Risk assessment method and system for safe navigation of ship

Abstract

The invention relates to a risk assessment method and system for safe navigation of a ship, and the method comprises the following steps: risk perception: obtaining the original data of the natural environment of a polar region, and carrying out the preprocessing; performing risk cognition: performing systematic analysis on risk influence factors of the navigation ship in the polar region water area by applying a system theory accident model and process, taking the risk influence factors as nodes, obtaining a transition rate between the nodes according to the preprocessed original data, and constructing a random Petri network model; risk control: sorting the nodes in the model according to the steady-state probability by using a Markov chain, and carrying out risk control processing on the nodes with low steady-state probability; and risk decision-making: generating an air route and a risk control scheme in the polar navigation water area. According to the method, the informatization technology is adopted to perceive risk factors to reveal polar region environment space features, the complex system safety theory is applied to research polar region water area ship navigation risks, and precise risk cognition is achieved by combining subjective and objective data fusion.

Description

Technical Field

[0001] The present invention relates to the field of water transportation technology, and in particular to a risk assessment method and system for safe navigation of ships. Background Art

[0002] In the polar shipping navigation risk assessment, with the advent of the information age, a large amount of accident data, environmental data, ship data and other information can be used in the study of maritime traffic risks, providing experts with large-scale and high-precision research basis.

[0003] The development of Arctic shipping routes has provided new options for the shipping industry. The risk of ship navigation in Arctic waters is currently a hot topic in the research of Arctic ship safety. There are many cutting-edge technologies and methods in the field of maritime traffic safety to study and analyze the risk of ship navigation, including traditional risk analysis techniques such as event tree analysis, fault tree analysis, quantitative risk assessment, and comprehensive safety assessment methods. When scholars identify risk factors for navigation in Arctic waters, they often combine fuzzy theory with the analytic hierarchy process to quantify the sources of danger based on expert judgment or accidents. In risk analysis and assessment, appropriate algorithms or models are selected according to different research objects and research scenarios, mainly Bayesian networks, accident trees, bowtie methods, system dynamics, etc.

[0004] Among them, the Formal Safety Assessment (FSA) method provides a structured, systematic and standardized method of risk management to control maritime traffic risks. The Formal Safety Assessment method has been widely used in the industry, and it uses the crew's experience as a basis to identify the dangerous sources that may affect the safety of ships and personnel during navigation. It uses different qualitative and quantitative analysis methods to rank the importance of dangerous sources, and takes factors with high probability of occurrence and serious consequences as the primary control options. It also formulates risk control plans based on the loss of personnel and property caused by these factors, and finally determines the decision-making plan to reduce the risk. Although this method can be effectively applied to the safety assessment of ship navigation in polar waters, it has the following shortcomings: First, the engineering problem is not solved enough, and the research results are only management countermeasures and strategies; second, the complexity of the ship traffic system is not involved, and the emergence from the mutation, synergy and chaos of the complexity of the system is not involved; third, the response to the numbers and data of the ship system is weak, and the knowledge discovery caused by machine learning cannot be emphasized, and too much depends on expert or empirical knowledge.

[0005] In summary, there is currently a lack of a risk assessment method and system for safe navigation of ships to solve or partially solve the above-mentioned problems. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a risk assessment method and system for safe navigation of ships, so as to solve or partially solve the problem of unsatisfactory digital and data response sensitivity of ship systems.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] One aspect of the present invention provides a method for risk assessment of safe navigation of a ship, comprising the following steps:

[0009] Step S1, risk perception: obtaining the original data of polar natural environment and preprocessing it;

[0010] Step S2, risk recognition: using the system theory accident model and process to systematically analyze the risk influencing factors of ships navigating in polar waters, taking the risk influencing factors as nodes, obtaining the transition rate between nodes based on the preprocessed raw data, and constructing a random Petri net model;

[0011] Step S3, risk control: for the random Petri net model, use the Markov chain to sort the nodes in the model according to the steady-state probability, and perform risk control processing on the nodes with low steady-state probability;

[0012] Step S4, risk decision: generate routes and risk control plans in polar navigable waters.

