A method for risk monitoring and control for offshore construction

Abstract

The application discloses a risk monitoring and control method for offshore construction, and particularly relates to the technical field of collision control of offshore construction. The method comprises the following steps: obtaining real-time position, speed, heading, tonnage, wind speed, wind direction, wave height, wave direction, flow rate, ship type, ship historical trajectory and ship behavior mode of a ship, preprocessing the obtained data, inputting the preprocessed multi-source data into a ship trajectory prediction model, outputting multiple possible trajectories and trajectory probabilities of a future time window, calculating collision probability and dynamic safety distance, improving the closeness between collision probability and real collision probability, improving the accuracy of collision probability analysis, quantifying the safety boundary of the ship and the construction area in real time, avoiding the failure of a fixed threshold in a complex environment, providing a dynamic judgment benchmark for risk level classification, improving the risk identification accuracy, and further more accurately matching response instructions to further reduce collision hazards.

Description

Technical Field

[0001] The present invention relates to the technical field of collision control for offshore construction, and more particularly to a risk monitoring and control method for offshore construction. Background Art

[0002] In the field of offshore construction, with the continuous advancement of marine resource development and offshore engineering construction, various offshore construction activities are becoming increasingly frequent, such as offshore oil and gas platform construction, submarine cable laying, offshore wind power installation, etc. These construction activities usually involve a large number of ships, marine engineering equipment and complex underwater operations. Due to the complexity of the marine environment and the limitations of construction space, collision accidents occur from time to time, posing a huge threat to the safety of personnel lives, the marine ecological environment and the construction projects themselves. Therefore, how to achieve collision monitoring and control has become a focus of offshore construction.

[0003] The existing anti-collision method for offshore construction is to arrange a large number of physical buffer facilities in the edge area of ​​offshore construction to absorb energy in the event of a collision to reduce the risk of collision accidents. At the same time, sensors are deployed to monitor the physical distance between the ship and the offshore construction area and the direction of the ship's movement. When there is a geometric overlap between the ship's movement direction and the offshore construction area and the physical distance is less than the safe distance, an early warning is issued, achieving the purpose of early warning and reducing collision risks.

[0004] However, the existing method still has some defects: the existing method judges whether there is geometric overlap between the ship and the construction area based on the current driving state of the ship, and then analyzes the collision probability. The navigation trajectory is relatively simple, and the final calculated collision probability is lower than the actual collision probability. The corresponding decision response may therefore lose accuracy, which is not conducive to reducing collision losses; an early warning is issued when there is geometric overlap and the physical distance is less than the safety distance. The safety distance used is a fixed set value. However, with the dynamic changes in the marine meteorological environment and the driving state of the ship, the fixed safety distance is difficult to accurately describe the collision risk between the ship and the offshore construction area. The safety distance between the ship and the offshore construction area should be dynamically analyzed to accurately identify the collision risk between the ship and the offshore construction area, more accurately match the response instructions, and further reduce the collision hazard. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a risk monitoring and control method for offshore construction to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a risk monitoring and control method for offshore construction, comprising the following steps:

[0007] S1. Offshore construction area deployment: Obtain historical environmental traffic data for the construction area and pre-process it. Perform risk analysis on the pre-processed data and develop a safety zoning plan and monitoring equipment deployment plan based on the analysis results.

[0008] S2. Monitoring equipment deployment control: Deploy monitoring equipment according to the generated monitoring equipment deployment plan, and execute monitoring tasks after confirming that the installation information is correct, the equipment performance is intact, and the communication link stability meets the requirements;

[0009] S3. Online monitoring of the construction area: Deployed monitoring equipment will be used to monitor ship data, visual environment data, meteorological data, and submarine obstacle data in the construction area online.

[0010] S4. Construction Sea Area Data Analysis: Based on online monitoring data of the construction sea area, ship trajectory prediction, collision probability calculation, and risk level classification are carried out, and risk level reports and early warning signals are output;

[0011] S5. Intelligent hierarchical response: After receiving the risk level report and early warning signal, it generates decision instructions based on the risk level and the preset rule base, and prioritizes the decision instructions from high to low risk level. It then distributes the instructions through multiple channels and backs up the response result report;

[0012] S6. Periodic indicator collection: Collect early warning accuracy indicators, response efficiency indicators, accident control indicators, operation reliability indicators, and environmental adaptability indicators within the preset period;

[0013] S7. Anti-collision effect evaluation: Calculate the warning accuracy coefficient, response efficiency coefficient, accident control capability coefficient, operation reliability coefficient, and environmental adaptability coefficient, and evaluate the anti-collision effect based on the calculation results;

[0014] S8. Dynamic update: When the anti-collision effect evaluation fails, the data will be temporarily updated. When the anti-collision effect evaluation passes, the data will be updated regularly. The updated static risk heat map and the qualified anti-collision effect index will be backed up to the control center.

