A method and system for dynamically assessing offshore wind power construction risks based on a Bayesian network, a computer device, and a storage medium

Abstract

The application provides a kind of offshore wind power construction risk dynamic assessment method, system based on bayesian network, computer equipment and storage medium, the method comprises the following steps: S1: risk factor identification is carried out to various operation activities in offshore wind power construction;S2: the risk factors identified, failure consequence are classified;S3: offshore wind power construction operation failure evolution model is constructed using bayesian network method;S4: according to risk factor data source, it is divided into two categories;S5: the occurrence probability calculated by mathematical statistics and fuzzy set theory is used as input, realizes the dynamic risk assessment of offshore wind power construction operation. Through detailed job safety analysis (JSA) method, the risk factors of various operation activities in offshore wind power construction are comprehensively identified and classified, the offshore wind power construction risk dynamic assessment model based on bayesian network is constructed, the occurrence probability of risk factor can be updated in real time, which helps to improve the safety and efficiency of offshore wind power construction.

Description

Technical Field

[0001] This invention relates to the field of offshore new energy engineering technology, specifically to a method, system, computer equipment, and storage medium for dynamic risk assessment of offshore wind power construction based on Bayesian networks. Background Technology

[0002] With the accelerating pace of global energy structure transformation, offshore wind power, as an important component of clean energy, has ushered in unprecedented development opportunities. Coastal provinces are ramping up the construction of offshore wind power projects, with an estimated new installed capacity exceeding 60 gigawatts, forming a new construction boom.

[0003] However, the implementation environment of offshore wind power projects is complex and changeable, easily affected by various factors such as marine meteorological conditions, hydrogeology, and engineering technology, posing significant challenges to construction. In recent years, due to the relatively late start of offshore wind power construction, related safety management experience and risk control technologies are relatively scarce, leading to several serious safety accidents. Furthermore, traditional methods for assessing safety risks in offshore wind power construction largely rely on qualitative, experience-based judgments, failing to achieve real-time tracking and quantitative analysis of dynamic risk changes. Especially during construction, various uncertainties intertwine and overlap, making traditional assessment methods ill-suited to this rapidly changing environment and unable to provide timely and effective support for safety management decisions. Summary of the Invention

[0004] The main objective of this invention is to provide a dynamic risk assessment method for offshore wind power construction based on Bayesian networks, addressing the aforementioned problems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A dynamic risk assessment method for offshore wind power construction based on Bayesian networks includes the following steps:

[0007] S1: Identify risk factors for various operational activities in offshore wind power construction;

[0008] S2: Classify the identified risk factors and categorize the consequences of failure;

[0009] S3: A failure evolution model for offshore wind power construction operations is constructed using the Bayesian network method. Based on the classification of risk factors and consequences, a Bayesian network model of individual operational influencing factors and consequences is established, and sub-Bayesian networks are added to the main Bayesian network to form a complete Bayesian network model.

[0010] S4: Based on the source of risk factor data, they are divided into two categories;

[0011] S5: Using the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory as input, the forward inference capability of Bayesian network is used to analyze the failure probability of offshore wind power construction operations, predict the completion probability of each step in the operation process, and transmit it in the constructed Bayesian network, thereby realizing dynamic risk assessment of offshore wind power construction operations.

[0012] While adopting the above technical solutions, the present invention may also adopt or combine the following technical solutions:

[0013] As a preferred technical solution of the present invention: the risk factor identification in step S1 includes six main processes: transportation, pile foundation construction, wind turbine installation, offshore substation construction, submarine cable laying, commissioning and trial operation.

[0014] As a preferred technical solution of the present invention: in step S2, the risk factors are classified into four aspects: personnel, environment, equipment and management.

[0015] As a preferred technical solution of the present invention: the failure consequences in step S2 are classified into six aspects: personnel injury and death, equipment damage, interruption of installation work, collision damage to the hull, capsizing and sinking of the hull, and fire.

[0016] As a preferred technical solution of the present invention: in step S4, for risk factors with clear data sources, statistical analysis methods are used to calculate their frequency of occurrence, and the calculation formula is as follows:

[0017]

[0018] In equation (1), N A This indicates the number of times a risk with a clearly defined data source occurs, where N represents the total number of risk factors with clearly defined data sources.

