CN115239110A - Navigation risk evaluation method based on improved TOPSIS method - Google Patents
Navigation risk evaluation method based on improved TOPSIS method Download PDFInfo
- Publication number
- CN115239110A CN115239110A CN202210825865.6A CN202210825865A CN115239110A CN 115239110 A CN115239110 A CN 115239110A CN 202210825865 A CN202210825865 A CN 202210825865A CN 115239110 A CN115239110 A CN 115239110A
- Authority
- CN
- China
- Prior art keywords
- risk
- index
- area
- accident
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000011156 evaluation Methods 0.000 title claims abstract description 42
- 238000011160 research Methods 0.000 claims abstract description 20
- 238000003064 k means clustering Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 58
- 239000013598 vector Substances 0.000 claims description 41
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000012502 risk assessment Methods 0.000 claims description 10
- 230000000694 effects Effects 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 9
- 206010039203 Road traffic accident Diseases 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 230000005484 gravity Effects 0.000 claims description 5
- 238000011835 investigation Methods 0.000 claims description 5
- 238000005452 bending Methods 0.000 claims description 3
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 230000036961 partial effect Effects 0.000 claims description 3
- 230000002441 reversible effect Effects 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 229910052731 fluorine Inorganic materials 0.000 claims description 2
- 125000001153 fluoro group Chemical group F* 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 4
- 238000007726 management method Methods 0.000 abstract description 2
- 238000013468 resource allocation Methods 0.000 abstract 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 10
- 238000005096 rolling process Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 4
- 230000002411 adverse Effects 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 101100057959 Mus musculus Atxn1l gene Proteins 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 231100001261 hazardous Toxicity 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000009360 aquaculture Methods 0.000 description 1
- 244000144974 aquaculture Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000414 obstructive effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
- 239000003643 water by type Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a navigation risk evaluation method based on an improved TOPSIS method, which is characterized in that a K-means clustering method based on minimum circle coverage is utilized to identify and divide accident-prone areas in a researched sea area, then traffic factors, ship factors and channel conditions are comprehensively considered to establish a relative navigation risk evaluation index system, index weights obtained by a hierarchical analysis method and an entropy method are combined through a game theory to obtain a combined weight of each index, finally, the relative navigation risk of each accident-prone area is evaluated by utilizing the improved TOPSIS method, the average risk is used as a risk threshold value to judge the position of each high-risk area, the accuracy of navigation risk evaluation is improved, a basis and a basis are provided for marine emergency rescue resource allocation research, a decision reference is provided for marine safety management work, and relevant managers can perform corresponding human and material resource allocation according to the calculated risk.
Description
Technical Field
The invention belongs to the technical field of navigation risk assessment of ships in sea areas, and particularly relates to a navigation risk assessment method based on an improved TOPSIS method.
Background
In recent years, with the continuous development of economy, the throughput of ports in China is rapidly increased, the problem of ship navigation safety is more and more emphasized, and the ship navigation safety is influenced by a plurality of factors; the ship navigation risk evaluation is to perform system analysis and screening on all factors influencing ship safety, determine key factors capable of reflecting the risk level, establish a risk evaluation index system, judge all indexes by means of a qualitative or quantitative model, obtain a risk value capable of reflecting the overall risk level of a ship, and provide decision support for ship safety management, wherein the main navigation safety hidden dangers in a sea area are as follows:
(1) Adverse weather and sea conditions effects
The hidden danger that the occurrence of traffic accidents on the sea can not be stopped forever is one of the factors causing the occurrence of adverse weather and sea conditions such as typhoon, big waves and the like. On one hand, when the ship sails under bad weather and sea conditions such as strong wind and big waves, the handling performance of the ship is very easily influenced, and particularly for small ships, the ship sails in danger under the condition, and ship overturning accidents are particularly easy to happen. On the other hand, if the ship has an accident in bad weather and sea conditions such as strong wind, strong waves and the like, the emergency rescue environment is more difficult, the help of the peripheral ship is difficult to obtain at the first time of the accident, and the implementation of emergency rescue under the bad conditions such as professional rescue force of a maritime administration and a rescue administration is also a great challenge, so that the emergency rescue efficiency is greatly reduced, the emergency rescue effect is greatly reduced, and the accident consequence is more serious.
(2) Navigation risk of passenger rolling ship and dangerous goods ship
The passenger-rolling ship is used as a main transport tool for transporting marine personnel, the lives of all the personnel on the ship are threatened once an accident occurs, and the region where a passenger-rolling route is located is one of the centers of gravity of the strait offshore safety supervision. And with the increase of the scale and the number of the distribution ships of each port and the increase of the demands of the port for entering and exiting, the density of the ships in the area where the passenger rolling route passes through is continuously increased, the probability of meeting the ships is greatly increased, and the probability of collision is increased accordingly. Furthermore, the strait includes a hazardous cargo wharf in addition to the passenger roller quay, and thus the proportion of hazardous cargo vessels in the strait is relatively high. Oil tankers, LNG ships and other dangerous goods ships are prone to causing large-scale sea area environmental pollution after accidents occur, and when emergency disposal is performed on the oil tankers, large-area traffic control needs to be performed, which can adversely affect the navigation efficiency of straits and the operation of each port. And dangerous goods ship is in danger and is easy to develop into accidents such as explosion, fire and the like, which not only threatens the personnel on the ship and the safety of the ship, but also threatens the ships in the peripheral sea area.