[0015] Technical effects and advantages of the present invention:

[0016] The application obtains the real-time position, speed, heading, tonnage, wind speed, wind direction, wave height, wave direction, flow rate, ship type, ship historical trajectory and ship behavior mode of the ship, removes radar positioning jitter through Kalman filtering after the above, converts the latitude and longitude into a local coordinate system to calculate the relative distance of the ship and the water surface facility or the seabed obstacle in the construction sea area, synchronizes the obtained multi-source data according to a unified timestamp, inputs the processed multi-source data into the constructed ship trajectory prediction model, and outputs multiple possible trajectories, trajectory probability and confidence interval in a future time window, so that the navigation trajectory is more comprehensive, and data support is provided for subsequent collision probability prediction.

[0017] 2、The application retrieves the ship predicted trajectory segment and the construction sea area polygon vertex coordinates, calculates the minimum circumscribed rectangle of the construction sea area polygon and the circumscribed rectangle of the ship predicted trajectory segment, if there is no intersection between the two, the trajectory is skipped, otherwise, the precise detection is entered, for each trajectory segment, each edge of the construction sea area polygon is traversed to detect whether the line segment intersects with the construction sea area edge, if yes, it is determined that the trajectory and the construction sea area exist intersection, otherwise, it is determined that the trajectory and the construction sea area do not exist intersection, after it is determined that the predicted trajectory and the construction sea area exist intersection, the time window matching is performed, the time interval when the predicted trajectory enters the construction sea area is recorded and the potential conflict trajectory is marked, the collision probability and the dynamic safety distance are calculated, the closeness between the collision probability and the real collision probability is improved, the accuracy of the collision probability analysis is improved, the calculated dynamic safety distance quantifies the safety boundary of the ship and the construction area in real time, the failure of the fixed threshold in the complex environment is avoided, the dynamic judgment benchmark for the risk level classification is provided, the risk identification accuracy is improved, and the response instruction is matched more accurately, and the collision hazard is further reduced. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 The method steps of the application.

[0019] Figure 2 The system structure block diagram of the application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0021] As Figure 1 shown, the embodiment provides a risk monitoring and control method for offshore construction, including the following steps:

[0022] S1, offshore construction area deployment: after preprocessing the historical environmental traffic data of the construction sea area, risk analysis is performed on the preprocessed data, and based on the analysis results, a safety zoning scheme and a monitoring equipment deployment scheme are set;

[0023] Further, the historical environmental traffic data of the construction sea area includes historical meteorological data, historical ship traffic data, obstacle data and historical collision accident data, wherein the historical meteorological data includes wind speed, wave height and visibility; the historical ship traffic data includes AIS historical trajectory information containing ship ID, speed, heading and tonnage, and ship type distribution information; the obstacle data includes seabed obstacle coordinates and water surface facility position coordinates; the historical collision accident refers to the report of the offshore construction collision accident.

[0024] Further, the risk analysis includes static risk analysis and probability-consequence matrix analysis, the static risk analysis obtains a static risk heat map of the offshore construction area, and the probability-consequence matrix analysis obtains a risk event resource allocation priority, and simultaneously based on the static risk heat map of the offshore construction area and the probability-consequence matrix, a safety zoning scheme and a monitoring equipment deployment scheme are set.

[0025] Specifically, the specific steps of cleaning and fusing the obtained data in this embodiment are as follows: the linear interpolation method is used to complete the position of the signal loss section in the AIS trajectory data, and the moving window mean value method is used to correct the abnormal value of the meteorological data; different sources of geographic data are uniformly converted into WGS-84 coordinate system; the ship trajectory and meteorological data are matched according to time stamp and geographical position using a space-time correlation algorithm (such as the correlation between the speed of ship A at a certain time and the local wind speed); multi-source data is layered and superimposed on the GIS platform, the basic layer data includes bathymetric data and channel boundary, the dynamic layer data includes ship trajectory heat data and wind speed vector field data, and the event layer data includes historical accident points and obstacle coordinates.

[0026] Specifically, the specific steps of static risk analysis are as follows:

[0027] S111, dividing the offshore construction area into a plurality of grids with a preset fixed area A grid ;

