CN116523210A - Maintenance period ship passing gate scheduling optimization method and system based on bald eagle algorithm - Google Patents
Maintenance period ship passing gate scheduling optimization method and system based on bald eagle algorithm Download PDFInfo
- Publication number
- CN116523210A CN116523210A CN202310384235.4A CN202310384235A CN116523210A CN 116523210 A CN116523210 A CN 116523210A CN 202310384235 A CN202310384235 A CN 202310384235A CN 116523210 A CN116523210 A CN 116523210A
- Authority
- CN
- China
- Prior art keywords
- ship
- lock
- scheduling
- algorithm
- population
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000005457 optimization Methods 0.000 title claims abstract description 63
- 241000511338 Haliaeetus leucocephalus Species 0.000 title claims abstract description 49
- 238000012423 maintenance Methods 0.000 title claims abstract description 46
- 230000015654 memory Effects 0.000 claims abstract description 68
- 238000010845 search algorithm Methods 0.000 claims abstract description 23
- 238000009826 distribution Methods 0.000 claims abstract description 20
- 230000007246 mechanism Effects 0.000 claims abstract description 14
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 10
- 230000008569 process Effects 0.000 claims description 25
- 230000003044 adaptive effect Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 241000894007 species Species 0.000 claims description 4
- 230000036621 balding Effects 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000002028 premature Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000005055 memory storage Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008521 reorganization Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000011144 upstream manufacturing Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/30—Flood prevention; Flood or storm water management, e.g. using flood barriers
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a ship passing gate scheduling optimization method and system based on a bald eagle algorithm in a maintenance period, wherein the method comprises the following steps: acquiring ship information, navigation requirements and navigation characteristics of a water transportation junction in a maintenance period; constructing a ship passing-gate scheduling model in a maintenance period by taking the minimum ship waiting time as an optimization target, and establishing constraint conditions met by ship passing-gate scheduling in the maintenance period; respectively solving three sub-problems of a ship lock dispatching lock distribution problem, a lock chamber arrangement problem and a time schedule optimization problem of the ship lock in the overhaul period based on the satisfied constraint conditions to obtain an initial ship lock dispatching scheme in the overhaul period; and performing iterative solution on the initial scheme of the ship brake scheduling in the overhaul period based on a balying search algorithm with elite group guidance, population memory intersection and a self-adaptive mechanism to obtain an optimized scheduling scheme. The invention can solve the problem that the existing method for solving the ship passing scheduling scheme by using the bald hawk search algorithm is easy to sink into local optimum, and makes up the defect of the overhaul period scheduling method.
Description
Technical Field
The invention relates to the technical field of intelligent management of water transportation, in particular to a maintenance period ship passing gate scheduling optimization method and system based on a balding algorithm.
Background
With the increase of the operation time of the water transportation hub, the ship lock needs to be periodically subjected to the operation of the navigation maintenance work to check and replace relevant facility equipment. During the navigation and maintenance period of the ship lock, the hub passing capacity can be greatly reduced, so that the ship traffic is seriously blocked, and a large amount of ship backlog is kept in the lock. With the development of social economy, the transportation demands of the water transportation hub are also rapidly increased, and the contradiction of insufficient passing capacity of the ship lock in the off-air overhaul period is further aggravated. In view of the above, the navigation pressure of the dam area is reduced, the adverse influence of ship lock overhaul on shipping is reduced, the problem of insufficient passing capacity in the ship lock overhaul period is solved or relieved, and the method has important practical significance and application value.
