A method and system for optimizing lockage scheduling of a ship in overhaul period based on bald eagle algorithm

Abstract

The application provides a kind of overhaul period ship lockage scheduling optimization method and system based on bald eagle algorithm, method includes: obtaining the ship information of water transport hub, navigation requirement, overhaul period navigation characteristic;With minimization ship lockage time as optimization goal, construct overhaul period ship lockage scheduling model, establish the constraint condition satisfied by overhaul period ship lockage scheduling;Based on the constraint condition satisfied, three sub-problems of ship lockage scheduling lock distribution problem, lock chamber arrangement problem, schedule optimization problem of overhaul period ship are solved respectively, and the initial scheme of overhaul period ship lockage scheduling is obtained;Overhaul period ship lockage scheduling initial scheme is iteratively solved based on bald eagle search algorithm with elite group guide, population memory crossover and self-adaptive mechanism, and the optimized scheduling scheme is obtained.The application can solve the problem of easy to fall into local optimum when using bald eagle search algorithm to solve ship lockage scheduling scheme at present, and make up the lack of overhaul period scheduling method.

Description

Technical Field

[0001] This invention relates to the field of intelligent water transport management technology, and more specifically, to a method and system for optimizing the scheduling of ships passing through locks during maintenance periods based on the Bald Eagle algorithm. Background Technology

[0002] With the increasing operational time of waterway transportation hubs, locks require regular maintenance shutdowns for inspection and replacement of related facilities and equipment. During lock maintenance shutdowns, the hub's throughput capacity decreases significantly, leading to severe traffic congestion and a large backlog of vessels waiting to pass through. With socio-economic development, the demand for transportation at waterway transportation hubs will continue to grow rapidly, further exacerbating the problem of insufficient throughput capacity during lock maintenance shutdowns. Therefore, alleviating navigation pressure in the dam area, reducing the adverse impact of lock maintenance on shipping, and resolving or mitigating the problem of insufficient throughput capacity during lock maintenance shutdowns are of significant practical importance and application value.

[0003] Existing research on ship traffic organization largely focuses on the joint navigation scheduling of dam areas or the optimization of lock chamber arrangement. It neglects the supply-demand imbalance of ship traffic caused by maintenance events and lacks methods for optimizing ship scheduling during lock maintenance. Furthermore, existing solution methods mostly rely on deterministic approaches and mathematical models for precise solutions, which are more suitable for solving basic problems. However, the lock scheduling problem involves multiple sub-problems with high coupling between them. Therefore, using swarm optimization algorithms such as the Vulture Search algorithm for lock scheduling would better meet the actual decision-making needs of lock scheduling operations. However, the basic Vulture Search algorithm is prone to getting trapped in local optima, experiencing rapid decline in population diversity, and exhibiting redundancy in search behavior. Therefore, optimizing the Vulture Search algorithm and applying it to the ship passage scheduling problem during maintenance can further enrich the theory of optimization problems and provide a reference for solving similar structural problems. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing a ship lock passage optimization method and system based on the vulture search algorithm. This method can solve the problem of easily getting trapped in local optima when using the vulture search algorithm to solve ship lock passage scheduling schemes, and make up for the shortcomings of the scheduling method during maintenance period.

[0005] According to a first aspect of the present invention, a method for optimizing the scheduling of ships passing through locks during maintenance periods based on the vulture algorithm is provided, comprising the following steps:

[0006] Step S1: Obtain vessel information, navigation requirements, and navigation characteristics during the maintenance period for water transport hubs;

[0007] Step S2: With minimizing the waiting time of ships in the lock as the optimization objective, construct a ship lock passage scheduling model during the maintenance period; establish the constraints to be satisfied by the ship lock passage scheduling during the maintenance period based on the scheduling model;

[0008] Step S3: Based on the constraints satisfied by the vessel lock passage scheduling during the maintenance period, solve the three sub-problems of the vessel lock passage scheduling during the maintenance period: lock allocation problem, lock chamber arrangement problem, and timetable optimization problem, to obtain the initial scheme for vessel lock passage scheduling during the maintenance period;

[0009] Step S4: Based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism, the initial scheme for the ship lock passage scheduling during the maintenance period is iteratively solved to obtain the optimized scheduling scheme.

[0010] Based on the above technical solution, the present invention can also be improved as follows.

[0011] Optionally, the vessel information includes: vessel arrival time, vessel length, vessel width, vessel name, vessel type, and vessel displacement;

[0012] Navigation requirements include: when a ship passes through a lock, in order to ensure the safe passage of two lock sessions, there must be a lock session interval between adjacent lock chambers;

[0013] The navigation characteristics during the maintenance period include: when one lock line is shut down for maintenance, the other line implements a strategy of one-way operation and timed reversal.

[0014] Optionally, in step S2, the maintenance period vessel passage scheduling model constructed with minimizing vessel waiting time at the lock as the optimization objective includes:

[0015] Step S21: Calculate the average waiting time for ships in the lock;

[0016] Step S22: Based on the calculated average waiting time of ships, construct the constraints that the ship passage scheduling during the maintenance period must satisfy.

[0017] Optionally, in step S21, the formula for calculating the average waiting time of a vessel is as follows:

[0018]

[0019] Where ΔT is the average waiting time for ships in the lock, and Δt is the average waiting time for ships in the lock. pq Let N represent the waiting time of ship p in phase q, N represent the total number of ships, p and q represent ships p in phase q, and S represent a group of scheduling units.

[0020] Optionally, in step S3, solving the three sub-problems of lock allocation, lock chamber arrangement, and ship passage planning during the maintenance period includes the following steps:

[0021] Step S31: Using the lock status and the generated scheduling unit, considering the lock area utilization rate and the corresponding requirements and constraints of the ship load balance among the three locks of Gezhouba Dam, determine the total number of lock openings and the corresponding lock opening schedule for each lock.

[0022] Step S32: Solve the gate cell arrangement based on the Bottom-Left algorithm;

[0023] Step S33: Generate a ship passage schedule based on lock allocation and lock chamber arrangement.

[0024] Optionally, in step S33, the algorithm for generating a ship passage schedule based on lock allocation and lock chamber arrangement includes:

[0025] Step S331: Set d = 1, d ij ∈{-1,1} represents the direction of transportation service for (i,j), where "1" represents upward and "-1" represents downward;

[0026] Step S332: Set i = 1, where i represents the i-th gate chamber;

[0027] Step S333: Set j = 1, where j represents the j-th gate chamber;

[0028] Step S334: If r ij =1, remove the corresponding transportation service (i, j) from the current queue to be assigned;

[0029] Step S335: If i = 1, t ij =t b r ij =1, meaning the planned gate passage time of transportation service (i,j) is equal to the start time of the planned period; otherwise, proceed to S337.

[0030] Step S336: Find the last ship to arrive at the anchorage from sp(i,j), assuming the arrival time of this ship is... r ij =1;

[0031] Step S337: If j <n i Proceed to step S338; otherwise, proceed to step S340.