(3) Influence of obstructions such as submerged reef and sunken ship and movement of fishing boat
The water depth span of the strait is large, and the ship is easy to have reef touch and grounding accidents at the severe water depth change point and the submerged reef. In addition, the strait sea area includes a plurality of shoals, and if the navigation mark is displaced in typhoon weather, misleading the ship driver, and driving into the shoal area, the ship is also likely to have grounding or bottom-touching accidents. In addition, due to severe topography fluctuation, complex flow state and large flow rate in individual areas of the strait, sometimes, after the ship has accidents such as self-sinking and overturning, the sunken ship cannot be processed in time, which may also have certain influence on the passing of the ship. The strait is the breeding development of aquaculture and fishery fishing industry, but the fishing boats are relatively small in size, and most fishing boats are not equipped with AIS systems according to requirements, other peripheral ships can only judge the trails of the fishing boats by radar and visual inspection methods sometimes, if the fishing boats are not concerned, collision accidents with the fishing boats are easily caused, generally speaking, people driving the fishing boats are fishermen who catch fish and are different from professional captain and drivers, the fishermen may not receive professional training, lack of certain safety consciousness and emergency capacity, and are easy to be confused when encountering emergency, so that the consequences are aggravated. In addition, by analyzing the accident dangerous situation of the strait, the number of accidents caused by factors such as fishing nets and fishing fences is found to be large, and in conclusion, the fishing boat activity is a great hidden danger of the navigation safety of the strait.
In the current research, a method for carrying out maritime navigation risk assessment aiming at the sea area navigation potential safety hazard does not exist; the current water area navigation risk research mainly takes the whole condition of an internal river water area or a single water area as a research object, the risk research of the sea area is less, the sea area and the internal river research still have certain differences, such as ship navigation rules, environment, research data availability and the like, and the specific condition of each sea area has differences, so that a proper index system and a research method are searched for researching the characteristics of the research object.
Disclosure of Invention
The invention aims to provide a navigation risk evaluation method based on an improved TOPSIS method aiming at the problems in the prior art.
In order to achieve the purpose, the invention adopts the technical scheme that:
a navigation risk evaluation method based on an improved TOPSIS method comprises the following steps:
s1, identifying and dividing accident-prone areas in a researched sea area by using a K-means clustering method based on minimum circle coverage;
s2, comprehensively considering traffic factors, ship factors and channel condition factors to establish a relative navigation risk evaluation index system;
s3, determining subjective weight vectors W of all indexes by using an analytic hierarchy process 1 ；
S4, determining objective weight of each index by using entropy weight methodVector W 2 ；
S5, combining the subjective weight vector and the objective weight vector by adopting a game theory method to obtain a combined weight vector W of each index * ；
And S6, evaluating the relative navigation risks of the accident-prone area by using an improved TOPSIS method of calculating the closeness by using the Mahalanobis distance instead of the Euclidean distance and the gray correlation degree, and judging the positions of the high-risk areas by taking the average risk as a risk threshold.
Specifically, step S1 includes the steps of:
s101, acquiring all marine traffic accident information of a research area within a period of time, extracting longitude and latitude coordinates of each accident point, and drawing a marine traffic accident spatial distribution map of the research area by utilizing ArcGis software;
s102, drawing a plurality of circles on an accident point space distribution map of the marine traffic accident in a research area based on the idea of minimum circle coverage so as to distinguish clustering areas, wherein one circle represents one clustering area, each clustering area at least comprises two accident points, and identifying and eliminating isolated accident noise points according to clustering results;
and S103, taking the number of all circles as a clustering number, taking the centers of the circles as clustering centers of the clustering areas respectively, and obtaining a final marine accident-prone area space division result by utilizing a K-means clustering method.
Specifically, in step S2, the relative navigation risk evaluation index system includes:
traffic factors including the following: vessel flow (i.e./hour), vessel average density (i.e./hr), and vessel density dispersion; the ship flow is converted standard ship flow, namely, the ship is converted according to a ship length conversion coefficient table and then counted; the unit of the ship density statistics is a grid enclosed by each longitude and each latitude on the chart;
marine factors including the following: ship type (%) and ship speed (section);
channel condition factors, including the following indices: influence of special areas, the number (number) of special points of a navigation channel and the number (number) of obstructive objects; the specific area effects include the overall effects of docks, fishing areas, anchorages, shoals, and reefs; the number of the special points of the navigation channel comprises the total number of the end parts of the navigation channel, the intersection area of the navigation channel and the bending part.
