Multi-level multi-line ship-lock coordinated dispatching method, system, terminal and medium

Abstract

The present invention discloses a method, system, terminal, and medium for coordinated scheduling of multi-stage, multi-line ship-locks. The method applies the MOEA / D decomposition technique to divide the coordinated scheduling problem of multi-stage, multi-line ship-locks into several scalar subproblems. Within each subproblem, an interactive solution process is designed that includes three modules: ship-lock order allocation, lock chamber scheduling, and lock order scheduling. These subproblems are solved using a large neighborhood search algorithm, a multiple optimal ship scheduling algorithm, and a critical path method, respectively. During the iteration process, the current Pareto front is maintained by an external non-dominated solution set (EP). Ultimately, the solutions in the non-dominated solution set (EP) and their corresponding solutions are output, and ship-lock coordinated scheduling is performed accordingly. The present invention can solve the coordinated scheduling problem of multi-stage, multi-line ship-locks with more general lock structures, while simultaneously considering lock chamber scheduling and ship speed optimization, providing a feasible strategy for alleviating traffic pressure on inland waterways.

Description

Technical Field

[0001] The present invention relates to the technical field of ship and lock scheduling, and in particular to a multi-step multi-line ship-lock coordinated scheduling method, system, terminal and medium. Background Art

[0002] Currently, there are few research methods targeting the general characteristics of multi-stage, multi-line ship-lock coordinated scheduling. Most existing methods are based on specialized cascade lock structures, such as single-stage lock scheduling, multi-stage lock scheduling with simplified two-dimensional ship packing constraints, or specialized lock scheduling modes, such as discretizing lock opening times. For the more general multi-stage, multi-line ship-lock coordinated scheduling problem, there is still no effective solution.

[0003] Existing scheduling schemes, including single-line ship coordinated scheduling (A), single-line ship-lock coordinated scheduling (B), multi-stage ship-lock coordinated scheduling considering special lock structures (C), and multi-stage, multi-line ship-lock coordinated scheduling (D), are not applicable to the more general multi-stage ship-lock coordinated scheduling problem. A detailed analysis is as follows:

[0004] Technology A takes the process of ships passing through a single lock as the object, and studies the impact of the speed optimization of a single ship on other ships from a game theory perspective. However, this method is only applicable to a single lock with a single lock chamber, and only one ship can pass through a lock. Lock scheduling is not considered. This method cannot be applied to the optimization of multi-level and multi-line ship-lock coordinated scheduling schemes.

[0005] Technology B established a mixed integer linear programming model based on lock order and time indexes to solve the ship-lock coordinated scheduling problem, where each lock contains only one lock chamber and all lock chambers have identical properties (size, operating time). The model's efficiency was tested and confirmed on 10 small-scale examples. However, beyond simple lock configurations, this method does not consider the placement of two-dimensional ships and cannot be applied to situations where each lock may contain multiple parallel lock chambers, meaning it cannot be applied to the optimization of multi-stage, multi-line ship-lock coordinated scheduling solutions.

[0006] Technology C established a mixed-integer nonlinear programming model for optimizing multi-stage ship lock scheduling and proposed a heuristic method to solve the problem. However, this model assumes continuous, uninterrupted operation of the locks and fixed start times, and fails to consider the two-dimensional placement of locks and ships, making it incompatible with actual scheduling scenarios. Both the modeling process and the solution design exploit this peculiarity, making it unsuitable for optimizing ship-lock coordinated scheduling in situations where locks can be opened flexibly.

[0007] D technology treats the joint scheduling problem of multi-stage, multi-line locks as a variant of the batch-based flexible job shop scheduling problem, and considers the lock chamber scheduling problem as a two-dimensional bin packing problem with additional constraints. A mixed-integer linear programming model for optimizing the joint scheduling of multi-stage, single-line locks was developed and solved using a commercial solver. However, this method cannot solve large-scale problems within a reasonable computational time, and its performance does not meet the requirements of practical scheduling. Summary of the Invention

[0008] The present invention provides a multi-stage and multi-line ship-lock coordinated scheduling method, system, terminal and medium based on a multi-objective decomposition framework and a large neighborhood search algorithm, so as to overcome the shortcomings of the effective solutions to the multi-stage and multi-line ship-lock coordinated scheduling problem in the background technology and solve the problem of coordinated scheduling of ships and locks.