[0028] S112, calculating ship flow density D ship , obstacle density D obstacle , meteorological risk coefficient R weather , historical accident frequency F accident and ship type risk coefficient R type , the ship flow density is represented as the ratio of the number of ships N ship passing through the grid in the historical data period to the grid area A grid , and the specific formula is: , obstacle density is expressed as the cumulative number of obstacles N in the grid during the historical data period obstacle and the grid area A grid The specific formula is: , the meteorological risk coefficient is expressed as the historical wind speed V of the i-th group in the historical data period wind,i With the set maximum wind speed V wmax The ratio of the value and the set weight coefficient a1 multiplied by the historical wave height H wave,i With the set maximum wave height H wmax The ratio of the value of the multiplication of the set weight coefficient a2 and 1 minus the historical visibility L visibility,i With the minimum visibility L set vmin The cumulative sum of the sum of the three and the multiplication of the set weight coefficient a3 is the average result. The specific formula is: , N q is the number of meteorological data groups in the historical data period, and the historical accident frequency is expressed as the total number of collision accidents in the grid during the historical data period, N accident The statistical number of years N of the historical data cycle y The specific formula is: , the ship type risk coefficient is expressed as the ship type weight b i and the number of ship types N ai The summary sum of the products is as follows: ;

[0029] S113. Calculate the normalized ship flow density D based on the minimum-maximum normalization formula ship * , obstacle density D obstacle * , meteorological risk factor R weather * , historical accident frequency F accident * and the ship type risk factor R type * , the specific formula is as follows: 、 、 、 、 , where D ship,min 、D ship,max 、D obstacle,min 、D obstacle,max 、R weather,min 、R weather,max 、F accident,min 、F accident,max 、R type,min 、R type,maxThe order is minimum and maximum values ​​of ship traffic density, minimum and maximum values ​​of obstacle density, minimum and maximum values ​​of meteorological risk factor, minimum and maximum values ​​of historical accident frequency, and minimum and maximum values ​​of ship type risk factor;

[0030] S114. Calculate the grid static risk value R grid , the specific formula is: , R grid ∈[0,1], divided into levels according to the threshold, that is, R grid <0.4 is low risk, 0.4≦R grid ≦0.7 is medium risk, 0.7 <R grid For high risk, c1, c2, c3, c4, and c5 are the weight coefficients of the corresponding risk factors;

[0031] S115. Apply GIS kernel density analysis and smoothing to generate a continuous risk distribution map, which is then colored according to risk values.

[0032] It should be specifically noted that in this embodiment, the maximum or minimum values ​​used are all derived from historical data. The weight coefficients of the three influencing factors of the meteorological risk coefficient can be calculated by the entropy weight method. The entropy weight method is an existing technology, so the specific calculation process is not given here. A method for determining the weight coefficient of the risk factor is now given. The specific steps are as follows:

[0033] A1. Collect p risk factor data from n samples to form a data matrix B (dimension n×p). Standardize each variable so that its mean is 0 and its standard deviation is 1. Then we have , μ j is the mean of the jth variable, σ j is the standard deviation of the j-th variable;

[0034] A2. Calculate the correlation coefficient matrix R (dimension p×p) of the standardized data matrix C. ;

[0035] A3. Perform eigendecomposition on R and obtain eigenvalues ​​λ1≧λ2≧...≧λ p and the corresponding eigenvectors v1, v2, ..., v p , calculate the variance contribution rate d of each principal component i , then , extract the top k principal components whose cumulative contribution rate exceeds the set threshold;

[0036] A4. Load L of the i-th principal component on variable j ij for , v ij For the jth element of the i-th eigenvector, construct a p×k loading matrix L;

[0037] A5. For each variable j, calculate its comprehensive score F on the first k principal components. j , then ;

[0038] A6. Normalize the comprehensive score to obtain the weight coefficient c of the j-th risk factor j ,Right now .

[0039] Specifically, it should be noted that the specific steps of the probability-consequence matrix analysis are as follows:

[0040] S121. Define three probability levels: high, medium, and low. For example, annual probability of occurrence ≥ 10% is classified as high, annual probability of occurrence between 1% and 10% is classified as medium, and annual probability of occurrence ≤ 1% is classified as low.

[0041] S122. Define four consequence levels: Level 1: no casualties and minor property damage; Level 2: minor injuries and partial equipment damage; Level 3: severe injuries and major equipment damage; and Level 4: death and severe structural damage.

[0042] S123. Define the high, medium, and low probabilities of first-level consequences as low risk, acceptable, and acceptable, respectively; define the high, medium, and low probabilities of second-level consequences as medium risk, low risk, and acceptable, respectively; define the high, medium, and low probabilities of third-level consequences as high risk, medium risk, and low risk, respectively; and define the high, medium, and low probabilities of fourth-level consequences as extremely high risk, high risk, and medium risk, respectively;

[0043] S124. When multiple risk probability events occur, they shall be sorted from high to low according to the consequence level. If the consequence levels are the same, they shall be sorted from high to low according to the probability level, and resources shall be allocated according to the order of arrangement.

