An auxiliary lockage method and system for an inland multi-line ship lock

Abstract

This invention provides an auxiliary lock passage method and system for multi-line inland waterway locks. The method includes: acquiring basic attributes and dynamic data of vessels to set priority coefficients and lock passage safety constraints; acquiring lock waiting area information and obtaining a candidate lock passage vessel dataset using dynamic data and lock waiting area information; performing lock passage scheduling on the candidate lock passage vessel dataset using lock passage safety constraints to obtain initial lock passage scheduling scheme particles and their corresponding fitness scores; and performing several rounds of iterative optimization on the initial lock passage scheduling scheme particles based on the fitness scores to obtain the final lock passage scheduling scheme particles. By predicting vessels waiting to pass through the lock using dynamic data and lock waiting area information, passive planning is transformed into active planning. By obtaining fitness scores and combining them with several rounds of iterative optimization to improve the scheme, and considering multiple constraint combinations, the lock passage scheduling of a large number of vessels can be completed automatically and intelligently, ensuring lock passage efficiency.

Description

Technical Field

[0001] This invention relates to the field of transportation control technology, and in particular to an auxiliary lock passage method and system for multi-line inland waterway locks. Background Technology

[0002] Water transport, also known as water transport, is a mode of transportation that uses ships as transport vehicles to carry goods and passengers along certain routes in waterways such as oceans, rivers, lakes, reservoirs, and canals.

[0003] A lock is a box-shaped hydraulic structure that allows ships to pass through water level differences in a waterway. It consists of a lock head, lock chamber, water conveyance system, and gates. By filling and emptying water to regulate the water level in the lock chamber, ships can vertically ascend and descend to cross water level differences.

[0004] Inland waterway locks are crucial for ensuring smooth waterway transportation. With the increasing volume of waterborne cargo, the efficiency of lock passage directly affects the overall transportation efficiency of the waterway. Currently, lock passage scheduling mainly relies on manual experience. However, manual scheduling struggles to quickly handle the combined optimization problems of a large number of vessels and various constraints (such as vessel type and size), resulting in low lock chamber space utilization, long average waiting times for vessels, and congestion during peak periods. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide an auxiliary lock passage method and system for multi-line inland waterway locks. This system addresses the technical problems in existing technologies where manual lock passage scheduling makes it difficult to quickly handle the combination and optimization of a large number of ships and various constraints, resulting in low lock chamber space utilization and long average waiting time for ships.

[0006] To achieve the above objectives, in a first aspect, embodiments of this application provide an auxiliary lock passage method for multi-line locks on inland waterways, comprising the following steps: The transportation information of the vessel is obtained, including basic attributes and dynamic data. Priority coefficients are set for the vessel based on the basic attributes, and lock passage safety constraints are set for the vessel based on the basic attributes. Obtain information about the lock-up area, and obtain a candidate lock-up vessel dataset using the dynamic data and the lock-up area information. The candidate lock-up vessel dataset includes several vessels waiting to pass through the lock. The candidate lock-passing vessel dataset is sorted by the lock-passing safety constraints to obtain several initial lock-passing scheduling scheme particles, and the fitness score of the initial lock-passing scheduling scheme particles is obtained. Based on the fitness score, the initial gate passage scheduling scheme particles are iteratively optimized in several rounds to obtain the final gate passage scheduling scheme particles, and gate passage arrangements are made based on the final gate passage scheduling scheme particles.

[0007] Furthermore, the basic attributes include cargo type, which includes salvage cargo, dangerous goods, general cargo, and no cargo. The step of setting the priority coefficient of the ship based on the basic attributes includes: If the cargo is emergency relief goods, then the priority coefficient of the vessel is set to 1; If the cargo is dangerous goods, the priority coefficient of the vessel is set to 0.8; If the type of cargo is general cargo, then the priority coefficient of the vessel is set to 0.5; If the type of cargo is no cargo, then the priority coefficient of the vessel is set to 0.3.

[0008] Furthermore, the basic attributes also include ship type and static data. The ship type includes passenger ships, self-unloading ships, container ships, and general cargo ships. The static data includes ship height, ship volume, and cargo weight. The step of setting lock passage safety constraints for ships using the basic attributes includes: The vessel's carrying status is obtained based on the cargo weight. The carrying status includes empty and loaded status. If the vessel's carrying status is loaded, the vessel is identified as a loaded vessel. If the vessel's carrying status is empty, the vessel is identified as an empty vessel. The ship's volume is compared with a volume threshold. If the ship's volume is greater than the volume threshold, the ship is classified as a large ship. The height difference between the two ships is determined based on the ship height. The height difference is compared with a difference threshold. If the height difference is greater than the difference threshold, the two ships are selected as a difference ship group. If the vessel type is a passenger ship, its preset position is set to the front row position. If the vessel type is a self-unloading ship, its preset position is set to a non-front row position. If the vessel is a large vessel or a heavily loaded vessel, its preset position is set to the shore side position. If the two vessels are different vessel groups, their preset positions are set to the first non-co-row position. If the vessel is empty, its preset position with the container ship is set to the second non-co-row position.