[0019] As a preferred technical solution of the present invention: In step S4, for risk factors without a clear data source, the probability is determined by expert evaluation and fuzzy set theory, and includes the following steps:

[0020] S41: Use expert fuzzy terminology to define the probability of risk factors occurring. Through questionnaires, invite front-line staff and experts to assess the probability of specific risk factors occurring. The experts invited to participate in the risk factor probability assessment need to be very familiar with the site conditions and the actual situation of each risk factor occurring at the construction site.

[0021] S42: Convert expert fuzzy terms into fuzzy numbers and aggregate the opinions of multiple experts. Depending on the actual situation of the project, three or more experts may be invited to assess the probability of the occurrence of risk factors. The analytic hierarchy process (AHP) is used to calculate the weights of the experts, and a weighted average method is employed to consider the opinions of multiple different experts. The formula is as follows:

[0022]

[0023] In equation (2), P i Let P be the fuzzy number aggregated for event i. i,j Let m represent the fuzziness of expert j's judgment on event i, and m be the total number of experts.

[0024] S43: Defuzzification process: Convert the aggregated fuzzy numbers into fuzzy probability scores (FPS), and convert the FPS into event occurrence probabilities. Use the weighted average method to aggregate the fuzzy numbers judged by different experts, and apply the maximum-minimum aggregation method for defuzzification process to convert the fuzzy numbers into probability values. The maximum and minimum fuzzy sets are as follows.

[0025]

[0026] The fuzzy probability score of the fuzzy number P is determined by the following formula:

[0027] FPS(P i ) = [FPS Right (P i +1-FPS Left (P i )] / 2 (5)

[0028] The fuzzy probability score is converted into a probability value using equations (3) and (4):

[0029]

[0030] k = 2.301 × [(1-FPS) / FPS] 1 / 3 (7).

[0031] A dynamic risk assessment system for offshore wind power construction based on Bayesian networks, the system comprising the following modules:

[0032] The risk factor identification module is used to identify risk factors that occur during transportation, pile foundation construction, wind turbine installation, offshore substation construction, submarine cable laying, commissioning and trial operation.

[0033] The risk factors and failure consequences classification module classifies risk factors into four aspects: personnel, environment, equipment and management, and failure consequences into six aspects: personnel injury and death, equipment damage, installation work interruption, hull collision damage, hull capsizing and sinking, and fire.

[0034] The probability calculation module is used to calculate the probability of occurrence of risk factors with clear data sources and risk factors without clear data sources;

[0035] The offshore wind power construction operation risk assessment module is used to assess the risk index of offshore wind power construction operations by combining the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory.

[0036] A computer device includes a memory and a processor, characterized in that: the memory stores a computer program, and the processor is configured to run the computer program to execute the aforementioned method for dynamic risk assessment of offshore wind power construction based on Bayesian networks.

[0037] A computer-readable storage medium is characterized in that: the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the aforementioned method for dynamic risk assessment of offshore wind power construction based on Bayesian networks.

[0038] This invention provides a method, system, computer equipment, and storage medium for dynamic risk assessment of offshore wind power construction based on Bayesian networks. It offers the following advantages: Through detailed job safety analysis (JSA), it comprehensively identifies and classifies risk factors for various operational activities in offshore wind power construction, categorizing them from four aspects: personnel, environment, equipment, and management. It also details the consequences of failure, thereby constructing a dynamic risk assessment model for offshore wind power construction based on Bayesian networks. This model can update the probability of risk factors occurring in real time, utilizing the forward inference capability of Bayesian networks for dynamic risk assessment, predicting the completion probability of each step in the operational process, improving the accuracy and real-time nature of construction risk assessment, and contributing to enhanced safety and efficiency in offshore wind power construction. Attached Figure Description

[0039] Figure 1 The flowchart shows the method for dynamic risk assessment of offshore wind power construction based on Bayesian networks provided by this invention.

[0040] Figure 2 This is a schematic diagram of the JSA analysis results in a specific embodiment of the present invention.

[0041] Figure 3This is a schematic diagram of a Bayesian network model for human error and management error in a specific embodiment of the present invention.