[0013] As a preferred technical solution, the pretreatment includes the following steps:

[0014] Step S101, for the original data of the natural environment, fill in the missing parts in the original data by binomial interpolation;

[0015] Step S102, using the extreme value standard method to process the filled original data.

[0016] As a preferred technical solution, in step S2, the process of using the system theory accident model and process to systematically analyze the risk influencing factors of ships navigating in polar waters includes the following steps:

[0017] Step S201, using the ship's own environmental monitoring equipment and hull monitoring equipment as sensors, the ship's command personnel and operators and the ship's automatic control algorithm as controllers, and the system for controlling the ship's speed and direction as an actuator, a system theoretical accident model is constructed and analyzed.

[0018] As a preferred technical solution, in step S2, the construction process of the random Petri net model includes the following steps:

[0019] Step S202, based on the system theory accident model and the analysis results of the process, the ship navigation process under the time-space sequence is modeled using the Petri net model;

[0020] Step S203, setting the initial transition value of the Petri net model.

[0021] As a preferred technical solution, the Petri net model is modeled as follows:

[0022] SPN=(P,T,F,W,M 0 ,λ)

[0023] Among them, SPN represents the six-tuple of the Petri net model, P represents the discrete state of the ship during the navigation process, that is, the library, T represents the ship state transition set, that is, the transition, represents the evolution direction of the state during the system movement, that is, the directed arc set, and W represents F→N + is the arc weight function, N + is a set of positive integers, M represents the set of all states generated by the ship during the conversion process, and λ represents the set of average implementation rates of ship state transitions.

[0024] As a preferred technical solution, the calculation process of the steady-state probability includes the following steps:

[0025] Step S301, based on the system theory accident model and the analysis results of the process, construct the isomorphism of the random Petri net model and the Markov chain;

[0026] Step S302, obtaining an evaluation index of the impact on navigation risk, and obtaining the transition rate of each node in the random Petri net model;

[0027] Step S303, based on the transition rate of each node in the random Petri net model, the steady-state probability value of each node is obtained.

[0028] As a preferred technical solution, in step S3, risk control processing is performed on nodes with low steady-state probability, including icebreaker assistance or route selection.

[0029] As a preferred technical solution, in step S4, generating routes and risk control plans in polar navigable waters includes the following steps:

[0030] Step S401, determining the navigation window period, route and navigation mode of the ship according to the ship's ice class;

[0031] Step S402, making decisions to recommend routes and risk control plans in polar navigable waters.

[0032] As a preferred technical solution, the natural environment raw data includes sea ice thickness, sea ice density, temperature, wind speed and wave height.

[0033] Another aspect of the present invention provides a risk assessment system for safe navigation of ships, comprising:

[0034] The risk perception module is used to obtain the original data of the polar natural environment and perform preprocessing;

[0035] The risk recognition module is used to systematically analyze the risk influencing factors of ships sailing in polar waters using the system theory accident model and process. The risk influencing factors are used as nodes, and the change rate between nodes is obtained based on the pre-processed raw data to construct a random Petri net model.

[0036] A risk control module is used to sort the nodes in the random Petri net model according to the steady-state probability by using a Markov chain, and perform risk control processing on the nodes with low steady-state probability;

[0037] Risk decision module, used to generate routes and risk control plans in polar navigable waters.

[0038] Compared with the prior art, the present invention has at least one of the following beneficial effects:

[0039] (1) Use information technology to perceive risk factors and reveal the spatial characteristics of the polar environment. Use interpolation algorithms to describe environmental risk characteristics, use information technology to intuitively present the monthly distribution of various indicators in the original data of the polar natural environment, and achieve a preliminary analysis of the risk situation, so as to better complete the qualitative risk modeling.

[0040] (2) Using the complex system safety theory to study the navigation risks of ships in polar waters. This paper uses the STAMP model under the complex system safety theory to explain the correlation and state change trend of internal and external factors of the navigation system of ships, and clarifies the functions of various stakeholders in the navigation system of ships in polar waters and their respective roles when risks are presented.