[0009] Furthermore, the dynamic data includes vessel position, vessel speed, rate of change of vessel speed, and vessel heading; the lock-waiting area information includes the lock-waiting area location, route restrictions, real-time water flow, and real-time weather conditions; the step of obtaining a candidate lock-waiting vessel dataset using the dynamic data and the lock-waiting area information, wherein the candidate lock-waiting vessel dataset includes several vessels waiting to pass through the lock, includes: The dynamic data and the information of the waiting area are input into a pre-trained LSTM prediction model to obtain the estimated arrival time through the LSTM prediction model. The estimated arrival time is compared with a time threshold. If the estimated arrival time is less than the time threshold, the vessel is selected as a vessel to be passed through the lock, and several vessels to be passed through the lock are combined into a candidate vessel dataset.

[0010] Furthermore, the step of obtaining the fitness score of the particles in the initial gate passage scheduling scheme includes: The utilization rate of a single lock chamber is obtained by the ship volume, and the lock load balance is obtained based on several single lock chamber utilization rates. Based on the initial lock passage scheduling scheme particles, the planned lock entry time of the vessel is obtained, and based on the estimated lock arrival time, the planned lock entry time and the priority coefficient, the total waiting time is obtained. The fitness score is obtained by measuring the lock load balance and the total waiting time.

[0011] Furthermore, the formula for obtaining the load balancing degree of the lock is: , in, This represents the lock load balance degree corresponding to the i-th initial lock passage scheduling scheme particle. This represents the single-cell utilization rate of the k-th gate chamber in the i-th initial gate scheduling scheme particle. This represents the average space utilization rate of all gate chambers in the i-th initial gate scheduling scheme particle. Indicates the total number of gate chambers; The formula for obtaining the total waiting time is: , in, This represents the total waiting time corresponding to the i-th initial gate passage scheduling scheme particle. This represents the priority coefficient of the d-th ship. This indicates the planned entry time of the d-th vessel. This indicates the estimated arrival time of the d-th vessel at the lock. Indicates the total number of ships; The formula for obtaining the fitness score is: , in, Let represent the fitness score of the particle in the i-th initial gate passage scheduling scheme. This indicates the maximum waiting time for the lock. , , All of these represent the balancing weight coefficients.

[0012] Furthermore, the step of performing several rounds of iterative optimization on the initial gate passage scheduling scheme particles based on the fitness score to obtain the final gate passage scheduling scheme particles includes: In the same round of iterative optimization, among different initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the globally optimal gate passage scheduling scheme particle. In different rounds of iterative optimization, among the same initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the historical best gate passage scheduling scheme particle. In the same round of iterative optimization, the score variance is obtained through the fitness score, and the inertia weight is obtained through the score variance. An update vector is obtained based on the global optimal gate crossing scheduling scheme particle, the historical optimal gate crossing scheduling scheme particle and the inertia weight. The initial gate crossing scheduling scheme particle is modified based on the update vector. If the change in fitness score corresponding to the particle of the global optimal gate passage scheduling scheme from the fitness score before the preset number of iterations is less than a preset threshold, then it is selected as the particle of the final gate passage scheduling scheme.

[0013] Furthermore, the formula for obtaining the inertia weight is: , in, This represents the inertia weight for the p-th iteration of optimization. This represents the maximum inertia weight. This represents the minimum inertia weight. This represents the variance of the score in the p-th iteration optimization. This indicates the preset maximum score variance.

[0014] Furthermore, the formula for obtaining the update vector is: , in, Let represent the update vector of the particle of the i-th initial gate passage scheduling scheme obtained from the iterative optimization in the p-th round, in the next round of iterative optimization. Let represent the update vector of the particle of the i-th initial gate passage scheduling scheme in the p-th iteration optimization, obtained based on the p-1-th iteration optimization. Let represent the historical best gate scheduling scheme particle corresponding to the i-th initial gate scheduling scheme particle in the p-th round of iterative optimization. Let represent the globally optimal gate passage scheduling scheme particle in the p-th round of iterative optimization. Let represent the i-th initial gate passage scheduling scheme particle in the p-th round of iterative optimization. , All represent learning factors. , Both represent random numbers between 0 and 1.

