Event chain-oriented offshore oil spill dynamic risk assessment method

Abstract

An event chain-oriented offshore oil spill dynamic risk assessment method comprises a data preparation stage, an event chain modeling stage and a dynamic risk assessment stage, and the data preparation stage comprises the steps of A, determining a risk assessment area and a risk assessment unit; the event chain modeling stage comprises the steps of B, oil spill event chain construction and node analysis; the dynamic risk assessment stage comprises the steps that C, from the perspective of quantitative risk assessment, the risk index of a current oil spill event, the vulnerability index of a disaster-bearing body and the risk index of a native event are calculated respectively, and oil spill native event risk assessment is executed; step D, oil spill secondary / derivative event risk assessment; step E, oil spill event chain risk assessment; and step F, carrying out oil spill event chain risk early warning grade division. The evaluation method analyzes and evaluates the influence object of the oil spill event and the risk of secondary / derivative events caused by further disaster due to disaster elements contained in the influence object or the disaster event from the perspective of event evolution.

Description

Technical Field

[0001] The present invention relates to the technical field of event chain risk assessment, and in particular to an event chain-oriented marine oil spill dynamic risk assessment method. Background Art

[0002] With the large-scale development of global offshore oil development, marine oil spills have become a major environmental risk to marine ecological security. These incidents not only exhibit the physical characteristics of rapid spread, widespread contamination, and lasting ecological damage, but also form a multidimensional disaster network due to a complex chain of events. For example, the initial spread of oil pollution can directly trigger primary secondary / derivative disasters such as damage to fishery resources and decline in coastal tourism. Long-term residual petroleum hydrocarbons, through bioaccumulation, can trigger secondary secondary / derivative disasters such as food chain collapse and regional ecological imbalance. They can even trigger social risks such as public opinion crises and a decline in government credibility, forming a multi-stage event chain of "physical pollution-ecological damage-social impact," the cumulative effect of which can lead to exponentially increasing disaster losses.

[0003] At present, my country's marine oil spill emergency response technology has made breakthroughs in areas such as satellite monitoring and trajectory prediction. However, there are still three core technical bottlenecks in risk assessment:

[0004] Lack of event chain analysis mechanism: Traditional methods regard oil spills as independent events and fail to build a full-chain transmission model of "oil spill incidents - secondary / derivative disasters - long-term impacts". As a result, it is impossible to identify key secondary / derivative event risk nodes such as fire and explosion, nuclear safety incidents, etc., making it difficult to achieve full-process risk management of the disaster chain.

[0005] Insufficient efficiency of multi-source data fusion: The current assessment system relies on single-dimensional data such as oil spill volume and distance to sensitive areas. It lacks in-depth integration of multi-dimensional factors such as real-time meteorological parameters (wind speed / tide), spatial distribution of ecologically sensitive areas (such as coral reefs / fishing grounds), and socio-economic value data (such as the number of tourists in tourist areas). This leads to deviations between risk rating results and the actual degree of harm.

[0006] Weak spatial emergency support capabilities: Existing technologies lack refined grid-based risk assessment capabilities and are unable to achieve accurate spatial mapping of "refined grid units—event chain risks—emergency strategies," resulting in a lack of clear spatial targeting and dynamic adaptability in emergency resource allocation.

[0007] Although existing research has explored some risk assessment techniques for oil spills (such as artificial neural network models, fuzzy Petri nets, and Bayesian networks), most methods do not incorporate the event chain evolution process into the assessment framework, lacking the real-time integration of multi-dimensional risk factors in spatial sea areas and precise risk-based emergency support. For example:

[0008] Although the disaster chain fuzzy Petri net model proposed in CN201910994504.2 involves event chain analysis, it does not realize the spatiotemporal dynamic coupling of risk factors. In contrast, the present invention introduces a spatiotemporal dynamic coupling factor (such as the distance probability P in formula (8)). d ), which solves the problem that the traditional Petri net model lacks description of the spatial diffusion process.

[0009] The Bayesian network model proposed in CN202310388009.3 focuses on single event risk prediction and does not construct a multi-level event chain conduction path. In contrast, the present invention constructs a multi-level event chain conduction path (such as Figure 2 ), breaking through the limitations of single event risk prediction.

[0010] Other related technologies (such as CN201610677219.4, CN202011226473.5, etc.) mostly focus on oil spill diffusion or resource layout, and lack a systematic assessment of the risks of the event chain.

[0011] Therefore, how to break through the limitations of the traditional oil spill emergency risk assessment model and establish a risk assessment method that covers the entire cycle of the event chain, integrates multi-source dynamic data, and supports spatial precision decision-making has become an urgent need in the field of marine oil spill emergency response. Summary of the Invention

[0012] In response to the core defects of the existing technology, such as the lack of event chain risk identification, insufficient multi-source data fusion, and low spatial assessment accuracy, the present invention provides an event chain-oriented marine oil spill dynamic risk assessment method. By constructing a three-dimensional assessment framework of "risk source-diffusion path-secondary event-spatial grid", it realizes the full-chain quantitative analysis and dynamic early warning control of oil spill risks.

[0013] The present invention provides an event chain-oriented marine oil spill dynamic risk assessment method, which includes three core stages: data preparation, event chain modeling, and dynamic risk assessment.

