058] In summary, the present application includes at least one of the following beneficial technical effects:

[0059] 1. This application obtains ice environment data and meteorological environment data, combines the dynamic changes of ship performance parameters, and comprehensively considers multiple risk factors such as icebergs, sea ice, waves, and meteorological conditions. It can adapt to the dynamics and uncertainty of the ice environment in real time and improve the accuracy of route planning. At the same time, according to the risk characteristics output by the risk prediction model, the original route information is dynamically adjusted to ensure that the ship can make timely adjustments and avoid potential risks, which has the effect of improving the safety and efficiency of navigation;

[0060] 2. This application conducts risk assessment on the current position of the ship and dynamically adjusts the collection frequency of the ship's performance parameters, so that data can be collected more frequently in high-risk areas, while the collection frequency can be appropriately reduced in low-risk areas, thereby improving the pertinence and effectiveness of data collection, while also avoiding unnecessary excessive data collection, saving storage space and transmission resources, and optimizing resource utilization;

[0061] 3. This application improves the accuracy of predicting the movement characteristics of icebergs by comprehensively considering the flow field, wind field, and the interaction between icebergs and sea ice; at the same time, by comprehensively considering a variety of key factors affecting ocean waves (such as wind speed, tidal height, current speed, etc.), it has the effect of significantly improving the accuracy of predicting ocean wave characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flow chart of an embodiment of an iceberg-based route avoidance method of the present application;

[0063] Figure 2 This is a flowchart of an implementation of step S10 in an embodiment of an iceberg-based route avoidance method of the present application;

[0064] Figure 3 This is a flowchart of an implementation of step S30 in an embodiment of an iceberg-based route avoidance method of the present application;

[0065] Figure 4 This is a flowchart of an implementation of step S40 in an embodiment of an iceberg-based route avoidance method of the present application;

[0066] Figure 5 This is a flowchart of an implementation of step S42 in an embodiment of an iceberg-based route avoidance method of the present application;

[0067] Figure 6 It is an implementation flow chart of step S43 in an embodiment of an iceberg-based route avoidance method of the present application. DETAILED DESCRIPTION

[0068] The following is combined with Figure 1-6 This application is described in further detail.

[0069] In one embodiment, if Figure 1 As shown, the present application discloses a route avoidance method based on icebergs, which specifically includes the following steps:

[0070] S10: Acquire original route information of the target ship, and identify the target route area associated with the original route information;

[0071] In this embodiment, the target ship is a ship that currently needs to plan an iceberg avoidance route; the original route information is the original sailing route of the target ship, which includes a starting point, an end point, a route, and a passing point by default; the target route area is a specific sea area associated with the original route information, and different route areas may involve different marine environments;

[0072] Specifically, the original route information of the target ship is obtained, and the target route area corresponding to the original route information is identified, that is, the route area included in the original route of the target ship is identified;

[0073] S20: Acquire ice area environment data and meteorological environment data of the target route area;

[0074] In this embodiment, the ice environment data is data describing the characteristics of the ice environment in the route area, which is specifically obtained through satellite remote sensing technology: high-resolution satellite images are obtained through satellite remote sensing technology to monitor key parameters such as the distribution position, shape, size and movement trajectory of icebergs in the ice area in real time; among them, satellites commonly used in satellite remote sensing technology include NOAA series satellites, RADARSAT, MODIS, etc., which can provide image data of different resolutions and bands: NOAA series satellites have excellent performance in monitoring the changes of icebergs in large-scale ice environments with their wide coverage and moderate resolution; RADARSAT, with its radar remote sensing technology, can penetrate clouds and realize all-weather monitoring of icebergs in ice environments; MODIS is known for its high spectral resolution, which can capture more detailed information and help to deeply analyze the characteristics of icebergs in ice environments; meteorological environment data is data reflecting the meteorological conditions and ocean surface conditions in the route area;

[0075] Furthermore, in some key sea areas, disposable buoys and meteorological observation stations will be deployed to continuously monitor icebergs in the ice zone and meteorological conditions;

[0076] Specifically, for the target route area, ice environment data reflecting the characteristics of the ice environment and meteorological environment data reflecting the meteorological conditions and ocean surface conditions are obtained;

[0077] S30: acquiring ship performance parameters of the target ship at intervals and converting them into a dynamic parameter sequence;