[0015] Secondly, embodiments of this application provide an auxiliary lock passage system for a multi-line inland waterway lock, applied to the auxiliary lock passage method for a multi-line inland waterway lock as described in the first aspect above. The system includes: The first acquisition module is used to acquire the transportation information of the vessel, which includes basic attributes and dynamic data. The basic attributes are used to set priority coefficients for the vessel and to set lock passage safety constraints for the vessel. The second acquisition module is used to acquire information about the lock-up area and to acquire a candidate lock-up vessel dataset through the dynamic data and the lock-up area information. The candidate lock-up vessel dataset includes several vessels waiting to pass through the lock. The analysis module is used to perform lock passage scheduling on the candidate lock passage vessel dataset through the lock passage safety constraints to obtain several initial lock passage scheduling scheme particles and obtain the fitness score of the initial lock passage scheduling scheme particles. The execution module is used to perform several rounds of iterative optimization on the initial gate passage scheduling scheme particles based on the fitness score to obtain the final gate passage scheduling scheme particles, and to arrange gate passage based on the final gate passage scheduling scheme particles.

[0016] Thirdly, embodiments of this application provide a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the auxiliary lock passage method for inland waterway multi-line locks as described in the first aspect above.

[0017] Fourthly, embodiments of this application provide a storage medium storing a computer program thereon, which, when executed by a processor, implements the auxiliary lock passage method for inland waterway multi-line locks as described in the first aspect above.

[0018] Compared with existing technologies, the beneficial effects of this invention are as follows: By setting the lock passage safety constraints, the fuzzy scheduling rules that rely on human experience are transformed into structured rules, providing a rule basis for obtaining subsequent lock passage schemes; by predicting the vessels waiting to pass through the lock using the dynamic data and the lock waiting area information, the problem of manual scheduling being unable to anticipate and dynamically adapt to changes in vessel navigation is solved, transforming passive planning into active planning and providing data support for generating efficient scheduling schemes; by obtaining the fitness score, the complex gearing problem is transformed into a multi-objective optimization problem, and the scheme is improved through several rounds of iterative optimization to obtain the final lock passage scheduling scheme particles. Under the premise of considering multiple constraint combinations, the lock passage gearing of a large number of vessels can be completed automatically and intelligently, ensuring lock passage efficiency. Attached Figure Description

[0019] Figure 1 This is a flowchart of the auxiliary lock passage method for a multi-line inland waterway lock in the first embodiment of the present invention; Figure 2 This is a structural block diagram of the auxiliary lock passage system of the inland waterway multi-line lock in the second embodiment of the present invention; The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0020] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0021] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0023] Please see Figure 1 The auxiliary lock passage method for multi-line ship locks in inland waterways provided in the first embodiment of the present invention includes the following steps: S10: Obtain the vessel's transportation information, which includes basic attributes and dynamic data. Set priority coefficients for the vessel based on the basic attributes and set lock passage safety constraints for the vessel based on the basic attributes. Understandably, the transportation information also includes lock passage declaration information. Based on the lock passage declaration information, it is determined whether the vessel needs to pass through the lock, and the basic attributes and dynamic data of the vessel that needs to pass through the lock are collected for subsequent lock passage scheduling. It should be noted that after obtaining the basic attributes and dynamic data, data preprocessing is required. The data preprocessing includes removing abnormal data (such as position jumps or speed changes caused by AIS signal loss) and standardization processing (normalizing vessel speed and other data to the [0,1] interval) to facilitate subsequent data use. The removal of abnormal data and data standardization processing are now widely used and will not be elaborated here.

[0024] The basic attributes include cargo type, which includes emergency cargo, dangerous cargo, general cargo, and no cargo. The basic attributes also include ship type and static data, which include passenger ships, self-unloading ships, container ships, and general cargo ships. The static data includes ship height, ship volume, and cargo weight.

[0025] Step S10 includes: S110: If the cargo is emergency relief goods, the priority coefficient of the vessel shall be set to 1. S120: If the cargo is dangerous goods, the priority coefficient of the vessel shall be set to 0.8; S130: If the type of cargo is general cargo, the priority coefficient of the vessel is set to 0.5; S140: If the type of cargo is no cargo, then set the priority coefficient of the vessel to 0.3; S150: Obtain the vessel's carrying status based on the cargo weight. The carrying status includes empty and loaded status. If the vessel's carrying status is loaded, the vessel is determined to be a loaded vessel. If the vessel's carrying status is empty, the vessel is determined to be an empty vessel. S160: Compare the ship's volume with a volume threshold. If the ship's volume is greater than the volume threshold, the ship is classified as a large ship. S170: Determine the height difference between the two ships based on the ship height, compare the height difference with a difference threshold, and if the height difference is greater than the difference threshold, select the two ships as a difference ship group. S180: If the vessel type is a passenger ship, its preset position is set to the front row position; if the vessel type is a self-unloading ship, its preset position is set to a non-front row position; if the vessel is a large vessel or a heavily loaded vessel, its preset position is set to the shore side position; if the two vessels are different vessel groups, their preset positions are set to the first non-co-row position; if the vessel is an empty vessel, its preset position with the container ship is set to the second non-co-row position. The lock passage safety constraints are as follows: 1. Passenger ships must be in the lead row; 2. Self-unloading ships must not be in the lead row; 3. Large ships and heavily loaded ships must be near the shore; 4. Two ships with a height difference greater than a threshold value must not be placed side by side; 5. Container ships must not be placed side by side with empty ships. In this embodiment, the threshold value is 4.5m. It should be noted that the lead row refers to the first row to enter a lock chamber.