[0041] (3) Integration of subjective and objective data for precise risk recognition. Maritime traffic risk research mainly relies on experienced sailors or highly qualified experts to identify and analyze hazardous sources. After the introduction of qualitative and quantitative models, due to the lack of accident data, risk assessment is achieved through expert scoring to quantify risk. The problem is that the data is too subjective, and the ranking of factors also requires the selection of a suitable weighting algorithm. The reliability and accuracy of the results need to be verified. Therefore, it is necessary to reduce the randomness and ambiguity of the data to a certain extent through data processing. Objective data is large in volume and easy to obtain. The combination of expert evaluation and objective data can reduce the subjectivity of risk analysis and further improve the feasibility and reliability of ship navigation risk decision-making.

[0042] (4) The digital and data response of the ship system is sensitive: By constructing a random Petri net model to evaluate the risk of safe navigation, combined with the system theory accident model and process analysis, the risk influencing factors are used as nodes of the random Petri net model, and the change rate between nodes is obtained based on the processed original data. By changing the change rate, dynamic risk assessment under different times and channels can be achieved. Finally, the comprehensive risk value of different ships sailing on a certain channel within a certain period of time can be obtained, which improves the sensitivity to digital and data response. The visualized data shows that it can not only show the performance of risk assessment of ship navigation plans under different routes and seasons, but also characterize the differences in risk levels during ship navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Flow chart of the risk assessment method for safe navigation of ships in the embodiment;

[0044] Figure 2 It is a visualization diagram of environmental data in the embodiment;

[0045] Figure 3 This is a schematic diagram of the navigation environment data distribution and risk perception of the Northeast Passage in the embodiment;

[0046] Figure 4 It is a schematic diagram of the STAMP model in the embodiment;

[0047] Figure 5 A schematic diagram of a random Petri net model in an embodiment;

[0048] Figure 6 It is a schematic diagram of a navigation risk control scheme isomorphic to a random Petri net and a Markov chain in the embodiment;

[0049] Figure 7 The random Petri net operation result in the embodiment;

[0050] Figure 8Schematic diagram of the navigation standards for the Arctic Northeast Passage for different ships in the embodiment;

[0051] Fig. 9 Schematic diagram of the golden waterway and golden window period in the embodiment;

[0052] Fig.10 Schematic diagram of the risk assessment system for safe navigation of ships in the embodiment. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0054] Example 1

[0055] In view of the problems existing in the aforementioned prior art, this embodiment provides a risk assessment method for safe navigation of ships, which aims to use information technology to perceive the risks of the navigation environment in polar waters, and use risk perception factors to achieve precise risk cognition under the fusion of subjective and objective data to obtain risk control plans. At the same time, the impact on the safety of polar ship navigation is used as the evaluation standard to form risk decisions on ship navigation routes for spatiotemporal planning in polar waters.

[0056] This method is based on polar environmental big data and takes the objective data visualization results as the analysis basis for the risk characteristics of polar ship navigation. It combines objective data such as accident reports, environmental information and ship parameters with subjective data of crew experience to establish an enhanced risk assessment model for ship navigation in polar waters, completes the ship navigation risk analysis under multi-factor coupling, and derives risk control plans and risk decision-making recommendations based on the spatiotemporal characteristics of risk perception.

[0057] The polar waters are complex and changeable. Compared with other sea areas, the main problems faced by ships in navigation are the harsh environment of sea ice, low temperature and other high-latitude characteristics. Currently, many countries have been monitoring and recording sea ice conditions. Through the website of the European Climate Data Store (CDS), sea ice density and thickness data with different accuracies in the past two decades can be directly obtained, as well as wind speed at 10m above sea level, temperature at 2m above sea level and significant height of combined wind waves and swells. The risk assessment method for ships in polar waters consists of four steps, such as Figure 1 As shown, the method comprises the following steps:

[0058] S1: Risk perception.

[0059] S101: Obtain the original polar natural environment data from the data website, and use MATLAB software and binomial interpolation algorithm to complete the missing parts of the original data, as shown in Formula 1.