[0044] Specifically, in this embodiment, it is important to note that the risk heat map is a comprehensive risk assessment based on spatial distribution, quantifying the static risk values ​​of different areas (such as vessel traffic, obstacle density, and meteorological conditions). It can visually display the comprehensive risk level of each area (e.g., high, medium, or low) through color or numerical values. However, it cannot distinguish specific risk types (e.g., collision, equipment failure, or person overboard), nor can it clearly define the dynamic characteristics of risk (e.g., the combined relationship between the probability of an event and the severity of its consequences). The probability-consequence matrix categorizes specific risk events based on the two dimensions of probability (likelihood of occurrence) and consequence (severity of impact), categorizing risk events as acceptable, low, medium, high, and extremely high, providing guidance on which events should be prioritized. It can clarify the dynamic characteristics of risk (e.g., the difference between "high probability and low consequences" and "low probability and high consequences") and match differentiated response strategies to different risk events (e.g., extremely high risks require immediate suspension of work, while low risks require only monitoring). While the risk heat map identifies "where risks are high," the probability-consequence matrix answers "which risks require priority." If the risk value of a certain area in the heat map is high, it may be dominated by "large ship traffic (high probability) - minor consequences (C1)". In fact, it is necessary to give priority to areas with medium risk values ​​but "low probability - catastrophic consequences (C4)".

[0045] Specifically, in this example, a probability-consequence matrix analysis can avoid misleading comprehensive risk values. For example, if a high comprehensive risk value is caused by "frequent directional changes by fishing boats (high probability)" in a certain area, but the consequence is only a minor collision (C1), its actual priority may be lower than another area, such as "uncontrolled merchant ship striking a platform (low probability but C4)." Using the probability-consequence matrix, high-consequence events can be individually labeled as extremely high risk, even if their comprehensive risk value is moderate. Extremely high risks (such as C4 consequences) require immediate resource deployment (e.g., deploying additional patrol vessels, upgrading protective equipment), while low risks (such as C1 consequences) require only routine monitoring. If resources are limited, prioritize "low probability, catastrophic consequences" (e.g., a platform anchor chain break caused by a typhoon) over "high probability, minor consequences" (e.g., a collision with a small floating object).

[0046] S2. Monitoring equipment deployment control: Deploy monitoring equipment according to the generated monitoring equipment deployment plan, and execute monitoring tasks after confirming that the installation information is correct, the equipment performance is intact, and the communication link stability meets the requirements;

[0047] S3. Online monitoring of the construction area: Deployed monitoring equipment will be used to monitor ship data, visual environment data, meteorological data, and submarine obstacle data in the construction area online.

[0048] What needs to be specifically explained in this embodiment is that the ship data of the online monitoring construction sea area includes ship dynamic information, ship identity information, ship navigation status information, real-time images of the construction sea area, wind condition data, wave data, construction sea area visibility data, seabed obstacle data and surface facility status data. The ship dynamic information includes the ship's latitude and longitude coordinates, speed, heading and tonnage; the ship identity information includes the maritime service mobile identification, ship registration information and ship type; the ship status information includes the starting point, destination, estimated arrival time, draft and navigation status mark; the wind condition data includes wind speed and wind direction; the wave data includes wave height and wave direction; the construction sea area visibility data refers to optical visibility; the seabed obstacle data includes the three-dimensional model coordinates and height of the obstacle; the surface facility status data refers to the position coordinates of the surface facility.

[0049] S4. Construction Sea Area Data Analysis: Based on online monitoring data of the construction sea area, ship trajectory prediction, collision probability calculation, and risk level classification are carried out, and risk level reports and early warning signals are output;

[0050] It should be specifically explained in this embodiment that the specific steps of ship trajectory prediction are as follows:

[0051] S411. Obtain the real-time position, speed, heading, tonnage, wind speed, wind direction, wave height, wave direction, current speed, ship type, ship historical trajectory, and ship behavior pattern of the ship;

[0052] S412. Use Kalman filtering to eliminate radar positioning jitter, convert longitude and latitude into a local coordinate system to calculate the relative distance between the ship and surface facilities or submarine obstacles in the construction area, and synchronize the acquired multi-source data with a unified timestamp.

[0053] S413: Input the multi-source data processed by S412 into the constructed ship trajectory prediction model, and output multiple possible trajectories, trajectory probabilities, and confidence intervals in the future time window.

[0054] Furthermore, the specific steps for calculating the collision probability are as follows:

[0055] S421, retrieve the predicted ship trajectory segment and the coordinates of the vertices of the construction sea area polygon;

[0056] S422, calculating the minimum enclosing rectangle of the construction sea area polygon and the enclosing rectangle of the ship's predicted trajectory segment. If the two do not intersect, skip the trajectory; otherwise, proceed to precise detection;

[0057] S423, for each trajectory line segment, traverse each edge of the construction sea area polygon, and detect whether the line segment intersects with the construction sea area edge. If so, it is determined that the trajectory and the construction sea area have an intersection; otherwise, it is determined that the trajectory and the construction sea area do not have an intersection;

[0058] S424: After determining that the predicted trajectory intersects the construction sea area, perform time window matching, record the time interval when the predicted trajectory enters the construction sea area, and mark the potential conflicting trajectory;

[0059] S425, calculate collision probability P a , the specific formula is: , e i is the probability of the i-th trajectory, I(·) is the indicator function, and the value is 1 if the trajectory i overlaps with the construction sea area, otherwise the value is 0;

[0060] S426, calculate dynamic safety distance L a , the specific formula is: , v a , t x 、a h , σ a They are the current speed of the ship, the response time of the ship, the braking acceleration of the ship, and the standard deviation of the ship positioning error.