[0009] In a first aspect, a multi-level multi-line ship-lock coordinated scheduling method is provided, comprising the following steps:

[0010] S1: Obtain ship information and lock attribute information;

[0011] S2: Construct an objective function that minimizes the ship's waiting time and total fuel consumption;

[0012] S3: Generate weight vector population {λ 1 ,λ 2 ,…,λ N}, each weight vector λ i Corresponding to a sub-problem; generate an initial solution for each sub-problem, and obtain the initial solution population X={X1,X2,…,X N Each solution in the solution population describes the allocation information of ships and locks; and initializes a reference point and an external non-dominated solution set EP that records the Pareto optimal solutions in the current population;

[0013] S4: Traverse each subproblem in the solution population X, and find the solution X contained in each subproblem p The destruction operator based on the taboo table and the Chebyshev penalty insertion operator are used to search and reconstruct the allocation relationship between ships and locks. A new solution is obtained through destruction and repair operations. Then, the ship-lock allocation scheme in the new solution is tested to see if it meets the lock chamber scheduling constraints. If it meets the constraints, the berthing position of each ship in each lock chamber in the new solution is recorded. For the feasible new solution X' p The critical path method based on triangular distribution is used to decide the ship's speed at each stage and the lock opening time. The time when the ship leaves the lock system and the speed of the ship at each stage in the lock system are obtained, and then the total stay time and total fuel consumption of the ship are calculated.

[0014] S5: Determine the new solution X' obtained for each sub-problemp Whether it is accepted, the accepted new solution will be called the current latest solution of the subproblem The latest solution to each subproblem Compare with its neighbor solutions, update the neighbor solutions, get a new solution population, and update the non-dominated solution set EP and reference point;

[0015] S6: Loop through steps S4 to S5 until the iteration termination condition is met, output the solutions in the non-dominated solution set EP and their corresponding solutions, and use them to perform ship-lock coordinated scheduling.

[0016] Furthermore, in step S2, the ship waiting time objective function f1 is expressed as follows:

[0017]

[0018] Where, π i represents the weight of ship i, Indicates the last lock that ship i needs to pass through, represents the time when ship i leaves the lock system; r i represents the time when ship i arrives at the lock system; Represents a collection of ships;

[0019] The ship's total fuel consumption objective function f2 is expressed as follows:

[0020]

[0021] Where, Ω i is the stage set of ship i, 0-1 variable τ ikv Indicates whether ship i is sailing at speed V in the kth stage, α i Indicates the ship fuel consumption coefficient, D ik represents the sailing distance of ship i in the kth stage, V i represents the speed set of ship i.

[0022] Furthermore, in step S3, the solution uses the following structure to describe the ship and lock allocation information:

[0023]

[0024] Where, Indicates the number of locks assigned to lock chamber m Ship stage collection.

[0025] Furthermore, in step S4, a multiple sorting optimal ship scheduling algorithm is used to detect whether the ship-lock order allocation scheme in the new solution satisfies the lock chamber scheduling constraints, specifically including:

[0026] In the process of inserting the removed ship phase using the Chebyshev penalty insertion operator, the solution obtained after the removed ship phase is inserted into phase k is checked using the multi-(length, width, area) sorting optimal ship row algorithm. Whether the lock chamber gear constraint is met, if so, record The berthing position of each ship in each lock chamber; otherwise The objective function value of will be set to infinity;

[0027] Record the optimal insertion position and the corresponding Chebyshev penalty for each removed ship phase;

[0028] Sort the Chebyshev penalties of the best insertion positions of all removed ship stages and then insert the removed ship stages into their best positions in turn according to the deep greedy strategy to obtain the new solution X' i .

[0029] Furthermore, in step S4, a critical path method based on triangular distribution is used to make decisions on the ship's speed and the lock opening time at each stage, specifically including:

[0030] The decision interval of ship speed is processed by triangular distribution, including: dividing the ship speed into several intervals from the lowest speed to the highest speed, assuming that the upper limit of an interval is b and the lower limit is a; according to the weight vector of the current sub-problem p Generate the most likely value This gives the expected speed in this interval for the current subproblem. x is the integration variable, where

[0031] The critical path method is used to determine the ship speed and lock opening time, including:

[0032] Determine the key locks and key ships, and construct a directed acyclic graph G = (A, V) to represent the current solution X' p , V represents the node represented by the lock, A represents the temporal relationship between the locks; calculate the earliest start time and the latest end time of each node, among which the node with the difference between the sum of the earliest start time and the lock operation time and the latest end time is 0 is the key lock, and the latest arriving ship in the key lock is the key ship;

[0033] Adjust the speed and calculate the contribution of each key ship to the objective function Among them, f1 is the ship waiting time objective function, f2 is the ship total fuel consumption objective function, v i represents the current speed of ship i, Indicates the speed at which ship i increases by one gear, and increases the speed range of ships with decreasing contribution values ​​until the critical route is about to change;

[0034] Perform time relaxation and decelerate non-critical ships whose waiting time is not 0, that is, reduce the speed range until all non-critical ships in the current lock reach the minimum speed or are about to affect the lock opening time;

[0035] The start time of each lock is calculated based on the final speed, and thus the objective function value is calculated.

[0036] Furthermore, in step S5, the new solution X' obtained by each sub-problem is determined. p Whether it is accepted, including:

[0037] Prioritize the solution of the current sub-problem that corresponds to the goal with a greater weight If f m (X' p ) <f m (X p ), the new solution will be accepted, f1 and f2 represent the objective functions of ship waiting time and total fuel consumption respectively;

[0038] If f m (X' p )>f m (X p ), X' p will be The probability of being accepted is is a predefined coefficient that controls the probability of accepting a new solution that is worse than the current solution. The accepted new solution will be called the current state-of-the-art solution to the subproblem.