Specifically, in step S3, the method for determining the subjective weight vector of each index by using the analytic hierarchy process includes:
according to the intention investigation result of experts, navigation employees, marine safety supervisors and other personnel for selecting the importance of the index, the geometric mean value of the intention investigation result is taken to obtain an index importance judgment matrix X = (X) ij ) n×n (ii) a Calculating the geometric mean value of each row element of the judgment matrix X
Wherein n is the number of indexes; x is the number of ij Representing the element of the ith row and the jth column in the judgment matrix;
will be provided with
Normalized to obtain w i ：
Calculating the maximum characteristic root lambda of the judgment matrix max ：
Wherein, w = (w) 1 ,w 2 ,...,w n ) T Is a weight vector;
respectively solving the CR values of the single-analysis sequencing and the total hierarchical sequencing by using the following formula:
wherein, CI is a consistency index; RI is a random consistency index;
if CR is<0.1, judging that the matrix has consistency, and further determining the subjective weight vector W 1 ＝(w 1 ,w 2 ,...,w n )。
Specifically, in step S4, the method for determining the objective weight vector of each index by using the entropy weight method includes:
assuming that the number of objects to be evaluated is m and the number of indexes is n, an initial evaluation matrix A = (a) is obtained by assigning each index of each object ij ) m×n And normalizing the signal to obtain a normalized matrix C = (C) ij ) m×n ；
Calculating the specific gravity f of the jth index in the ith evaluation object according to the standard matrix C ij ：
Calculating information entropy e contained in each evaluation index j ：
Calculating the entropy weight w of each evaluation index j ：
That is, objective weight vector W of each index determined by entropy weight method is obtained 2 ＝(w j ) T ,j＝1,2,...,n。
Specifically, in step S5, the method for obtaining the combined weight vector of each index by using the game theory method includes:
for the subjective weight vector W 1 And objective weight vector W 2 Performing linear combination weighting;
based on the idea of game theory, the following combination coefficient equations are established:
wherein, a 1 And a 2 The combination coefficients are the proportions of the subjective weight and the objective weight in the combination weight respectively;
the above equation set is based on the traditional game theory idea, the obtained combination coefficient may be negative, therefore, in order to ensure that the linear combination coefficient is positive, the combination coefficient is optimized and improved, and the following improved game theory model is obtained:
by constructing a Lagrange function and taking a partial derivative:
wherein λ is a Lagrangian multiplier;
the result is normalized, and the combination weight coefficient obtained by the improved game theory model is as follows:
will be provided with
Substituting the following formula to obtain a combined weight vector:
wherein the content of the first and second substances,
and
respectively representing the proportion of the improved normalized subjective weight and the improved normalized objective weight in the combined weight.
Specifically, step S6 includes the steps of:
s601, assuming that the number of objects to be evaluated is m and the number of indexes is n, assigning values to each index of each accident-prone area in the research sea area to obtain an initial evaluation matrix A = (a) ij ) m×n ：
S602, determines the index attribute from the influence relationship of each index on the risk level, and performs index normalization on the reverse index using the following equation to obtain a normalization matrix B = (B) ij ) m×n ：
S603, each index is normalized according to the following formula,
obtain the normalized matrix C = (C) ij ) m×n ；
Recombining the combined weight vectors W * To obtain the weighted normalized matrix D = (D) ij ) m×n ；
Wherein d is ij ＝c ij ·w s (1≤i≤m,1≤j≤n,1≤s≤n)；
S604, respectively determining a maximum risk set and a minimum risk set calculated based on the Mahalanobis distance and the grey correlation degree according to the standardized matrix and the weighted normalized matrix;
wherein, C + 、C - Respectively determining a maximum risk set and a minimum risk set which are obtained by calculation based on Mahalanobis distance and gray correlation degree calculation according to the standardized matrix; d + 、D - Respectively determining a maximum risk set and a minimum risk set which are obtained by calculation based on the Mahalanobis distance and calculation based on the grey correlation degree according to the weighted normalized matrix;
s605, obtaining the Mahalanobis distance between each accident-prone area and the maximum risk set and the minimum risk set by using a Mahalanobis distance calculation formula;
wherein, the first and the second end of the pipe are connected with each other,
and is
Sigma is a sample covariance matrix;
s606, calculating the grey correlation coefficient of each accident easy area and each index of the maximum risk set and the minimum risk set according to the following formula;
wherein ρ is a resolution;
calculating the grey correlation degree of each accident susceptibility area and the maximum risk set and the minimum risk set by using the following formula;
s607, the normalized Mahalanobis distance and the gray correlation degree are linearly combined to form a combined distance
And with
Wherein α and β are respectively a combination coefficient, and α + β =1;
s608, the closeness CC of each accident easy-to-occur area and the maximum risk set is obtained according to the following formula i Will CC i As a relative navigation risk value of each accident-prone area; the larger the relative navigation risk value is, the higher the relative navigation risk of the accident-prone area is;
wherein, CC is more than or equal to 0 i ≤1；
S609, according to the relative navigation risk evaluation result of each accident-prone area, taking the average value of the relative navigation risk values of all accident-prone areas as a risk threshold value
If the relative navigation risk value of the accident-prone area is larger than the risk threshold value, the area is a high risk area, otherwise, the area is a general risk area.
Compared with the prior art, the invention has the beneficial effects that: the method comprises the steps of identifying and dividing accident-prone areas in a researched sea area by using a K-means clustering method based on minimum circle coverage, then comprehensively considering traffic factors, ship factors and channel conditions to establish a relative navigation risk evaluation index system, combining index weights obtained by a analytic hierarchy process and an entropy value method through a game theory to further obtain a combined weight of each index, and finally replacing Euclidean distance with a Mahalanobis distance to calculate the closeness by using a gray relevance improved TOPSIS method.
Drawings
Fig. 1 is a schematic flow chart of a navigation risk evaluation method based on the improved TOPSIS method.
FIG. 2 is a spatial distribution diagram of marine accidents in the Queen State channel mouth jurisdiction in an embodiment of the present invention.
Fig. 3 is a schematic diagram of preliminary division of regions of easy accident occurrence at the johnson state strait sea in the embodiment of the present invention.