[0061] Furthermore, the specific steps for risk level classification are as follows:

[0062] S431. Extract static risk value R grid With collision probability P a ;

[0063] S432. Calculate the environmental correction factor P h , the specific formula is: , V wind is the real-time wind speed, H wave is the real-time wave height, L visibility For real-time visibility, e V 、e H 、e L is the weight coefficient of wind speed, wave height and visibility, which can be calculated and updated by entropy weight method;

[0064] S433. Calculate the dynamic risk coefficient R d , the specific formula is: , f1f2f3 are the weight coefficients of static risk value, collision probability and environmental correction factor respectively, which can be calculated and updated by entropy weight method;

[0065] S434、R d ∈[0,1], divided into levels according to the threshold, that is, R d <0.3 is low risk, 0.3≦R d ≦0.6 is medium risk, 0.6 <R d High risk, which is triggered directly when the minimum distance between the ship and the construction sea area is less than the dynamic safety distance;

[0066] S435. Automatically compare the calculated dynamic risk coefficient with the risk threshold, and output a warning signal based on the comparison result.

[0067] S5. Intelligent hierarchical response: After receiving the risk level report and early warning signal, it generates decision instructions based on the risk level and the preset rule base, and prioritizes the decision instructions from high to low risk level. It then distributes the instructions through multiple channels and backs up the response result report;

[0068] S6. Periodic indicator collection: Collect early warning accuracy indicators, response efficiency indicators, accident control indicators, operation reliability indicators, and environmental adaptability indicators within the preset period;

[0069] Furthermore, the warning accuracy index includes the ship collision warning accuracy rate X Ua , obstacle misrecognition rate X Ub And the false negative rate X Uc The response efficiency index includes the average response time coefficient X Ea And the instruction execution success rate X Eb The accident control indicators include the collision accident rate in the construction area X Ra and dynamic safety distance compliance rate X Rb ; Operation reliability indicators include monitoring equipment online rate X Ta , Communication link availability coefficient X Tb And the average fault tolerance recovery coefficient X Tc Environmental adaptability indicators include the success rate of multi-objective conflict handling X Ga and severe weather warning effectiveness X Gb .

[0070] In this embodiment, it should be specifically explained that the ship collision warning accuracy rate X Ua It is defined as the proportion of correctly warned ship collision events, and the calculation formula is: , Mz qyj 、M zyj The numbers are the number of actual collisions that were warned by the system and the number of all warnings triggered by the system (including false alarms). The obstacle misidentification rate XUb is defined as the proportion of non-obstacles (such as waves and fish schools) mistakenly identified as obstacles by radar / sonar. The calculation formula is: , M wpz 、M zjcz The order is the number of misjudged obstacles, the total number of detected obstacles; the missed alarm rate X Uc It is defined as the proportion of actual collisions that were not warned by the system. The calculation formula is: , M lbpz 、M sjpzThey are the number of missed collisions and the number of actual collisions, respectively. The average response timeliness coefficient XEa is defined as the normalized result of the average time taken from the generation of risk warning to the execution of collision avoidance measures. The calculation formula is: , N2 is the number of effective warning events, t bi , t ai The time when the i-th anti-collision measure is completed and the time when the i-th risk warning is generated are respectively; the instruction execution success rate X Eb It is defined as the ratio of successful execution of anti-collision instructions (such as ship steering, anchor chain adjustment), and the calculation formula is: , M cgzx 、M zzl The number of successful executions of anti-collision commands and the total number of anti-collision commands are in turn; the collision accident rate in the construction area is X Ra It is defined as the frequency of collision accidents occurring within a unit of construction time, and the calculation formula is: , M pzsg , t zsc The order is the number of collision accidents, the total construction time; the dynamic safety distance compliance rate X Rb It is defined as the proportion of the ship's final avoidance distance that meets the dynamic safety threshold. The calculation formula is: , M dbbr 、M zbr The order is the number of avoidance events that meet the standards, the total number of avoidance events; the online rate of monitoring equipment X Ta It is defined as the percentage of uptime of key equipment such as radar, AIS, and sonar. The calculation formula is: , t sbzc is the normal time of the equipment; the communication link availability coefficient X Tb It is defined as the proportion of time that the communication link operates without failure, and the calculation formula is: , t txzc is the normal communication duration; the average fault tolerance recovery capability coefficient X Tc It is defined as the normalized average recovery time from device failure to backup system takeover, calculated as: , N3 is the number of equipment failure events, t dj , t cj are the jth equipment recovery time and the jth equipment failure time respectively; the success rate of multi-objective conflict processing X Ga It is defined as the success rate of handling conflicts with ≥3 ships or obstacles at the same time, and is calculated as: , M acg 、M act The number of successful multi-target processing events and the number of multi-target conflict events are respectively; the efficiency of severe weather warning is X Gb It is defined as the accuracy of the system warning when the wave height is greater than 3 meters or the visibility is less than 1 km. The calculation formula is: , M etzy, M zety The number of correct severe weather early warning, the total number of severe weather early warning.