[0042] Figure 4 This is a schematic diagram of a Bayesian network model for harsh environments in a specific embodiment of the present invention.

[0043] Figure 5 This is a schematic diagram of a Bayesian network model for command equipment failure in a specific embodiment of the present invention.

[0044] Figure 6 This is a schematic diagram of a Bayesian network model for the failure of hoisting-related equipment in a specific embodiment of the present invention.

[0045] Figure 7 This is a schematic diagram of a Bayesian network model for hull factor failure in a specific embodiment of the present invention.

[0046] Figure 8 This is a schematic diagram of a Bayesian network model for the failure of a special lifting tool for blade installation in a specific embodiment of the present invention.

[0047] Figure 9 This is a schematic diagram of a Bayesian network model for electrical equipment failure in a specific embodiment of the present invention.

[0048] Figure 10 This represents the completion probability of each step in the wind turbine installation operation in a specific embodiment of the present invention.

[0049] Figure 11 This represents the probability of occurrence of each type of operational influencing factor in a specific embodiment of the present invention.

[0050] Figure 12 This represents the probability of consequences arising from installation operation failure in a specific embodiment of the present invention. Detailed Implementation

[0051] The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

[0052] like Figure 1 As shown, a dynamic risk assessment method for offshore wind power construction based on Bayesian networks includes the following steps:

[0053] S1: Identify risk factors for various operational activities in offshore wind power construction;

[0054] In step S1, the offshore wind turbine installation process is selected as a specific implementation case. Through JSA analysis, the risk factors in the offshore wind turbine installation process include equipment failure, construction personnel error, harsh environment, and organizational and management errors. The possible consequences of failure in the operation process include personal injury, equipment damage, hull collision damage, hull capsizing and sinking, fire, work stoppage and production stoppage, adverse impact on the next operation, and leaving hidden dangers in subsequent normal production.

[0055] S2: Classify the identified risk factors and categorize the consequences of failure;

[0056] In step S2, for the selected construction operation case, based on the JSA analysis results of the wind turbine installation process, the identified harmful factors, consequences, and critical operation steps are categorized, such as... Figure 2 As shown. The risk factors in the work process are categorized into four types: worker error, management error, adverse environment, and equipment failure. The consequences of work process failures can be divided into six aspects: personnel injury, equipment damage, installation interruption, ship collision damage, ship capsizing and sinking, and fire. Table 1 below illustrates the influencing factors and consequences of each process in the hub + engine room installation mode:

[0057] Table 1

[0058]

[0059]

[0060] S3: A failure evolution model for offshore wind power construction operations is constructed using the Bayesian network method. Based on the classification of risk factors and consequences, a Bayesian network model of individual operational influencing factors and consequences is established, and sub-Bayesian networks are added to the main Bayesian network to form a complete Bayesian network model.

[0061] In step S3, based on the classification of risk causal factors and consequences, a Bayesian network model of individual operational influencing factors and consequences is established. The conditional probability table of the sub-Bayesian networks is determined using OR gate logic. The offshore wind power engineering construction operation process is directly mapped into the main Bayesian network according to the correspondence between nodes and operation steps. Then, the sub-Bayesian networks of operational influencing factors and consequences are added to the main Bayesian network to form a complete offshore wind power engineering construction operation Bayesian network model.

[0062] For the selected construction case studies, individual factors involved in the installation of offshore wind turbines during construction included non-compliance with regulations, lack of qualifications, improper personal operation, and insufficient safety awareness (failure to wear appropriate safety protective equipment). Management errors included the absence of monitoring personnel or inadequate monitoring, lack of pre-construction training, plans not conforming to site conditions, failure to conduct pre-operation safety briefings, and an unreasonable personnel structure. Offshore wind turbine installation requires the participation of multiple workers; both personnel and management factors exist throughout the entire operation. Therefore, both factors can occur at every step of the operation and adversely affect personnel safety, equipment, and the overall operation.

[0063] The offshore wind turbine installation process is directly transformed into a main Bayesian network, and the sub-Bayesian networks of operational influencing factors are added to the main Bayesian network. Figures 3-9 A Bayesian network model is used for the installation and construction of offshore wind turbines. The connections between work steps and their consequences represent the logical relationship between the failure of a work step and its resulting scenarios. Nodes representing work-influencing factors are connected to the work step nodes they affect, while work step nodes are connected to the nodes representing the potential consequences of their failure. Work step nodes are assigned two states: Yes and No. Yes indicates that the work step is completed, while No indicates that the work step is not completed.