[0014] The data preparation phase includes the following steps:

[0015] Step A: Risk assessment area and risk assessment unit division. Obtain oil spill information, oil spill monitoring data (oil spill location, oil spill volume, oil film distribution), and oil spill prediction data (oil particle trajectories). Dynamically set the risk warning distance based on the oil spill volume in the oil spill monitoring data and the predicted duration of the oil spill drift prediction simulation. Combine these data to generate a risk assessment area and divide the risk assessment units.

[0016] The event chain modeling includes the following steps:

[0017] Step B: Construction of oil spill event chain and node analysis: Based on GIS spatial analysis, the types of hazard-prone objects are identified and an oil spill event chain including secondary / derivative events is constructed.

[0018] The risk assessment phase includes the following steps:

[0019] Step C: Oil spill primary event risk assessment. From the perspective of quantitative risk assessment, calculate the hazard index of the current oil spill event, the vulnerability index of the hazard-bearing body, and the oil spill primary event risk index, and perform the oil spill primary event risk assessment;

[0020] Step D: Risk assessment of secondary / derivative events of oil spills;

[0021] Step E: Oil spill event chain risk assessment. Comprehensively assess the oil spill event chain risk by integrating the primary oil spill event risk and the secondary / derivative event risk;

[0022] Step F: Classification of risk warning levels of the oil spill incident chain.

[0023] The present invention provides an event chain-oriented marine oil spill dynamic risk assessment calculation process as follows: Figure 1 shown.

[0024] Furthermore, step A determines the risk assessment area and risk assessment units, including obtaining oil spill information, oil spill monitoring data and oil spill prediction data, determining the risk assessment area, and dividing the risk assessment units, wherein:

[0025] (1) Obtain oil spill incident information, oil spill monitoring data and oil spill prediction data.

[0026] Oil spill incident information refers to information such as the oil spill location, oil spill volume, and oil characteristics obtained through incident reporting; oil spill monitoring data includes data such as the spatial distribution of the oil film and the oil film thickness distribution obtained through on-site oil spill monitoring or satellite remote sensing monitoring and analysis; oil spill prediction data is obtained by calculating the oil particle drift and diffusion model. The input parameters include real-time wind speed (such as a resolution of 10 minutes), tidal data (such as a resolution of 1 hour), and sea surface temperature field (such as a resolution of 1km). The oil film diffusion trajectory is simulated by the particle release frequency (such as 100 particles per second). The model output can obtain the oil particle volume and distribution position data (oil particle trajectory) at each moment within the prediction time range.

[0027] (2) Determine the risk assessment area.

[0028] The risk assessment area is the area where oil spill risk assessment calculations are carried out, representing the area that may be affected by oil spills within the current period of time.

[0029] The risk assessment area is dynamically set based on the oil spill monitoring prediction area and risk warning distance.

[0030] The risk warning distance refers to the distance between the center of the residual oil spill and the outermost edge of the warning area, which is used to delineate the sensitive disaster-bearing body for risk assessment and early warning in order to carry out risk assessment and early warning.

[0031] The risk warning distance is calculated based on 24 hours and is dynamically set according to the amount of oil spill and the predicted duration. The specific calculation method is shown in Table 1 Risk Warning Distance Setting Method.

[0032] Among them, the oil spill volume is mainly set with reference to the event classification in the "Offshore Petroleum Exploration and Development Oil Spill Emergency Plan". At the same time, based on the statistics of oil spill volume in oil spill incidents in previous years, the classification standard of 50t oil spill volume below 100t is added.

[0033] The risk assessment area is the combined area of ​​the sea area within the risk warning distance range and the maximum distribution area of ​​oil spill monitoring prediction.

[0034] Table 1 Risk warning distance setting method

[0035]

[0036] (3) Divide the risk assessment units.

[0037] The risk assessment unit is the smallest unit for conducting risk assessments and displaying risk assessment results. The risk assessment unit is based on a grid structure within the risk assessment area. Based on the residual oil spill volume and user visualization requirements, the risk assessment area is dynamically divided into M rows x N columns of regular grid cells. The grid cell size is the smallest unit for displaying risk calculation results and also represents the decision-making unit for implementing oil spill response. Theoretically, the grid cell size should be no less than 500m x 500m.

[0038] Furthermore, step B includes oil spill event chain construction and event chain node analysis based on the current event.

[0039] Oil spills are typically secondary or derivative events of other safety incidents, with oil as a contributing factor to the hazard chain of preceding events. As the spill occurs and spreads across the ocean, the chain of events undergoes further changes. This study examines the threat posed by oil slicks to the marine environment and hazard-bearing structures from the perspective of supporting emergency management. Therefore, it is necessary to construct and analyze the chain following the oil spill.

[0040] Furthermore, an oil spill event chain is constructed. The occurrence and development of an oil spill at sea is a dynamically changing process. The drift and diffusion of the oil film affects different hazard-bearing bodies, generating a drifting oil spill primary event. After interacting with the hazard-bearing body, the contained disaster elements further trigger subsequent different secondary / derivative events, forming a complex disaster network. However, the goal of risk warning is to break the chain and reduce disasters according to the size of the risk borne by the hazard-bearing body. Therefore, the present invention focuses on sorting out and constructing an oil spill secondary event chain, and considers the oil spill chain risk from two aspects: the primary event risk faced by the hazard-bearing body and the secondary / derivative event set risk that may be triggered by it. At the same time, in the present invention, the oil spill event chain is only used for model construction and only needs to be constructed once. Therefore, the present invention must first analyze all the evolution directions and consequences of the oil spill event based on the type of hazard-bearing body that the oil spill event may threaten, and form a complete event chain for the oil spill event.