[0071] S7, collision avoidance effect evaluation: calculate the warning accuracy coefficient, response efficiency coefficient, accident control ability coefficient, operation reliability coefficient and environmental adaptability coefficient, and evaluate the collision avoidance effect based on the calculation results;

[0072] Further, the specific steps of collision avoidance effect evaluation are as follows:

[0073] S71, calculate the warning accuracy coefficient Y Ua , the specific formula is: To avoid the denominator being 0, the denominator in the formula is compensated by 1;

[0074] S72, calculate the response efficiency coefficient Y Ea , the specific formula is: ;

[0075] S73, calculate the accident control ability coefficient Y Ra , the specific formula is: To avoid the denominator being 0, the denominator in the formula is compensated by 1;

[0076] S74, calculate the operation reliability coefficient Y Ta , the specific formula is: ;

[0077] S75, calculate the environmental adaptability coefficient Y Ga , the specific formula is: ;

[0078] S76, compare the calculated warning accuracy coefficient, response efficiency coefficient, accident control ability coefficient, operation reliability coefficient and environmental adaptability coefficient with the corresponding expected value respectively, if the number of indicators with calculated value lower than expected value is more than half, it is determined that the collision avoidance effect evaluation is unqualified, otherwise it is determined that the collision avoidance effect is qualified.

[0079] S8, dynamic update: when the collision avoidance effect evaluation is unqualified, enter data temporary update, when the collision avoidance effect evaluation is qualified, change to data timing update, backup the updated static risk heat map and qualified collision avoidance effect index to the control center.

[0080] Further, data temporary update includes updating the static risk heat map, safety zone division scheme, monitoring equipment deployment scheme and ship trajectory prediction model, the specific calculation formula of the updated static risk value R grid,new is: , λ s , R grid,old , M xsg , M zts , Mgfx 、M zyj The order is historical data attenuation factor, static risk value before update, number of new collision accidents, total number of days, average daily high-risk warning times, and total warning times. When the anti-collision effect is determined to be qualified, the anti-collision effect index ZP is calculated. The specific formula is: ,Y Ub 、Y Eb 、Y Rb 、Y Tb 、Y Gb They are the expected value of the early warning accuracy coefficient, the expected value of the response efficiency coefficient, the expected value of the accident control capability coefficient, the expected value of the operation reliability coefficient and the expected value of the environmental adaptability coefficient.

[0081] It should be specifically noted in this embodiment that the expected values, set values, and preset values ​​used are all selected based on actual operation requirements and are not limited to specific values ​​here.

[0082] like Figure 2 The embodiment shown provides a collision monitoring and control system for offshore construction, including an offshore construction area deployment plan generation module, a monitoring equipment deployment control module, a construction sea area online monitoring module, a construction sea area ship trajectory prediction module, a construction sea area ship collision probability analysis module, a construction sea area risk level classification module, an intelligent hierarchical response module, an indicator data collection module, an anti-collision effect evaluation module, a dynamic update module and a database. The offshore construction area deployment plan generation module, the monitoring equipment deployment control module, the construction sea area online monitoring module, the construction sea area ship trajectory prediction module, the construction sea area ship collision probability analysis module, the construction sea area risk level classification module, the intelligent hierarchical response module, the indicator data collection module, the anti-collision effect evaluation module and the dynamic update module are connected in sequence. The offshore construction area deployment plan generation module is connected to the construction sea area risk level classification module. The construction sea area online monitoring module is connected to the construction sea area ship collision probability analysis module and the construction sea area risk level classification module. All modules in the system are connected to the database.