[0078] In this embodiment, the ship performance parameters are various parameters reflecting the ship's operating status; the dynamic parameter sequence is a set of ship performance parameters arranged in time sequence;

[0079] Specifically, during the navigation process of the ship, the ship performance parameters of the target ship are obtained at intervals and converted into a dynamic parameter sequence, thereby reflecting the real-time state of the ship during the navigation process;

[0080] S40: inputting the ice area environment data and the meteorological environment data into the pre-trained ocean numerical model to simulate and predict the characteristics of the waves and the characteristics of the icebergs;

[0081] In this embodiment, the ocean numerical model is a pre-trained mathematical model for simulating the ocean environment and simulating and predicting the wave state. Specifically, the ocean environment in the region is simulated based on the ice area environment data and the meteorological environment data, so as to simulate and predict the wave state and iceberg state in the region and output the wave characteristics and iceberg characteristics; the wave characteristics are parameters reflecting the characteristics of the wave state; the iceberg characteristics are parameters reflecting the state characteristics of the iceberg and the changing state of the iceberg on the sea surface;

[0082] Specifically, the ice zone environment data and meteorological environment data are input into the pre-trained ocean numerical model to simulate and predict the characteristics of waves and icebergs, so as to evaluate the dynamic changes of the ocean and ice zone environment;

[0083] S50: inputting the original route information and the dynamic parameter sequence into the pre-trained navigation numerical model to simulate and predict the navigation characteristics;

[0084] In this embodiment, the navigation numerical model is a pre-trained mathematical model for simulating the performance state of a ship sailing on the ocean, specifically, based on the original route information and the dynamic parameter sequence, the performance state of the ship's navigation is simulated and predicted and the navigation characteristics are output; the navigation characteristics are parameters that reflect the navigation state of the ship;

[0085] Specifically, the original route information and dynamic parameter sequence are input into the pre-trained navigation numerical model to simulate and predict the navigation characteristics, so as to evaluate the changes in the navigation state of the ship during the navigation process;

[0086] S60: inputting the time-series wave characteristics and iceberg characteristics output by the ocean numerical model, the time-series navigation characteristics output by the navigation numerical model, and the original route information into the risk prediction model;

[0087] In this embodiment, the risk prediction model is pre-trained for combining time-series wave characteristics, iceberg characteristics, navigation characteristics and original route information for risk prediction, and outputs corresponding risk characteristics when predicting that the ship has the potential to collide with an iceberg or the distance to the iceberg is less than a preset range value when sailing on the original route; the risk characteristics are information reflecting the potential risks of the ship sailing on the original route and corresponding parameters;

[0088] Specifically, the time-series wave characteristics and iceberg characteristics output by the ocean numerical model, the time-series navigation characteristics output by the navigation numerical model, and the original route information are input into the risk prediction model to enable it to conduct comprehensive risk prediction and assessment;

[0089] S70: When receiving the risk characteristics output by the risk prediction model, dynamically adjusting the original route information based on the risk characteristics;

[0090] In this embodiment, dynamically adjusting the original route information based on the risk characteristics is to dynamically adjust the original route of the target ship accordingly based on the potential risk reflected by the risk characteristics. For example, if the target route area associated with the original route corresponding to the target ship includes five sub-areas A, B, C, D, and E, and the risk characteristics reflect that the original route of the target ship in area C has a potential risk of colliding with an iceberg, the original route of the target ship in area C is dynamically adjusted;

[0091] Furthermore, the original route information is dynamically adjusted based on the risk characteristics and preset adjustment rules, wherein the preset adjustment rules include:

[0092] Priority is given to adjusting part of the original route associated with the sub-area within the same sub-area. Specifically, if the target route area associated with the original route corresponding to the target ship includes five sub-areas A, B, C, D, and E, and the risk characteristics show that the original route of the target ship in area C has a potential risk of colliding with an iceberg, then the part of the original route associated with area C is dynamically adjusted within the scope of area C;