[0041] Based on the technical requirements of risk early warning, the present invention screens out the important types of hazard-bearing bodies of marine oil spill incidents, and constructs an oil spill event chain (such as the first-level secondary / derivative events and the lower-level secondary / derivative events) based on typical oil spill cases at home and abroad using probability statistics and Bayesian network training methods. Figure 2 ), the correspondence table between event chain node events and secondary / derivative events is shown in Table 2.

[0042] Table 2. Correspondence between typical event chain node events and secondary / derivative events

[0043]

[0044] Furthermore, the hazard-bearing bodies and assessable event chains of oil spill incidents are determined through spatial analysis methods. This includes using GIS spatial analysis methods to overlay the risk assessment area with marine functional sea use data (such as protected areas, aquaculture areas, ports, etc.), extracting the hazard-bearing bodies that intersect with the risk assessment area, and analyzing the sea use types of the hazard-bearing bodies; Figure 2 The oil spill event chain node establishes the mapping relationship between the sea use type of the disaster-bearing body and the secondary / derivative events, and constructs the secondary event chain structure.

[0045] A primary oil spill event refers to an event in which oil may drift to the vicinity of the spatial location of the hazard-bearing object after the spill occurs and cause damage to the hazard-bearing object. It is the first level of damage that may be caused after an oil spill occurs.

[0046] Furthermore, the risk assessment of primary oil spill incidents includes the hazard assessment of primary oil spill incidents, the vulnerability assessment of the hazard-bearing body and the risk assessment of primary oil spill incidents.

[0047] (1) Risk assessment of primary oil spill incidents

[0048] The risk of an oil spill is determined by the density of the oil spilled and the average residual oil volume per square kilometer in the assessment unit, and is expressed as the oil spill risk index H. i Calculate according to formula (1) to formula (3):

[0049]

[0050] Where:

[0051] H i ——The hazard index value of the risk assessment unit grid, which is related to the total oil spill volume and the oil spill thickness of the risk assessment unit grid. The maximum value is 10. If it exceeds 10, it needs to be artificially truncated. It is dimensionless;

[0052] Q oil ——The total residual oil spill volume of the oil spill incident, in tons (t);

[0053] T u ——Oil spill thickness value per square kilometer of risk assessment unit grid, in microns

[0054] (μm);

[0055] H T ——the oil spill thickness index value of the risk assessment unit grid, dimensionless;

[0056] N——the total number of particles predicted for oil spill drift, in pieces;

[0057] n——The number of predicted particles contained in the current risk assessment unit grid, in pieces.

[0058] ρ——Density of oil spilled, in kilograms per cubic meter (kg / m 3 );

[0059] s——risk assessment unit grid area, in square meters (m 2 ).

[0060] The ArcGIS Pro natural breakpoint classification method (Jenks Natural Breaks) was used to automatically classify the hazard level of the original oil spill incident into four levels based on the inflection points of the hazard index distribution. The levels were divided according to the hazard index scores, and the risk colors were marked as red, orange, yellow, and blue. The specific classification method is shown in Table 3.

[0061] Table 3 Correspondence table of risk levels of primary oil spill incidents

[0062] Hazard level Level IV Level III Level II Level I range (1,2] (2,4] (4,6] (6,10] color blue yellow orange red RGB settings RGB(0,0,255) RGB(255,255,0) RGB(255,128,0) RGB(255,0,0)

[0063] The vulnerability of a disaster-prone body refers to the degree of vulnerability and lack of recovery capacity of a disaster-prone body due to its own weaknesses in structure, function, management, etc. when it suffers from natural disasters or man-made disasters.

[0064] Furthermore, an oil spill vulnerability assessment index system is established to evaluate the extent to which the oil spill affects the hazard-bearing body. This includes three specific steps:

[0065] (1) Using the analytic hierarchy process (AHP), a theoretical framework for the vulnerability assessment index system is established, which includes five levels: target layer, criterion layer, factor layer, assessment object index layer, and scheme layer.

[0066] The index system takes oil spill environmentally sensitive resources (hazard-bearing bodies) as the evaluation objects and aims to comprehensively measure the vulnerability of marine oil spill sensitive resources; the element layer starts from the perspectives of sensitivity, exposure / resilience and risk management capabilities, integrates ecological, socioeconomic and management knowledge, selects three basic indicators: social and humanistic values, environmental attributes, and management needs, and refines them into five secondary indicators: functional type, economic value, importance, self-purification capacity, and social prosperity according to the vulnerability evaluation characteristics of the hazard-bearing bodies; the evaluation object indicator layer further refines the secondary indicators into third-level and fourth-level indicators according to the type and attributes of the hazard-bearing bodies; the scheme layer establishes a vulnerability assessment method to realize the calculation of vulnerability index for specific hazard-bearing bodies.

[0067] (2) Establish a disaster-prone body assessment index system for specific assessment objects, and refine the assessment indicators according to the type and attributes of the oil spill disaster-prone body, including the classification of disaster-prone body sensitivity and the setting of vulnerability assessment indicators.