[0039] Furthermore, in step S5, the latest solution of each sub-problem Each solution is compared with its neighbor solutions and updated, including:

[0040] The latest solution to each sub-problem Solve X with its neighbors j ,j∈T to compare and calculate the neighbor solution and Chebyshev value of and if but Where T is the neighbor solution set of the current subproblem solution, and it is determined by calculating the Euclidean distance between the weight vectors of each subproblem.

[0041] Secondly, a multi-level, multi-line ship-lock coordinated dispatching system is provided, comprising:

[0042] Data acquisition module, used to obtain ship information and lock attribute information;

[0043] An objective function building module, used to build an objective function that minimizes ship waiting time and total ship fuel consumption;

[0044] Initialization module, used to generate the weight vector population {λ 1 ,λ 2 ,…,λ N}, each weight vector λ i Corresponding to a sub-problem; generate an initial solution for each sub-problem, and obtain the initial solution population X={X1,X2,…,X N Each solution in the solution population describes the allocation information of ships and locks; and initializes a reference point and an external non-dominated solution set EP that records the Pareto optimal solutions in the current population;

[0045] The loop iteration module is used to traverse each sub-problem in the solution population X, and for each sub-problem contained in the solution X p The destruction operator based on the taboo table and the Chebyshev penalty insertion operator are used to search and reconstruct the allocation relationship between ships and locks. A new solution is obtained through destruction and repair operations. Then, the ship-lock allocation scheme in the new solution is tested to see if it meets the lock chamber scheduling constraints. If it meets the constraints, the berthing position of each ship in each lock chamber in the new solution is recorded. For the feasible new solution X' p , the critical path method based on triangular distribution is used to make decisions on the ship's speed at each stage and the opening time of the lock; the time when the ship leaves the lock system and the speed of the ship at each stage in the lock system are obtained, and then the total stay time and total fuel consumption of the ship are calculated; the new solution X' obtained by each sub-problem is determined p Whether it is accepted, the accepted new solution will be called the current latest solution of the subproblem The latest solution to each subproblem Compare them with their neighbor solutions, update the neighbor solutions, obtain a new solution population, and update the non-dominated solution set EP and the reference point; loop through the above iterative process until the iteration termination condition is met, and output the solutions in the non-dominated solution set EP and their corresponding solutions;

[0046] The scheduling module is used to perform ship-lock coordinated scheduling based on the solutions in the non-dominated solution set EP and their corresponding solutions.

[0047] In a third aspect, an electronic terminal is provided, comprising:

[0048] one or more memories having computer programs or instructions stored thereon;

[0049] One or more processors are used to load and execute the computer program or instructions to implement the multi-level multi-line ship-lock coordinated scheduling method as described above.

[0050] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program or instruction is stored. When the computer program or instruction is called by a computer, the multi-level multi-line ship-lock coordinated scheduling method as described above is implemented.

[0051] The present invention proposes a multi-stage and multi-line ship-lock coordinated dispatching method, system, terminal and medium, which has the following beneficial effects:

[0052] In the multi-stage multi-line ship-lock coordinated scheduling (SLCP) problem, due to the interdependence between ship navigation and locks, optimizing them as an integrated system can maximize the overall operational efficiency and economic benefits of water transportation under the growing demand. The present invention designs a multi-objective heuristic algorithm that can effectively solve large-scale SLCP instances and output a set of multi-objective Pareto front solutions within a reasonable time, thereby improving the diversity of solutions. Compared with the existing technology, the present invention can solve the multi-stage multi-line ship-lock coordinated scheduling problem of a more general lock structure, while taking into account the optimization of lock chamber gearing and ship speed, providing a feasible strategy for alleviating the traffic pressure of inland waterways, serving the sustainable development of inland shipping, and providing theoretical and methodological support for the decision-making of lock operation scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 1 is a schematic diagram of coordinated scheduling of a cascade ship-lock system comprising n ship locks provided by an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of the lock chamber gear shift constraint provided by an embodiment of the present invention;

[0056] Figure 3 Schematic diagram of a multi-objective large neighborhood search algorithm (MOLNS) provided by an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the processing relationship between different gates provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without inventive effort are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, the multi-step multi-line ship-lock structure provided in the embodiment of the present invention includes a plurality of ship locks and a river channel, wherein the plurality of ship locks are arranged on the river channel and are arranged in sequence in the upstream and downstream directions of the river channel, and the ship locks include a plurality of lock chambers, wherein the plurality of lock chambers are arranged in parallel in the river channel, and the plurality of lock chambers are independent structures of each other. The river channel includes tributaries, and the tributaries are arranged between adjacent ship locks and connected to the river channel. The plurality of lock chambers are of different types. There are two-way navigation ship flows with different temporal and spatial distributions on the waterway, and the ships have different routes, i.e., they can enter or leave the cascade ship lock system through the main stream or tributaries. The declared information of the ship includes length, width, draft, tonnage, speed range, distance from the first ship lock it needs to pass through, etc. The lock operator formulates a ship-lock coordinated scheduling plan based on the lock attribute parameters and ship declaration information. It needs to decide the lock information (which ships to serve), lock chamber gear (the two-dimensional spatial layout of ships in the lock chamber), lock start time and recommended speed for each section of the ship for each lock, and feed back the recommended speed to the ship.