[0032] Step S338: Setting r ij =1, j=j+1;

[0033] Step S339: If t ij -t b >T, d = -d, return to step S332;

[0034] Step S340: If i < 4, set i = i + 1 and return to step S333; otherwise, stop the algorithm and generate a ship lock passage plan table.

[0035] Optionally, in step S4, the iterative solution of the ship lock passage planning problem in the maintenance period ship lock passage scheduling model based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism includes the following steps:

[0036] Step S41: Determine the individual dimension D, the search space range [lb, ub], and the initial values ​​of the parameters controlling the positional changes and spiral trajectory of the vulture's flight in the algorithm, namely the initial values ​​m0, a0, and R0;

[0037] Step S42: Construct an initial population based on the number of ships and their entry order, and calculate the fitness value of individual bald eagles with the goal of minimizing the average waiting time of ships.

[0038] Step S43: Calculate the fitness value of each bald eagle according to the fitness function. Randomly select elite individuals from the top 5% of the current population's fitness values ​​and denot them as Pre.

[0039] Step S44: Calculate Pmean based on the average position of the vulture population distribution during the search process, update the position of the vulture population, and retain the current global optimal solution;

[0040] Step S45: Randomly retain the results of the space explored by individuals during the search process to form a population memory bank with a maximum size of N;

[0041] Step S46: Randomly select different individuals P from the union of the current population and the population memory set, respectively. r1 and P r2 Update the population position and retain the optimal solution for the current stage;

[0042] Step S47: Select a target individual and its corresponding individuals in the memory set, perform cross-operations on each dimension to construct a new individual, calculate its fitness value and compare it with the globally optimal individual, and retain the individual with the better fitness value;

[0043] Step S48: Determine whether the algorithm has reached the maximum number of iterations T. If so, the algorithm terminates and obtains the final information on the lock chamber arrangement, lock number arrangement, and ship passage timetable. Otherwise, proceed to step S43.

[0044] Optionally, in step S44, the formula for updating the position in the search space is as follows:

[0045]

[0046] In the above formula, This represents the updated d-dimensional state of the i-th bald eagle; This refers to an elite individual randomly selected from the elite group; P represents the dimension corresponding to the average position of the bald eagle flock distribution during the search process. i d This represents the current position of the i-th bald eagle in dimension d; r is a random number taking values ​​in (0,1); m represents a parameter that controls the magnitude of position changes.

[0047] Optionally, step S45, constructing the population memory bank, includes:

[0048] Step S351: By retaining a certain number of individual search memories, a population memory bank is formed. When an individual updates its position, an individual is randomly selected from the population memory bank to participate in the current individual's position update.

[0049] Step S352: Set the population memory size to be the same as the population size. In each iteration, if the population memory set size exceeds the threshold N, randomly remove individuals to maintain the population memory set size unchanged.

[0050] According to a second aspect of the present invention, a lock passage scheduling optimization system based on the vulture algorithm during maintenance period is provided, comprising:

[0051] The acquisition module is used to acquire information on vessels, navigation requirements, and navigation characteristics during the maintenance period at water transport hubs.

[0052] A construction module is used to build a vessel lock passage scheduling model during the maintenance period with the optimization objective of minimizing vessel waiting time in the lock; and to establish the constraints that the vessel lock passage scheduling during the maintenance period must satisfy based on the scheduling model.

[0053] The initial scheme establishment module is used to solve three sub-problems of the ship lock passage scheduling during the maintenance period, namely the lock allocation problem, the lock chamber arrangement problem, and the timetable optimization problem, based on the constraints satisfied by the ship lock passage scheduling during the maintenance period, so as to obtain the initial scheme of ship lock passage scheduling during the maintenance period.

[0054] The optimization iteration module is used to iteratively solve the initial scheme for ship lock passage scheduling during the maintenance period based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism, so as to obtain the optimized scheduling scheme.

[0055] The technical effects and advantages of this invention are as follows:

[0056] This invention provides a method and system for optimizing ship lock passage scheduling during maintenance periods based on the Bald Eagle algorithm. For the first time, it extends the application of the traditional Bald Eagle search algorithm to ship scheduling optimization problems through improvements. It solves the problems of easily getting trapped in local optima and excessively long search times when using the basic Bald Eagle search algorithm to solve ship lock passage scheduling schemes. It can stably obtain better ship lock passage scheduling schemes, improving ship lock passage efficiency during maintenance periods. In terms of algorithm improvement, fitness functions are constructed based on the shortest average waiting time for ships and the maximum number of ships passing through locks within a maintenance cycle, respectively, to optimize ship scheduling schemes. To prevent optimization from getting trapped in local optima, an elite population guidance and population memory crossover mechanism are introduced for algorithm optimization. This retains the dominant population, enabling the algorithm to converge quickly while reducing the high-intensity guidance of a single individual, overcoming the limitations of the search phase. The algorithm achieves self-adaptation by improving core parameters, dynamically adjusting parameters according to the problem to alleviate premature convergence. Redundant search mechanisms are eliminated to reduce ineffective searches.

[0057] Meanwhile, addressing the practical problems in the ship lock passage scheduling process during maintenance periods, and under the constraints of ship lock passage scheduling optimization problems such as safety time intervals, timed reversal, lock chamber constraints, and lock frequency balance, this paper takes into account the interests of both management and ship owners. With the goal of minimizing the average waiting time of ships, an adaptive vulture search algorithm with elite group guidance and population memory crossover is adopted to stably obtain a better ship scheduling scheme, improve the efficiency of ship lock passage operations during maintenance periods, provide a new method for solving the ship traffic organization optimization problem, and provide auxiliary decision-making for ship lock passage scheduling during maintenance periods.

[0058] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0059] Figure 1 The flowchart of the ship lock passage scheduling optimization method based on the vulture search algorithm provided in the embodiments of the present invention is as follows:

[0060] Figure 2 This is a diagram illustrating the mapping model between individual codes and ship formation provided in an embodiment of the present invention.

[0061] Figure 3 A comparative table of solutions to the dynamic scheduling problem provided in the embodiments of the present invention;

[0062] Figure 4 A schematic diagram of the dynamic scheduling of the Three Gorges North Line Ship Lock chambers provided in this embodiment of the invention;

[0063] Figure 5This is a diagram showing the dynamic scheduling of ship passage through the locks during a maintenance cycle, provided as an embodiment of the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] The Three Gorges-Gezhouba Dam project plays a crucial role in the development of the Yangtze River Economic Belt, making a significant contribution to the economic development of the middle and upper reaches of the Yangtze River. However, with the increasing operational time of the Three Gorges-Gezhouba Dam, the lock-related facilities and equipment require periodic maintenance and closures. During these maintenance periods, the dam's throughput capacity decreases significantly, leading to severe traffic congestion and a large backlog of ships waiting to pass through the locks. Optimizing the ship passage scheduling process during maintenance periods and improving the efficiency of ship navigation scheduling at the Three Gorges-Gezhouba Dam would have high practical significance for the economic development of the Yangtze River region. Therefore, this invention, based on an analysis of the current navigation scheduling situation at the Three Gorges Dam, constructs a ship passage scheduling model during maintenance periods with the goal of minimizing ship waiting time. It optimizes the Bald Eagle Search algorithm to solve the model, providing a solution to improve the efficiency of ship passage during maintenance periods.