[0068] First, hazard-prone objects were classified by sensitivity based on their type. Their sensitivity to oil spills was divided into three levels: highly sensitive, sensitive, and low. These sensitivity values ​​were assigned 10, 7, and 3, respectively. Coastal nuclear power plants, protected areas, coastlines, and key activity areas, which pose significant secondary risk and economic and ecological damage, were classified as highly sensitive. The economically valuable fisheries and tourism industries were classified as moderately sensitive. Other oil platforms, resource development areas, and traffic-intensive areas were classified as low-sensitivity. Second, vulnerability assessment indicators for hazard-prone objects of varying sensitivity were established, with secondary, tertiary, and quaternary indicators refined based on the type and attributes of the hazard-prone objects. Specific settings are shown in Tables 4-6.

[0069] Table 4 Vulnerability evaluation indicators and index value settings for highly sensitive disaster-bearing bodies

[0070]

[0071] Table 5 Vulnerability evaluation indexes and index value settings for sensitive hazard-bearing bodies

[0072]

[0073]

[0074] Table 6 Vulnerability evaluation indexes and index value settings for low-sensitivity disaster-bearing bodies

[0075]

[0076] (3) Carry out vulnerability assessment of disaster-prone objects and comprehensive vulnerability assessment of evaluation units, and calculate the vulnerability index of disaster-prone objects and comprehensive vulnerability index of evaluation units respectively.

[0077] The vulnerability index of the disaster-bearing body is calculated according to formula (4).

[0078]

[0079] Where:

[0080] V k ——the vulnerability index of the kth hazard-bearing body;

[0081] S——sensitivity value of the hazard-bearing body;

[0082] Wx i ——The i-th secondary indicator value corresponding to the attribute of the disaster-prone body;

[0083] Wy i ——The i-th third-level indicator value corresponding to the attributes of the disaster-prone body;

[0084] Wz i ——The i-th fourth-level indicator value corresponding to the attributes of the disaster-prone body.

[0085] The natural breakpoint classification method of ArcGIS Pro was used to automatically divide the vulnerability level of the hazard-prone body into four levels based on the inflection points of the vulnerability index distribution. The levels were divided according to the value of the vulnerability index, and the risk colors were marked as red, orange, yellow, and blue. The specific classification method is shown in Table 7.

[0086] Table 7 Correspondence table of vulnerability levels of hazard-bearing bodies

[0087] Vulnerability Level Level IV Level III Level II Level I range (0-2] (2,3.5] (3.5,5.3] (5.3,10] color blue yellow orange red RGB settings RGB(0,0,255) RGB(255,255,0) RGB(255,128,0) RGB(255,0,0)

[0088] The comprehensive vulnerability of the assessment unit is comprehensively analyzed according to the vulnerability of the hazard-bearing bodies within the assessment unit. If the assessment unit contains multiple hazard-bearing bodies, the vulnerability of the hazard-bearing bodies within the assessment unit should first be ranked from large to small according to the risk value, and then the comprehensive vulnerability index value of the assessment unit should be calculated according to formula (5):

[0089]

[0090] Where:

[0091] V——comprehensive vulnerability value of risk assessment unit;

[0092] n——the number of hazard-bearing bodies in the risk assessment unit;

[0093] V j ——Vulnerability index of the disaster-bearing body with risk value ranking j within the risk assessment unit.

[0094] Furthermore, for the risk assessment of oil spill incidents,

[0095] The primary event of the oil spill event chain is the first-level event that occurs when the current oil spill event interacts with the hazard-bearing body. The primary event risk is the expected evaluation of the possible impact of the current oil spill event on the risk assessment unit where the hazard-bearing body is located. It is represented by the risk index of the risk assessment unit where the hazard-bearing body is located and is calculated according to formula (6):

[0096] R=H×V (6)

[0097] Where:

[0098] R——Oil spill risk index value of the risk assessment unit, dimensionless;

[0099] H——Oil spill hazard index value of the risk assessment unit, dimensionless;

[0100] V is the grid vulnerability index value of the risk assessment unit, dimensionless.

[0101] Furthermore, for the classification of risk levels of oil spill incidents,

[0102] According to the calculation formula and value range of the oil spill risk index value of the risk assessment unit, the natural breakpoint classification method of ArcGIS Pro is used to automatically divide the risk level of the oil spill event chain into four levels based on the inflection points of the risk index distribution. The levels are divided according to the size of the risk index score, and the risk colors are marked as red, orange, yellow, and blue respectively. The specific classification method is shown in Table 8.

[0103] Table 8 Correspondence table of risk levels of oil spill event chain

[0104] Risk Level Level IV Level III Level II Level I range (1-10] (10,22] (22,39] (39,100] color blue yellow orange red RGB settings RGB(0,0,255) RGB(255,255,0) RGB(255,128,0) RGB(255,0,0)

[0105] Furthermore, step D, secondary / derivative event risk estimation includes:

[0106] (1) Analysis of the types and consequences of secondary / derivative events of oil spills.

[0107] Secondary / derivative events of marine oil spills are first environmental pollution events, and further evolve into public opinion events, economic loss events, traffic events, poisoning events, and safety events, depending on the degree of interaction between the oil spill and the hazard-bearing body. The consequences of these events are strongly correlated with the hazard-bearing body and the type of secondary / derivative events. However, given that most secondary / derivative events of marine oil spills have local characteristics and are more closely related to the hazard-bearing body, this patent will evaluate the consequence index of the oil spill chain based on the type of oil spill hazard-bearing body and the type of secondary / derivative events at the first level of the oil spill chain, and compare and assign values ​​based on the sensitivity of different hazard-bearing bodies to oil spills and the corresponding consequence indexes of secondary / derivative events. The list of consequence indices corresponding to specific hazard-bearing bodies and secondary / derivative events is as follows.