Fig. 4 is a schematic diagram of the final division of the region at which accidents are likely to occur at the jone state strait sea in the embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it is to be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the present embodiment provides a navigation risk evaluation method based on the improved TOPSIS method, and takes a sea area within the range of the south-shore estuary of the seif-hou strait as a case to evaluate the navigation risk of a ship; the method comprises the following specific steps:
s1, identifying and dividing accident-prone areas in a research sea area by using a K-means clustering method based on minimum circle coverage, and specifically comprising the following steps:
s101, according to the statistical data of emergency search and rescue of the johnson state strait from 2012 to 2020, disclosed by the haikou office, extracting the available accidents 61, and importing the longitude and latitude into the ArcGIS software to obtain a spatial distribution map of the marine accidents in the strait prefecture, as shown in fig. 2, where the black dots in the map are accident points.
S102, on the basis of the marine accident spatial distribution map, drawing a plurality of unequal minimum circles capable of covering accident points, and primarily dividing the easy-to-occur region of the marine accident of the Johnson channel as shown in fig. 3, so that 1 isolated accident point can be obtained, and after the isolated accident point is removed, 11 easy-to-occur regions of the accident can be obtained, wherein the easy-to-occur regions are respectively P1-P11, and the position of the center of each circle is the initial clustering center of the accident event region.
S103, utilizing Matlab software to input clustering numbers, initial clustering center coordinates and other parameters, and performing K-means clustering on each accident coordinate point in the sea area to obtain the final easy accident occurrence area in the Johnson channel at sea as shown in FIG. 4, wherein the related information of each easy accident occurrence area is shown in Table 1.
TABLE 1 JONG HOUN CANCHA FISH AREA ACCESS EMERGENCY AREA INFORMATION TABLE
As can be seen from the schematic diagram of the accident-prone area, the accident-prone area identified by the K-means cluster covered by the smallest circle basically comprises important harbor areas such as Ma village, xiu hong Kong and the like which need important attention in the jurisdiction and the water area where the passenger rolling route is located, and the explanation method is suitable for identifying and dividing the accident-prone area in the jurisdiction of the Johnson channel.
S2, comprehensively considering traffic factors, ship factors and channel condition factors to establish a relative navigation risk evaluation index system;
the sea area relative navigation risk indexes and the meanings thereof are shown in the following table 2:
TABLE 2 sea area relative navigation risk index and its meanings
According to the table, the relative navigation risk evaluation index system comprises 3 first-level index factor layers and 8 second-level index factor layers, which are respectively:
traffic factors including the following: vessel flow (i.e./hour), vessel average density (i.e./hr), and vessel density dispersion; the ship flow is converted standard ship flow, namely, the ship is converted according to a ship length conversion coefficient table and then counted; the unit of the ship density is counted as a grid enclosed by each longitude and each latitude on the chart;
marine factors including the following: ship type (%) and ship speed (section);
channel condition factors including the following: influence of special areas, the number (number) of special points of a channel and the number (number) of obstacles; the special area effects include the overall effects of docks, fishing areas, anchorages, shoals, and reefs; the number of the special points of the navigation channel comprises the total number of the end parts of the navigation channel, the intersection area of the navigation channel and the bending part.
Relevant index data are observed and obtained through channels such as a ship communication network, a treasure ship network, a sea-shore maritime office and the like, and according to collected data, various index data of each accident-prone area are obtained through arrangement and are shown in a table 3:
TABLE 3 index parameters of each accident-prone region of the Johnson strait
S3, determining subjective weight vector W of each index by using an analytic hierarchy process 1 ；
According to the intention investigation results of experts, navigation practitioners, marine safety monitoring personnel and the like for selecting the importance of the index, the geometric mean value of the intention investigation results is taken to obtain an index importance judgment matrix X = (X) ij ) n×n (ii) a Calculating the geometric mean value of each row element of the judgment matrix X
Wherein n is the number of indexes; x is a radical of a fluorine atom ij Representing the element of the ith row and the jth column in the judgment matrix;
will be provided with
Normalized to obtain w i ：
Calculating the maximum characteristic root lambda of the judgment matrix max ：
Wherein, w = (w) 1 ,w 2 ,...,w n ) T Is a weight vector;
respectively solving the CR values of the single-analysis sequencing and the total hierarchical sequencing by using the following formula:
wherein, CI is a consistency index; RI is a random consistency index;
if CR is<0.1, judging that the matrix has consistency, and further determining the subjective weight vector W 1 ＝(w 1 ,w 2 ,...,w n )。
S4, determining objective weight vector W of each index by using entropy weight method 2 ；
Assuming that the number of objects to be evaluated is m and the number of indexes is n, an initial evaluation matrix A = (a) is obtained by assigning each index of each object ij ) m×n And normalizing the signal to obtain a normalized matrix C = (C) ij ) m×n ；
Calculating the specific gravity f of the jth index in the ith evaluation object according to the standard matrix C ij ：
Calculating information entropy e contained in each evaluation index j ：
Calculating the entropy weight w of each evaluation index j ：
That is, objective weight vector W of each index determined by entropy weight method is obtained 2 ＝(w j ) T ,j＝1,2,...,n。
S5, combining the subjective weight vector and the objective weight vector by adopting a game theory method to obtain a combined weight vector W of each index * ；
For the subjective weight vector W 1 And objective weight vector W 2 Linear combination weighting is carried out:
based on the idea of game theory, the following combination coefficient equations are established:
wherein, a 1 And a 2 For combining coefficients, i.e. subjective weight and objective weight, respectively, in a combined weightThe specific gravity of the weight;
the above equation set is based on the traditional game theory idea, the obtained combination coefficient may be negative, therefore, in order to ensure that the linear combination coefficient is positive, the combination coefficient is optimized and improved, and the following improved game theory model is obtained:
by constructing a Lagrange function and calculating a partial derivative:
wherein λ is a Lagrange multiplier;
the results are normalized, and the combination weight coefficients obtained by the improved game theory model are as follows:
will be provided with
Substituting the following formula to obtain a combined weight vector:
wherein, the first and the second end of the pipe are connected with each other,
to improve the proportion of the normalized subjective weights in the combined weights,
has a value of 0.5423;
to improve the proportion of the normalized objective weights in the combined weights,
has a value of 0.4577;
the evaluation indexes of the accident-prone areas are weighted as shown in the following table 4:
TABLE 4 evaluation index weights
| x 1 | x 2 | x 3 | x 4 | x 5 | x 6 | x 7 | x 8 | |
| Subjective weighting | 0.030 | 0.078 | 0.034 | 0.200 | 0.086 | 0.356 | 0.137 | 0.078 |
| Objective weight | 0.065 | 0.089 | 0.234 | 0.154 | 0.043 | 0.180 | 0.107 | 0.129 |
| Combining weights | 0.046 | 0.083 | 0.126 | 0.179 | 0.066 | 0.275 | 0.123 | 0.102 |
As can be seen from table 4, of the 8 indices, the two indices of the special area influence and the ship type account for a greater proportion, on the one hand, this is related to the accident characteristics of the seo zhou strait, which can be calculated from the collected accident statistics, and the total number of the statistical accidents occurring in the waters near the anchor, the wharf, and in addition, the influence of the wharf, the fishing area, the anchor, the reef, and the shoal is considered as an index, which also deepens the importance of this index to some extent.