[0060] For the convenience of subsequent description, the symbols used below are explained as follows:

[0061] Lock collection;

[0062] Ship muster;

[0063] the collection of locks that the ship needs to pass through;

[0064] Ω: the set of all ship phases;

[0065] Ω(q): a subset of Ω, representing the set of ship stages assigned to lock q;

[0066] Ω m (q): a subset of Ω, representing the set of ship stages that are assigned to lock chamber m at lock time q;

[0067] π i : weight of ship i;

[0068] r i : the time when ship i arrives at the lock system;

[0069] D ik : The sailing distance of ship i in the kth stage;

[0070] α i : Fuel consumption coefficient of ship i.

[0071] The present invention designs a multi-objective large neighborhood search algorithm (MOLNS) for the multi-stage multi-line ship-lock coordinated scheduling problem (SLCP). The algorithm applies the MOEA / D decomposition technique to divide the SLCP into several scalar subproblems. The subproblems are interconnected by the Euclidean distance of the weight vector. The nearest subproblems of each subproblem are called its neighbors. In each subproblem, the present invention designs an interactive solution process consisting of three modules: ship-lock allocation, lock chamber scheduling and lock scheduling, which are solved by large neighborhood search (LNS), multiple optimal ship scheduling (MOBF) and critical path method (CPM) respectively. During the iteration process, the current Pareto front will be maintained by an external group (called external Pareto front (EP)).

[0072] (1) Ship-lock allocation: The ship allocation problem is the main problem of SLCP, which determines which lock the ship will be processed by during its navigation. In the present invention, SLCP is regarded as a variant of the flexible workshop scheduling problem based on batch processing, that is, the ship that needs to pass through a series of locks is mapped to the situation where a series of operations need to be processed for the job. The ship is considered as an "job", and the lock corresponds to the "batch" in the flexible job workshop scheduling. Each "machine" (lock chamber) may have a different structure and size. In order to realize this idea, the present invention adopts the following solution structure to describe the allocation information of ships and locks:

[0073]

[0074] In the formula, each Contains the ship stage required for a lock service, m represents the lock chamber, The present invention uses a large-scale neighborhood search operator to implement this process to quickly search for a better solution.

[0075] (2) Lock chamber packing: Different from the traditional two-dimensional packing problem, the placement of ships in the lock chamber needs to consider not only the width of the ship, but also the length and area of ​​the ship and whether the mooring constraints between ships are met. The new solution obtained by MOLNS after removing and inserting the solution structure may violate the ship placement constraints. Therefore, the present invention uses the Multiple Ordering Optimal Ship Placement (MOBF) algorithm to perform constraint detection on the new solution. Solutions that do not meet the constraints will not be accepted.

[0076] (3) Lock scheduling: This invention introduces the concept of engineering project management and regards the fuel of the ship as a resource required for workpiece processing. The lock scheduling sub-problem is regarded as a resource scheduling problem whose goal is to use as few resources as possible to achieve the shortest possible construction period. The critical path method (CPM) is introduced to determine the key stage, and the triangular distribution is introduced to adjust the speed.

[0077] The following is a detailed description of a multi-stage multi-line ship-lock coordinated scheduling method provided by an embodiment of the present invention. Figure 3 As shown, the following steps are included:

[0078] S1: Obtain ship information and lock attribute information. The ship information includes ship size, ship lock route, ship arrival time at the cascade lock system, speed range, etc. The lock attribute information includes lock size, one-time service operation time, reversal time, etc.

[0079] S2: Construct the objective function of minimizing the ship waiting time and the total fuel consumption of the ship.

[0080] S21: Minimize vessel waiting time. Minimizing vessel waiting time is often a top priority for navigation dispatchers and ship owners, especially when lock capacity is limited. This objective can be described by the vessel's detention time within the entire cascade lock system, i.e., the time difference between a vessel's arrival at the originating lock and its departure from the destination lock. The vessel waiting time objective function f1 is expressed as follows:

[0081]

[0082] Where, π i represents the weight of ship i, which represents the importance of ship i; Indicates the last lock that ship i needs to pass through, represents the time when ship i leaves the lock system; r i represents the time when ship i arrives at the lock system; Represents a collection of ships.