[0026] S20: Obtain lock-up area information, and obtain a candidate lock-up vessel dataset through the dynamic data and the lock-up area information. The candidate lock-up vessel dataset includes several vessels waiting to pass through the lock. The dynamic data includes ship position, ship speed, ship speed change rate, and ship heading. The lock waiting area information includes the lock waiting area location, route restrictions, real-time water flow, and real-time weather conditions.

[0027] Step S20 includes: S210: Input the dynamic data and the information of the waiting area into the pre-trained LSTM prediction model to obtain the estimated arrival time through the LSTM prediction model; The LSTM prediction model can be trained based on historical dynamic data, historical waiting area information, and historical actual arrival time. The historical dynamic data and historical waiting area information are used as input values ​​for the untrained LSTM prediction model, and the historical actual arrival time is used as the output value for the untrained LSTM prediction model, so that the trained model can obtain the expected arrival time. The LSTM prediction model is now widely used and will not be described in detail here.

[0028] S220: Compare the estimated arrival time with a time threshold. If the estimated arrival time is less than the time threshold, select the vessel as the vessel to be passed through the lock and combine several vessels to be passed through the lock into a candidate vessel dataset. Preferably, the time threshold is 105 min to 135 min; in this embodiment, the time threshold is 120 min. The candidate lock-passing vessel dataset S can be represented as {S1, S2, ..., Sm}, where m represents the number of vessels waiting to pass through the lock.

[0029] S30: The candidate lock-passing vessel dataset is sorted according to the lock-passing safety constraints to obtain several initial lock-passing scheduling scheme particles. The initial fitness score of the initial lock-passing scheduling scheme particles is obtained. The initial lock-passing scheduling scheme particle corresponding to the highest initial fitness score is selected as the stage lock-passing scheduling scheme particle, and the single-round lock-passing prediction is completed. Based on the aforementioned lock passage safety constraints, various combinations of ship lock passage can be arranged, namely, several initial lock passage scheduling scheme particles. , can be characterized as ,in, This indicates that the b-th vessel is assigned to the a-th lock chamber. This indicates the vessel entry sequence for the a-th lock chamber. It should be noted that all vessel positions are unique, ensuring that each vessel is assigned to only one entry position within one lock chamber.

[0030] Step S30 includes: S310: Obtain the single lock chamber utilization rate through the ship volume, and obtain the lock load balance based on several single lock chamber utilization rates; The utilization rate of a single lock chamber is obtained by dividing the total volume of all vessels in the chamber by the volume of the lock chamber itself. The formula for obtaining the lock load balance is as follows: , in, This represents the lock load balance degree corresponding to the i-th initial lock passage scheduling scheme particle. This represents the single-cell utilization rate of the k-th gate chamber in the i-th initial gate scheduling scheme particle. This represents the average space utilization rate of all gate chambers in the i-th initial gate scheduling scheme particle. This indicates the total number of gate chambers.

[0031] S320: Based on the initial lock passage scheduling scheme particles, obtain the planned lock entry time of the vessel, and based on the estimated lock arrival time, the planned lock entry time and the priority coefficient, obtain the total waiting time; Understandably, after obtaining the initial lock passage scheduling scheme particles, the planned entry time for each vessel can be estimated based on historical lock passage data. The formula for obtaining the total waiting time is: , in, This represents the total waiting time corresponding to the i-th initial gate passage scheduling scheme particle. This represents the priority coefficient of the d-th ship. This indicates the planned entry time of the d-th vessel. This indicates the estimated arrival time of the d-th vessel at the lock. This indicates the total number of ships.

[0032] S330: Obtain an initial fitness score based on the lock load balance and the total waiting time. The formula for obtaining the initial fitness score is: , in, Let represent the initial fitness score of the particle in the i-th initial gate passage scheduling scheme. This indicates the maximum waiting time for the lock. , , All represent the balance weight coefficients. By introducing the fitness score, the gate scheduling scheme is comprehensively considered from multiple perspectives, ensuring that the optimal scheme can be obtained comprehensively and accurately in the future.