[0060]

[0061] where x 0 ,x 1 ,…,x n For a given discrete data point, f(x 0 ),f(x 1 ),…,f(x n ) are the function values ​​corresponding to these points. n represents the degree of the interpolation polynomial, and x is the point to be interpolated.

[0062] S102: Use the extreme value normalization method to perform normalization processing to eliminate dimensional differences and retain the original data relationship, such as formula 2, and then complete the visualization of objective environmental data such as sea ice thickness, sea ice density, temperature, and wind speed.

[0063] Sea ice density W:

[0064]

[0065] W' is the normalized sea ice density value, expressed in %, W is the sea ice density in the assessment area for each month, and W min , W max are the minimum and maximum values ​​of sea ice concentration during the assessment period.

[0066] 2m temperature T:

[0067]

[0068] T is the original temperature value of the evaluation unit, T min is the minimum temperature of the unit;

[0069] 10m wind speed V:

[0070]

[0071] From the perspective of energetics, the wind intensity in the evaluation unit is determined by the wind speed. V' is the wind intensity index, V is the wind speed in the evaluation area for each month, and V min This is the minimum standard value for strong winds. According to the Beaufort wind scale, the minimum standard for strong winds is 10.8m / s.

[0072] Wave height P:

[0073]

[0074] Based on the characteristics of the visualization map, the polar waters’ non-navigable channels throughout the year, channels navigable during part of the year, and channels navigable throughout the year are determined.

[0075] Step 2 S2: Risk perception.

[0076] S201: Use the System-Theoretic Accident Model and Processes (STAMP) to systematically analyze the risk factors affecting ships sailing in polar waters. The system is mainly divided into three parts: sensors, controllers, and actuators. Sensors mainly include environmental monitoring equipment and hull monitoring equipment on the ship itself; controllers mainly include various commanders and operators on board, as well as ship automatic control algorithms; actuators include systems that control the speed and direction of the ship, such as the main propulsion system, power system, and cooling water system.

[0077] S202: Based on the theoretical analysis of the accident model and process of system theory, a control process model of the risk system of polar water ship navigation is constructed, and a quantitative risk assessment of polar water ship navigation is proposed using the Stochastic Petri net, forming the mechanism of the ship navigation process under the time-space sequence. This method takes the risk influencing factors as the nodes of the stochastic Petri net model, obtains the transition rate between nodes based on the processed original data (these data include both objective data from the environment and subjective data from expert experience), and realizes dynamic risk assessment under different times and channels by changing the transition rate. Finally, the comprehensive risk value of different ships sailing in a certain channel within a certain time period can be obtained.

[0078] The specific mathematical assumptions are as follows: SPN = (P, T, F, W, M 0 ,λ) is represented as a sextuple.

[0079] P = {P1, P2, ..., Pm}, m>0 represents the discrete state of the ship during the navigation process;

[0080] T = {T1, T2, …, Tn}, n is the number of transitions n>0, representing the set of ship state transitions, which can be expressed as the connection between complex system states;

[0081] It indicates the evolution direction of the state during the system movement, and there is a set of directed arcs between places and transitions;

[0082] W means F→N + is the arc weight function, N + is a set of positive integers; M 0Represents the initial state identification. The evolution of the state over time is a random process, also called the identification process.

[0083] M={M 1 ,M 2 ,…,M j}, j>0, which represents the set of all states generated by the ship during the transformation process.

[0084] λ={λ 1 ,λ 2 ,…,λ n} represents the average implementation rate set of ship state transitions. In the model, each transition T has a corresponding λ.

[0085] Step 3 S3: Risk control plan.

[0086] S301: Performance analysis of the random Petri net model in S202. Using the isomorphic properties of the state space and the Markov chain, we can achieve:

[0087] (1) A transition is enabled if each of its input places (there is a directed arc from the place to the transition) contains at least one label.

[0088] (2) The activation of an enabling transition will cause the mark in each of its input places to decrease, and the mark in each of its output places (where there is a directed arc to the place) to increase.