[0108] Table 9 List of oil spill disaster-affected body types and secondary / derivative event consequence indices

[0109]

[0110]

[0111] (2) Estimation of the probability of secondary / derivative events.

[0112] The probability of secondary / derivative events is closely related to the risk of the primary oil spill event and the distance between the oil spill and the disaster-affected body. This patent temporarily refers to them as risk probability and distance probability. Distance probability, from the perspective of emergency rescue, indirectly calculates the relative risk probability of the disaster-affected body by estimating the distance between the disaster-affected body and the oil spill. Assuming that the probability of the disaster-affected body being affected within the spatial range of the oil spill prediction is 1, then the probability of the disaster-affected body being affected in the area between the monitoring prediction range and the risk warning area is 0 to 1. The size of this probability value is nonlinearly positively correlated with the distance. The calculation formula is as shown in formula (8). Risk probability is based on the triggering conditions of the event chain to calculate the probability of secondary / derivative events. If the risk of the primary oil spill event is large and the consequences are more serious, then the probability of triggering the lower-level secondary / derivative events is greater, otherwise it is smaller. The calculation formula is as shown in formula (9).

[0113] The calculation formula for the probability of occurrence of secondary / derivative events is shown in formula (7):

[0114] P=P d ×P R (7)

[0115]

[0116] Where P represents the probability of the oil spill incident affecting the current disaster-bearing body.

[0117] P d ——Indicates the probability of distance affecting the probability.

[0118] P R ——Indicates the risk impact probability of the primary oil spill incident. Generally speaking, the greater the risk index value of the primary oil spill incident, the higher the possibility of secondary / derivative incidents.

[0119] d——indicates the closest distance between the maximum distribution range predicted by the monitoring of the primary oil spill event and the disaster-stricken body.

[0120] When the distance between the original oil spill event and the disaster-stricken object is less than or equal to d0, the oil spill event will definitely pose a threat to the disaster-stricken object. In this patent, this value is set to the current event prediction simulation distance. Since the risk warning area is expanded based on the maximum distribution surface of the oil spill monitoring prediction, this value is 0.

[0121] D0——represents the risk warning distance of the original oil spill event. As mentioned above, this variable is related to the amount of oil spilled and the currently set prediction time.

[0122] R l1 ——Indicates the risk level of the primary oil spill incident.

[0123] Furthermore, the oil spill incident chain risk calculation is carried out.

[0124] The event chain-oriented dynamic risk assessment of oil spills first calculates the event chain risk of each type of hazard-bearing body in the risk assessment area. Secondly, with the grid-based risk warning as the goal, the oil spill event chain risk of each risk assessment unit is calculated.

[0125] (1) Risk estimation method for oil spill hazard-bearing bodies.

[0126] The chain risk calculation formula for each type of hazard-bearing body is as follows:

[0127] R l =R l1 +W×R l2 (10)

[0128] R l1 =H l1 ×V l1 (11)

[0129] R l2 =P l2 ×D l2 (12)

[0130] Among them, R l It represents the chain risk value centered on the grid hazard-bearing body. Since the probability of occurrence of primary events in a single chain risk is much greater than the probability of secondary / derivative events, for a single series event on the same hazard-bearing body, the overall risk of the primary event is greater than the risk of the secondary / derivative event, that is, R l1is the original event risk, and its value is determined by the risk of the original event H l1 , Vulnerability of the disaster-bearing body V l1 Calculated, R l2 is the secondary / derivative event risk, whose value is determined by the probability of occurrence of the secondary / derivative event P l2 and consequence index D l2 W is the normalized weight of the primary event and the secondary / derivative event, and is artificially set to 5 here based on the value ratio.

[0131] (2) Dynamic risk assessment method of oil spill based on event chain.

[0132] The event chain-oriented oil spill risk assessment aims at grid-based risk warning. Therefore, it is necessary to calculate the total event chain risk of each risk assessment unit. This risk is composed of the single chain risks of all hazard-bearing objects in the assessment unit. The calculation formula is as follows:

[0133]

[0134] Where R represents the risk of the event chain in each risk assessment unit, n is the number of grid-based hazard-prone objects in the assessment unit, and R1, R2, R3, etc. are the values ​​of the chain risk centered on the hazard-prone object in the assessment unit grid, sorted by size, satisfying R1 ≥ R2 ≥ R3.... This is dimensionless.

[0135] Further, step F, oil spill incident chain risk warning level classification

[0136] According to the calculation formula and value range of the oil spill risk index value of the risk assessment unit, the natural breakpoint classification method of ArcGIS Pro is used to automatically divide the oil spill event chain risk level into four levels based on the inflection points of the risk index distribution. The risk levels are marked as red, orange, yellow and blue respectively. The specific classification method is shown in the table below.

[0137] Table 10 Correspondence table of risk levels of oil spill event chain

[0138] Risk Level Level IV Level III Level II Level I range (1-15] (15,28] (28,47] (47,150] color blue yellow orange red RGB settings RGB(0,0,255) RGB(255,255,0) RGB(255,128,0) RGB(255,0,0)

[0139] The event chain-oriented marine oil spill dynamic risk assessment method of the present invention has the following advantages:

[0140] (1) Full cycle coverage of the event chain: For the first time, “oil spill incident – ​​secondary / derivative incident – ​​social impact” is included in a unified assessment framework, realizing risk estimation of the full cycle of the event chain from “accident occurrence – damage formation – impact spread”.