[0093] When it is impossible to adjust part of the original route associated with the sub-region in the same sub-region, the sub-region is set as a risk sub-region, and at the same time, the corresponding replacement sub-region is matched based on the preset area matching rule, and the replacement sub-region is replaced by the risk sub-region in the original route information, so as to adjust the original route information, specifically: if the target route area associated with the original route corresponding to the target ship includes five sub-regions A, B, C, D, and E, and the risk characteristics reflect that part of the original route of the target ship in area C has a potential risk of collision with an iceberg, and it is impossible to adjust part of the original route associated with area C in area C, the corresponding replacement sub-region is matched based on the preset area matching rule, for example, area G is a replacement sub-region, then part of the original route associated with area C is adjusted to area G, and the adjusted target route area includes five sub-regions A, B, G, D, and E, and the adjusted route is still continuous;

[0094] Wherein, matching the corresponding replacement sub-region based on the preset region matching rule includes: identifying the region location information of the risk sub-region and the adjacent sub-region, and matching another sub-region that can still be adjacent to the adjacent sub-region as the replacement sub-region based on the region location information, and re-performing steps S20 to S70 based on the replacement sub-region;

[0095] Among them, the preset region matching rule gives priority to matching replacement sub-regions that are less than or equal to the number of risk sub-regions;

[0096] Furthermore, when the preset region matching rule cannot match the replacement sub-regions that are less than or equal to the number of risk sub-regions, the number of replacement sub-regions is gradually increased until a successful match is achieved;

[0097] Specifically, based on the risk characteristics received from the risk prediction model output, the original route information is dynamically adjusted to enable ships to avoid icebergs when sailing in ice areas, thereby ensuring the safety of their navigation.

[0098] In one embodiment, if Figure 2 As shown, step S10 includes the steps of:

[0099] S11: Divide the route area set based on a preset division rule to obtain a plurality of route sub-areas;

[0100] S12: constructing an R-tree index using the obtained several route sub-areas as nodes;

[0101] S13: converting the original route information into continuous coordinate points, and quickly querying each coordinate point based on the R-tree index;

[0102] S14: integrating the target route sub-areas obtained from the quick query result to obtain a target route area, wherein the target route sub-area includes: a route sub-area intersecting with at least one coordinate point and a route sub-area including at least one coordinate point;

[0103] In this embodiment, the preset division rule is a division method for dividing the entire route area set into multiple route sub-areas based on actual needs (such as geographical location, marine environment, etc.); the route area set is a set of navigable geographical areas covered by all possible routes; the route sub-area is the smallest unit area obtained after dividing the route area set by the preset division rule; the R-tree index is an efficient index structure for multidimensional data, specifically a number of route sub-areas for equivalent geographic spatial data; the coordinate point is a point describing the geographical location, specifically a point including coordinate information such as longitude and latitude; the quick query is to determine the route sub-area to which the coordinate point belongs; the target route area is a complete route area integrated from the route sub-areas in the quick query result, wherein the target route sub-area includes: a route sub-area intersecting with at least one coordinate point and a route sub-area including at least one coordinate point;

[0104] Specifically, the entire route area set is divided into several smaller route sub-areas based on a preset division rule, and an index is constructed using the route sub-areas as nodes through an R-tree; the original route information is converted into a series of continuous coordinate points, and each coordinate point is quickly queried through the R-tree index to identify the route sub-areas that contain or intersect at least one coordinate point, and the identified route sub-areas are integrated to form the final target route area; by accurately identifying all route areas related to the original route of the ship, accurate spatial location information is provided for subsequent analysis and decision-making.

[0105] In one embodiment, if Figure 3 As shown, step S30 includes the steps of:

[0106] S31: Acquire ship performance parameters of the target ship based on the frequency corresponding to the preset dynamic acquisition rule;

[0107] S32: allocating a timestamp of the accuracy preset by the dynamic collection rule to each group of acquired ship performance parameters;

[0108] S33: converting the ship performance parameters into a dynamic parameter sequence in time sequence;

[0109] In this embodiment, the dynamic collection rule is a pre-set rule for setting the frequency of collecting ship performance parameters; the timestamp is a mark for recording the time corresponding to the ship performance parameters, and the precision is the accuracy of the timestamp;

[0110] Specifically, the ship performance parameters of the target ship are collected based on the frequency corresponding to the preset dynamic collection rules, and each group of collected ship performance parameters is assigned an accurate timestamp corresponding to the current dynamic collection rule preset, and all ship performance parameters with timestamps are sorted and converted into a dynamic parameter sequence based on time sequence; by adopting dynamic collection rules and accurate timestamps, the collected ship performance parameters have a high degree of accuracy and reliability.

