A Method for Calculating the Probability of Ship Collision with Bridge Based on Big Data Mining and Improved State Transition Algorithm

Abstract

This invention proposes a method for calculating the probability of a ship colliding with a bridge based on big data mining and an improved state transition algorithm. It adaptively adjusts the translation factor of the state transition algorithm and uses AIS data to fit the standard deviation of a Gaussian function model of ship traffic flow, thereby improving the AASHTO specification model recommended by the International Association for Bridge Engineering (IABTE). This invention can more accurately calculate the probability of a ship colliding with a bridge, providing a more scientifically sound and practically valuable reference for the safety management of bridge authorities, maritime departments, and bridge construction units.

Description

Technical Field

[0001] This invention belongs to the field of ship navigation safety technology, specifically relating to a method for calculating the probability of a ship colliding with a bridge based on big data mining and an improved state transition algorithm. Background Technology

[0002] How to avoid collisions with bridge piers during navigation is a crucial issue related to ship navigation safety. Existing technologies have proposed various technical means to solve this problem, such as Chinese patents: CN202010691545 - Bridge Collision Warning System, Method, Device and Storage Medium, CN202111564382 - Ship Collision Avoidance Method, etc.

[0003] A common problem with current technologies is that they typically use the AASHTO specification model recommended by the International Association for Bridge Engineering. However, the AASHTO model uses the vessel length (LOA) instead of the standard deviation σ of the Gaussian distribution of traffic flow, which introduces estimation errors. Summary of the Invention

[0004] To more accurately calculate the probability of a ship colliding with a bridge, this invention proposes a method based on big data mining and an improved state transition algorithm. This method adaptively adjusts the translation factor of the state transition algorithm and uses AIS data to fit the standard deviation of a Gaussian function model of ship traffic flow, thereby improving the AASHTO specification model recommended by the International Association for Bridge Engineering (IABTE). This invention can more accurately calculate the probability of a ship colliding with a bridge, providing a more scientifically sound and practically valuable reference for the safety management of bridge authorities, maritime departments, and bridge construction units.

[0005] The present invention specifically adopts the following technical solution:

[0006] A method for calculating the probability of a ship colliding with a bridge based on big data mining and an improved state transition algorithm, characterized by the following steps:

[0007] Step S1: Input the AIS data of ships crossing the bridge and calculate the frequency of ship crossings at each location;

[0008] Step S2: Fit a Gaussian function using the frequency obtained in step S1, and obtain the standard deviation that minimizes the fitting error based on the improved state transition algorithm;

[0009] Step S3: Based on the standard deviation of the Gaussian function of traffic flow obtained in step S2, the ship's width, the distance between the channel centerline and the bridge axis, and the width of the bridge piers, calculate the probability of a ship colliding with the bridge.

[0010] Furthermore, in step S2, the Gaussian model is:

[0011]

[0012] The objective function is to minimize the sum of squared deviations.

[0013]

[0014] Where N is the total amount of data;

[0015] Specifically, the following steps are included:

[0016] Step S21: Initialize the population, each solution represents x 0i =σ i ;

[0017] Step S22: Rotation Transformation:

[0018] x k+1 =x k +αRx k (3)

[0019] Where R is a random number, α is a rotation factor, and k is the number of iterations;

[0020] Step S23: Translation transformation:

[0021] x k+1 =x k +β k R t (x k -x k-1 (4)

[0022] Where R is a random number, β k R is the translation factor. t It is a random number;

[0023] Step S24: Scaling Transformation:

[0024] x k+1 =x k +γR e x k (5)

[0025] Where R is a random number, γ is a translation factor, and R e It is a random number;

[0026] Step S25: Axis Transformation:

[0027] x k+1 =x k +δR a x k (6)

[0028] Where R is a random number, δ is the axis factor, and R a It is a random number;

[0029] Step S26: Perform a hyperbolic tangent function transformation on the translation factor, as shown below:

[0030]

[0031] The coefficient λ>0, and T is the maximum number of iterations;

[0032] Step S27: Calculate the objective function f(x) according to equation (2). k+1 ); such as f(x) k+1 ) <f(x k If x is used, then x k+1 Replace x k ;

[0033] Step S27: k = k + 1. If k > T, end the iteration and output the optimal standard deviation σ of the Gaussian distribution of traffic flow.