[0066] Understandably, given the deficiencies in the background technology, this invention, taking the Three Gorges-Gezhouba Dam as an example, proposes an optimization method for ship lock passage scheduling during maintenance periods based on the Bald Eagle algorithm, specifically as follows: Figure 1 As shown, the optimization method includes the following steps:

[0067] Step S1: Obtain vessel information, navigation requirements, and navigation characteristics during the maintenance period for water transport hubs;

[0068] In this embodiment of the invention, the ship information includes: ship arrival time, ship length, ship width, ship name, ship type, and ship displacement;

[0069] Navigation requirements include: When ships pass through the Three Gorges Dam ship locks, a lock interval must be maintained between adjacent lock chambers to ensure safe passage between lock cycles. For example, under normal circumstances, the minimum interval for one lock cycle on the South Route of the Three Gorges Dam (Level 5 operation) is 90 minutes, and the minimum interval for one lock cycle on the North Route is also 90 minutes. Within the locks, the maximum ship speed entering and exiting the locks is 1 m / s, and the maximum ship speed between two adjacent locks is 0.6 m / s. The ship lift can generally only carry one ship at a time, and the time interval between ships transporting ships in the same direction is 30 minutes. The speed of ships entering and exiting the ship lift must not exceed 1 m / s.

[0070] The navigation characteristics during the maintenance period include: when one line of the Three Gorges Dam ship lock is shut down for maintenance, the other line implements a strategy of one-way operation with timed reversals.

[0071] Step S2: With minimizing the waiting time of ships in the lock as the optimization objective, construct a ship lock passage scheduling model during the maintenance period; establish the constraints to be satisfied by the ship lock passage scheduling during the maintenance period based on the scheduling model;

[0072] It should be noted that the construction of the vessel lock passage scheduling model during the maintenance period, with the optimization objective of minimizing vessel waiting time, includes:

[0073] Step S21: Calculate the average waiting time for vessels in the lock; the calculation formula is as follows:

[0074]

[0075] In the above formula, ΔT is the average waiting time for ships in the lock, and Δt is the average waiting time for ships in the lock. pq Let p represent the waiting time of ship p in phase q, N represent the total number of ships; p and q represent ships p in phase q; S represents a set of scheduling units; S = {(p,q)|1≤p≤N,1≤q≤q(p);}

[0076] Step S22: Based on the calculated average waiting time of ships, construct the constraints that the ship passage scheduling during the maintenance period must satisfy; the constraints are as follows:

[0077] Constraint 1:

[0078] Constraint 1 shows the minimum time interval constraint for two adjacent lock cycles of a lock. During the reversing cycle, when the Gezhouba single-stage lock operates in the same direction, and when the Three Gorges five-stage lock operates in the opposite direction during the reversing, an additional lock switching operation is required.

[0079] Where, r ij It is a conditional variable, r ij ∈{0,1}, if the transportation service (i,j) is running, r ij =1, otherwise r ij =0; t ij t is the planned gate passage time for transportation service (i, j). ij ∈R+. i and j represent the j-th gate operation in the i-th gate chamber, respectively; j-1 represents the previous gate operation; t i(j-1) This represents the planned passage time (i, j) for the (j-1)th gate transport service in the i-th gate chamber;

[0080] Indicates the time interval for the use of lock i; Indicates the switching time of lock i;

[0081] dij Indicates the transport service direction of the i-th gate chamber at the j-th gate; d i(j-1) Indicates the transport service direction of the (j-1)th gate in the i-th gate chamber; d ij ∈{-1,1} represents the direction of transportation service for (i,j), where "1" indicates upward and "-1" indicates downward.

[0082] Constraint 2: z ijpq ·t b ≤z ijpq ·t ij ≤z ijpq ·t e ;

[0083] Constraint 2 represents a constraint on the gate opening time; the planned end time should not be earlier than the earliest start time of a given time period. Where Z... ijpq To determine whether a scheduling unit (p,q) is transported by a transport service (i,j), z ijpq ∈{0,1}, if the scheduling unit (p,q) is transmitted by the transport service (i,j), then Z ijpq =1; otherwise Z ijpq =0; t b , t e Indicates the beginning and end of the planning period;

[0084] Constraint 3: z ijpq ·r ij =z ijpq ;

[0085] Constraint 3 states that every scheduled lock cycle must be executed. Since the Gezhouba Dam has three locks, which typically operate in the same pattern, the distribution of vessels across the three locks should be balanced.

[0086] Constraint 4:

[0087] Constraint 4 represents the lock balance of Gezhouba Dam locks 1, 2, and 3. λ1, λ2, and λ3 represent the optimal working load balance rates of locks 1, 2, and 3; n i ε represents the number of times lock chamber i is opened; n3, n4, and n5 represent the unbalanced square difference of the three lock chambers of Gezhouba Dam; n3, n4, and n5 represent locks 1, 2, and 3 of Gezhouba Dam.

[0088] Constraint 5:

[0089] Constraint 5 ensures that each scheduling unit is transferred by only one gate chamber. (p,q) Let S represent the set of available lock chambers for transferring vessel p in stage q; let S represent a set of scheduling units; S = {(p,q)|1≤p≤N,1≤q≤q(p);

[0090] Constraint 6:

[0091] Constraint 6 ensures that the scheduling unit can only be transferred from an available lock; L represents the lock service set, L = {(i,j)|1≤i≤5,1≤j≤n} i}, where (i,j) represents the transport service of the j-th lock of lock i; i = 1, 2, 3, 4, 5, 6 represent the Three Gorges North Line Lock, the ship lift, the Gezhouba No. 1, 2, 3 locks, and the Three Gorges South Line Lock, respectively.

[0092] Constraint 7:

[0093] Constraint 7 means that the time a vessel enters the lock cannot be earlier than the time the vessel arrives at the anchorage. t represents the arrival time of the scheduling unit (p, q); ij Indicates the planned gate passage time for (i,j); Indicates the arrival time of the scheduling unit (p,q);

[0094] Constraint 8: d ij ·z ijpq =v ij ·z ijpq (i,j)∈L,(p,q)∈S;

[0095] Constraint 8 ensures that the ship's course aligns with the lock's operating direction. ij This indicates the shipping direction of the scheduling unit (p, q), where "1" represents upward and "-1" represents downward.