[0141] (2) Intelligent fusion of multi-source data: Integrate satellite remote sensing, field monitoring, marine environmental dynamics models, and socioeconomic data, and realize dynamic integration and application through the spatiotemporal data engine to improve assessment accuracy.

[0142] (3) Precise spatial emergency support: The grid-based risk assessment method realizes risk assessment based on spatial units, which can assist in real-time early warning and rapid emergency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0143] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0144] Figure 1 It is a flow chart of dynamic risk assessment calculation of marine oil spills oriented to the event chain.

[0145] Figure 2 It is a chain of marine oil spill incidents.

[0146] Figure 3 It is an index system for evaluating the vulnerability of oil spill-affected bodies.

[0147] Figure 4 It is a predicted path map of oil particle drift in a simulated oil spill incident in Liaodong Bay.

[0148] Figure 5 It is a risk warning area and risk assessment unit division map for a simulated oil spill incident in Liaodong Bay.

[0149] Figure 6 This is the distribution map of disaster-prone objects in the simulated oil spill incident in Liaodong Bay.

[0150] Figure 7 This is the primary event hazard level distribution map of the simulated oil spill incident in Liaodong Bay.

[0151] Figure 8 This is the distribution map of vulnerability levels of hazard-bearing bodies in a simulated oil spill event in Liaodong Bay.

[0152] Figure 9 This is the primary event risk level distribution map of the simulated oil spill incident in Liaodong Bay.

[0153] Figure 10 This is the event chain risk level distribution map of the simulated oil spill incident in Liaodong Bay. DETAILED DESCRIPTION

[0154] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0155] An event chain-oriented marine oil spill dynamic risk assessment method includes data preparation stage, event chain modeling stage and risk assessment stage.

[0156] The data preparation phase includes the following steps:

[0157] Step A: Determine the risk assessment area and assessment unit.

[0158] (1) Obtain oil spill monitoring and prediction data.

[0159] This implementation case uses a hypothetical oil spill event. The spill's location, time of occurrence, oil type, spill volume, and hazard-bearing volume and distribution are all assumed. The oil spill prediction model uses real-world marine environmental forecast information released by the North Sea Marine Environmental Forecasting Department to obtain information on oil particle drift and dispersion. The oil spill event chain risk assessment is performed according to the algorithmic steps.

[0160] Assume that at 15:00 on February 23, 2025, an oil spill alarm is received, indicating that a ship collision has occurred near the location (121.116, 40.334) in the Liaodong Bay area, causing the loss of approximately 200 tons of crude oil (density 1000 kg / m 3 ) leak, assuming that there is no relevant oil spill monitoring data, the oil particle drift prediction model is connected to the marine environment forecast product of the sea area to calculate the oil particle drift prediction path of the oil spill accident 48 hours later (such as Figure 4 ).

[0161] This paper assumes that the oil spill event chain only considers the sea surface diffusion stage after the oil spill occurs, does not involve complex scenarios such as submarine pipeline leakage, and does not consider the long-term ecological impact of the oil spill. The case will intercept the oil particle drift prediction information within 24 hours after the oil spill occurs to verify the event chain risk assessment method.

[0162] (2) Determine the risk assessment area.

[0163] The risk assessment area is the combined area of ​​the risk warning distance range and the maximum distribution area of ​​the oil spill monitoring forecast. Based on the aforementioned risk warning distance setting method, a 200-ton crude oil spill corresponds to a 24-hour warning distance of 13 km. The risk warning area for this oil spill is formed by expanding the maximum distribution area of ​​the oil spill monitoring forecast to the surrounding sea area with a radius of 13 km.

[0164] (3) Divide the risk assessment units.

[0165] The risk warning area is divided into grids according to user needs. In order to display the oil spill risk assessment results in a refined manner, this patent divides the risk assessment unit grid size into 500m×500m.

[0166] See also Figure 5 Risk warning area and risk assessment unit division map for the simulated oil spill incident in Liaodong Bay.

[0167] The event chain modeling phase includes the following steps:

[0168] Step B: Construction of oil spill event chain and analysis of event chain nodes based on current events.

[0169] (1)Reference Figure 2 Construct the oil spill event chain of this incident.

[0170] (2) According to the spatial analysis method, the disaster-prone objects with spatial intersection between the maximum distribution area of ​​oil spill monitoring prediction and the risk warning area are analyzed and determined as the risk assessment objects of this oil spill incident. Figure 6 The table below shows the relative position of the current oil spill monitoring prediction range, risk warning area and disaster-prone objects.

[0171] Table 11 Information on disaster-prone objects in the Liaodong Bay oil spill incident

[0172]

[0173]

[0174] See also Figure 6 Distribution map of disaster-prone bodies in the Liaodong Bay oil spill incident.

[0175] The risk assessment phase includes the following steps:

[0176] Step C: Risk assessment of the primary oil spill incident.

[0177] (1) Hazard assessment of primary oil spill incidents.

[0178] A primary oil spill event refers to an event in which the oil may drift to the vicinity of the spatial location of the hazard-bearing object and cause damage to the hazard-bearing object after the current oil spill occurs. It is the first level of damage event that may be caused after an oil spill occurs.