Existing marine traffic organizations have focused on optimization of joint navigation scheduling or gate room scheduling in dam areas. Neglecting the contradiction between ship traffic supply and demand caused by overhaul events, the ship scheduling optimization method during ship lock overhaul is not considered. In addition, in the research of the existing solving method, a deterministic method is mostly used for carrying out accurate solving by relying on a mathematical model, and the method is more suitable for effectively solving the basic problem. The ship lock scheduling problem comprises a plurality of sub-problems, and the problems have high coupling, so that the algorithm design research of the ship lock scheduling problem is developed by adopting the cluster intelligent optimization algorithm such as the balk-eagle search optimization algorithm, and the actual ship lock scheduling operation decision requirement can be met. However, the basic bald hawk search algorithm is easy to fall into a local extremum, the population diversity is rapidly reduced, and the search behavior is redundant and the like. Therefore, the bald eagle search algorithm is optimized and applied to the ship passing scheduling problem in the overhaul period, so that the theory of the optimization problem can be further enriched, and a reference is provided for solving the similar structure problem.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a ship passing optimization method and a ship passing optimization system based on a balk search algorithm, which can solve the problem that the existing method for solving the ship passing scheduling scheme by simply using the balk search algorithm is easy to fall into local optimum, and make up for the defect of the overhaul period scheduling method.
According to a first aspect of the invention, a maintenance period ship passing gate scheduling optimization method based on a bald eagle algorithm is provided, and the method comprises the following steps:
step S1: acquiring ship information, navigation requirements and navigation characteristics of a water transportation junction in a maintenance period;
step S2: constructing a ship passing-gate scheduling model in a maintenance period by taking the minimum ship waiting time as an optimization target; establishing constraint conditions met by ship passing-gate scheduling in a maintenance period according to the scheduling model;
step S3: respectively solving three sub-problems of a ship lock distribution problem, a lock chamber arrangement problem and a time table optimization problem of the ship lock scheduling in the overhaul period based on constraint conditions met by the ship lock scheduling in the overhaul period, so as to obtain an initial scheme of the ship lock scheduling in the overhaul period;
step S4: and performing iterative solution on the initial scheme of the ship brake scheduling in the overhaul period based on a balying search algorithm with elite group guidance, population memory intersection and a self-adaptive mechanism to obtain an optimized scheduling scheme.
On the basis of the technical scheme, the invention can also make the following improvements.
Optionally, the ship information includes: ship-to-anchor time, ship length, ship width, ship name, ship type, ship displacement;
the navigation requirements include: when a ship passes through a lock, in order to ensure the passing safety of two locks, the lock interval between adjacent lock chambers is required to be ensured;
the navigation features of the overhaul period comprise: when one line of the ship lock is shut down for overhauling, the other line implements a strategy of unidirectional operation and timing reversing.
Optionally, in step S2, the constructed overhaul period ship brake passing scheduling model with the minimum ship brake waiting time as an optimization target includes:
step S21: calculating the average time to be turned on of the ship;
step S22: and constructing constraint conditions met by the ship passing-gate scheduling in the overhaul period according to the calculated average ship waiting time.
Optionally, in step S21, the formula for calculating the average time to be shut down of the ship is as follows:
wherein Deltat is the average time of waiting for the ship, deltat pq The waiting time of the ship p in the q stage is represented, N represents the total number of ships, and p and q represent the ship p in the q stage; s denotes a set of scheduling units.
Optionally, in step S3, the solving of the three sub-problems of the ship lock allocation problem, the lock room arrangement problem, and the ship lock planning problem of the ship lock scheduling in the overhaul period includes the following steps:
Step S31: determining the total number of times of opening each ship lock and corresponding scheduling of the times by using the ship lock state and the generated scheduling unit and considering the corresponding requirements and constraints of ship load balance among the three ship locks of the Ge Zhou dam;
step S32: performing brake chamber programming solution based on a Bottom-Left algorithm;
step S33: a ship lock schedule is generated based on the ship lock allocation and the lock room arrangement.