[0096] Constraint 9:

[0097] Constraint 9 indicates that stage q can only proceed after the preceding stage q-1 is completed;

[0098] Constraint 10:

[0099] Constraint 10 ensures that the hull of any vessel entering the lock chamber must not extend beyond the lock chamber boundary. pq l pq W represents the width and length of ship p in stage q, respectively; ij ,L ij Indicates the available width and length of lock i at lock j; x pq ∈R* represents the x-coordinate value of the upper right corner of the lock chamber where ship p is placed during its stage q; y pq ∈R* represents the ordinate value of the upper right corner of the lock chamber where the ship p is placed during its stage q;

[0100] Constraint 11:

[0101] Constraint 11 ensures that two ships scheduled in the same lock do not overlap, where Z ijmn ∈{0,1}, if the scheduling unit (m,n) is transmitted by the transport service (i,j), then Z ijmn =1; otherwise Z ijmn =0;

[0102] Z ijmn Z indicates whether the scheduling unit (m,n) is transported by the transportation service (i,j). If the scheduling unit (m,n) is transported by the transportation service (i,j), then Z... ijmn =1; otherwise Z ijmn =0、 For step function, w mn l mn This represents the width and length of ship m at stage n; x mn ∈R* represents the x-coordinate value of the upper right corner of the lock chamber where ship m is placed during its stage n; y mn ∈R* represents the ordinate value of the upper right corner of the lock chamber where ship m is placed during its stage n; x pq ∈R* represents the x-coordinate value of the upper right corner of the lock chamber where ship m is placed during its stage n; y pq ∈R* represents the vertical coordinate value of the upper right corner of the lock chamber where the ship m is placed in its stage n;

[0103] Step S23: Determine whether the scheduling unit Z is transported by the transportation service based on the constraints. ijpq Planned gate passage time t for transportation services ij .

[0104] Step S3: Solve the three sub-problems of ship lock passage scheduling during the maintenance period: lock allocation problem, lock chamber arrangement problem, and ship passage planning problem, respectively, to obtain the initial scheme for ship lock passage scheduling during the maintenance period;

[0105] Furthermore, the solution to the three sub-problems of lock allocation, lock chamber arrangement, and lock passage planning during the maintenance period includes the following steps:

[0106] Step S31: Using the lock status and the generated scheduling unit, considering the lock area utilization rate and the corresponding requirements and constraints of the ship load balance among the three locks of Gezhouba Dam, determine the total number of lock openings and the corresponding lock opening schedule for each lock.

[0107] Specifically, regarding the utilization rate of lock chamber area, considering the demand for ship passage and the lock's throughput capacity, the system sets a range for the utilization rate. For example, the Three Gorges Dam lock is ≥75%, the Gezhouba No. 1 lock is ≥65%, and the Gezhouba No. 2 lock is ≥75%. Regarding lock frequency balance, the lock frequency balance ratio reflects the balance of workload across locks. Lock management departments aim for a balanced workload across each lock. For instance, when all three locks at Gezhouba are operating normally, the lock frequency ratio should be as close as possible to λ1:λ2:λ3 = 16:20:42.

[0108] The number of service visits for each lock is calculated based on the declared vessel type and the total lock area utilization rate. During lock maintenance, the total number of transfers at the Three Gorges Dam North Line Lock and the Three Gorges Ship Lift can be calculated as follows:

[0109]

[0110] in, The total number of times the Three Gorges North Line Ship Lock serves in the direction d, where d = 1, -1, where "1" represents upward and "-1" represents downward; The total number of times the Three Gorges ship lift serves in the direction d; η1 represents the total area of ​​ships applying for passage through the Three Gorges Dam in the direction d; η1 is an estimated value of the area utilization rate of the Three Gorges North Line Ship Lock, which is taken as 75% in this paper; L1 and W1 represent the length and width of the Three Gorges North Line Lock Chamber, respectively; L2 and W2 represent the length and width of the ship lift, respectively.

[0111] Considering the workload balance of the three ship locks at Gezhouba Dam, the total number of lock cycles can be calculated as follows:

[0112]

[0113] in, These represent the total number of services provided by Gezhouba Dam's No. 1, No. 2, and No. 3 ship locks in the direction of d. The total area of ​​vessels applying for passage through the Three Gorges Dam in direction d; η3 and η4 are estimated values ​​of the area utilization rate of Gezhouba No. 1 and No. 2 locks, respectively, taken as 70% and 75%; L3 and W3 represent the length and width of Gezhouba No. 1 lock chamber, respectively; L4 and W4 represent the length and width of Gezhouba No. 2 lock chamber, respectively; L5 and W5 represent the length and width of Gezhouba No. 3 lock chamber, respectively.

[0114] Step S32: Solve the gate cell arrangement based on the Bottom-Left algorithm;

[0115] It should be noted that the arrangement problem of the lock chamber can be described by a two-dimensional Packing model, which is an NP-complete problem. The lock chamber is a rigid rectangular space, that is, ships shall not exceed the boundary of the lock chamber during ship arrangement, and safe distances shall be maintained between adjacent ships inside the chamber and between ships and the inner wall of the lock chamber. To simplify the ship arrangement problem, ships and the lock are abstracted into rectangles with different lengths and widths, so the process of ship arrangement can be regarded as a process of filling a large rectangle with small rectangles. The specific steps are as follows:

[0116] Step S321: determining the locking order of ships;

[0117] Step S322: determining the set of arrangeable points for ships entering the lock; let the set of points available for ship arrangement in the lock chamber be ap(i,j,A), A represents the sequence number of arrangeable points in the lock chamber, when A=1, set ap(i,j,A)=(0,0);

[0118] Step S323: selecting the v-th ship from the sequence of ships to be arranged (v<N, the ship sequence number is smaller than the number of available waiting ships);

[0119] Step S324: setting v=1.

[0120] Step S325: setting an auxiliary variable a, a=1.

[0121] Step S326: checking whether ship v can be arranged at the available point ap(i,j,A) of lock i according to the ship arrangement constraints in constraint (10) and constraint (11). If the constraints are satisfied, perform step S327; otherwise, perform step S328.

[0122] Step S327: setting z ijpq =1. Delete the point from ap(i,j,n) and add two new points ((x pq ,y pq +w pq ),(x pq +l pq ,y pq )), set A=A+2. Then go to S329.

[0123] Step S328: if a<A, set a=a+1, and return to step S326; otherwise, set z ijpq =0. Then proceed to step S329.

[0124] Step S329: if v<N, set v=v+1, and return to step S326; otherwise, stop.

[0125] Step S33: Generate a ship passage schedule based on lock allocation and lock chamber arrangement.

[0126] It should be noted that, assuming the set of scheduling units (p, q) waiting for lock operations (i, j) is sp(i, j); the process of generating a ship passage schedule based on lock allocation and lock chamber arrangement includes the following steps:

[0127] Step S331: Set d = 1, d ij ∈{-1,1} represents the direction of transportation service for (i,j), where "1" represents upward and "-1" represents downward;

[0128] Step S332: Set i = 1, where i represents the i-th gate chamber;

[0129] Step S333: Set j = 1, where j represents the j-th gate chamber;

[0130] Step S334: If r ij =1, remove the corresponding transportation service (i, j) from the current queue to be programmed.