[0083] The offshore construction area deployment plan generation module obtains historical environmental traffic data of the construction sea area and performs preprocessing, performs risk analysis based on the preprocessed data, and sets a safety zoning plan and a monitoring equipment deployment plan based on the analysis results;

[0084] The monitoring device deployment control module deploys the monitoring devices according to the generated monitoring device deployment plan, and executes the monitoring task after confirming that the installation information is correct, the device performance is intact, and the communication link stability meets the requirements;

[0085] The construction sea area online monitoring module monitors the ship data, visual environment data, meteorological data and seabed obstacle data of the construction sea area online through the deployed monitoring equipment;

[0086] The ship trajectory prediction module in the construction sea area predicts the ship trajectory based on the online monitoring data of the construction sea area;

[0087] The construction sea area ship collision probability analysis module is used to analyze the construction sea area ship collision probability;

[0088] The construction sea area risk level classification module is used to classify the construction sea area risk levels;

[0089] After receiving the risk level report and warning signal, the intelligent hierarchical response module generates decision instructions based on the risk level and the preset rule base, and prioritizes the decision instructions from high to low according to the risk level. It then distributes the instructions through multiple channels and backs up the response result report;

[0090] The indicator data collection module is used to collect early warning accuracy indicators, response efficiency indicators, accident control indicators, operation reliability indicators and environmental adaptability indicators within a preset period;

[0091] The anti-collision effect evaluation module calculates the warning accuracy coefficient, response efficiency coefficient, accident control capability coefficient, operation reliability coefficient and environmental adaptability coefficient based on the warning accuracy indicators, response efficiency indicators, accident control indicators, operation reliability indicators and environmental adaptability indicators within a preset period, and evaluates the anti-collision effect based on the calculation results;

[0092] The dynamic update module enters temporary data update when the anti-collision effect evaluation fails, and switches to regular data update when the anti-collision effect evaluation passes, and backs up the updated static risk heat map and qualified anti-collision effect index to the control center;

[0093] The database is used to store data information of all modules in the system.

[0094] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A risk monitoring and control method for offshore construction, characterized by: The following steps are involved: S1. Offshore construction area deployment: Obtain historical environmental traffic data for the construction area and pre-process it. Perform risk analysis on the pre-processed data and develop a safety zoning plan and monitoring equipment deployment plan based on the analysis results. The risk analysis includes static risk analysis and probability-consequence matrix analysis. The static risk analysis obtains a static risk heat map of the offshore construction area, and the probability-consequence matrix analysis obtains the priority of risk event resource allocation. At the same time, the safety zoning plan and monitoring equipment deployment plan are set based on the static risk heat map and probability-consequence matrix of the offshore construction area. The specific steps of the static risk analysis are as follows: S111. Divide the offshore construction area into a plurality of grids of predetermined fixed areas; S112. Calculate ship traffic density, obstacle density, meteorological risk factor, historical accident frequency, and ship type risk factor; S113. Calculate the normalized ship flow density, obstacle density, meteorological risk factor, historical accident frequency, and ship type risk factor based on the minimum-maximum normalization formula; S114. Calculate the grid static risk value R grid , R grid ∈[0,1], divided into levels according to the threshold, that is, R grid <0.4 is low risk, 0.4≦R grid ≦0.7 is medium risk, 0.7 <R grid For high risk, c1, c2, c3, c4, and c5 are the weight coefficients of the corresponding risk factors; S115. Apply GIS kernel density analysis and smoothing to generate a continuous risk distribution map, which is colored by risk value. S2. Monitoring equipment deployment control: Deploy monitoring equipment according to the generated monitoring equipment deployment plan, and execute monitoring tasks after confirming that the installation information is correct, the equipment performance is intact, and the communication link stability meets the requirements; S3. Online monitoring of the construction area: Deployed monitoring equipment will be used to monitor ship data, visual environment data, meteorological data, and submarine obstacle data in the construction area online. S4. Construction Sea Area Data Analysis: Based on online monitoring data of the construction sea area, ship trajectory prediction, collision probability calculation, and risk level classification are carried out, and risk level reports and early warning signals are output; The specific steps of calculating the collision probability are as follows: S421, retrieve the predicted ship trajectory segment and the coordinates of the vertices of the construction sea area polygon; S422, calculating the minimum enclosing rectangle of the construction sea area polygon and the enclosing rectangle of the ship's predicted trajectory segment. If the two do not intersect, skip the trajectory; otherwise, proceed to precise detection; S423, for each trajectory line segment, traverse each edge of the construction sea area polygon, and detect whether the line segment intersects with the construction sea area edge. If so, it is determined that the trajectory and the construction sea area have an intersection; otherwise, it is determined that the trajectory and the construction sea area do not have an intersection; S424: After determining that the predicted trajectory intersects the construction sea area, perform time window matching, record the time interval when the predicted trajectory enters the construction sea area, and mark the potential conflicting trajectory; S425, calculate collision probability P a , the specific formula is: , e i is the probability of the i-th trajectory, I(·) is the indicator function, and the value is 1 if the trajectory i overlaps with the construction sea area, otherwise the value is 0; S426, calculate dynamic safety distance L a , the specific formula is: , v a , t x 、a h , σ a They are the current speed of the ship, the response time of the ship, the braking acceleration of the ship, and the standard deviation of the ship positioning error; S5. Intelligent hierarchical response: After receiving the risk level report and early warning signal, it generates decision instructions based on the risk level and the preset rule base, and prioritizes the decision instructions from high to low risk level. It then distributes the instructions through multiple channels and backs up the response result report; S6. Periodic indicator collection: Collect early warning accuracy indicators, response efficiency indicators, accident control indicators, operation reliability indicators, and environmental adaptability indicators within a preset period; S7. Anti-collision effect evaluation: Calculate the warning accuracy coefficient, response efficiency coefficient, accident control capability coefficient, operation reliability coefficient, and environmental adaptability coefficient, and evaluate the anti-collision effect based on the calculation results; S8. Dynamic update: When the anti-collision effect evaluation fails, the data will be temporarily updated. When the anti-collision effect evaluation passes, the data will be updated regularly. The updated static risk heat map and the qualified anti-collision effect index will be backed up to the control center.