[0033] S40: Based on the fitness score, perform several rounds of iterative optimization on the initial gate passage scheduling scheme particles to obtain the final gate passage scheduling scheme particles, and arrange gate passage based on the final gate passage scheduling scheme particles; Step S40 includes: S410: In the same round of iterative optimization, among different initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the globally optimal gate passage scheduling scheme particle. Suppose there are three initial gate passage scheduling scheme particles: X1, X2, and X3. During the first round of iterative optimization, the fitness score corresponding to X2 is the highest. Therefore, X2 is the globally optimal gate passage scheduling scheme particle during the first round of iterative optimization.

[0034] S420: In different rounds of iterative optimization, among the same initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the historical best gate passage scheduling scheme particle. Taking X1 as an example, after the first round of iterative optimization, it is updated to X. 1new During the second round of iterative optimization, if the fitness score corresponding to X1 is higher than X... 1new If X1 is still selected as the historically optimal gate passage scheduling scheme particle, and the fitness score corresponding to X1 is lower than that of X... 1new Then X 1new The particle selected is the historically optimal gate passage scheduling scheme.

[0035] S430: In the same round of iterative optimization, obtain the score variance through the fitness score, obtain the inertia weight through the score variance, obtain the update vector based on the global optimal gate crossing scheduling scheme particle, the historical optimal gate crossing scheduling scheme particle and the inertia weight, and modify the initial gate crossing scheduling scheme particle based on the update vector. In the same round of iterative optimization, the average fitness score of all initial gate passage scheduling scheme particles is obtained. Then, the average of the squared differences between each fitness score and the average score is obtained, which is the score variance.

[0036] The formula for obtaining the inertia weight is: , in, This represents the inertia weight for the p-th iteration of optimization. This represents the maximum inertia weight, in this embodiment, The value is 0.9. In this embodiment, the minimum inertia weight is represented. The value is 0.4. This represents the variance of the score in the p-th iteration optimization. This indicates the preset maximum score variance. This can be a preset value or set to the maximum value of the score variance during the iterative optimization process in previous rounds. Adaptive inertial weights are obtained based on the score variance. When the score variance is large, the search range can be expanded to avoid premature contraction. When the score variance is small, a fine local search can be performed near the optimal solution, thus stably and accurately approximating the global optimum. This significantly improves the probability and efficiency of the algorithm finding the global optimum or a high-quality approximate solution, thereby generating a better gate-crossing scheme.

[0037] The formula for obtaining the update vector is: , in, Let represent the update vector of the particle of the i-th initial gate passage scheduling scheme obtained from the iterative optimization in the p-th round, in the next round of iterative optimization. Let represent the update vector of the particle of the i-th initial gate passage scheduling scheme in the p-th iteration optimization, obtained based on the p-1-th iteration optimization. Let represent the historical best gate scheduling scheme particle corresponding to the i-th initial gate scheduling scheme particle in the p-th round of iterative optimization. Let represent the globally optimal gate passage scheduling scheme particle in the p-th round of iterative optimization. Let represent the i-th initial gate passage scheduling scheme particle in the p-th round of iterative optimization. , All represent learning factors. In this embodiment, during the initial stage of iterative optimization, The value is 2.5. The value is 1.5, which is used in the later stages of iterative optimization. The value is 1.5. The value is 2.5. , Both represent random numbers between 0 and 1.

[0038] After obtaining the update vector, the instruction is decoded based on the update vector, and a swap operation is performed on the particles. A common method for this swap operation is to select based on the update vector. The positions of the two ships are swapped or selected based on the update vector. When a vessel is moved from one position to another, it should be noted that the modification process must still comply with the aforementioned lock passage safety constraints.

[0039] S440: If the change in fitness score corresponding to the global optimal gate passage scheduling scheme particle from the fitness score before the preset number of iterations is less than a preset threshold, then it is selected as the final gate passage scheduling scheme particle. In this embodiment, the preset number of rounds is 3, and the preset threshold is 0.005. That is, assuming the current round is the 50th round, if the change in fitness score between the globally optimal gate passage scheduling scheme particle selected after this round of optimization and the globally optimal gate passage scheduling scheme particle selected in the 48th round is less than 0.005, then the iterative optimization ends, and the globally optimal gate passage scheduling scheme particle of this round is selected as the final gate passage scheduling scheme particle.