[0111] In one embodiment, before step S31, the following steps are performed:

[0112] S301: inputting ice area environment data and meteorological environment data of the target route area into a pre-trained risk assessment model, so that the risk assessment model outputs the risk level of each route sub-area in the target route area;

[0113] S302: Acquire the current location information of the target ship, and match the corresponding preset dynamic collection rules based on the risk level associated with the current location information;

[0114] In this embodiment, the risk assessment model is a mathematical model for assessing the risks of each route sub-area contained in the target route area in combination with the ice area environment data and the meteorological environment data of the target route area, and outputting the corresponding risk level, where the risk level is an indicator describing the risk degree of the route sub-area; the current location information is information describing the current geographical location of the target ship; the risk level associated with the current location information is matched with the corresponding preset dynamic collection rules, specifically: based on the risk level corresponding to the route sub-area where the current location information is located, the preset corresponding collection frequency of the ship performance parameters is matched;

[0115] Furthermore, the risk assessment model is trained and optimized through the marine biological information and seabed topography information of the route area set, making the risk assessment more comprehensive;

[0116] Furthermore, when the risk assessment model conducts risk assessment on the ice environment data and meteorological environment data of the target route area, it obtains the current time and season information, and performs risk assessment based on the time and season information matching the corresponding pre-stored marine biological information, marine productivity information and human activity pattern information;

[0117] Among them, the marine productivity information is the data information of salt cycle and primary productivity in the ocean; the human activity pattern information is seasonal human activities, including fishing, shipping and tourism, which have the potential to cause seasonal pressure on the marine environment;

[0118] Specifically, the ice area environmental data and meteorological environmental data of the target route area are input into a pre-trained risk assessment model, so that the model analyzes and calculates the risk level of each route sub-area in the target route area according to the input environmental data, identifies the current position information of the target ship and determines the route sub-area and its corresponding risk level based on the current position information, and matches the risk level with the preset corresponding dynamic collection rules to determine the collection frequency of the ship performance parameters; by conducting a risk assessment on the current position of the ship, the collection frequency of the ship performance parameters is dynamically adjusted, so that data can be collected more frequently in high-risk areas, while the collection frequency can be appropriately reduced in low-risk areas, thereby improving the pertinence and effectiveness of data collection, while avoiding excessive data collection when it is not necessary, saving storage space and transmission resources, and optimizing resource utilization.

[0119] In one embodiment, the ocean numerical model includes an extraction layer, a first simulation layer, and a second simulation layer. Figure 4 As shown, step S40 includes the steps of:

[0120] S41: the extraction layer extracts ice zone sequence features, meteorological sequence features and associated sequence features from the ice zone environment data and meteorological environment data based on time series;

[0121] S42: The first simulation layer simulates and predicts iceberg characteristics based on ice zone sequence characteristics and associated sequence characteristics, and outputs iceberg characteristics;

[0122] S43: The second simulation layer simulates and predicts the wave characteristics based on the meteorological sequence characteristics and the associated sequence characteristics, and outputs the wave characteristics;

[0123] In this embodiment, the extraction layer is a component layer in the ocean numerical model for extracting features of ice zone environment data and meteorological environment data based on time sequence; the ice zone sequence feature is a feature reflecting the change of the environment and iceberg state in the ice zone over time; the meteorological sequence feature is a feature reflecting the change of the meteorological environment and wave state over time; the association sequence feature is a feature reflecting the interaction or association between the ice zone environment, iceberg state, meteorological environment and wave state and changing over time; the first simulation layer is a component layer in the ocean numerical model for simulating and predicting the iceberg state in the region and outputting the iceberg features, and the second simulation layer is a component layer in the ocean numerical model for simulating and predicting the wave state in the region and outputting the wave features;

[0124] Specifically, the ocean numerical model includes an extraction layer, a first simulation layer and a second simulation layer. The extraction layer extracts corresponding ice zone sequence characteristics, meteorological sequence characteristics and associated sequence characteristics from the input ice zone environment data and meteorological environment data based on time series analysis. The first simulation layer simulates and predicts iceberg characteristics based on the ice zone sequence characteristics and associated sequence characteristics output by the extraction layer and outputs the iceberg characteristics. The second simulation layer simulates and predicts ocean wave characteristics based on the meteorological sequence characteristics and associated sequence characteristics output by the extraction layer and outputs the ocean wave characteristics.