[0089] The initiation of each transition specifies a random time delay parameter, which usually requires that the associated continuous-time random variable be exponentially distributed or Weibull distributed.

[0090] S302: Based on the data isomorphism of the risk of ship navigation in polar waters obtained in the previous steps, the evaluation results (i.e., the quantitative degree of risk perception) can be obtained through model operation and calculation, and the nodes in the model are sorted by steady-state probability. The high and low steady-state probability indicates the risk impact of the node on the navigation process of ships. Control plans are taken for nodes with greater risks affecting the navigation of different types of ships, such as whether icebreaker assistance or route selection is needed, and further risk re-cognition is carried out to ensure that the resulting risk control plan is reasonable and feasible.

[0091] Step 4 S4: Risk decision.

[0092] S401: Different navigation standards for ships in polar waters. Due to the different ice classes of ships, the navigation standards for ships in polar waters require the determination of the window period for ship navigation, routes and navigation methods for safe navigation.

[0093] S402: Combining risk perception, risk cognition and risk control plans, make decisions and recommend the best routes and risk control plans in polar navigable waters for ships in different ice classes and different navigation windows.

[0094] The implementation process of this method is explained below using the Northeast Passage from June to November 2022 as an example. First, data preprocessing is performed on wind, waves, temperature, and sea ice density. The data sources and data accuracy are shown in Table 1.

[0095] Table 1 Data medium

[0096]

[0097] Based on the obtained four types of NC data, formulas (1)-(5) are used for interpolation processing, and the polar coordinate projection is visualized. The temperature at 2 m above sea level, sea ice density, wind speed at 10 m above sea level, and effective wave height are visualized as follows: Figure 2 As shown, Figure 2 The upper middle part is the data for July 2022, and the lower part is the data for October 2022. From left to right are sea ice density (in percentage), temperature at 2 meters above sea level (in °C), wind speed at 10 meters above sea level (in m / s), and effective wave height (in m).

[0098] According to the extracted latitude and longitude data of the Northeast Passage in the Arctic, the navigation environment data such as temperature, sea ice density, wind speed and wave height in the database are searched and extracted based on Matlab software to obtain the corresponding navigation environment risk index data value at each coordinate position on the Northeast Passage. For the obtained data values, the distribution of various indicators of the Arctic route in different time periods is displayed by drawing a box plot, such as Figure 3 The distribution of the values ​​of various navigation environment factors in the Arctic Northeast Passage in July is shown. The upper figure shows the risk assessment results calculated under the navigation environment and the routes of ships, and the lower figure is the navigation environment risk map of the entire region.

[0099] By analyzing the environmental information of the Arctic Northeast Passage and combining it with information such as ship structure and personnel allocation, a STAMP model for the navigation risk of ships in Arctic waters is constructed. Figure 4 The model is mainly divided into three parts, namely, sensor, controller and processor. The sensor is mainly composed of ship monitoring equipment and systems, the processor includes shipping companies, captains, crew members and management, and the controller is various hardware systems that control the ship's heading and speed.

[0100] The random Petrie network technology is introduced, and based on the weighted analysis method, different navigation environment risk indicators and temperature, sea ice density, wind speed, and wave height are assigned weights of [0.25 0.35 0.15 0.25] to calculate the navigation environment risk value of the Northeast Passage in the Arctic waters. Figure 5 As shown. Combining the comprehensive environmental risk results and expert opinions, the initial transition value of the stochastic Petri net model was set. The original data of environmental risk indicators of the Northeast Passage in different months were extracted by taking the median method, and then converted into the transition rate values ​​between the corresponding indicator impacts of the stochastic Petri net model.