S6, evaluating the relative navigation risks of the accident-prone area by using an improved TOPSIS method of calculating the closeness by using the Mahalanobis distance instead of the Euclidean distance and the gray correlation degree, and judging the positions of the high-risk areas by taking the average risk as a risk threshold;
step S6 includes the steps of:
s601, assuming that the number of objects to be evaluated is m and the number of indexes is n, assigning values to each index of each accident-prone area in the research sea area to obtain an initial evaluation matrix A = (a) ij ) m×n ；
S602, determines the index attribute from the influence relationship of each index on the risk level, and performs index normalization processing on the reverse index using the following formula to obtain a normalization matrix B = (B) ij ) m×n ：
S603, standardizing each index according to the following formula;
combine Table 3 to obtain the normalized matrix C = (C) ij ) m×n ；
Recombining the combined weight vectors W * Obtaining a weighted normalization matrix D = (D) ij ) m×n ；
Wherein d is ij ＝c ij ·w s (1≤i≤m,1≤j≤n,1≤s≤n)；
S604, respectively determining a maximum risk set and a minimum risk set for the Mahalanobis distance calculation and the gray correlation calculation according to the standardized matrix and the weighted normalized matrix;
wherein, C + 、C - Respectively determining a maximum risk set and a minimum risk set which are obtained by calculation based on Mahalanobis distance and gray correlation degree calculation according to the standardized matrix; d + 、D - Respectively determining a maximum risk set and a minimum risk set which are obtained by calculation based on the Mahalanobis distance and calculation based on the grey correlation degree according to the weighted normalized matrix;
the maximum risk set and the minimum risk set of the normalized matrix and the weighted norm matrix are shown in table 5 below:
TABLE 5 maximum Risk set, minimum Risk set of normalization matrix and weighting criteria matrix
| x 1 | x 2 | x 3 | x 4 | x 5 | x 6 | x 7 | x 8 | |
| C + | 1.688 | 2.939 | 4.264 | 4.192 | 1.287 | 4.681 | 2.444 | 3.385 |
| C - | 0.291 | 0.419 | 0.000 | 0.000 | 0.680 | 0.000 | 0.000 | 0.000 |
| D + | 0.078 | 0.245 | 0.537 | 0.749 | 0.085 | 1.289 | 0.301 | 0.344 |
| D - | 0.013 | 0.035 | 0.000 | 0.000 | 0.045 | 0.000 | 0.000 | 0.000 |
S605, obtaining the Mahalanobis distance between each accident-prone area and the maximum risk set and the minimum risk set by using a Mahalanobis distance calculation formula;
wherein the content of the first and second substances,
and is
Sigma is a sample covariance matrix;
s606, calculating the grey correlation coefficient of each accident easy area and each index of the maximum risk set and the minimum risk set according to the following formula;
where ρ is a resolution, and ρ =0.5 in this embodiment;
calculating the grey correlation degree of each accident-prone area and the maximum risk set and the minimum risk set by using the following formula;
s607, the normalized Mahalanobis distance and the gray correlation degree are linearly combined to form a combined distance
And
where α and β are combination coefficients, and α + β =1, the present embodiment takes α = β =0.5.