[0040] Preferably, after obtaining the particles of the final lock passage scheduling scheme, the data can be pushed to the vessel waiting to pass through the lock via an app. Furthermore, the transportation information can be collected in real time to update the estimated arrival time in conjunction with the LSTM prediction model. If the difference between the updated estimated arrival time and the previous estimated arrival time is greater than 30 minutes, the vessel is contacted for arrival confirmation. If the vessel reports that it will arrive on time, there is no need to iterate again. If the vessel reports a delay, the scheme is readjusted based on the updated estimated arrival time, following the steps described above.

[0041] By setting the lock passage safety constraints, the fuzzy scheduling rules that rely on human experience are transformed into structured rules, providing a rule foundation for obtaining subsequent lock passage schemes. Predicting the number of vessels waiting to pass through the lock using dynamic data and lock waiting area information solves the problem that manual scheduling cannot proactively and dynamically adapt to changes in vessel navigation, transforming passive planning into proactive planning and providing data support for generating efficient scheduling schemes. By obtaining the fitness score, the complex queuing problem is transformed into a multi-objective optimization problem, and the scheme is improved through several rounds of iterative optimization, obtaining the final lock passage scheduling scheme particles. Considering multiple constraint combinations, the lock passage queuing of a large number of vessels can be completed automatically and intelligently, ensuring lock passage efficiency.

[0042] Please see Figure 2 The second embodiment of the present invention provides an auxiliary lock passage system for a multi-line inland waterway lock. This system is applied to the auxiliary lock passage method for a multi-line inland waterway lock described in the above embodiments, and will not be repeated hereafter. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0043] The system includes: The first acquisition module 10 is used to acquire the transportation information of the vessel, which includes basic attributes and dynamic data. The basic attributes are used to set priority coefficients for the vessel and to set lock passage safety constraints for the vessel. The first acquisition module 10 includes: The first unit is used to set the priority coefficient of the ship to 1 if the cargo is emergency relief goods; 0.8 if the cargo is dangerous goods; 0.5 if the cargo is general cargo; and 0.3 if there is no cargo. The second unit is used to obtain the ship's carrying status based on the cargo weight. The carrying status includes empty and loaded status. If the ship's carrying status is loaded, the ship is identified as a loaded ship. If the ship's carrying status is empty, the ship is identified as an empty ship. The third unit is used to compare the ship's volume with a volume threshold. If the ship's volume is greater than the volume threshold, the ship is identified as a large ship. The fourth unit is used to determine the height difference between two ships based on the ship height, compare the height difference with a difference threshold, and if the height difference is greater than the difference threshold, then the two ships are selected as a difference ship group. The fifth unit is used to set the preset position of a passenger ship to the front row position if the ship type is a passenger ship, to a non-front row position if the ship type is a self-unloading ship, to the shore side position if the ship is a large ship or a heavily loaded ship, to the first non-same row position if the two ships are different ship groups, and to the second non-same row position if the ship is an empty ship and the preset position of the container ship is set. The second acquisition module 20 is used to acquire information about the lock-up area and to acquire a candidate lock-up vessel dataset through the dynamic data and the lock-up area information. The candidate lock-up vessel dataset includes a number of vessels waiting to pass through the lock. The second acquisition module 20 includes: The sixth unit is used to input the dynamic data and the information of the waiting area into a pre-trained LSTM prediction model, so as to obtain the expected arrival time through the LSTM prediction model. The seventh unit is used to compare the estimated arrival time with a time threshold. If the estimated arrival time is less than the time threshold, the vessel is selected as a vessel to be passed through the lock, and several vessels to be passed through the lock are combined into a candidate vessel dataset. Analysis module 30 is used to perform lock passage scheduling on the candidate lock passage vessel dataset through the lock passage safety constraints to obtain several initial lock passage scheduling scheme particles and obtain the fitness score of the initial lock passage scheduling scheme particles. The analysis module 30 includes: The eighth unit is used to obtain the single lock chamber utilization rate through the ship volume, and to obtain the lock load balance based on several single lock chamber utilization rates. The ninth unit is used to obtain the planned entry time of the vessel based on the initial lock passage scheduling scheme particles, and to obtain the total waiting time based on the estimated arrival time, the planned entry time and the priority coefficient. The tenth unit is used to obtain a fitness score through the lock load balance degree and the total waiting time. The execution module 40 is used to perform several rounds of iterative optimization on the initial gate passage scheduling scheme particles based on the fitness score to obtain the final gate passage scheduling scheme particles, and to perform gate passage arrangements based on the final gate passage scheduling scheme particles. The execution module 40 includes: Unit 11 is used to select the initial gate passage scheduling scheme particle corresponding to the highest fitness score from different initial gate passage scheduling scheme particles in the same round of iterative optimization as the globally optimal gate passage scheduling scheme particle. The twelfth unit is used for iterative optimization in different rounds. Among the same initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the historical best gate passage scheduling scheme particle. The thirteenth unit is used to obtain the score variance through the fitness score and the inertia weight through the score variance in the same round of iterative optimization, obtain the update vector based on the global optimal gate crossing scheduling scheme particle, the historical optimal gate crossing scheduling scheme particle and the inertia weight, and modify the initial gate crossing scheduling scheme particle based on the update vector. The fourteenth unit is used to select the particle corresponding to the global optimal gate passage scheduling scheme as the final gate passage scheduling scheme particle if the change value between the fitness score corresponding to the particle and the fitness score before the preset number of iterations is less than a preset threshold.