[0083] S22: Minimize the total fuel consumption in the lock. This is a goal that takes into account the green economy. The carbon emissions of a ship's navigation are directly proportional to the fuel consumption of the ship during navigation. Reducing carbon emissions actually requires reducing the fuel consumption of the ship. The fuel consumption of a ship is affected by multiple external factors, among which the ship's speed and load are important factors, and it has highly nonlinear characteristics. The fuel consumption of a ship per unit time is proportional to the cube of the ship's speed. Assume that the time required for a ship to travel a distance D at a constant speed v is D / v. Therefore, for a ship i with a travel distance D and a speed v, its total fuel consumption can be expressed as α i Dv 2 , where α iis the ship fuel consumption coefficient. Therefore, the ship's total fuel consumption objective function f2 is expressed as follows:

[0084]

[0085] Where, Ω i is the stage set of ship i. The introduction of speed will produce nonlinear constraints, so the 0-1 variable τ is introduced. ikv To discretize and linearize the objective function, τ ikv Indicates whether ship i is sailing at speed v in the kth stage, D ik represents the sailing distance of ship i in the kth stage, V i represents the speed set of ship i,

[0086] S3: The multi-objective optimization problem is divided into several vector sub-problems using the Chebyshev decomposition strategy, generating a series of uniformly distributed weight vectors to form a weight vector population {λ 1 ,λ 2 ,…,λ N}, each weight vector For each weight vector, the Euclidean distance between weight vectors is calculated, and the solution represented by the nearest weight vector is regarded as the neighbor solution (T).

[0087] Generate an initial solution for each sub-problem and obtain the initial solution population X={X1,X2,…,X N Specifically, for each initialization solution (individual), it is obtained by the following method: the ship is divided into several ship stages according to the ship passage route obtained in S1, and a separate lock number is assigned to each ship stage; each lock number is assigned to a feasible lock chamber, and the number of lock times assigned to each lock chamber is calculated and balanced during the allocation process, and the lock times of each lock are allocated to each lock chamber using the capacity ratio allocation method, thereby obtaining the initialization solution. Among them, the use of the capacity ratio allocation method to allocate the lock times of each lock to each lock chamber specifically includes: for n lock chambers in a certain lock, calculating the capacity (or area) Sn of each lock chamber and the capacity ratio Sn / sum(Sn) of each lock chamber; when allocating lock times, a roulette wheel method is used to allocate with the capacity ratio as the probability.

[0088] When initializing, let the optimal solution of each subproblem be And calculate the corresponding objective function Initialization reference point In addition, an external non-dominated solution set EP is initialized to record the Pareto optimal solutions in the current population.

[0089] S4: Traverse each subproblem in the solution population X, and find the solution X contained in each subproblem p The destruction operator based on the taboo table and the Chebyshev penalty insertion operator are used to search and reconstruct the allocation relationship between ships and locks. A new solution is obtained through destruction and repair operations. Then, the multiple sorting optimal ship arrangement algorithm (MOBF) is used to detect whether the ship-lock allocation scheme in the new solution meets the lock chamber scheduling constraints. If the constraints are met, the berthing position of each ship in each lock chamber in the new solution is recorded. The objective function value of the solution that violates the constraints will be set to infinity; for the feasible new solution X' p The critical path method (CPM) based on triangular distribution is used to decide the ship's speed at each stage and the opening time of the locks; the time when the ship leaves the lock system and the ship's speed at each stage in the lock system are obtained, and then the total stay time of the ship and the total fuel consumption of the ship are calculated.

[0090] More specifically, step S4 includes the following steps:

[0091] S41, traverse the solution population X, start the inner iteration. The solution X of the current sub-problem i , generate a random number ρ as the number of ship stages that need to be removed.

[0092] S42, using the random destruction operator based on the taboo table to remove ρ X p The ship phase that does not exist in the taboo table is temporarily solved. When a ship phase is removed, it is recorded in the taboo table and will not be removed in the next phase within the artificially set taboo period.

[0093] S43. Insert the removed ship phase using the Chebyshev penalty insertion operator. Insert the removed ship phase into the best possible position. The possible position can be an existing lock in the same direction as the inserted ship phase or a newly generated lock. Use the Multiple Ordered Optimal Ship Placement (MOBF) algorithm to check the solution obtained after inserting the removed ship phase into phase k. Whether the lock chamber gear constraint is met, the lock chamber gear constraint diagram is as follows Figure 2 As shown, if the constraints are met, record The berthing position of each ship in each lock chamber is obtained, and a new solution X' is obtained. p ;otherwise The objective function value of will be set to infinity; if If there are ships that cannot fit into the lock chamber, The lock chamber gear constraint is not satisfied; the optimal insertion position and Chebyshev penalty are recorded for each removal stage Represents the set of stages to be removed, χ mRepresents the minimum value of the current m-th objective. ; Sort the Chebyshev penalties of the best insertion positions of all stages and then insert the removal stages into their best positions in turn according to the deep greedy strategy to obtain the new solution X' p .