[0064] S4: Based on the source of risk factor data, they are divided into two categories;

[0065] For risk factors with clear data sources, statistical analysis methods are used to calculate their frequency of occurrence. The calculation formula is as follows:

[0066]

[0067] In equation (1), N A This indicates the number of times a risk with a clearly defined data source occurs, where N represents the total number of risk factors with clearly defined data sources.

[0068] For risk factors without a clear data source, their probabilities are determined using expert evaluation and fuzzy set theory, including the following steps:

[0069] S41: Use expert fuzzy terminology to define the probability of risk factors occurring. Through questionnaires, invite front-line staff and experts to assess the probability of specific risk factors occurring. The experts invited to participate in the risk factor probability assessment need to be very familiar with the site conditions and the actual situation of each risk factor occurring at the construction site.

[0070] S42: Convert expert fuzzy terminology into fuzzy numbers and aggregate the opinions of multiple experts. Depending on the actual situation of the project, three or more experts may be invited to assess the probability of risk factors occurring. The analytic hierarchy process (AHP) is used to calculate the weights of the experts. A weighted average method is used to consider the opinions of multiple different experts, as shown in the following formula:

[0071]

[0072] In equation (2), P i Let P be the fuzzy number aggregated for event i. i,j Let m represent the fuzziness of expert j's judgment on event i, and m be the total number of experts.

[0073] S43: Defuzzification process: Convert the aggregated fuzzy numbers into fuzzy probability scores (FPS), and convert the FPS into event occurrence probabilities. Use the weighted average method to aggregate the fuzzy numbers judged by different experts, and apply the maximum-minimum aggregation method for defuzzification process to convert the fuzzy numbers into probability values. The maximum and minimum fuzzy sets are as follows.

[0074]

[0075] The fuzzy probability score of the fuzzy number P is determined by the following formula:

[0076] FPS(P i ) = [FPS Right (P i +1-FPS Left (P i )] / 2 (5)

[0077] The fuzzy probability score is converted into a probability value using equations (3) and (4):

[0078]

[0079] k = 2.301 × [(1-FPS) / FPS] 1 / 3 (7).

[0080] S5: Using the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory as input, the forward inference capability of Bayesian network is used to analyze the failure probability of offshore wind power construction operations, predict the completion probability of each step in the operation process, and transmit it in the constructed Bayesian network, thereby realizing dynamic risk assessment of offshore wind power construction operations.

[0081] In step S5, based on the constructed Bayesian network model of failure evolution of offshore wind turbine installation and construction operations, the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory is used as input. The forward inference capability of the Bayesian network is used to conduct failure probability analysis of offshore wind power construction operations, predict the completion probability of each step in the operation process, and assess the probability of occurrence of operation influencing factors and failure consequences.

[0082] Since some risk factors in offshore wind power construction operations change dynamically over time, it is necessary to continuously update their prior probability of occurrence and transmit it in the constructed Bayesian network to achieve dynamic risk assessment of offshore wind power construction operations.

[0083] This invention constructs a likelihood function by accumulating precursor data over time, continuously updates the prior probability of nodes based on the Bayesian formula in equation (8), and uses a Bayesian network structure to propagate the dynamic update probability of basic events, thereby obtaining the dynamic probability of key nodes in construction operations and the state of failure consequences changing over time:

[0084]

[0085] In equation (8), P(n) is the prior probability of a node, L(data|n) is the likelihood function, and P(n|data) is the probability after the update.

[0086] The construction of the likelihood function is generally based on the observation data of the basic risk factor within a certain period. Assuming that the number of times the risk factor is observed within a certain period is s, and that its likelihood function follows a beta distribution, based on the principle of equation (8), its updated probability can be expressed as:

[0087]

[0088] In equation (9), a and n are the prior occurrences of the risk factor within one cycle and the cumulative occurrences over the total cycle.