[0044] The present invention also provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the auxiliary lock passage method for inland waterway multi-line locks as described in the above technical solution.

[0045] The present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the auxiliary lock passage method for inland waterway multi-line locks as described in the above technical solution.

[0046] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0047] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method of assisted lockage of an inland waterway multi-lane lock, characterized in that Includes the following steps: The transportation information of the vessel is obtained, including basic attributes and dynamic data. Priority coefficients are set for the vessel based on the basic attributes, and lock passage safety constraints are set for the vessel based on the basic attributes. Obtain information about the lock-up area, and obtain a candidate lock-up vessel dataset using the dynamic data and the lock-up area information. The candidate lock-up vessel dataset includes several vessels waiting to pass through the lock. The candidate lock-passing vessel dataset is sorted by the lock-passing safety constraints to obtain several initial lock-passing scheduling scheme particles, and the fitness score of the initial lock-passing scheduling scheme particles is obtained. Based on the fitness score, the initial gate passage scheduling scheme particles are iteratively optimized in several rounds to obtain the final gate passage scheduling scheme particles, and gate passage arrangements are made based on the final gate passage scheduling scheme particles.

2. The method according to claim 1, characterized in that, The basic attributes include cargo type, which includes salvage cargo, dangerous goods, general cargo, and no cargo. The step of setting priority coefficients for ships based on the basic attributes includes: If the cargo is emergency relief goods, then the priority coefficient of the vessel is set to 1; If the cargo is dangerous goods, the priority coefficient of the vessel is set to 0.8; If the type of cargo is general cargo, then the priority coefficient of the vessel is set to 0.5; If the type of cargo is no cargo, then the priority coefficient of the vessel is set to 0.

3.

3. The auxiliary lock passage method for multi-line inland waterway locks according to claim 1, characterized in that, The basic attributes also include ship type and static data. The ship type includes passenger ships, self-unloading ships, container ships, and general cargo ships. The static data includes ship height, ship volume, and cargo weight. The step of setting lock passage safety constraints for ships using the basic attributes includes: The vessel's carrying status is obtained based on the cargo weight. The carrying status includes empty and loaded status. If the vessel's carrying status is loaded, the vessel is identified as a loaded vessel. If the vessel's carrying status is empty, the vessel is identified as an empty vessel. The ship's volume is compared with a volume threshold. If the ship's volume is greater than the volume threshold, the ship is classified as a large ship. The height difference between the two ships is determined based on the ship height. The height difference is compared with a difference threshold. If the height difference is greater than the difference threshold, the two ships are selected as a difference ship group. If the vessel type is a passenger ship, its preset position is set to the front row position. If the vessel type is a self-unloading ship, its preset position is set to a non-front row position. If the vessel is a large vessel or a heavily loaded vessel, its preset position is set to the shore side position. If the two vessels are different vessel groups, their preset positions are set to the first non-co-row position. If the vessel is empty, its preset position with the container ship is set to the second non-co-row position.

4. The auxiliary lock passage method for multi-line inland waterway locks according to claim 3, characterized in that, The dynamic data includes vessel position, vessel speed, rate of change of vessel speed, and vessel heading. The lock-waiting area information includes the lock-waiting area location, route restrictions, real-time water flow, and real-time weather conditions. The step of obtaining a candidate lock-waiting vessel dataset using the dynamic data and the lock-waiting area information, wherein the candidate lock-waiting vessel dataset includes several vessels waiting to pass through the lock, includes: The dynamic data and the information of the waiting area are input into a pre-trained LSTM prediction model to obtain the estimated arrival time through the LSTM prediction model. The estimated arrival time is compared with a time threshold. If the estimated arrival time is less than the time threshold, the vessel is selected as a vessel to be passed through the lock, and several vessels to be passed through the lock are combined into a candidate vessel dataset.