[0125] In one embodiment, if Figure 5 As shown, step S42 includes the steps of:

[0126] S421: extracting a first key vector from the associated sequence features and the ice zone sequence features, wherein the first key vector includes a flow field velocity vector, a wind field velocity vector, and a velocity vector of the interaction between the iceberg and the sea ice within a preset range;

[0127] S422: Calculate the velocity vector and position vector of the iceberg based on the extracted first key vector: in, is the position vector of the iceberg, is the flow field velocity vector, is the wind velocity vector, is the velocity vector of the iceberg interacting with the sea ice within its preset range;

[0128] S423: calculating the motion characteristics of the iceberg based on the velocity vector and the position vector of the iceberg, and outputting the calculation result as the iceberg characteristics;

[0129] In this embodiment, the first key vector is a set of preset vectors that have a key impact on the simulation and prediction of iceberg characteristics, including a flow field velocity vector, a wind field velocity vector, and a velocity vector of the interaction between the iceberg and the sea ice within its preset range; the flow field velocity vector is a vector reflecting the direction and magnitude of the ocean flow field velocity; the wind field velocity vector is a vector reflecting the direction and magnitude of the wind field velocity; the velocity vector of the interaction between the iceberg and the sea ice within its preset range is a vector reflecting the direction and magnitude of the velocity generated by the interaction between the iceberg and the surrounding sea ice; the velocity vector of the iceberg movement is a vector reflecting the direction and magnitude of the iceberg movement velocity; the position vector of the iceberg is a vector reflecting the position (longitude, latitude, etc.) of the iceberg in the ocean area; the motion characteristics of the iceberg are characteristics reflecting the motion characteristics of the iceberg, including motion trajectory, speed change, etc.;

[0130] Specifically, the first key vector is extracted from the associated sequence characteristics and the ice zone sequence characteristics: the flow field velocity vector, the wind field velocity vector and the velocity vector of the interaction between the iceberg and the sea ice within a preset range, and the velocity vector of the iceberg is calculated based on the three extracted velocity vectors; at the same time, the position vector of the iceberg is calculated based on its velocity vector, that is, the position of the iceberg at a specific point in time; the motion characteristics of the iceberg are further calculated based on the velocity vector and the position vector of the iceberg, and the calculated results are output as iceberg characteristics; the accuracy of the prediction of the motion characteristics of the iceberg is improved by comprehensively considering the flow field, wind field and the interaction between the iceberg and the sea ice.

[0131] In one embodiment, if Figure 6 As shown, step S43 includes the steps of:

[0132] S431: extracting a second key vector from the association sequence features and the meteorological sequence features, wherein the second key vector includes wind speed, tidal height, and ocean current speed;

[0133] S432: Calculate the wave spectral density based on the extracted second key vector:

[0134] Where S(ω) is the wave spectral density, and β are dimensionless constants, g is the acceleration of gravity, U is the wind speed, and ω is the angular frequency;

[0135] S433: Calculate and correct the wave spectral density based on the wave spectral density and the second key vector:

[0136] S modified (ω) = S (ω)·f (T tide ,V current ,ω), where S modified (ω) is the modified wave spectral density, S(ω) is the wave spectral density, f(T tide ,V current ,ω) is the correction function of the wave spectral density, T tide is the tidal height, V current is the ocean current velocity, ω is the angular frequency;

[0137] S434: obtaining wave height characteristics, period characteristics, and propagation characteristics based on the modified wave spectral density, and outputting them as wave characteristics;

[0138] In this embodiment, the second key vector is a set of preset vectors that have a key influence on the simulation and prediction of wave characteristics, including wind speed, tidal height and current speed; wind speed is a physical quantity reflecting the speed and direction of wind; tidal height is a physical quantity reflecting the height of seawater rising and falling due to tidal action; current speed is a physical quantity reflecting the speed and direction of seawater flow; wave spectral density is a physical quantity reflecting the distribution of wave capacity with frequency; dimensionless constant is a preset constant used to adjust the dimension or value in the formula; angular frequency is a physical quantity reflecting the speed of wave fluctuation, which is inversely proportional to the period; corrected wave spectral density is the wave spectral density after combining the influence of tides and currents; the correction function of wave spectral density is a function used to correct the wave spectral density, so that the wave spectral density is corrected in combination with the influence of tides and currents; wave height characteristics are characteristics reflecting the height of wave fluctuations; period characteristics are characteristics reflecting the period of wave fluctuations; propagation characteristics are characteristics reflecting the direction of wave propagation;