2. The risk monitoring and control method for offshore construction according to claim 1, characterized in that: The specific steps of the probability-consequence matrix analysis are as follows: S121. Define three probability levels: high, medium, and low; S122. Define four consequence levels: Level 1: no casualties and minor property damage; Level 2: minor injuries and partial equipment damage; Level 3: severe injuries and major equipment damage; and Level 4: death and severe structural damage. S123. Define the high, medium, and low probabilities of first-level consequences as low risk, acceptable, and acceptable, respectively; define the high, medium, and low probabilities of second-level consequences as medium risk, low risk, and acceptable, respectively; define the high, medium, and low probabilities of third-level consequences as high risk, medium risk, and low risk, respectively; and define the high, medium, and low probabilities of fourth-level consequences as extremely high risk, high risk, and medium risk, respectively; S124. When multiple risk probability events occur, they shall be sorted from high to low according to the consequence level. If the consequence levels are the same, they shall be sorted from high to low according to the probability level, and resources shall be allocated according to the order of arrangement.

3. The risk monitoring and control method for offshore construction according to claim 1, characterized in that: The specific steps of risk level classification are as follows: S431. Extract static risk value R grid With collision probability P a ; S432. Calculate the environmental correction factor P h , the specific formula is: , V wind is the real-time wind speed, H wave is the real-time wave height, L visibility For real-time visibility, e V 、e H 、e L is the weight coefficient of wind speed, wave height and visibility, which can be calculated and updated by entropy weight method; S433. Calculate the dynamic risk coefficient R d , the specific formula is: , f1f2f3 are the weight coefficients of static risk value, collision probability and environmental correction factor respectively, which can be calculated and updated by entropy weight method; S434、R d ∈[0,1], divided into levels according to the threshold, that is, R d <0.3 is low risk, 0.3≦R d ≦0.6 is medium risk, 0.6 <R d High risk, which is triggered directly when the minimum distance between the ship and the construction sea area is less than the dynamic safety distance; S435. Automatically compare the calculated dynamic risk coefficient with the risk threshold, and output a warning signal based on the comparison result.

4. The risk monitoring and control method for offshore construction according to claim 1, characterized in that: The historical environmental traffic data of the construction sea area includes historical meteorological data, historical ship traffic data, obstacle data and historical collision accident data, among which the historical meteorological data includes wind speed, wave height and visibility; the historical ship traffic data includes AIS historical track information containing ship ID, speed, heading and tonnage, and ship type distribution information; the obstacle data includes the coordinates of seabed obstacles and the location coordinates of surface facilities; and historical collision accidents refer to reports of offshore construction collision accidents.

5. The risk monitoring and control method for offshore construction according to claim 1, characterized in that: The warning accuracy indicators include ship collision warning accuracy, obstacle misidentification rate and missed alarm rate; the response efficiency indicators include average response timeliness coefficient and instruction execution success rate; the accident control indicators include construction zone collision accident rate and dynamic safety distance compliance rate; Operational reliability indicators include the online rate of monitoring equipment, communication link availability coefficient, and average fault tolerance recovery coefficient; environmental adaptability indicators include the success rate of multi-target conflict processing and the effectiveness of severe weather warning.

6. The risk monitoring and control method for offshore construction according to claim 1, characterized in that: After calculating the warning accuracy coefficient, response efficiency coefficient, accident control capability coefficient, operation reliability coefficient and environmental adaptability coefficient, the anti-collision effect evaluation process will compare the calculated warning accuracy coefficient, response efficiency coefficient, accident control capability coefficient, operation reliability coefficient and environmental adaptability coefficient with the corresponding expected values. If more than half of the indicators with calculated values ​​lower than the expected values ​​are judged to be unqualified, otherwise the anti-collision effect is judged to be qualified.

Images