[0179] According to the aforementioned oil spill primary event hazard assessment formula, the oil spill primary event hazard index of each risk assessment unit is calculated. It includes three steps: first, calculate the amount of oil spill represented by each oil particle, second, count the number of oil particles on each risk assessment unit, calculate the oil film thickness by the amount of oil spill represented by the oil particles and the area of ​​the risk assessment unit, and finally, calculate the hazard index of each assessment unit based on the total amount of oil spilled in the current accident and the oil film thickness index of each assessment unit. The hazard distribution map of this oil spill primary event is shown in Figure 7 As can be seen from the figure, among all the 6408 risk assessment units included in the risk warning area, there are 83 level I dangerous units, 57 level II units, 122 level III units, 0 level IV units, and the rest are risk assessment units that have not reached the classification level.

[0180] (2) Vulnerability assessment of disaster-prone bodies.

[0181] The vulnerability of a disaster-prone body refers to the vulnerability and lack of recovery capacity of a disaster-prone body due to its own structural, functional, and management weaknesses when it is hit by natural or man-made disasters. This implementation case first obtains the vulnerability index of the disaster-prone body included in the risk warning area according to the disaster-prone body vulnerability evaluation index and index value setting table. Secondly, the vulnerability index of the disaster-prone body is calculated according to formula (4), and a disaster-prone body vulnerability distribution map is drawn ( Figure 8 ), and finally, the consequence index of the assessable event chain was obtained based on Table 1. Based on the vulnerability distribution map of hazard-prone bodies, statistics show that among the hazard-prone bodies potentially affected by this oil spill, one protected area is classified as Level I vulnerability, and the other two are classified as Level II vulnerability. Of the 13 open-sea aquaculture sites, three are classified as Level II vulnerability, and the remaining nine are all classified as Level III vulnerability. One anchorage is classified as Level IV vulnerability. Of the three enclosed aquaculture sites, one is classified as Level II vulnerability, and the other two are classified as Level III vulnerability.

[0182] (3) Risk assessment of primary oil spill incidents.

[0183] The primary risk of an oil spill is the expected assessment of the possible impact of the oil spill on the basic assessment unit where the hazard-bearing body is located, and is represented by the risk index of the risk assessment unit where the hazard-bearing body is located.

[0184] In this implementation case, the risk of the primary oil spill event was assessed through three steps. First, the vulnerability index of the hazard-bearing body was mapped to each risk assessment unit through spatial relationships. Second, according to formula (6), the hazard index and vulnerability index of each risk assessment unit were multiplied to obtain the primary event risk index of each risk assessment unit.

[0185] like Figure 9 In this oil spill incident, there were 7 risk assessment units whose original event risks reached Level II, 14 risk assessment units whose risks reached Level III, and 304 risk assessment units whose risks reached Level IV.

[0186] Step D: Risk estimation of secondary / derivative events of oil spills.

[0187] (1) Analysis of secondary / derivative events of oil spills.

[0188] Based on the spatial analysis results of the risk warning area, the types of hazard-bearing objects that may be affected within 24 hours include protected areas, enclosed aquaculture areas, open aquaculture areas, anchorages, etc. Among them, part of the sea area of ​​two enclosed aquaculture areas and one anchorage overlaps with the oil spill monitoring and warning area, and will be greatly affected by this oil spill incident. Other hazard-bearing objects outside the risk warning area will not be affected within 24 hours.

[0189] Table 12 Analysis of disaster-bearing bodies in risk warning of Liaodong Bay oil spill incident

[0190] Sea use type Quantity (pieces) Number of warnings <![CDATA[Area (km 2 )]]> Consequence Index Protected area sea area 3 2 6.53 9 Sea enclosure for aquaculture 3 3 16.15 8 Open sea aquaculture 15 13 47.62 8 anchorage 1 1 5.83 5

[0191] (2) Estimation of the probability of secondary / derivative events.

[0192] The probability of secondary / derivative events is closely related to the risk of the primary oil spill and the distance between the oil spill and the hazard-bearing body. This patent defines it as risk probability and distance probability. The probability of secondary / derivative events is the product of risk probability and distance probability. In this implementation case, the risk probability of each risk assessment unit is first calculated based on the primary event risk. Secondly, the GIS spatial analysis method is used to calculate the closest distance from each hazard-bearing body to the risk warning area. According to the distance probability estimation formula, the distance probability of the risk assessment unit related to each hazard-bearing body is obtained. Finally, the risk probability is multiplied by the distance probability to obtain the probability of secondary / derivative events.

[0193] (3) Calculation of secondary / derivative event risks.

[0194] According to the formula, calculate the secondary / derivative event risk of each risk assessment unit.

[0195] Step E: Risk assessment of the oil spill event chain.

[0196] Based on the primary event risk and secondary / derivative event risk of each risk assessment unit, the risk value of each risk assessment unit is calculated using the oil spill event chain risk calculation formula.

[0197] Step F: Classification of risk warning levels of the oil spill incident chain.

[0198] Using the oil spill event chain risk level correspondence table, each risk assessment unit is assigned a risk warning level. As shown in the figure, for this oil spill incident, there are 10 risk assessment units with event chain risk reaching Level II, 11 with Level III, and 304 with Level IV.

[0199] See also Figure 10 Event chain risk level distribution map of the simulated oil spill incident in Liaodong Bay.