Optionally, in step S33, the algorithm for generating a ship lock schedule based on the ship lock allocation and the lock room arrangement includes:
step S331: set d=1, d ij E { -1,1} represents the transport service direction of (i, j), "1" represents upward, "-1" represents downward;
step S332: setting i=1, i representing the i-th chamber;
step S333: setting j=1, j representing a j-th chamber;
step S334: if r ij =1, deleting the corresponding transport service (i, j) from the current pending queue;
step S335: if i=1, t ij =t b ,r ij =1, i.e. the planned transit time of the transport service (i, j) is equal to the planned period start time, otherwise, proceeding to S337;
step S336: finding the vessel that finally arrives at the anchor from sp (i, j), assuming that the arrival time of this vessel isr ij =1;
Step S337: if j <n i Go to step S338; otherwise, go to step S340;
step S338: setting upr ij =1,j=j+1;
Step S339: if t ij -t b >T, d= -d, return to step S332;
step S340: if i <4, let i=i+1, return to step S333; otherwise, stopping the algorithm and generating a ship passing schedule.
Optionally, in step S4, the iterative solution of the ship brake planning problem in the overhauling period ship brake scheduling model based on the bald eagle search algorithm with elite group guidance, population memory crossover and self-adaptive mechanism includes the following steps:
step S41: determining an individual dimension D, search space range [ lb, ub ]]And the initial value of the parameter for controlling the position change of the flying bald eagle and the spiral track in the algorithm, namely the initial value m 0 ,a 0 And R is 0 ;
Step S42: constructing an initial population according to the number of the ships and the ship gate sequence, and calculating individual fitness values of the bald eagle population by taking the average time to be gate of the ships as a target;
step S43: calculating the fitness value of each bald eagle according to the fitness function, randomly extracting elite individuals from elite groups ranked in the top 5% of the current population fitness value, and marking as Pre;
step S44: calculating Pmeas according to the average position of the bald eagle group distribution in the searching process, updating the position of the bald eagle group, and reserving the current global optimal solution;
Step S45: randomly preserving the results of the individual searching space in the searching process to form a population memory bank with the maximum scale of N;
step S46: randomly selecting different individuals P from the union set of the current species cluster and the population memory set respectively r1 And P r2 Updating the population position, and reserving the optimal solution at the current stage;
step S47: selecting a target individual and the corresponding memory set individuals to perform cross operation dimension by dimension, constructing a new individual, calculating the fitness value of the new individual, comparing the fitness value with the global optimal individual, and reserving the individual with better fitness value;
step S48: judging whether the algorithm reaches the maximum iteration number T, if so, terminating the algorithm to obtain final ship brake passing chamber arrangement, brake order arrangement and ship brake passing schedule information; otherwise, the process goes to step S43.
Optionally, in the step S44, the formula for updating the location in the selected search space is as follows:
in the above-mentioned method, the step of,representing the d-dimensional condition of the ith bald eagle after updating; />Representing elite individuals randomly extracted from elite groups; />The corresponding dimension of the average position of the bald eagle group distribution in the searching process; p (P) i d Indicating the current position of the ith bald eagle d dimension; r is a random number with a value of (0, 1); m represents a parameter controlling the magnitude of the position change.
Optionally, the step S45 of constructing a population memory library includes:
step S351: forming a population memory library by reserving a certain scale of individual search memories, and randomly extracting individuals from the population memory library to participate in the current individual position updating when the individuals perform position updating;
step S352: setting the size of the population memory to be the same as the size of the population, and randomly removing individuals to maintain the size of the population memory library unchanged if the size of the population memory exceeds a threshold value N during each iteration.
According to a second aspect of the present invention, there is provided a bald eagle algorithm based overhaul period ship passing gate scheduling optimization system, comprising:
the acquisition module is used for acquiring ship information, navigation requirements and navigation characteristics of the water transportation junction in a maintenance period;
the construction module is used for constructing a ship passing-gate scheduling model in a maintenance period by taking the minimum ship waiting time as an optimization target; establishing constraint conditions met by ship passing-gate scheduling in a maintenance period according to the scheduling model;
the initial scheme establishing module is used for respectively solving three sub-problems of a ship lock distribution problem, a lock chamber arrangement problem and a time schedule optimization problem of the ship lock scheduling in the overhaul period based on constraint conditions met by the ship lock scheduling in the overhaul period to obtain an initial scheme of the ship lock scheduling in the overhaul period;
And the optimization iteration module is used for carrying out iteration solution on the initial dispatching scheme of the ship pass in the overhaul period based on a balying search algorithm with elite group guidance, population memory intersection and self-adaption mechanism, so as to obtain an optimized dispatching scheme.