[0131] Step S335: If i = 1, t ij =t b r ij =1, meaning the planned gate passage time of transportation service (i,j) is equal to the start time of the planned period; otherwise, proceed to S337.

[0132] Step S336: Find the last ship to arrive at the anchorage from sp(i,j), assuming the arrival time of this ship is... r ij =1;

[0133] Step S337: If j <n i Proceed to step S338; otherwise, proceed to step S33X.

[0134] Step S338: Setting r ij =1, j=j+1;

[0135] Step S339: If t ij -t b >T, d = -d, return to step S332;

[0136] Step S33X: If i < 4, set i = i + 1 and return to step S333; otherwise, stop the algorithm and generate a ship lock passage plan table.

[0137] Step S4: Based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism, the ship lock passage planning problem in the maintenance period ship lock passage scheduling model is solved iteratively to obtain the optimized scheduling scheme.

[0138] It should be noted that while the elite group guidance strategy can accelerate convergence by guiding updates through the optimal individual, relying solely on a single individual can easily lead to excessive learning intensity and loss of population diversity. If the optimal solution found by the algorithm during the optimization process is a local extremum, the population is prone to getting trapped in the local optimal neighborhood, leading to premature convergence and other problems. For problems with a single information source, randomly selecting the top-level individual to replace the single optimal solution is one solution. In the first-stage position update of the adaptive vulture search algorithm, this paper proposes a learning method based on elite group guidance. First, a certain proportion of elite individuals are selected to construct an elite group. Then, when the target individual updates in each dimension, the corresponding dimension information of elite individuals is randomly selected from the elite group to guide the update, thus forming a new collective guidance method for the entire elite group to update individuals, avoiding excessive learning intensity from a single source and strengthening inter-population communication. The update method is shown below:

[0139]

[0140] in, This represents the updated d-dimensional state of the i-th bald eagle; This refers to an elite individual randomly selected from the elite group; This represents the dimension corresponding to the average position of the vulture flock distribution during the search process; This represents the current position of the i-th bald eagle in dimension d; r is a random number with a value in (0,1); m represents a parameter that controls the magnitude of position changes, and its range is (1.5,2).

[0141] P re It involves randomly recombining elite individuals from the top 5% of the current population's fitness values. This update model preserves the dominant group, enabling the algorithm to converge quickly, while reducing the high-intensity guidance from a single individual, thus overcoming the limitations of the first search phase.

[0142] The population memory crossover strategy includes: (1) Population memory construction: When the population of the vulture search optimization algorithm updates its position, it will temporarily store the information of new individuals. If the newly generated vulture individual is an invalid search, it will not be saved. Although the invalid discarding method will reduce the algorithm's running cost, it will miss the potential effective information of individual cognition. This paper proposes a position update strategy of constructing population memory to participate in the search. By retaining a certain scale of individual search memory, a population memory bank is formed. When an individual updates its position, an individual is randomly selected from the population memory bank to participate in the current individual position update. The population memory bank fully retains the results of the space explored by the individual during the search process, allowing each individual to have the opportunity to retain its own potential effective information. It also provides additional search information through the difference between the population memory set and the current population, providing the individual with more search possibilities and alleviating problems such as premature convergence. The individual updates its position under the guidance of the population memory bank and the position mean information, as shown below:

[0143] P i,new =P i +x(i)*(P i -P mean )+y(i)*(P r1 -P r2 )

[0144] Where P r1 ,P r2 These are different individuals randomly selected from the union of the current population and the population memory set. To avoid the population memory storage space becoming too large, the size of the population memory is set to be the same as the population size. In each iteration, if the size of the population memory set exceeds the threshold N, individuals are randomly removed to maintain the population memory pool size unchanged.

[0145] (2) Crossover and recombination strategy based on population memory; The Bald Eagle Search optimization algorithm gradually slows down or even stagnates in the later stages of the search, making it difficult to escape local optima. The main reason is the lack of new and effective information, which causes the population to fall into a local predicament. This paper proposes a crossover and recombination strategy based on population memory fusion to maintain population diversity. Each target individual and its corresponding individuals in the memory set are selected and crossover operations are performed dimension by dimension to construct a new individual. The operation is defined as follows:

[0146]

[0147] Where cr is a uniformly random number in (0,1), CR∈(0,1) is the crossover rate, representing the new individual. The proportion of elements copied from the population memory set. drand is a random natural number between 1 and D. This represents the value of the i-th individual after the d-th dimension crossover. and Let represent the value of the d-th dimension of the i-th individual selected from the population memory set or the current population, respectively. The PM-C strategy improves population diversity by preserving the memory of "invalid solutions" and introducing differential information, effectively addressing the second limitation of the vulture search optimization algorithm. Furthermore, the third limitation is also resolved by removing search redundancy in the third stage during the dive phase.

[0148] Furthermore, adaptive strategies can fully utilize effective information from the evolutionary process to adaptively adjust parameters, endowing individuals with the necessary capabilities at different stages and improving the algorithm's search performance. In the improved vulture search optimization algorithm, m, a, R, and CR have a significant impact on the group's search behavior. During the selection phase, m controls the search step size in the first stage. A larger m value results in a larger vulture flight step size, allowing for exploration of large spaces; a smaller m value allows for deeper exploration of local areas. In the vulture search optimization algorithm, m is a fixed value, which can easily lead to a mismatch between search requirements and parameters, affecting search capability. a and R determine the spiral trajectory and the density of individual distribution, respectively. With a fixed population size, a larger a value results in more spiral cycles and a larger search range, but weakens the deep search around the individual; a larger R value results in a more dispersed individual distribution and a larger search range.

[0149] To effectively improve the search behavior of vulture flocks, this invention employs an adaptive parameter adjustment strategy to meet the varying search needs of vultures at different stages. The parameters m, a, and R are perturbed using the following formulas to help vultures find optimal values ​​more quickly. This allows individuals to transition between dispersion and aggregation within the population space, ensuring search randomness, maintaining population diversity, and preventing the algorithm from getting trapped in local optima.

[0150]

[0151]

[0152]

[0153] In the formula, t represents the current iteration number, T represents the maximum iteration number, and m0, a0, and R0 represent the initial values ​​of each parameter. In this embodiment, it is recommended that m0 = 3, a0 = 10, and R0 = 1.