S608, the closeness CC between each accident-prone area and the maximum risk set is obtained according to the following formula i Will CC i As a relative navigation risk value of each accident-prone area; the larger the relative navigation risk value is, the higher the relative navigation risk of the accident-prone area is represented;
wherein, CC is more than or equal to 0 i ≤1；
S609, according to the relative navigation risk evaluation result of each accident-prone area, taking the average value of the relative navigation risk values of all accident-prone areas as a risk threshold value
If the relative navigation risk value of the accident-prone area is greater than the risk threshold value, the area is a high risk area, otherwise, the area is a general risk area;
the normalized degree of association between mahalanobis distance and gray and the calculation results of the relative navigation risk evaluation are shown in table 6 below:
table 6 relative navigation risk evaluation calculation result table
| P 1 | P 2 | P 3 | P 4 | P 5 | P 6 | P 7 | P 8 | P 9 | P 10 | P 11 | |
| d(C + ) | 0.951 | 0.674 | 1.033 | 1.370 | 1.069 | 0.616 | 1.072 | 1.115 | 1.074 | 0.986 | 1.040 |
| d(C - ) | 1.003 | 1.711 | 0.841 | 0.761 | 0.801 | 2.027 | 0.749 | 0.669 | 0.714 | 0.915 | 0.809 |
| r + | 0.914 | 1.002 | 1.015 | 1.164 | 1.007 | 1.218 | 0.949 | 0.949 | 0.946 | 0.919 | 0.916 |
| - | 1.078 | 0.936 | 0.968 | 0.853 | 0.979 | 0.827 | 1.045 | 1.061 | 1.059 | 1.085 | 1.108 |
| H + | 1.917 | 2.712 | 1.856 | 1.925 | 1.808 | 3.245 | 1.698 | 1.617 | 1.660 | 1.835 | 1.725 |
| H - | 2.029 | 1.610 | 2.001 | 2.223 | 2.049 | 1.443 | 2.117 | 2.177 | 2.133 | 2.071 | 2.148 |
| CC | 0.486 | 0.628 | 0.481 | 0.464 | 0.469 | 0.692 | 0.445 | 0.426 | 0.438 | 0.470 | 0.445 |
According to the table, a total of 11 accident-prone areas are identified, the relative navigation risk values of the areas are evaluated to be 0.486, 0.628, 0.481, 0.464, 0.469, 0.692, 0.445, 0.426, 0.438, 0.470 and 0.445, and the accident-prone areas are sorted according to the relative navigation risk values from large to small as: p 6 、P 2 、P 1 、P 3 、P 10 、P 5 、P 4 、P 11 、P 7 、P 9 、P 8 . According to the calculation result, the risk threshold value of the research water area is 0.495, and the relative navigation risk value of each area is compared with the risk threshold value, so that two high-risk areas are obtained, wherein the two high-risk areas are P 6 、P 2 . Wherein, P 2 、P 6 The anchor areas are all located in the area of the port, the number of the anchor areas is large, the passenger rolling ships are close to each other, the density of the ships is high, the distribution of the ships is relatively centralized, and in addition, the anchor areas influence the high-density passenger rolling ships, so that the index values are high, and the risk values are ranked in the front. P is 4 Is also positioned in the water area of the port, but because the water area is relatively wide and is less influenced by special areas such as anchorage ground, shoals and the like, the risk level of the area is higher than that of P 6 、P 2 The water area is low. And P is 1 Near the west mouth reporting line of the strait in the Johnson State, the ship flow is high, a plurality of tracks are intersected, and the periphery is influenced by shallow regions and fishing boats, so that P 1 The risk of (c) is also at a higher level. P 3 And P 5 The ship is positioned near the warning area and is positioned at the intersection of the east-west passenger rolling ship and the north-south ship track, the daily ship flow is large, the passenger ship proportion is relatively high, and the risk is at the middle-upper level. P is 7 、P 8 Compared with other areas, the influence of special points and special areas is small, and the relative risk is low although the index values of the ship flow and the ship average speed are high. P 9 、P 10 、P 11 Are all located near the water channel in the Johnson channel, where P 10 Nearby shoal and track intersection, and P 11 Is positioned at the east of the Haxia in Qiongzhou, is greatly influenced by wind and waves, is easy to displace in typhoon seasons, and is used for shipsNeed to pass through P 11 In and out of the channel in the Qiongzhou channel, thus, P 10 And P 11 To compare P 9 The risk is high.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. A navigation risk evaluation method based on an improved TOPSIS method is characterized by comprising the following steps:
s1, identifying and dividing accident-prone areas in a researched sea area by using a K-means clustering method based on minimum circle coverage;
s2, comprehensively considering traffic factors, ship factors and channel condition factors to establish a relative navigation risk evaluation index system;
s3, determining subjective weight vector W of each index by using an analytic hierarchy process 1 ；
S4, determining objective weight vector W of each index by using entropy weight method 2 ；
S5, combining the subjective weight vector and the objective weight vector by adopting a game theory method to obtain a combined weight vector W of each index * ；
And S6, evaluating the relative navigation risk of the accident-prone area by using the improved TOPSIS method of calculating the closeness by using the Mahalanobis distance instead of the Euclidean distance and the gray correlation degree, and judging the position of each high-risk area by taking the average risk as a risk threshold.
2. The navigable risk assessment method based on the improved TOPSIS method according to claim 1, characterized in that step S1 comprises the following steps:
s101, acquiring all marine traffic accident information in a period of time in a research area, extracting longitude and latitude coordinates of each accident point, and drawing a marine traffic accident space distribution map of the research area by utilizing ArcGIS software;
s102, drawing a plurality of circles on an accident point space distribution map of the marine traffic accident in a research area based on the idea of minimum circle coverage so as to distinguish clustering areas, wherein one circle represents one clustering area, each clustering area at least comprises two accident points, and identifying and eliminating isolated accident noise points according to clustering results;
and S103, taking the number of all circles as a clustering number, taking each circle center as a clustering center of each clustering area, and obtaining a final marine accident-prone area space division result by utilizing a K-means clustering method.