[0139] Specifically, the second key vector is extracted from the associated sequence features and the meteorological sequence features: wind speed, tidal height and current speed, and the wave spectral density is calculated based on the wind speed in the extracted second key vector. At the same time, in order to further combine the influence of tidal height and current speed on waves, the wave spectral density is corrected based on a correction function including tidal height and current speed. The corrected wave spectral density further improves the accuracy of the actual wave conditions. The wave height characteristics, period characteristics and propagation characteristics of the waves are calculated based on the corrected wave spectral density, and the calculated characteristics are output as wave characteristics. By comprehensively considering a variety of key factors affecting waves (such as wind speed, tidal height, current speed, etc.), the prediction accuracy of wave characteristics is significantly improved.

[0140] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0141] In one embodiment, a route avoidance system based on icebergs is provided, and the route avoidance system based on icebergs corresponds one-to-one to the route avoidance method based on icebergs in the above embodiment. The route avoidance system based on icebergs includes:

[0142] A route area acquisition module, used to acquire original route information of a target ship and identify a target route area associated with the original route information;

[0143] An environmental data acquisition module, used to acquire ice area environmental data and meteorological environmental data of the target route area;

[0144] A dynamic parameter sequence acquisition module is used to obtain the ship performance parameters of the target ship at intervals and convert them into a dynamic parameter sequence;

[0145] A first data input module is used to input the ice area environment data and the meteorological environment data into a pre-trained ocean numerical model to simulate and predict the characteristics of waves and icebergs;

[0146] A second data input module is used to input the original route information and the dynamic parameter sequence into a pre-trained navigation numerical model to perform navigation characteristic simulation prediction;

[0147] A feature input module, used to input the time-series wave features and iceberg features output by the ocean numerical model, the time-series navigation features output by the navigation numerical model, and the original route information into the risk prediction model;

[0148] A route information adjustment module, configured to dynamically adjust the original route information based on the risk characteristics output by the risk prediction model when receiving the risk characteristics;

[0149] Optionally, the route area acquisition module includes:

[0150] A division submodule, used to divide the route area set based on a preset division rule to obtain a plurality of route sub-areas;

[0151] An R-tree index building submodule is used to build an R-tree index using the obtained several route sub-areas as nodes;

[0152] The fast query submodule is used to convert the original route information into continuous coordinate points and perform fast query on each coordinate point based on the R-tree index;

[0153] An integration submodule, used to integrate the target route sub-areas obtained from the quick query result to obtain the target route area, wherein the target route sub-area includes: a route sub-area intersecting with at least one coordinate point and a route sub-area including at least one coordinate point;

[0154] Optionally, the dynamic parameter sequence acquisition module includes:

[0155] The ship performance parameter acquisition submodule is used to acquire the ship performance parameters of the target ship based on the frequency corresponding to the preset dynamic acquisition rule;

[0156] A timestamp allocation submodule, used to allocate a timestamp of a precision corresponding to the preset dynamic acquisition rule to each group of acquired ship performance parameters;

[0157] A conversion submodule, used for converting the ship performance parameters into a dynamic parameter sequence in time sequence;

[0158] Optionally, also include:

[0159] A risk assessment input module is used to input ice area environmental data and meteorological environmental data of the target route area into a pre-trained risk assessment model, so that the risk assessment model outputs the risk level of each route sub-area in the target route area;

[0160] A dynamic acquisition module is used to obtain the current location information of the target ship and match the corresponding preset dynamic acquisition rules based on the risk level associated with the current location information;

[0161] Optionally, the ocean numerical model includes an extraction layer, a first simulation layer, and a second simulation layer, and further includes:

[0162] An extraction layer module is used to extract ice zone sequence features, meteorological sequence features and associated sequence features from ice zone environment data and meteorological environment data respectively based on time series;