[0154] Furthermore, the crossover probability (CR) determines the individual crossover probability. A higher crossover probability can indeed help particles escape their current local predicament, but it may also increase the optimization difficulty and complexity of the algorithm. Therefore, Qin et al. proposed an adaptive difference algorithm to adaptively select the value of CR. This paper updates CR based on this adaptive strategy, making... Generate a CR value for each individual in the current population, and denote the CR that successfully promotes position updates in each generation as the SCR. Calculate the mean of the SCR based on the results recorded over a period of time. Repeat the above steps with the newly generated normal distribution mean (SCR) and a variance of 0.1, where oCR = 0.5. This process can learn an appropriate range of CR values, thus adapting to different problems.

[0155] In this embodiment of the invention, the bald eagle search algorithm is essentially an optimization algorithm for real-valued space, where the feasible region of each individual is a continuous space. Therefore, the information of each individual still belongs to the continuous space throughout the entire population optimization process. The goal of the aforementioned timetable optimization problem, however, is to obtain a reasonable scheduling plan, i.e., the gate sequence and ship arrangement matrix, which is a non-numerical optimization problem in discrete space. Therefore, to achieve an effective mapping between individual information and ship arrangement schemes, this invention designs an encoding and decoding scheme for individual information based on a combination of individual position information and arrangement schemes, building upon continuous-space encoding.

[0156] Furthermore, the iterative solution of the ship passage planning problem in the maintenance period ship passage scheduling model based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism includes the following steps:

[0157] Step S41: Parameter initialization. This includes determining the individual dimension D, the search space range [lb, ub], and the initial values ​​of the parameters controlling the positional changes and spiral trajectory of the vulture's flight in the algorithm, namely the initial values ​​of m, a, and R: m0, a0, and R0.

[0158] Step S42: Determine the encoding and decoding method for individual bald eagles, construct an initial population based on the number of ships and the order in which they enter the lock, and calculate the fitness value of individual bald eagles with the goal of minimizing the average waiting time of ships.

[0159] Furthermore, the method for determining the encoding and decoding of individual bald eagles is as follows:

[0160] The location information of any individual can be represented as P = (p1, p2, ..., p...). n ), where n is the number of ships waiting to pass through the lock. i represents the lock chamber selected for ship passage. Sort the elements in P in ascending order to obtain a sequence similar to p. i The corresponding array S. Based on array S, with p i The original index number is used to construct another corresponding sequence O, which is continuously adjusted during algorithm iterations as the individual positions are updated. Sequence O is then associated with the corresponding ship lock passage order in the ship scheduling problem, thus constructing a mapping model between the solution space of the optimization objective and the representation of individual positions.

[0161] Based on the above ideas, to facilitate the representation of individual position information and its contained order information, it is labeled in the form of a two-dimensional array (see Table 1 below). The first dimension of the array is used to record the position information of the individual in continuous space (the lock chamber selected by the ship for passage), and the second dimension records the order of the elements in the position information (the lock sequence of the selected lock chamber). After each position movement of the individual, the position is reordered to obtain a new order O, thereby realizing the adjustment of the lock sequence of the ship.

[0162] Table 1 Two-dimensional individual design

[0163] Positional order (S) <![CDATA[S1]]> <![CDATA[S2]]> <![CDATA[S3]]> …… <![CDATA[S j ]]>

[0164] Based on the above encoding rules, combined with Figure 2 As shown, an array S based on the arrangement of elements in P can be obtained. The index of each element in P is used as the number of the vessel waiting to pass through the lock. The vessel entry sequence O can be determined based on array S. Based on the determined loading order, the vessel passage scheduling scheme A can be determined according to the lock sequence arrangement algorithm and the lock chamber allocation algorithm.

[0165] Step S43: Determine the elite population and select elite individuals. Based on the fitness values ​​of the individuals in the population, select the bald eagles that rank in the top 5% of the current population fitness values ​​as the elite population, and then randomly select an elite individual from them, denoted as P. re ;

[0166] Step S44: Select the search space and calculate P based on the average location of the vulture flock distribution during the search process. mean Update the location of the vulture population, retaining the global optimal solution for the current stage;

[0167] The formula for updating the position in the search space is as follows:

[0168]

[0169]

[0170] In the above formula, This represents the updated d-dimensional state of the i-th bald eagle; This refers to an elite individual randomly selected from the elite group; This represents the dimension corresponding to the average position of the vulture flock distribution during the search process; This represents the current position of the i-th bald eagle in dimension d; r is a random number with a value in (0,1); m represents a parameter that controls the magnitude of position changes, and its range is (1.5,2).

[0171] Step S45: Construct a population memory bank, randomly retaining the optimization records of individuals during the search process, forming a population memory bank with a maximum size of N;

[0172] Furthermore, the method for constructing the population memory bank includes:

[0173] Step S451: By retaining a certain number of individual search memories, a population memory bank is formed. When an individual updates its position, an individual is randomly selected from the population memory bank to participate in the current individual's position update.

[0174] Step S452: To avoid the population memory storage space becoming too large, the population memory size is set to be the same as the population size. In each iteration, if the population memory set size exceeds the threshold N, individuals are randomly removed to maintain the population memory pool size unchanged.

[0175] Step S46: Search for spatial prey, randomly select different individuals P from the union of the current population and the population memory set respectively. r1 and P r2 Update the population position and retain the global optimal solution for the current stage;

[0176] Furthermore, the prey update formula for the search space is as follows:

[0177] P i，new =P i +x(i)*(P i -P mean )+y(i)*(P r1 -P r2 )

[0178] Here, x(i) and y(i) are in a spiral relationship to guide the vulture's flight, and their values ​​are all in (-1, 1). The polar coordinate model is represented as follows:

[0179]

[0180] xr(i)=r(i)*sin[θ(i)], yr(i)=r(i)*cos[θ(i)]

[0181] θ(i)=a*π*rand

[0182] r(i)=θ(i)+R*rand

[0183] Where θ(i) and r(i) are the polar angle and polar radius of the helical equation, respectively; a and R are parameters that control the helical trajectory, with values ​​of (5, 10) and (0.5, 2), respectively.

[0184] Step S47: Crossover and recombination of population memory. Select a target individual and its corresponding memory set individuals and perform crossover operations dimensionally to construct a new individual. Calculate the fitness value and compare it with the globally optimal individual. Retain the individual with the better fitness value.

[0185] Furthermore, the crossover and recombination operation of population memory in step S47 is defined as follows:

[0186] For each target individual and its corresponding individual in the memory set, perform cross-operations dimension by dimension to construct a new individual:

[0187]

[0188] Where cr is a uniformly random number in (0, 1), CR is the crossover rate, and represents the new individual. The proportion of elements copied from the population memory set. d rand It is a random natural number between 1 and D. This represents the value of the i-th individual after the d-th dimension crossover. and represents the value of the d-th dimension of the i-th individual selected from the population memory set or the current population, respectively.

[0189] Step S48: Determine whether the algorithm has reached the maximum number of iterations T. If so, the algorithm terminates and obtains the final information on the lock chamber arrangement, lock number arrangement, and ship passage timetable. Otherwise, return to step S43.