[0034] Furthermore, in step S3, based on the standard deviation σ of the Gaussian function of the traffic flow and the ship's width B... p The distance d between the channel centerline and the bridge axis, and the width B of the bridge pier. m Calculate the geometric probability PG of a ship colliding with a bridge:

[0035]

[0036] Where Φ represents the distribution function of the standard normal distribution.

[0037] Furthermore, after step S3, the following is also included:

[0038] Step S4: Calculate the probability of ship yaw;

[0039] Step S5: Based on the calculation results of Step S3 and Step S4, calculate the annual ship collision probability of the bridge pier.

[0040] Further, in step S4, the ship's yaw probability PA is calculated:

[0041] PA = BR(R) B (R) C (R) XC (R) D (9)

[0042] Where BR is the basic rate, and R B R is the correction factor for the bridge location. C It is a correction factor parallel to the shipping route; R XC R is the correction factor for the crossflow perpendicular to the shipping route. D This is a correction factor for ship traffic density. For ordinary ships, BR = 0.6 × 10⁻⁶. –4 For barges, BR = 1.2 × 10⁻⁶–4 For bridges located in a straight region, R B =1.0. For bridges located in the transition zone, R = 1.0. B It can be calculated using the following formula:

[0043]

[0044] Where θ is the turning angle (degrees). For bridges located in turning or curvature areas, R B It can be calculated using the following formula:

[0045]

[0046] Water flow correction factor R parallel to the ship transport path in the waterway C It can be calculated using the following formula:

[0047]

[0048] Where V C This is the current component (nodal) parallel to the ship's path. The crosscurrent correction factor R is perpendicular to the ship's transport path in the channel. XC It can be calculated using the following formula:

[0049] R XC = (1+V) XC (13)

[0050] Where V XC This is the current component perpendicular to the ship's path (knot). When ships rarely meet, pass, or overtake near bridges, R... D =1.0. When ships occasionally meet, overtake, or pass each other near a bridge, R D =1.3. When ships frequently meet, overtake, or pass each other near bridges, R D =1.6.

[0051] Step 5: Calculate the annual ship collision probability of the bridge piers as follows:

[0052] P = Q·PA·PG(14)

[0053] Q represents the navigation volume of the ship type.

[0054] Compared to existing technologies, this invention and its preferred scheme adaptively adjust the translation factor of the state transition algorithm and use AIS data to fit the standard deviation of a Gaussian function model of ship traffic flow, thereby improving the AASHTO specification model recommended by the International Association for Bridge Engineering. This invention can more accurately calculate the probability of a ship colliding with a bridge, providing a more scientifically based and practically valuable reference for the safety management of bridge authorities, maritime departments, and bridge construction units. Attached Figure Description

[0055] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0056] Figure 1 This is a schematic diagram of the overall algorithm flow of an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of AIS frequency statistics according to an embodiment of the present invention;

[0058] Figure 3 This is a geometric probability diagram of a ship colliding with a bridge according to an embodiment of the present invention;

[0059] Figure 4 This is a convergence curve diagram of the adaptive state transition algorithm according to an embodiment of the present invention. Detailed Implementation

[0060] To make the features and advantages of this patent more apparent and understandable, specific embodiments are provided below for detailed explanation:

[0061] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0062] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0063] like Figure 1 As shown, the specific implementation process of the ship-bridge collision probability calculation method based on big data mining and improved state transition algorithm provided by the present invention includes the following steps:

[0064] Step 1: Input the AIS data of ships crossing the bridge, and calculate the ship crossing frequency yk at each location xk, such as... Figure 2 As shown.

[0065] Step 2: Fit a Gaussian function using a frequency array, reduce the error using an improved state transition algorithm, and find the standard deviation σ that minimizes the fitting error.