[0089] Based on the probability range of accident consequences obtained from real-time dynamic risk assessment, the probability range is divided into 5 intervals according to the uniform distribution method, corresponding to 5 levels of risk for offshore wind power construction operations, as shown in Table 2. Based on the fuzzy membership relationship and the principle of maximum finite probability of several consequence states, the risk level of offshore wind power construction operations is determined.

[0090] Table 2

[0091] probability interval Risk level <0.20 Low 0.20<R<0.30 lower 0.30<R<0.50 middle 0.5<R<0.6 higher 0.6< high

[0092] Based on the established model, probabilistic update analysis can also be performed. Failure of critical operational steps or the occurrence of some undesirable consequence can be used as evidence to perform diagnostic inference within the constructed Bayesian network model, obtaining the posterior probability of the basic risk factors under the given evidence (undesirable accident consequence). The calculation formula is as follows:

[0093]

[0094] In equation (10), the joint probability distribution P(U), Pa(A1,A2,A3…) of the variable set U={A1,A2,A3…} i ) represents the parent node of the node set U = A1, A2, A3, ...};

[0095] In equation (11), P(U,E) represents the joint probability of risk factor U and evidence E occurring simultaneously; P(E) represents the marginal probability of evidence E.

[0096] When new data or evidence is observed, the posterior probability of the basic risk factors can be calculated using equation (11) based on Bayesian theory.

[0097] Figure 3 Each node is assigned two states: Yes indicates that the factor has occurred, and No indicates that the factor has not occurred.

[0098] Figure 4 The adverse environmental conditions include heavy fog, thunderstorms, strong winds, fast-flowing water, large waves, and limited operating space. These unfavorable environmental conditions can affect most operational procedures. Therefore, adverse environmental conditions may increase the overall risk of the installation operation.

[0099] Figure 5-9 Equipment failure has a very negative impact on the entire installation operation. Equipment failures during the installation process can be categorized into command equipment failure, hoisting equipment failure, hull failure, blade installation lifting equipment failure, and electrical equipment failure. Sling and guide rope failure, construction equipment failure, and positioning failure are considered intermediate points.

[0100] Figure 10 The probability of completion for each step in the wind turbine installation operation is 0.671 for the nacelle + hub installation mode in offshore wind turbine installation operations.

[0101] Figure 11 For each type of operation, the probability of occurrence of influencing factors is as follows: harsh environment has the highest probability during installation operations, while the probability of hoisting equipment failure is slightly lower. The probability of worker error falls between that of hoisting equipment failure and management error.

[0102] Figure 12The probabilities of consequences arising from installation failures are ranked from highest to lowest as follows: shutdown > fire > personal injury > equipment damage > ship collision damage > ship capsizing and sinking. The probabilities of "shutdown," "equipment damage," and "personal injury" are all between 0.3 and 0.4, and according to Table 2, the risk level of the construction operation can be classified as medium.

[0103] A dynamic risk assessment system for offshore wind power construction based on Bayesian networks, the system comprising the following modules:

[0104] The risk factor identification module is used to identify risk factors that may occur during transportation, pile foundation construction, wind turbine installation, offshore substation construction, submarine cable laying, commissioning and trial operation.

[0105] The risk factors and failure consequences classification module categorizes risk factors into four aspects: personnel, environment, equipment and management, and failure consequences into six aspects: personnel injury and death, equipment damage, installation work interruption, hull collision damage, hull capsizing and sinking, and fire.

[0106] The probability calculation module is used to calculate the probability of occurrence of risk factors with clear data sources and risk factors without clear data sources.

[0107] The offshore wind power construction operation risk assessment module is used to assess the risk index of offshore wind power construction operations by combining the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory.

[0108] A computer device includes a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to execute the aforementioned method for dynamic risk assessment of offshore wind power construction based on Bayesian networks.

[0109] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned method for dynamic risk assessment of offshore wind power construction based on Bayesian networks.