[0094] S44. Use triangular distribution to process the decision interval of ship speed. Specifically, divide the ship speed into several intervals from the lowest speed to the highest speed. Assume that the upper limit of an interval is b and the lower limit is a; according to the weight vector of the current sub-problem p Generate the most likely value From this we can get the expected speed in this interval under the current subproblem x is the integration variable, where

[0095] S45. Use the critical path method to determine the ship speed and lock opening time. First, determine the key locks and key ships, such as Figure 4 As shown, construct a directed acyclic graph G = (A, V) to represent the current solution X' p , V represents the node represented by the gate, A represents the time sequence relationship between the gates; secondly, calculate the earliest start and latest end time of each node, among which the node with the difference of the sum of the earliest start time and the gate operation time and the latest end time is 0 is the key gate, and the latest arriving ship in the key gate is the key ship. Make speed adjustments and calculate the contribution value of each key ship to the objective function where v i represents the current speed of ship i, Indicates the speed of ship i increasing by one gear. The speed range of ships with the largest contribution value is increased from the smallest. When all ships with positive contribution values ​​reach the maximum speed or the critical route changes due to the speed increase, the process stops. Time relaxation is performed, and the waiting time of non-critical ships is decelerated. That is, the speed range is reduced until all non-critical ships in the current lock reach the minimum speed or are about to affect the lock opening time. The start time of each lock is calculated based on the final speed, and the objective function value f(X' p ).

[0096] S5: Determine the new solution X' obtained for each sub-problem p Whether it is accepted, the accepted new solution will be called the current latest solution of the subproblem The latest solution to each subproblem They are all compared with their neighbor solutions, and the neighbor solutions are updated to obtain a new solution population, and the non-dominated solution set EP and the reference point χ are updated.

[0097] More specifically, step S5 includes the following steps:

[0098] S51: Use simulated annealing to consider whether the sub-problem accepts the new solution X' p Prioritize the solution of the current sub-problem that corresponds to the goal with a greater weight. If f m (X' p ) <f m (X p ), the new solution will be accepted; if f m (X' p )>f m (X p ), X' p will be The probability of being accepted is is a predefined coefficient that controls the probability of accepting a new solution that is worse than the current solution. The accepted new solution will be called the current state-of-the-art solution to the subproblem.

[0099] S52: After completing the inner iteration, the latest solution of each sub-problem All are solved with their neighbors j ,j∈T to compare and calculate the neighbor solution X j and The Chebyshev value g te (X j )and if but Where T is the set of neighboring solutions to the current subproblem solution, and is determined by calculating the Euclidean distance between the weight vectors of each subproblem. The Chebyshev value is calculated using the following formula: χ m Indicates the minimum value of the current m-th target.

[0100] S53: Update the non-dominated solution set EP and the reference point χ.

[0101] S6: Loop through steps S4 to S5 until the iteration termination condition is met, output the solutions in the non-dominated solution set EP and their corresponding solutions, and use them to perform ship-lock coordinated scheduling.

[0102] It should be noted that a set of multi-objective Pareto frontiers is ultimately output, and the optimal solution and optimal scheduling scheme with corresponding weights are selected according to the decision maker's current preference (preference for minimizing ship waiting time and minimizing total ship fuel consumption) to perform ship-lock coordinated scheduling.

[0103] The above embodiment provides a multi-stage multi-line ship-lock coordinated scheduling method, which has the following beneficial effects: in the multi-stage multi-line ship-lock coordinated scheduling (SLCP) problem, due to the mutual dependence between ship navigation and locks, optimizing them as an integrated system can maximize the overall operational efficiency and economic benefits of water transportation under the growing demand. The present invention designs a multi-objective heuristic algorithm that can effectively solve large-scale SLCP instances and output a set of multi-objective Pareto front solutions within a reasonable time, thereby improving the diversity of solutions. Compared with the existing technology, the present invention can solve the multi-stage multi-line ship-lock coordinated scheduling problem of a more general lock structure, while taking into account the optimization of lock chamber gears and ship speeds, providing a feasible strategy for alleviating the traffic pressure of inland waterways, serving the sustainable development of inland shipping, and providing theoretical and methodological support for the decision-making of lock operation scheduling.

[0104] The embodiment of the present invention further provides a multi-level multi-line ship-lock coordinated dispatching system, comprising:

[0105] Data acquisition module, used to obtain ship information and lock attribute information;

[0106] An objective function building module, used to build an objective function that minimizes ship waiting time and total ship fuel consumption;

[0107] Initialization module, used to generate the weight vector population {λ 1 ,λ 2 ,…,λ N}, each weight vector λ i Corresponding to a sub-problem; generate an initial solution for each sub-problem, and obtain the initial solution population X={X1,X2,…,X N Each solution in the solution population describes the allocation information of ships and locks; and initializes a reference point and an external non-dominated solution set EP that records the Pareto optimal solutions in the current population;

[0108] The loop iteration module is used to traverse each sub-problem in the solution population X, and for each sub-problem contained in the solution X p The destruction operator based on the taboo table and the Chebyshev penalty insertion operator are used to search and reconstruct the allocation relationship between ships and locks. A new solution is obtained through destruction and repair operations. Then, the ship-lock allocation scheme in the new solution is tested to see if it meets the lock chamber scheduling constraints. If it meets the constraints, the berthing position of each ship in each lock chamber in the new solution is recorded. For the feasible new solution X' p , the critical path method based on triangular distribution is used to make decisions on the ship's speed at each stage and the opening time of the lock; the time when the ship leaves the lock system and the speed of the ship at each stage in the lock system are obtained, and then the total stay time and total fuel consumption of the ship are calculated; the new solution X' obtained by each sub-problem is determinedp Whether it is accepted, the accepted new solution will be called the current latest solution of the subproblem The latest solution to each subproblem Compare them with their neighbor solutions, update the neighbor solutions, obtain a new solution population, and update the non-dominated solution set EP and the reference point; loop through the above iterative process until the iteration termination condition is met, and output the solutions in the non-dominated solution set EP and their corresponding solutions;

[0109] The scheduling module is used to perform ship-lock coordinated scheduling based on the solutions in the non-dominated solution set EP and their corresponding solutions.