[0101] Based on the qualitative analysis results of the STAMP model, the isomorphism between the stochastic Petri net and the Markov chain of the navigation risk of ships in Arctic waters was constructed. Figure 6 As shown. Through the evaluation and scoring of the impact of ship hardware equipment and onboard personnel on the risk of ship navigation in Arctic waters by multiple experts, the change rates of the remaining nodes were obtained. By running the Pipe software, the steady-state probability value of each node can be directly obtained, among which the comprehensive risk value of ship navigation from June to November is [0.18223 0.179330.1642 0.15816 0.17716 0.18583]. Figure 7 As shown in the figure, the upper left figure shows the navigation risk level of ships in the Northeast Passage I in polar waters in different months, the upper right figure shows the navigation risk level of ships in the Northwest Passage I in polar waters in different months, the middle left figure shows the navigation risk level of ships in the Northeast Passage II in polar waters in different months, the middle right figure shows the navigation risk level of ships in the Northwest Passage II in polar waters in different months, and the lower middle figure shows the navigation risk level of ships at different locations in the Northwest Passage I in polar waters in September. From the data visualization diagram, it can be seen that the comprehensive navigation risk of the Northeast Passage in August and September 2022 is lower than that of the rest of the months. The steady-state probability of the environmental factor nodes is generally higher than that of other nodes, among which the steady-state probability of sea ice density is the highest, indicating that ships need to pay special attention to sea ice factors when sailing in the Northeast Passage.

[0102] According to the risk perception results, the risk control standard is established, that is, when the comprehensive risk value is lower than 0.17, the navigation risk level is considered to be low; when the comprehensive risk value is between 0.17 and 0.18, the navigation risk level is considered to be medium; when the comprehensive risk value is higher than 0.18, the navigation risk level is considered to be high. Figure 8As shown, the two columns on the left are data from 2013, and the two columns on the right are data from 2020. ICC(max) is the annual maximum sea ice density, ICC(min) is the annual minimum sea ice density, ICM(max) is the annual maximum sea ice thickness, ICM(min) is the annual minimum sea ice thickness, ICF(max) is the annual maximum sea ice composite, and ICF(min) is the annual minimum sea ice composite. Therefore, when ships sail in the Arctic Northeast Passage, the navigation risk level in August and September is low, and the sea ice coverage is low, so they can sail freely. Fig. 9 As shown in the figure, the shaded area is the ice control area, and the straight line represents the optional route. In general, it can be considered that the navigation risk level in July and October is medium, and there is sea ice cover in some navigation areas. Icebreakers can be selected to assist navigation depending on the situation. The navigation risk in June and November is high, and most waters are covered with sea ice. It is recommended to use icebreakers to assist navigation.

[0103] This method has the following beneficial effects:

[0104] (1) Use information technology to perceive risk factors and reveal the spatial characteristics of the polar environment.

[0105] An interpolation algorithm is used to describe the environmental risk characteristics of polar waters within 6 months, and information technology is used to intuitively present the monthly distribution of wind, waves, temperature and sea ice conditions in polar waters, to achieve a preliminary analysis of the risk situation and better complete qualitative risk modeling.

[0106] (2) Apply complex system safety theory to study the navigation risks of ships in polar waters.

[0107] This embodiment uses the STAMP model under the complex system safety theory to illustrate the correlation and state change trend of internal and external factors of the navigation ship system, and clarifies the functions of various stakeholders in the polar waters ship navigation system and their respective roles when risks arise.

[0108] (3) Integration of subjective and objective data for accurate risk recognition

[0109] The research on maritime traffic risks mainly relies on experienced sailors or highly qualified experts to identify and analyze the sources of danger. After the introduction of qualitative and quantitative models, due to the lack of accident data, risk assessment is achieved through expert scoring to quantify risks. The problem is that the data is too subjective, and the ranking problem between factors also requires the selection of a suitable weight algorithm. The reliability and accuracy of the results need to be verified. This embodiment reduces the randomness and ambiguity of the data to a certain extent through data processing. With the rapid development of information technology, the volume of objective data is large and easy to obtain. The combination of expert evaluation and objective data can reduce the subjectivity of risk analysis and further improve the feasibility and reliability of ship navigation risk decision-making.

[0110] Example 2

[0111] Based on Example 1, see Fig.10 (Workflow and system operation requirements), this embodiment provides a risk assessment system for safe navigation of ships, including:

[0112] The risk perception module is used to obtain the original data of the polar natural environment and perform preprocessing;

[0113] The risk recognition module is used to systematically analyze the risk influencing factors of ships sailing in polar waters using the system theory accident model and process. The risk influencing factors are used as nodes, and the change rate between nodes is obtained based on the pre-processed raw data to construct a random Petri net model.