[0190] To further illustrate this, the embodiments of the present invention are constructed using actual vessel navigation data from the Three Gorges-Gezhouba cascade hydropower project. When there is no congestion upstream and downstream of the project, vessel lock passage scheduling is dynamic. In this case, the priority is to minimize vessel waiting time, which requires considering vessel arrival times when formulating lock chamber scheduling schemes. The scheduling time range is selected from 24 hours of waiting vessel data from May 20, 2020 to May 21, 2020 (as shown in Table 2).

[0191] Table 2 Dynamic Scheduling Cases (Uplink)

[0192]

[0193] Figure 3 A comparative table of solutions to the dynamic scheduling problem provided in the embodiments of the present invention; by Figure 3The results show that the adaptive vulture search algorithm outperforms the vulture search optimization algorithm. The fitness value decreases rapidly with increasing iterations, reducing the average waiting time for ships by approximately 3 hours. However, the vulture search optimization algorithm gets trapped in a local optimum around generation 60. This is because the parameters in the position update formula are fixed, preventing the population from adaptively adjusting to the optimization process. This leads to severe loss of population diversity in the later stages, hindering the search for new solutions and resulting in premature convergence. In contrast, the adaptive parameter adjustment strategy, while increasing the population's search range and global search capability in the early iterations, allows the population to quickly converge to the optimal neighborhood. As the iteration count increases, the adaptive parameter adjustment incorporates random perturbations, strengthening the algorithm's local search capability, maintaining population diversity, avoiding local optima, and improving convergence accuracy. Therefore, the adaptive vulture search algorithm is effective in solving ship scheduling problems.

[0194] Taking the dynamic scheduling experiment results of the Three Gorges North Line Ship Lock as an example, a schematic diagram of the lock chamber arrangement for some lock sessions is selected. The safe distance between ships is included in the ship dimensions, as shown in the following figures. Figure 4 As shown in Table 3, the dynamic scheduling plan for single locks on the northbound route of the Three Gorges Dam achieves an average lock chamber area utilization rate of over 75%, effectively utilizing lock chamber resources. The average waiting time for vessels is effectively reduced, meeting the vessel waiting time limit requirements. The entire scheduling algorithm generates a scheduling plan for all operating locks within a maintenance cycle, as shown in [the table]. Figure 5 The diagram shown illustrates a dynamic scheduling plan for ship passage through locks during a maintenance cycle, as provided in an embodiment of the present invention. Due to the numerous lock operations scheduled for the Gezhouba No. 3 ship lock and the Three Gorges Dam ship lift, only the entire cycle schedule for these two locks is displayed.

[0195] Table 3. Summary of Dynamic Scheduling Plan for Vessels Passing Through the Three Gorges Dam on the Northern Route

[0196]

[0197]

[0198] Furthermore, this invention also provides a lock passage scheduling optimization system based on the vulture algorithm during the maintenance period, comprising:

[0199] The acquisition module is used to acquire information on vessels, navigation requirements, and navigation characteristics during the maintenance period at water transport hubs.

[0200] A construction module is used to build a vessel lock passage scheduling model during the maintenance period with the optimization objective of minimizing vessel waiting time in the lock; and to establish the constraints that the vessel lock passage scheduling during the maintenance period must satisfy based on the scheduling model.

[0201] The initial scheme establishment module is used to solve three sub-problems of the ship lock passage scheduling during the maintenance period, namely the lock allocation problem, the lock chamber arrangement problem, and the timetable optimization problem, based on the constraints satisfied by the ship lock passage scheduling during the maintenance period, so as to obtain the initial scheme of ship lock passage scheduling during the maintenance period.

[0202] The optimization iteration module is used to iteratively solve the maintenance period ship lock passage scheduling model based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism, so as to obtain the optimized scheduling scheme.

[0203] It is understood that the maintenance period vessel lock passage scheduling optimization system based on the bald eagle algorithm provided by this invention corresponds to the maintenance period vessel lock passage scheduling optimization method based on the bald eagle algorithm provided in the foregoing embodiments. The relevant technical features of the maintenance period vessel lock passage scheduling optimization system based on the bald eagle algorithm can be referred to the relevant technical features of the maintenance period vessel lock passage scheduling optimization method based on the bald eagle algorithm, and will not be repeated here.

[0204] In summary, the embodiments of the present invention have the following technical effects:

[0205] (1) A mathematical model for vessel lock passage scheduling during the Three Gorges Gezhouba cascade hubs under maintenance was constructed. Based on the analysis of the characteristics of vessel lock passage scheduling during maintenance, a mathematical model for vessel lock passage scheduling during maintenance was constructed. The model sets the optimization objective as minimizing vessel waiting time and proposes several relevant constraints on lock operation and lock chamber arrangement.

[0206] (2) Research on Improved Solution Algorithm. To address the problem of ship lock passage scheduling during maintenance periods, this paper analyzes and improves the basic Bald Eagle Search algorithm, which is prone to getting trapped in local optima, rapidly declining population diversity, and redundant search behavior. Mechanisms such as elite population guidance and population memory crossover are introduced to optimize the algorithm. By improving core parameters to achieve algorithm adaptability and eliminating redundant search mechanisms, an Adaptive Bald Eagle Search algorithm (ABES) with elite population guidance and population memory crossover is proposed.

[0207] (3) Model Solving and Instance Verification. This paper divides the ship passage scheduling problem during the lock maintenance period into three sub-problems: lock allocation problem, lock chamber scheduling problem, and timetable optimization problem. The adaptive vulture search algorithm is used to solve the ship passage scheduling problem during the maintenance period, and experimental results are given. In the experiment, the throughput capacity of the two dams, the average waiting time of ships, the average lock chamber area utilization rate, and the ship throughput are compared during a maintenance cycle to verify the effectiveness of the model and algorithm.

[0208] The improved adaptive vulture search algorithm was compared with seven other intelligent optimization algorithms on the CEC2013 function test set. Both experimental results and Wilcoxon signed-rank test results show that the improved algorithm has stronger overall optimization performance and significantly improved robustness. In the model solution, experimental verification was conducted using actual historical navigation data from the Three Gorges-Gezhouba cascade hubs. Comparing the optimization results of the vulture search algorithm before and after the improvement, it was found that within one scheduling cycle, the average waiting time for ships can be reduced by about 3 hours, and the average lock chamber area utilization rate can reach over 75%. Experimental results indicate that the model and algorithm proposed in this study can be used for ship passage scheduling during lock maintenance periods, effectively reducing ship waiting time and improving passage efficiency.