Claims

1. A dynamic risk assessment method for marine oil spills oriented to event chains, including data preparation stage, event chain modeling stage and dynamic risk assessment stage. in, The data preparation phase includes the following steps: Step A: Determine the risk assessment area and risk assessment unit, obtain oil spill information, oil spill monitoring data (oil spill location, oil spill volume, oil film distribution) and oil spill prediction data (oil particle trajectory), dynamically set the risk warning distance based on the oil spill volume in the oil spill monitoring data and the predicted duration of the oil spill drift prediction simulation, merge to generate the risk assessment area, and divide the risk assessment unit; The event chain modeling phase includes the following steps: Step B: Construction of oil spill event chain and node analysis: identification of hazard-bearing body types based on GIS spatial analysis, and construction of an oil spill event chain including secondary / derivative events; The dynamic risk assessment phase includes the following steps: Step C, oil spill primary event risk assessment, starting from the perspective of quantitative risk assessment, calculate the hazard index of the current oil spill event, the vulnerability index of the hazard-bearing body, and the primary event risk index, and perform the oil spill primary event risk assessment; Step D: Risk assessment of secondary / derivative events of oil spills; Step E: Risk assessment of the oil spill chain, which comprehensively assesses the risks of the primary oil spill event and the risks of secondary / derivative events; Step F: Classification of risk warning levels of the oil spill incident chain.

2. The risk assessment method according to claim 1, characterized in that: Step A, determining the risk assessment area and assessment unit, includes obtaining oil spill monitoring and prediction data, determining the risk assessment area, and dividing the risk assessment unit.

3. The risk assessment method according to claim 1, characterized in that: Step B: Determine the hazard-prone objects and assessable event chains of the oil spill incident through spatial analysis, and use GIS spatial analysis to obtain the types of hazard-prone objects included in the risk assessment area.

4. The risk assessment method according to claim 1, characterized in that: Step C, the steps of oil spill primary incident risk assessment include oil spill primary incident hazard assessment, hazard-bearing body vulnerability assessment and oil spill primary incident risk assessment.

5. The risk assessment method according to claim 4, characterized in that: The hazard level of a primary oil spill event is determined by the density of the oil spilled and the average amount of oil spilled per square kilometer in the assessment unit, and is expressed as the oil spill hazard index. It is calculated using formulas (1) to (3): Where: H i ——The hazard index value of the risk assessment unit grid, which is related to the total risk of oil spill incidents. The amount of oil spilled is related to the oil spill thickness of the evaluation unit grid, with a maximum value of 10. If it exceeds 10, it needs to be truncated artificially and is dimensionless; Q oil ——the residual oil spill volume of the oil spill incident, in tons (t); T u ——Oil spill thickness value per square kilometer of risk assessment unit grid, in microns (μm); H T ——the oil spill thickness index value of the risk assessment unit grid, dimensionless; N——the total number of particles predicted for oil spill drift, in pieces; n——The number of predicted particles contained in the current risk assessment unit grid, in pieces. ρ——Density of oil spilled, in kilograms per cubic meter (kg / m 3 ); s——risk assessment unit grid area, in square meters (m 2 ).

6. The risk assessment method according to claim 1, characterized in that: Establishing an oil spill hazard vulnerability assessment index system to evaluate the extent to which the hazard-bearing body is affected by the oil spill includes three steps: (1) Using the analytic hierarchy process, a theoretical framework for the vulnerability assessment index system is established, which includes five levels: target level, criterion level, factor level, assessment object indicator level, and scheme level. (2) Establish an evaluation index system for specific hazard-bearing bodies, and refine the evaluation indexes according to the type and properties of the oil spill hazard-bearing body, including the hazard-bearing body sensitivity classification and vulnerability assessment index setting; (3) Carry out vulnerability assessment of disaster-prone objects and comprehensive vulnerability assessment of evaluation units, and calculate the vulnerability index of disaster-prone objects and comprehensive vulnerability index of evaluation units respectively.

7. The risk assessment method according to claim 6, characterized in that: The vulnerability index of the disaster-bearing body is calculated according to formula (4). Where: V k ——Vulnerability index of the kth hazard-bearing body; S——sensitivity value of the hazard-bearing body; Wx i ——The i-th secondary indicator value corresponding to the attribute of the disaster-prone body; Wy i ——The i-th third-level indicator value corresponding to the attributes of the disaster-prone body; Wz i ——The i-th fourth-level indicator value corresponding to the attributes of the disaster-prone body.

8. The risk assessment method according to claim 1, characterized in that: For the classification of risk levels of primary oil spill events, the natural breakpoint method is used to divide the risk levels of primary oil spill events into 4 levels based on the calculation formula and value range of the oil spill risk index value of the risk assessment unit. The levels are divided according to the size of the risk index score, and the risk colors are marked as red, orange, yellow and blue respectively.

9. The risk assessment method according to claim 1, characterized in that: The risk estimation of secondary / derivative events includes: (1) analysis of the types and consequences of oil spill secondary / derivative events, and (2) estimation of the probability of occurrence of secondary / derivative events.

10. The risk assessment method according to claim 1, characterized in that: The event chain-oriented dynamic risk assessment of oil spills first calculates the event chain risk of each type of hazard-bearing body in the risk assessment area. Secondly, with the grid-based risk warning as the goal, the oil spill event chain risk of each risk assessment unit is calculated.

Images