[0114] The risk control module is used to sort the nodes in the random Petri net model according to the steady-state probability using the Markov chain, and to perform risk control on the nodes with low steady-state probability;

[0115] Risk decision module, used to generate routes and risk control plans in polar navigable waters.

[0116] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A risk assessment method for safe navigation of ships, characterized in that: The steps include: Step S1, risk perception: obtaining the original data of the polar natural environment and performing preprocessing; Step S2, risk recognition: using the system theory accident model and process to systematically analyze the risk influencing factors of ships navigating in polar waters, taking the risk influencing factors as nodes, obtaining the transition rate between nodes based on the preprocessed raw data, and constructing a random Petri net model; Step S3, risk control: for the random Petri net model, use the Markov chain to sort the nodes in the model according to the steady-state probability, and perform risk control processing on the nodes with low steady-state probability; Step S4, risk decision: generate routes and risk control plans in polar navigable waters.

2. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: In the step S1, the preprocessing includes the following steps: Step S101, for the original data of the natural environment, fill in the missing parts of the original data by binomial interpolation; Step S102, using the extreme value standard method to process the filled original data.

3. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: In step S2, the process of using the system theory accident model and process to systematically analyze the risk influencing factors of ships navigating in polar waters includes the following steps: Step S201, using the ship's own environmental monitoring equipment and hull monitoring equipment as sensors, the ship's command personnel and operators and the ship's automatic control algorithm as controllers, and the system for controlling the ship's speed and direction as an actuator, a system theoretical accident model is constructed and analyzed.

4. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: In step S2, the construction process of the random Petri net model includes the following steps: Step S202, based on the system theory accident model and the analysis results of the process, the ship navigation process under the time-space sequence is modeled using the Petri net model; Step S203, setting the initial transition value of the Petri net model.

5. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: The Petri net model is modeled as: SPN=(P,T,F,W,M0,λ) Among them, SPN is a six-tuple representing the Petri net model, P represents the discrete state of the ship during the navigation process, that is, the library, T represents the ship state transition set, that is, the transition, represents the evolution direction of the state during the system movement, that is, the directed arc set, and W represents F→N + is the arc weight function, N + is a set of positive integers, M represents the set of all states generated by the ship during the conversion process, and λ represents the set of average implementation rates of ship state transitions.

6. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: The calculation process of the steady-state probability comprises the following steps: Step S301, based on the system theory accident model and the analysis results of the process, construct the isomorphism of the random Petri net model and the Markov chain; Step S302, obtaining an evaluation index of the impact on navigation risk, and obtaining the transition rate of each node in the random Petri net model; Step S303, based on the transition rate of each node in the random Petri net model, the steady-state probability value of each node is obtained.

7. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: In step S3, risk control processing is performed on nodes with low steady-state probability, including icebreaker assistance or route selection.

8. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: In step S4, generating routes and risk control plans in polar navigable waters includes the following steps: Step S401, determining the navigation window period, route and navigation mode of the ship according to the ship's ice class; Step S402, making decisions to recommend routes and risk control plans in polar navigable waters.

9. A risk assessment method for safe navigation of a ship according to claim 1, characterized in that: The natural environment raw data include sea ice thickness, sea ice density, temperature, wind speed and wave height.

10. A risk assessment system for safe navigation of ships, characterized in that: include: The risk perception module is used to obtain the original data of the polar natural environment and perform preprocessing; The risk recognition module is used to systematically analyze the risk influencing factors of ships sailing in polar waters using the system theory accident model and process. The risk influencing factors are used as nodes, and the change rate between nodes is obtained based on the pre-processed raw data to construct a random Petri net model. A risk control module is used to sort the nodes in the random Petri net model according to the steady-state probability by using a Markov chain, and perform risk control processing on the nodes with low steady-state probability; Risk decision module, used to generate routes and risk control plans in polar navigable waters.

Images