3. The navigable risk assessment method based on the improved TOPSIS method according to claim 1, characterized in that in step S2, the relative navigable risk assessment index system comprises:
traffic factors including the following: ship flow, ship average density and ship density dispersion;
marine factors including the following: the ship type and ship speed;
channel condition factors including the following: influence of special areas, the number of special points of a channel and the number of obstacles; the special area effects include the overall effects of docks, fishing areas, anchorages, shoals, and reefs; the number of the special points of the navigation channel comprises the total number of the end parts of the navigation channel, the intersection area of the navigation channel and the bending part.
4. The navigable risk assessment method based on the improved TOPSIS method according to claim 1, characterized in that in step S3, the method of determining the subjective weight vector of each index by using the analytic hierarchy process is:
according to the intention investigation result of the importance of the selected index, obtaining an index importance judgment matrix X = (X) ij ) n×n (ii) a Calculating the geometric mean value of each row element of the judgment matrix X
Wherein n is the number of indexes; x is a radical of a fluorine atom ij Elements representing the ith row and the jth column in the judgment matrix;
will be provided with
Normalized to obtain w i ：
Calculating the maximum characteristic root lambda of the judgment matrix max ：
Wherein w = (w) 1 ,w 2 ,...,w n ) T Is a weight vector;
respectively obtaining CR values of single-level sequencing and total-level sequencing by using the following formula:
wherein, CI is a consistency index; RI is a random consistency index; CR is the consistency ratio;
if CR is<0.1, judging that the matrix has consistency, and further determining the subjective weight vector W 1 ＝(w 1 ,w 2 ,...,w n )。
5. The navigable risk assessment method based on the improved TOPSIS method according to claim 1, characterized in that in step S4, the method for determining the objective weight vector of each index by using the entropy weight method comprises:
assuming that the number of objects to be evaluated is m and the number of indexes is n, an initial evaluation matrix A = (a) is obtained by assigning each index of each object ij ) m×n The normalized and normalized matrix is processed to obtain normalized matrix C = (C) ij ) m×n ；
Wherein, a ij And c ij Respectively representing the elements of the ith row and the jth column in the initial evaluation matrix and the standardized evaluation matrix;
calculating the specific gravity f of the jth index in the ith evaluation object according to the standard matrix ij ：
Calculating information entropy e contained in each evaluation index j ：
Calculating the entropy weight w of each evaluation index j ：
That is, objective weight vector W of each index determined by entropy weight method is obtained 2 ＝(w j ) T ,j＝1,2,...,n。
6. The navigable risk assessment method based on the improved TOPSIS method according to claim 1, characterized in that in step S5, the method of obtaining the combined weight vector of each index by using a game theory method is as follows:
for the subjective weight vector W 1 And an objective weight vector W 2 Performing linear combination empowerment;
based on the idea of game theory, the following combination coefficient equations are established:
wherein, a 1 And a 2 The combination coefficients are the proportions of the subjective weight and the objective weight in the combination weight respectively;
in order to ensure that the linear combination coefficient is positive, the combination coefficient is optimized and improved to obtain the following improved game theory model:
by constructing a Lagrange function and taking a partial derivative:
wherein λ is a Lagrange multiplier;
the results are normalized, and the combination weight coefficients obtained by the improved game theory model are as follows:
will be provided with
Substituting the following equation to obtain the combined weight vector W * ＝{w 1 ,w 2 ,...w s }, s =1,2, · n is:
wherein the content of the first and second substances,
and
respectively representing the proportion of the improved normalized subjective weight and the improved normalized objective weight in the combined weight.
7. The navigable risk assessment method based on the improved TOPSIS method according to claim 1, characterized in that step S6 comprises the following steps:
s601, assuming that the number of objects to be evaluated is m and the number of indexes is n, assigning values to each index of each accident-prone area in the research sea area to obtain an initial evaluation matrix A = (a) ij ) m×n ：
S602, determines the index attribute from the influence relationship of each index on the risk level, and performs index normalization processing on the reverse index using the following formula to obtain a normalization matrix B = (B) ij ) m×n ：
S603, each index is normalized according to the following formula,
obtain the normalized matrix C = (C) ij ) m×n ；
Recombined combined weight vector W * Obtaining a weighted normalization matrix D = (D) ij ) m×n ；
Wherein d is ij ＝c ij ·w s (1≤i≤m,1≤j≤n,1≤s≤n)；
Wherein w s A weight representing the s-th index;
s604, respectively determining a maximum risk set and a minimum risk set calculated based on the Mahalanobis distance and the grey correlation degree according to the standardized matrix and the weighted normalized matrix;
wherein, C + 、C - Respectively determining a maximum risk set and a minimum risk set which are obtained by calculation based on the Mahalanobis distance and gray correlation degree according to the standardized matrix; d + 、D - Respectively determining a maximum risk set and a minimum risk set which are obtained by calculation based on the Mahalanobis distance and calculation based on the grey correlation degree according to the weighted normalized matrix;
s605, obtaining the Mahalanobis distance between each accident-prone area and the maximum risk set and the minimum risk set by using a Mahalanobis distance calculation formula;
wherein, the first and the second end of the pipe are connected with each other,