[0110] It should be understood that the functional unit modules in various embodiments of the present invention can be concentrated in one processing unit, or each unit module can exist physically separately, or two or more unit modules can be integrated into one unit module, and can be implemented in the form of hardware or software.

[0111] An embodiment of the present invention further provides an electronic terminal, including:

[0112] one or more memories having computer programs or instructions stored thereon;

[0113] One or more processors are used to load and execute the computer program or instructions to implement the multi-level multi-line ship-lock coordinated scheduling method as described above.

[0114] An embodiment of the present invention further provides a computer-readable storage medium having a computer program or instruction stored thereon, which, when called by a computer, implements the multi-level multi-line ship-lock coordinated scheduling method as described above.

[0115] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0116] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0117] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0118] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0120] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A multi-level multi-line ship-lock coordinated scheduling method, characterized in that: The steps include: S1: Obtain ship information and lock attribute information; S2: Construct an objective function that minimizes the ship's waiting time and total fuel consumption; S3: Generate weight vector population {λ 1 ,λ 2 , ..., λ N }, each weight vector λ i Corresponding to a sub-problem; generate an initial solution for each sub-problem, and obtain the initial solution population X={X1,X2,...,X N Each solution in the solution population describes the allocation information of ships and locks; and initializes a reference point and an external non-dominated solution set EP that records the Pareto optimal solutions in the current population; S4: Traverse each subproblem in the solution population X, and find the solution X contained in each subproblem p The destruction operator based on the taboo table and the Chebyshev penalty insertion operator are used to search and reconstruct the allocation relationship between ships and locks. A new solution is obtained through destruction and repair operations. Then, the ship-lock allocation scheme in the new solution is tested to see if it meets the lock chamber scheduling constraints. If the constraints are met, the berthing positions of each ship in each lock chamber in the new solution are recorded. For the feasible new solution X′ p The critical path method based on triangular distribution is used to decide the ship's speed at each stage and the lock opening time. The time when the ship leaves the lock system and the speed of the ship at each stage in the lock system are obtained, and then the total stay time and total fuel consumption of the ship are calculated. S5: Determine the new solution X′ obtained for each sub-problem p Whether it is accepted, the accepted new solution will be called the current latest solution of the subproblem The latest solution to each subproblem Compare with its neighbor solutions, update the neighbor solutions, get a new solution population, and update the non-dominated solution set EP and reference point; S6: Loop through steps S4 to S5 until the iteration termination condition is met, output the solutions in the non-dominated solution set EP and their corresponding solutions, and use them to perform ship-lock coordinated scheduling.

2. The multi-stage multi-line ship-lock coordinated scheduling method according to claim 1 is characterized in that: In step S2, the ship waiting time objective function f1 is expressed as follows: Where, π i represents the weight of ship i, Indicates the last lock that ship i needs to pass through, represents the time when ship i leaves the lock system; r i represents the time when ship i arrives at the lock system; Represents a collection of ships; The ship's total fuel consumption objective function f2 is expressed as follows: Where, Ω i is the stage set of ship i, 0-1 variable τ ikv Indicates whether ship i is sailing at speed v in the kth stage, α i Indicates the ship fuel consumption coefficient, D ik represents the sailing distance of ship i in the kth stage, V i represents the speed set of ship i.

3. The multi-stage multi-line ship-lock coordinated scheduling method according to claim 1, characterized in that: In step S3, the solution uses the following structure to describe the ship and lock allocation information: Where, Indicates the number of locks assigned to lock chamber m Ship stage collection.

4. The multi-stage multi-line ship-lock coordinated scheduling method according to claim 1, characterized in that: In step S4, a multiple sorting optimal ship scheduling algorithm is used to detect whether the ship-lock order allocation scheme in the new solution satisfies the lock chamber scheduling constraints, specifically including: In the process of inserting the removed ship phases using the Chebyshev penalty insertion operator, each time a ship phase is inserted, the solution obtained after the removed ship phase is inserted into phase k is checked using the multi-sort optimal ship arrangement algorithm. Whether the lock chamber gear constraint is met, if so, record The berthing position of each ship in each lock chamber; otherwise The objective function value of will be set to infinity; Record the optimal insertion position and the corresponding Chebyshev penalty for each removed ship phase; Sort the Chebyshev penalties of the best insertion positions of all removed ship phases and then insert the removed ship phases into their best positions in turn according to the deep greedy strategy to obtain the new solution X′ i .