[0066] The Gaussian model is:

[0067]

[0068] The objective function is to minimize the sum of squared deviations.

[0069]

[0070] Where N is the total amount of data.

[0071] The improved state transition algorithm includes the following steps:

[0072] Step 2-1: Initialize the population, each solution represents x 0i =σ i .

[0073] Step 2-2: Rotation Transformation:

[0074] x k+1 =x k +αRx k (3)

[0075] Where R is a random number, α is a rotation factor, and k is the number of iterations.

[0076] Steps 2-3: Translation transformation:

[0077] x k+1 =x k +β k R t (x k -x k-1 (4)

[0078] Where R is a random number, β k Rt is the translation factor, and Rt is a random number.

[0079] Steps 2-4: Scaling Transformation:

[0080] x k+1 =x k +γR e x k (5)

[0081] Where R is a random number, γ is a translation factor, and R e It is a random number.

[0082] Steps 2-5: Axis Transformation:

[0083] x k+1 =x k +δR a x k (6)

[0084] Where R is a random number, δ is the axis factor, and R a It is a random number.

[0085] Steps 2-6: Perform a hyperbolic tangent function transformation on the translation factor, as shown below:

[0086]

[0087] The coefficient λ>0, and T is the maximum number of iterations.

[0088] Steps 2-7: Calculate the objective function f(x) according to equation (2). k+1 ). For example, f(x) k+1 ) <f(x k If x is used, then x k+1 Replace x k .

[0089] Step 2-7: k = k + 1. If k > T, end the iteration and output the optimal standard deviation σ of the Gaussian distribution of traffic flow.

[0090] The convergence curve of the adaptive state transition algorithm is as follows: Figure 4 As shown.

[0091] Step 3: Use as follows Figure 3 The geometric probability model of a ship colliding with a bridge is shown, based on the standard deviation σ of the Gaussian function of traffic flow and the ship's width B. p The distance d between the channel centerline and the bridge axis, and the width B of the bridge pier. m Calculate the geometric probability PG of a ship colliding with a bridge:

[0092]

[0093] Where Φ represents the distribution function of the standard normal distribution.

[0094] This information can also be used to obtain the annual probability of ship collisions with bridge piers, which is of greater general concern, to assist relevant construction, management, and maintenance departments in optimizing bridges or providing warnings.

[0095] Step 4: Calculate the ship's yaw probability PA.

[0096] PA = BR(R) B (R) C (R) XC (R) D (9)

[0097] Where BR is the basic rate, and R B R is the correction factor for the bridge location. C It is a correction factor parallel to the shipping route; R XC R is the correction factor for the crossflow perpendicular to the shipping route. D This is a correction factor for ship traffic density. For ordinary ships, BR = 0.6 × 10⁻⁶. –4For barges, BR = 1.2 × 10⁻⁶ –4 For bridges located in a straight region, R B =1.0. For bridges located in the transition zone, R = 1.0. B It can be calculated using the following formula:

[0098]

[0099] Where θ is the turning angle (degrees). For bridges located in turning or curvature areas, R B It can be calculated using the following formula:

[0100]

[0101] Water flow correction factor R parallel to the ship transport path in the waterway C It can be calculated using the following formula:

[0102]

[0103] Where V C This is the current component (nodal) parallel to the ship's path. The crosscurrent correction factor R is perpendicular to the ship's transport path in the channel. XC It can be calculated using the following formula:

[0104] R XC = (1+V) XC (13)

[0105] Where V XC This is the current component perpendicular to the ship's path (knot). When ships rarely meet, pass, or overtake near bridges, R... D =1.0. When ships occasionally meet, overtake, or pass each other near a bridge, R D =1.3. When ships frequently meet, overtake, or pass each other near bridges, R D =1.6.

[0106] Step 5: Calculate the annual ship collision probability of the bridge piers as follows:

[0107] P = Q·PA·PG(14)

[0108] Where Q represents the number of navigation trips (ship trips / year) for the representative ship type.