and is provided with
Sigma is a sample covariance matrix;
s606, calculating the grey correlation coefficient of each accident easy area and each index of the maximum risk set and the minimum risk set according to the following formula;
wherein ρ is a resolution;
calculating the grey correlation degree of each accident susceptibility area and the maximum risk set and the minimum risk set by using the following formula;
s607, the normalized Mahalanobis distance and the gray correlation degree are linearly combined to form a combined distance H i + And H i - ；
Wherein α and β are combination coefficients, respectively, and α + β =1;
s608, the closeness CC between each accident-prone area and the maximum risk set is obtained according to the following formula i Will CC i The relative navigation risk value of each accident-prone area is used; the larger the relative navigation risk value is, the higher the relative navigation risk of the accident-prone area is;
wherein, CC is more than or equal to 0 i ≤1；
S609, according to the relative navigation risk evaluation result of each accident-prone area, taking the average value of the relative navigation risk values of all accident-prone areas as a risk threshold value
If the relative navigation risk value of the accident-prone area is larger than the risk threshold value, the area is a high risk area, otherwise, the area is a general risk area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210825865.6A CN115239110A (en) | 2022-07-14 | 2022-07-14 | Navigation risk evaluation method based on improved TOPSIS method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210825865.6A CN115239110A (en) | 2022-07-14 | 2022-07-14 | Navigation risk evaluation method based on improved TOPSIS method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115239110A true CN115239110A (en) | 2022-10-25 |
Family
ID=83674392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210825865.6A Pending CN115239110A (en) | 2022-07-14 | 2022-07-14 | Navigation risk evaluation method based on improved TOPSIS method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115239110A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777304A (en) * | 2023-08-16 | 2023-09-19 | 中海油能源发展股份有限公司采油服务分公司 | Ship emergency refute rescue management system and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017032210A1 (en) * | 2015-08-24 | 2017-03-02 | 国家电网公司 | Cluster analysis based power transmission line mountain fire risky area division method |
| CN114565253A (en) * | 2022-02-21 | 2022-05-31 | 武汉理工大学 | Passenger-rolling ship navigation risk assessment method based on combined empowerment-cloud model |
-
2022
- 2022-07-14 CN CN202210825865.6A patent/CN115239110A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017032210A1 (en) * | 2015-08-24 | 2017-03-02 | 国家电网公司 | Cluster analysis based power transmission line mountain fire risky area division method |
| CN114565253A (en) * | 2022-02-21 | 2022-05-31 | 武汉理工大学 | Passenger-rolling ship navigation risk assessment method based on combined empowerment-cloud model |
Non-Patent Citations (5)
| Title |
|---|
| 唐欣文;陈海群;高崇阳;邵辉;: "可拓优度在港口航道通航风险评价中的应用" * |
| 张文芬等: "基于最小圆覆盖的海上突发事件空间聚类研究" * |
| 王英等: "基于改进TOPSIS的受灾点需求紧迫性分级研究" * |
| 钟庆云: "东营港区水上交通安全风险评价与网格化研究" * |
| 马全党;江福才;王群朋;王晨;: "基于改进TOPSIS法的船舶通航环境风险评价模型" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777304A (en) * | 2023-08-16 | 2023-09-19 | 中海油能源发展股份有限公司采油服务分公司 | Ship emergency refute rescue management system and method |
| CN116777304B (en) * | 2023-08-16 | 2023-11-21 | 中海油能源发展股份有限公司采油服务分公司 | Ship emergency refute rescue management system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhen et al. | Maritime anomaly detection within coastal waters based on vessel trajectory clustering and Naïve Bayes Classifier | |
| Cakir et al. | An analysis of severity of oil spill caused by vessel accidents | |
| Silveira et al. | A multicriteria outranking approach for ship collision risk assessment | |
| CN104050329B (en) | A kind of method for detecting Collision Risk Index | |
| Li et al. | AIS data-based decision model for navigation risk in sea areas | |
| CN109543715A (en) | A kind of ship air route is extracted and the method for track deviation detection | |
| CN111507586A (en) | Early warning and grading method and system for overwater construction safety risks in navigable water area | |
| CN111613094A (en) | Port water area ship traffic risk early warning method | |
| CN115081731B (en) | Emergency resource optimal configuration method based on sea area accident characteristics | |
| CN115294804B (en) | Submarine cable safety early warning method and system based on ship state monitoring | |
| CN111428916B (en) | Navigation path planning method for rescue vessel at sea | |
| CN115239110A (en) | Navigation risk evaluation method based on improved TOPSIS method | |
| CN116402143A (en) | Intelligent ship index system construction and evaluation method | |
| Zhou et al. | Macroscopic collision risk model based on near miss | |
| CN114331042A (en) | Ship collision risk analysis method | |
| Yücel et al. | Determination of risk factors caused by ships in port planning | |
| Wang et al. | Analysis of navigation characteristics of inland watercraft based on DBSCAN clustering algorithm | |
| CN113177713B (en) | Ship collision risk calculation method based on accident data mining | |
| Zhou | Spatial risk assessment of maritime transportation in offshore waters of China using machine learning and geospatial big data | |
| Sun et al. | Collision Risk Hotspot Identification Based on AIS Data and DBSCAN Algorithm | |
| Li et al. | The Severity of Maritime Traffic Accidents Prediction Method Based on Bilayer GA-SVM | |
| 한재림 | A Spatio-Temporal Variation Pattern Analysis of Fishing Activity in the Jeju Sea of Korea Using V-pass data | |
| Han et al. | Data processing research on the comprehensive risk assessment model for overboard-man at sea | |
| CN117829613A (en) | Marine accident influence factor analysis method, system, intelligent terminal and medium | |
| Meng | Research on navigation safety evaluation of Jiaxing Port |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20221025 |
|
| RJ01 | Rejection of invention patent application after publication |