[0163] The first simulation layer module is used to simulate and predict iceberg characteristics based on ice zone sequence characteristics and associated sequence characteristics, and output iceberg characteristics;

[0164] The second simulation layer module is used to simulate and predict the wave characteristics based on the meteorological sequence characteristics and the associated sequence characteristics, and output the wave characteristics;

[0165] Optionally, the first simulation layer module includes:

[0166] A first key vector extraction submodule is used to extract a first key vector from the associated sequence features and the ice zone sequence features, wherein the first key vector includes a flow field velocity vector, a wind field velocity vector, and a velocity vector of the interaction between the iceberg and the sea ice within a preset range;

[0167] The vector calculation submodule is used to calculate the velocity vector and position vector of the iceberg based on the extracted first key vector: in, is the position vector of the iceberg, is the flow field velocity vector, is the wind velocity vector, is the velocity vector of the iceberg interacting with the sea ice within its preset range;

[0168] An iceberg feature output submodule is used to calculate the motion features of the iceberg based on the velocity vector and position vector of the iceberg, and output the calculation results as iceberg features;

[0169] Optionally, the second simulation layer module includes:

[0170] A second key vector extraction submodule, used to extract a second key vector from the associated sequence features and the meteorological sequence features, wherein the second key vector includes wind speed, tidal height and ocean current speed;

[0171] The wave spectral density calculation submodule is used to calculate the wave spectral density based on the extracted second key vector:

[0172] Where S(ω) is the wave spectral density, and β are dimensionless constants, g is the acceleration of gravity, U is the wind speed, and ω is the angular frequency;

[0173] The modified ocean wave spectral density calculation submodule is used to calculate the modified ocean wave spectral density based on the ocean wave spectral density and the second key vector:

[0174] S modified (ω) = S (ω)·f (T tide ,V current ,ω), where S modified (ω) is the modified wave spectral density, S(ω) is the wave spectral density, f(T tide ,V current ,ω) is the correction function of the wave spectral density, T tide is the tidal height, V current is the ocean current velocity, ω is the angular frequency;

[0175] The wave characteristic output submodule is used to obtain wave height characteristics, period characteristics and propagation characteristics based on the modified wave spectral density, and output them as wave characteristics.

[0176] For the specific definition of an iceberg-based route avoidance system, please refer to the definition of an iceberg-based route avoidance method above, which will not be repeated here. Each module in the above-mentioned iceberg-based route avoidance system can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0177] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A route avoidance method based on icebergs, characterized by: Includes steps: Acquire original route information of the target ship, and identify the target route area associated with the original route information; Acquiring ice area environment data and meteorological environment data of the target route area; Obtaining the ship performance parameters of the target ship at intervals and converting them into a dynamic parameter sequence; Inputting the ice area environment data and the meteorological environment data into a pre-trained ocean numerical model to simulate and predict the characteristics of waves and icebergs; Inputting the original route information and the dynamic parameter sequence into a pre-trained navigation numerical model to simulate and predict navigation characteristics; Inputting the time-series wave characteristics and iceberg characteristics output by the ocean numerical model, the time-series navigation characteristics output by the navigation numerical model, and the original route information into the risk prediction model; When the risk characteristics output by the risk prediction model are received, the original route information is dynamically adjusted based on the risk characteristics.

2. The iceberg avoidance method according to claim 1, characterized in that: The step of obtaining the original route information of the target ship and identifying the target route area associated with the original route information comprises the steps of: The route area set is divided based on a preset division rule to obtain a plurality of route sub-areas; Building an R-tree index with the obtained plurality of route sub-areas as nodes; Convert the original route information into continuous coordinate points, and quickly query each coordinate point based on the R-tree index; The target route sub-areas obtained from the quick query results are integrated to obtain the target route area, wherein the target route sub-areas include: route sub-areas intersecting with at least one of the coordinate points and route sub-areas including at least one of the coordinate points.

3. The iceberg avoidance method according to claim 1, characterized in that: The step of obtaining the ship performance parameters of the target ship at intervals and converting them into a dynamic parameter sequence comprises the steps of: Acquire the ship performance parameters of the target ship based on the frequency corresponding to the preset dynamic acquisition rule; Assign a timestamp of the accuracy corresponding to the preset dynamic acquisition rule to each group of acquired ship performance parameters; The ship performance parameters are converted into dynamic parameter sequences in time sequence.