5. The auxiliary lock passage method for multi-line inland waterway locks according to claim 4, characterized in that, The step of obtaining the fitness score of the particles in the initial gate crossing scheduling scheme includes: The utilization rate of a single lock chamber is obtained by the ship volume, and the lock load balance is obtained based on several single lock chamber utilization rates. Based on the initial lock passage scheduling scheme particles, the planned lock entry time of the vessel is obtained, and based on the estimated lock arrival time, the planned lock entry time and the priority coefficient, the total waiting time is obtained. The fitness score is obtained by measuring the lock load balance and the total waiting time.

6. The auxiliary lock passage method for multi-line inland waterway locks according to claim 5, characterized in that, The formula for obtaining the load balance of the lock is: ， in, This represents the lock load balance degree corresponding to the i-th initial lock passage scheduling scheme particle. This represents the single-cell utilization rate of the k-th gate chamber in the i-th initial gate scheduling scheme particle. This represents the average space utilization rate of all gate chambers in the i-th initial gate scheduling scheme particle. Indicates the total number of gate chambers; The formula for obtaining the total waiting time is: ， in, This represents the total waiting time corresponding to the i-th initial gate passage scheduling scheme particle. This represents the priority coefficient of the d-th ship. This indicates the planned entry time of the d-th vessel. This indicates the estimated arrival time of the d-th vessel at the lock. Indicates the total number of ships; The formula for obtaining the fitness score is: ， in, Let represent the fitness score of the particle in the i-th initial gate passage scheduling scheme. This indicates the maximum waiting time for the lock. , , All of these represent the balancing weight coefficients.

7. The auxiliary lock passage method for multi-line inland waterway locks according to claim 1, characterized in that, The step of performing several rounds of iterative optimization on the initial gate passage scheduling scheme particles based on the fitness score to obtain the final gate passage scheduling scheme particles includes: In the same round of iterative optimization, among different initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the globally optimal gate passage scheduling scheme particle. In different rounds of iterative optimization, among the same initial gate passage scheduling scheme particles, the initial gate passage scheduling scheme particle corresponding to the highest fitness score is selected as the historical best gate passage scheduling scheme particle. In the same round of iterative optimization, the score variance is obtained through the fitness score, and the inertia weight is obtained through the score variance. An update vector is obtained based on the global optimal gate crossing scheduling scheme particle, the historical optimal gate crossing scheduling scheme particle and the inertia weight. The initial gate crossing scheduling scheme particle is modified based on the update vector. If the change in fitness score corresponding to the particle of the global optimal gate passage scheduling scheme from the fitness score before the preset number of iterations is less than a preset threshold, then it is selected as the particle of the final gate passage scheduling scheme.

8. The auxiliary lock passage method for multi-line inland waterway locks according to claim 7, characterized in that, The formula for obtaining the inertia weight is: ， in, This represents the inertia weight for the p-th iteration of optimization. This represents the maximum inertia weight. This represents the minimum inertia weight. This represents the variance of the score in the p-th iteration optimization. This indicates the preset maximum score variance.

9. The auxiliary lock passage method for inland waterway multi-line ship locks according to claim 7, characterized in that, The formula for obtaining the update vector is: ， in, Let represent the update vector of the particle of the i-th initial gate passage scheduling scheme obtained from the iterative optimization in the p-th round, in the next round of iterative optimization. Let represent the update vector of the particle of the i-th initial gate passage scheduling scheme in the p-th iteration optimization, obtained based on the p-1-th iteration optimization. Let represent the historical best gate scheduling scheme particle corresponding to the i-th initial gate scheduling scheme particle in the p-th round of iterative optimization. Let represent the globally optimal gate passage scheduling scheme particle in the p-th round of iterative optimization. Let represent the i-th initial gate passage scheduling scheme particle in the p-th round of iterative optimization. , All represent learning factors. , Both represent random numbers between 0 and 1.

10. An auxiliary lock passage system for a multi-line inland waterway lock, applied to the auxiliary lock passage method for a multi-line inland waterway lock as described in any one of claims 1 to 9, characterized in that, The system includes: The first acquisition module is used to acquire the transportation information of the vessel, which includes basic attributes and dynamic data. The basic attributes are used to set priority coefficients for the vessel and to set lock passage safety constraints for the vessel. The second acquisition module is used to acquire information about the lock-up area and to acquire a candidate lock-up vessel dataset through the dynamic data and the lock-up area information. The candidate lock-up vessel dataset includes several vessels waiting to pass through the lock. The analysis module is used to perform lock passage scheduling on the candidate lock passage vessel dataset through the lock passage safety constraints to obtain several initial lock passage scheduling scheme particles and obtain the fitness score of the initial lock passage scheduling scheme particles. The execution module is used to perform several rounds of iterative optimization on the initial gate passage scheduling scheme particles based on the fitness score to obtain the final gate passage scheduling scheme particles, and to arrange gate passage based on the final gate passage scheduling scheme particles.

Images