[0109] This patent is not limited to the above-described preferred embodiments. Anyone can derive other various methods for calculating the probability of a ship colliding with a bridge based on big data mining and improved state transition algorithms under the guidance of this patent. All equivalent changes and modifications made within the scope of this patent application shall fall within the scope of this patent.

Claims

1. A method for calculating the probability of a ship colliding with a bridge based on big data mining and improved state transition algorithm, characterized in that, The method comprises the following steps: Step S1: inputting AIS data of ships passing through the bridge, and obtaining ship passing frequency at each position by statistics; Step S2: fitting a Gaussian function by using the frequency obtained in step S1, and obtaining a standard deviation that minimizes fitting error based on an improved state transition algorithm; Step S3: calculating the probability of ship collision with the bridge according to the standard deviation of the traffic flow Gaussian function obtained in step S2, the ship width, the distance between the channel center line and the bridge axis, and the bridge pier width; In step S2, the Gaussian model is: (1) An objective function is established to minimize the deviation square sum: (2) Where N is the total amount of data; Specifically, the method comprises the following steps: Step S21: initialize population, each solution represents x 0i = σ i ; Step S22: rotation transformation: (3) Where R is a random number, a is a rotation factor, and k is the number of iterations; Step S23: translation transformation: (4) where β k is a translation factor, R t is a random number; Step S24: scaling transformation: (5) where γ is a translation factor, R e is a random number; Step S25: axis transformation: (6) where δ is the axial factor, R a is a random number; Step S26: hyperbolic tangent function transformation is performed on the translation factor, as shown below: (7) The coefficient λ>0, and T is the maximum number of iterations; Step S27: Calculate the objective function f(x k+1 ) according to (2); if f(x k+1 )<f(x k ), replace x k+1 with x k ; Step S28: k=k+1, if k>T, end iteration, and output the best standard deviation σ of the traffic flow Gaussian distribution; In step S3, the geometric probability PG of a ship colliding with the bridge is calculated based on the traffic flow Gaussian function standard deviation σ, the ship width B p , the distance d of the channel center line from the bridge axis, and the pier width B m . (8) Where Φ represents the distribution function of the standard normal distribution.

2. The method for calculating the ship-bridge collision probability based on big data mining and improved state transition algorithm according to claim 1, characterized in that, After step S3, the method further comprises: Step S4: calculating the ship yawing probability; Step S5: calculating the annual ship collision probability of the bridge pier according to the calculation results of steps S3 and S4.

3. The ship-bridge collision probability calculation method based on big data mining and an improved state transition algorithm according to claim 2, characterized in that: In step S4, the ship yawing probability PA is calculated as: PA = BR(R B )(R C )(R XC )(R D ) (9) where BR is the basic rate, R B is the correction factor for the bridge location, R C is the correction factor parallel to the ship transportation path; R XC is the correction factor for cross current perpendicular to the ship transportation path; R D is the correction factor for the ship traffic density; for ordinary ships, BR = 0.6 x 10 –4 ; for barges, BR = 1.2 x 10 –4 ; for bridges located in the straightaway area, R B = 1.0; for bridges located in the transition area, R B is calculated by the following equation: (10) where θ is the angle of rotation; for bridges located at a turn or curve, R B is calculated by the following equation: (11) A flow correction factor R parallel to the shipping path of the vessel in the waterway C is calculated by the following equation: (12) where V C is the current component parallel to the ship's path; the cross current correction factor R XC is calculated by the following formula: R XC = (1+V XC ) (13) where V XC is the vertical component of current perpendicular to the path of the vessel; R D = 1.0 when there are few encounters, passes or over-takes of vessels near the bridge; R D = 1.3 when vessels occasionally encounter or over-take each other near the bridge; R D = 1.6 when vessels frequently encounter or over-take each other near the bridge. Step 5: the annual ship collision probability of the bridge pier is calculated as: (14) Where Q represents the traffic volume of the ship type.

Images