[0209] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing ship lock passage scheduling during maintenance periods based on the Bald Eagle algorithm, characterized in that, Includes the following steps: Step S1: Obtain vessel information, navigation requirements, and navigation characteristics during the maintenance period for water transport hubs; Step S2: With minimizing the waiting time of ships in the lock as the optimization objective, construct a ship lock passage scheduling model during the maintenance period, and establish the constraints to be satisfied by the ship lock passage scheduling during the maintenance period based on the scheduling model; Step S3: Based on the constraints satisfied by the vessel lock passage scheduling during the maintenance period, solve the three sub-problems of the vessel lock passage scheduling during the maintenance period: lock allocation problem, lock chamber arrangement problem, and timetable optimization problem, to obtain the initial scheme for vessel lock passage scheduling during the maintenance period; Step S4: Based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanisms, the initial scheme for ship lock passage scheduling during the maintenance period is iteratively solved to obtain the optimized scheduling scheme, including: Step S41: Determine the individual dimension D, the search space range [lb, ub], and the initial values ​​of the parameters controlling the positional changes and spiral trajectory of the vulture's flight in the algorithm, namely the initial values ​​m0, a0, and R0; Step S42: Construct an initial population based on the number of ships and their entry order, and calculate the fitness value of individual bald eagles with the goal of minimizing the average waiting time of ships. Step S43: Calculate the fitness value of each bald eagle according to the fitness function, and randomly select elite individuals from the top 5% of the current population's fitness ranking, denoted as P. re ; Step S44: Calculate the corresponding dimension P based on the average position of the vulture flock distribution during the search process. mean Update the location of the bald eagle population, retaining the current global optimum; calculate the corresponding dimension P based on the average location of the bald eagle population distribution during the search process. mean The formula is as follows: In the above formula, This represents the updated d-dimensional state of the i-th bald eagle; This refers to an elite individual randomly selected from the elite group; This represents the dimension corresponding to the average position of the vulture flock distribution during the search process; This represents the current position of the i-th bald eagle in dimension d; r is a random number taking values ​​in (0, 1); m is a parameter that controls the magnitude of position changes; Step S45: Randomly retain the results of individuals exploring the space during the search process, forming a population memory bank with a maximum size of N; wherein, constructing the population memory bank includes: Step S351: By retaining a certain number of individual search memories, a population memory bank is formed. When an individual updates its position, an individual is randomly selected from the population memory bank to participate in the current individual's position update. Step S352: Set the population memory size to be the same as the population size. In each iteration, if the population memory set size exceeds the threshold N, randomly remove individuals to keep the population memory set size unchanged. Step S46: Randomly select different individuals P from the union of the current population and the population memory set, respectively. r1 and P r2 Update the population position and retain the optimal solution for the current stage; Step S47: Select a target individual and its corresponding individuals in the memory set, perform cross-operations on each dimension to construct a new individual, calculate its fitness value and compare it with the globally optimal individual, and retain the individual with the better fitness value; Step S48: Determine whether the algorithm has reached the maximum number of iterations T. If so, the algorithm terminates and obtains the final information on the lock chamber arrangement, lock number arrangement, and ship passage timetable. Otherwise, proceed to step S43.

2. The method for optimizing ship lock passage scheduling during maintenance period based on the vulture algorithm according to claim 1, characterized in that, The vessel information includes: vessel arrival time, vessel length, vessel width, vessel name, vessel type, and vessel displacement; Navigation requirements include: when a ship passes through a lock, in order to ensure the safe passage of two lock sessions, there must be a lock session interval between adjacent lock chambers; The navigation characteristics during the maintenance period include: when one lock line is shut down for maintenance, the other line implements a strategy of one-way operation and timed reversal.

3. The method for optimizing ship lock passage scheduling during maintenance period based on the vulture algorithm according to claim 1, characterized in that, In step S2, the maintenance period vessel passage scheduling model, constructed with the optimization objective of minimizing vessel waiting time in the locks, establishes the following constraints for maintenance period vessel passage scheduling: Step S21: Calculate the average waiting time for ships in the lock; Step S22: Based on the calculated average waiting time of ships, construct the constraints that the ship passage scheduling during the maintenance period must satisfy.

4. The method for optimizing ship lock passage scheduling during maintenance period based on the vulture algorithm according to claim 3, characterized in that, In step S21, the formula for calculating the average waiting time of a ship in the lock is as follows: in, This represents the average waiting time for ships in the locks. Let N represent the waiting time of ship p in phase q, N represent the total number of ships, p and q represent ship p in phase q, and S represent a group of scheduling units.

5. The method for optimizing ship lock passage scheduling during maintenance period based on the vulture algorithm according to claim 1, characterized in that, In step S3, solving the three sub-problems of lock allocation, lock chamber arrangement, and ship passage planning during the maintenance period includes the following steps: Step S31: Determine the total number of lock openings for each lock and the corresponding lock opening schedule; Step S32: Solve the gate cell arrangement based on the Bottom-Left algorithm; Step S33: Generate a ship passage schedule based on lock allocation and lock chamber arrangement.

6. The method for optimizing ship lock passage scheduling during maintenance period based on the vulture algorithm according to claim 5, characterized in that, In step S33, the algorithm for generating a ship passage schedule based on lock allocation and lock chamber arrangement is as follows: Step S331: Set d=1, Indicates the direction of transportation service at (i, j), where "1" indicates upward and "-1" indicates downward; Step S332: Set i=1, where i represents the i-th gate chamber; Step S333: Set j=1, where j represents the j-th gate chamber; Step S334: If r ij =1, remove the corresponding transportation service (i, j) from the current queue to be programmed; Step S335: If i=1, r ij =1, meaning the planned gate passage time of transportation service (i,j) is equal to the start time of the planned period; otherwise, proceed to S337. Step S336: From Find the last ship to arrive at the anchorage, assuming its arrival time is... r ij =1; Step S337: If j < Proceed to step S338; Otherwise, proceed to step S340; Step S338: Setting ,,r ij =1, j=j+1; Step S339: If d=-d, return to step S332; Step S340: If i < 4, set i = i + 1 and return to step S333; otherwise, stop the algorithm and generate a ship lock passage plan table.

7. A lock passage scheduling optimization system for vessels under maintenance based on the vulture algorithm, used to execute the lock passage scheduling optimization method for vessels under maintenance based on the vulture algorithm as described in any one of claims 1 to 6, characterized in that, include: The acquisition module is used to acquire information on vessels, navigation requirements, and navigation characteristics during the maintenance period at water transport hubs. A construction module is used to build a vessel lock passage scheduling model during the maintenance period with the optimization objective of minimizing vessel waiting time in the lock; and to establish the constraints that the vessel lock passage scheduling during the maintenance period must satisfy based on the scheduling model. The initial scheme establishment module is used to solve three sub-problems of the ship lock passage scheduling during the maintenance period, namely the lock allocation problem, the lock chamber arrangement problem, and the timetable optimization problem, based on the constraints satisfied by the ship lock passage scheduling during the maintenance period, so as to obtain the initial scheme of ship lock passage scheduling during the maintenance period. The optimization iteration module is used to iteratively solve the initial scheme for ship lock passage scheduling during the maintenance period based on the vulture search algorithm with elite group guidance, population memory crossover and adaptive mechanism, so as to obtain the optimized scheduling scheme.

Images