4. The iceberg avoidance method according to claim 3, characterized in that: Before the step of acquiring the ship performance parameters of the target ship based on the frequency corresponding to the preset dynamic acquisition rule, the following steps are performed: Inputting ice area environmental data and meteorological environmental data of the target route area into a pre-trained risk assessment model, so that the risk assessment model outputs the risk level of each route sub-area in the target route area; The current location information of the target ship is obtained, and a corresponding preset dynamic collection rule is matched based on the risk level associated with the current location information.

5. The iceberg avoidance method according to claim 1, characterized in that: The ocean numerical model includes an extraction layer, a first simulation layer and a second simulation layer. The step of inputting ice area environment data and meteorological environment data into the pre-trained ocean numerical model to simulate and predict wave characteristics and iceberg characteristics includes the following steps: The extraction layer extracts ice zone sequence features, meteorological sequence features and associated sequence features from the ice zone environment data and meteorological environment data respectively based on time series; The first simulation layer simulates and predicts iceberg characteristics based on ice zone sequence characteristics and associated sequence characteristics, and outputs iceberg characteristics; The second simulation layer simulates and predicts the wave characteristics based on the meteorological sequence characteristics and the associated sequence characteristics, and outputs the wave characteristics.

6. The iceberg-based route avoidance method according to claim 5, characterized in that: The first simulation layer simulates and predicts iceberg characteristics based on ice zone sequence characteristics and associated sequence characteristics, and outputs the iceberg characteristics, including the steps of: Extracting a first key vector from the associated sequence features and the ice zone sequence features, wherein the first key vector includes a flow field velocity vector, a wind field velocity vector, and a velocity vector of the interaction between the iceberg and the sea ice within a preset range; Calculate the velocity vector and position vector of the iceberg based on the extracted first key vector: in, is the position vector of the iceberg, is the flow field velocity vector, is the wind field velocity vector, is the velocity vector of the iceberg interacting with the sea ice within its preset range; The motion characteristics of the iceberg are calculated based on the velocity vector and the position vector of the iceberg, and the calculation results are output as iceberg characteristics.

7. The iceberg-based route avoidance method according to claim 5, characterized in that: The second simulation layer simulates and predicts the wave characteristics based on the meteorological sequence characteristics and the associated sequence characteristics, and outputs the wave characteristics, including the steps of: Extracting a second key vector from the association sequence features and the meteorological sequence features, wherein the second key vector includes wind speed, tidal height, and ocean current speed; Calculate the wave spectral density based on the extracted second key vector: Where S(ω) is the wave spectral density, and β are dimensionless constants, g is the acceleration of gravity, U is the wind speed, and ω is the angular frequency; Calculate the corrected wave spectral density based on the wave spectral density and the second key vector: S modified (ω) = S (ω)·f (T tide ,V current ,ω), where S modified (ω) is the modified wave spectral density, S(ω) is the wave spectral density, f(T tide ,V current ,ω) is the correction function of the wave spectral density, T tide is the tidal height, V current is the current velocity, ω is the angular frequency; The wave height characteristics, period characteristics and propagation characteristics are obtained based on the modified wave spectral density and output as wave characteristics.

8. An iceberg-based route avoidance system, characterized in that: include: A route area acquisition module, used to acquire original route information of a target ship and identify a target route area associated with the original route information; An environmental data acquisition module, used to acquire ice area environmental data and meteorological environmental data of the target route area; A dynamic parameter sequence acquisition module is used to obtain the ship performance parameters of the target ship at intervals and convert them into a dynamic parameter sequence; A first data input module is used to input the ice area environment data and the meteorological environment data into a pre-trained ocean numerical model to simulate and predict the characteristics of waves and icebergs; A second data input module is used to input the original route information and the dynamic parameter sequence into a pre-trained navigation numerical model to perform navigation characteristic simulation prediction; A feature input module, used to input the time-series wave features and iceberg features output by the ocean numerical model, the time-series navigation features output by the navigation numerical model, and the original route information into the risk prediction model; The route information adjustment module is used to dynamically adjust the original route information based on the risk characteristics output by the risk prediction model when the risk characteristics are received.

Images