5. The multi-stage multi-line ship-lock coordinated scheduling method according to claim 1, characterized in that: In step S4, the critical path method based on triangular distribution is used to make decisions on the ship's speed and the lock opening time at each stage, specifically including: The decision interval of ship speed is processed by triangular distribution, including: dividing the ship speed into several intervals from the lowest speed to the highest speed, assuming that the upper limit of an interval is b and the lower limit is a; according to the weight vector of the current sub-problem p Generate the most likely value Thus, the expected speed E(X) in this interval under the current sub-problem is obtained; The critical path method is used to determine the ship speed and lock opening time, including: Determine the key locks and key ships, and construct a directed acyclic graph G = (A, V) to represent the current solution X' p , V represents the node represented by the lock, A represents the temporal relationship between the locks; calculate the earliest start time and the latest end time of each node, among which the node with the difference between the sum of the earliest start time and the lock operation time and the latest end time is 0 is the key lock, and the latest arriving ship in the key lock is the key ship; Adjust the speed and calculate the contribution of each key ship to the objective function Among them, f1 is the ship waiting time objective function, f2 is the ship total fuel consumption objective function, v i represents the current speed of ship i, Indicates the speed at which ship i increases by one gear, and increases the speed range of ships with decreasing contribution values ​​until the critical route is about to change; Perform time relaxation and decelerate non-critical ships whose waiting time is not 0, that is, reduce the speed range until all non-critical ships in the current lock reach the minimum speed or are about to affect the lock opening time; The start time of each lock is calculated based on the final speed, and thus the objective function value is calculated.

6. The multi-stage multi-line ship-lock coordinated scheduling method according to claim 1, characterized in that: In step S5, determine the new solution X′ obtained for each sub-problem p Whether it is accepted, including: Prioritize the solution of the current sub-problem that corresponds to the goal with a greater weight If f m (X′ p ) <f m (X p ), the new solution will be accepted, f1 and f2 represent the objective functions of ship waiting time and total fuel consumption respectively; If f m (X′ p )>f m (X p ), X′ p will be The probability of being accepted is is a predefined coefficient that controls the probability of accepting a new solution that is worse than the current solution. The accepted new solution will be called the current state-of-the-art solution to the subproblem.

7. The multi-stage multi-line ship-lock coordinated scheduling method according to claim 1, characterized in that: In step S5, the latest solution of each sub-problem Each solution is compared with its neighbor solutions and updated, including: The latest solution to each sub-problem Solve X with its neighbors j , j∈T to compare and calculate the neighbor solution and The Chebyshev value g te (X j )and if but Where T is the neighbor solution set of the current subproblem solution, and it is determined by calculating the Euclidean distance between the weight vectors of each subproblem.

8. A multi-level multi-line ship-lock coordinated dispatching system, characterized in that: include: Data acquisition module, used to obtain ship information and lock attribute information; An objective function building module, used to build an objective function that minimizes ship waiting time and total fuel consumption of the ship; Initialization module, used to generate the weight vector population {λ 1 ,λ 2 , ..., λ N }, each weight vector λ i Corresponding to a sub-problem; generate an initial solution for each sub-problem, and obtain the initial solution population X={X1,X2,...,X N Each solution in the solution population describes the allocation information of ships and locks; and initializes a reference point and an external non-dominated solution set EP that records the Pareto optimal solutions in the current population; The loop iteration module is used to traverse each sub-problem in the solution population X, and for each sub-problem contained in the solution X p The destruction operator based on the taboo table and the Chebyshev penalty insertion operator are used to search and reconstruct the allocation relationship between ships and locks. A new solution is obtained through destruction and repair operations. Then, the ship-lock allocation scheme in the new solution is tested to see if it meets the lock chamber scheduling constraints. If it meets the constraints, the berthing position of each ship in each lock chamber in the new solution is recorded. For the feasible new solution X' p , the critical path method based on triangular distribution is used to make decisions on the ship's speed and the opening time of the locks at each stage; Get the time when the ship leaves the lock system and the speed of the ship at each stage in the lock system, and then calculate the total stay time and total fuel consumption of the ship; determine the new solution X' obtained for each sub-problem p Whether it is accepted, the accepted new solution will be called the current latest solution of the subproblem The latest solution to each subproblem Compare them with their neighbor solutions, update the neighbor solutions, obtain a new solution population, and update the non-dominated solution set EP and the reference point; loop through the above iterative process until the iteration termination condition is met, and output the solutions in the non-dominated solution set EP and their corresponding solutions; The scheduling module is used to perform ship-lock coordinated scheduling based on the solutions in the non-dominated solution set EP and their corresponding solutions.

9. An electronic terminal, characterized in that: include: one or more memories having computer programs or instructions stored thereon; One or more processors are used to load and execute the computer program or instructions to implement the multi-level multi-line ship-lock coordinated scheduling method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is called by a computer, the multi-level multi-line ship-lock coordinated scheduling method according to any one of claims 1 to 7 is implemented.

Images