[0110] The above specific embodiments are used to explain and illustrate the present invention, and are only preferred embodiments of the present invention, not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A dynamic risk assessment method for offshore wind power construction based on Bayesian networks, characterized in that, Includes the following steps: S1: Identify risk factors for various operational activities in offshore wind power construction; S2: Classify the identified risk factors and categorize the consequences of failure; S3: A failure evolution model for offshore wind power construction operations is constructed using the Bayesian network method. Based on the classification of risk factors and failure consequences, a Bayesian network model for a single operational risk factor and failure consequence is established, and the sub-Bayesian networks are added to the main Bayesian network to form a complete Bayesian network model. S4: Based on the source of risk factor data, they are divided into two categories; S5: Using the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory as input, the forward reasoning ability of Bayesian network is used to analyze the failure probability of offshore wind power construction operations, predict the completion probability of each step in the operation process, and transmit it in the constructed Bayesian network, thereby realizing dynamic risk assessment of offshore wind power construction operations. In step S2, risk factors are categorized into four aspects: personnel, environment, equipment, and management. The failure consequences in step S2 are categorized into six aspects: personnel injury and death, equipment damage, interruption of installation work, collision damage to the hull, capsizing and sinking of the hull, and fire. In step S4, for risk factors with clear data sources, statistical analysis methods are used to calculate their frequency of occurrence. The calculation formula is as follows: In equation (1), N A This indicates the number of times a risk with a clearly defined data source occurs, where N represents the total number of risk factors with clearly defined data sources. In step S4, for risk factors without a clear data source, the probability is determined using expert evaluation and fuzzy set theory, and includes the following steps: S41: Use expert fuzzy terminology to define the probability of risk factors occurring. Through questionnaires, invite front-line staff and experts to assess the probability of specific risk factors occurring. The experts invited to participate in the risk factor probability assessment need to be very familiar with the site conditions and the actual situation of each risk factor occurring at the construction site. S42: Convert expert fuzzy terms into fuzzy numbers and aggregate the opinions of multiple experts. Depending on the actual situation of the project, three or more experts may be invited to assess the probability of the occurrence of risk factors. The analytic hierarchy process (AHP) is used to calculate the weights of the experts, and a weighted average method is employed to consider the opinions of multiple different experts. The formula is as follows: In equation (2), P i Let P be the fuzzy number aggregated for event i. i,j Let m be the fuzzy number of fuzzy expert j's judgment on event i, and m be the total number of experts. S43: Defuzzification process: Convert the aggregated fuzzy numbers into fuzzy probability scores (FPS), and convert the FPS into event occurrence probabilities. Use the weighted average method to aggregate the fuzzy numbers judged by different experts. Apply the maximum-minimum aggregation method for defuzzification process, converting the fuzzy numbers into probability values. The maximum and minimum fuzzy sets are as follows. The fuzzy probability score of the fuzzy number P is determined by the following formula: FPS(P i )=[FPS Right (P i )+1-FPS Left (P i )] / 2 (5) The fuzzy probability score is converted into a probability value using equations (3) and (4): k=2.301×[(1-FPS) / FPS] 1 / 3 (7)。 2. The method for dynamic risk assessment of offshore wind power construction based on Bayesian networks according to claim 1, characterized in that: The risk factor identification in step S1 includes six main processes: transportation, pile foundation construction, wind turbine installation, offshore substation construction, submarine cable laying, commissioning and trial operation.

3. A dynamic risk assessment system for offshore wind power construction based on Bayesian networks, characterized in that: The assessment system is based on the Bayesian network-based dynamic risk assessment method for offshore wind power construction as described in claim 1, and includes the following modules: The risk factor identification module is used to identify risk factors that occur during transportation, pile foundation construction, wind turbine installation, offshore substation construction, submarine cable laying, commissioning and trial operation. The risk factors and failure consequences classification module classifies risk factors into four aspects: personnel, environment, equipment and management, and failure consequences into six aspects: personnel injury and death, equipment damage, installation work interruption, hull collision damage, hull capsizing and sinking, and fire. The probability calculation module is used to calculate the probability of occurrence of risk factors with clear data sources and risk factors without clear data sources; The offshore wind power construction operation risk assessment module is used to assess the risk index of offshore wind power construction operations by combining the probability of occurrence of basic risk factors calculated by mathematical statistics and fuzzy set theory.

4. A computer device, comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to perform the Bayesian network-based dynamic risk assessment method for offshore wind power construction as described in any one of claims 1-2.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for dynamic risk assessment of offshore wind power construction based on Bayesian networks as described in any one of claims 1-2.

Images