The invention has the technical effects and advantages that:
according to the ship passing dispatching optimization method and system based on the balk algorithm, the application field of the traditional balk search algorithm is expanded to the ship dispatching optimization problem through improvement for the first time, the problems that a local optimal value is easy to fall into when a ship passing dispatching scheme is solved by using the basic balk search algorithm, and the search time is too long can be solved, the optimal ship passing dispatching scheme can be obtained stably, and the ship passing efficiency in the overhaul period is improved. In the aspect of algorithm improvement, according to the problem characteristics, an adaptability function based on the shortest average ship waiting time and the largest ship passing number in a maintenance period is respectively constructed and used for optimizing a ship scheduling scheme. In order to prevent optimization from falling into local optimization, elite population guidance is introduced, algorithm optimization is performed by a population memory crossover mechanism, the dominant population is reserved, so that the algorithm is quickly converged, single-body high-strength guidance is reduced, and the limitation of a search stage is overcome. The algorithm self-adaption is realized by improving the core parameters, the parameters are dynamically adjusted according to the problems, and the condition of premature convergence of the algorithm is relieved. And eliminating a redundant search mechanism and reducing invalid searches.
Meanwhile, aiming at the actual problems in the ship passing scheduling process in the overhaul period, under the condition of meeting the ship passing scheduling optimization problem constraints such as safety time interval, timing reversing, brake chamber constraint, brake order balancing and the like, the interest requirements of a management party and a ship party are considered, the average time to wait for a ship is taken as the target, a better ship scheduling scheme can be stably obtained by adopting a self-adaptive bald hawk searching algorithm with elite group guiding and population memory crossing, the ship passing operation efficiency in the overhaul period is improved, a novel method is provided for solving the ship traffic organization optimization problem, and an auxiliary decision is provided for ship passing scheduling in the overhaul period.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
FIG. 1 is a flowchart of a ship passing-gate scheduling optimization method based on a bald eagle search algorithm provided by an embodiment of the invention;
FIG. 2 is a map of an individual code and marine layout map provided in an embodiment of the present invention;
FIG. 3 is a diagram of a comparison table of solving a dynamic scheduling problem according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a dynamically scheduled three gorges north line lock chamber arrangement provided by an embodiment of the present invention;
FIG. 5 is a diagram of a dynamic dispatch gate schedule on a maintenance cycle vessel according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The three gorges-Ge Zhou dam junction engineering plays a role in the construction of the Yangtze river economic zone and makes great contribution to the upstream economic development in the Yangtze river. However, as the operating duration of the three gorges Ge Zhou dam hub increases, the ship lock related facilities and equipment need to be periodically subjected to the off-shore maintenance work. During the navigation and maintenance period of the ship lock, the hub passing capacity can be greatly reduced, so that the ship traffic is seriously blocked, and a large amount of ship backlog is kept in the lock. If the ship passing gate scheduling process in the overhaul period can be optimized, the navigation scheduling efficiency of the three gorges-Ge Zhou dam hub ship is improved, and the method has high practical significance for the economic development around the Yangtze river. Therefore, on the basis of analyzing the current navigation scheduling condition of the three gorges junction, the invention builds the ship passing-gate scheduling model in the overhaul period by taking the minimum ship waiting time as the target, optimizes the bald eagle searching algorithm to carry out model solving, and provides a solution for improving the ship passing-gate efficiency in the overhaul period.
It can be appreciated that based on the defects in the background technology, the embodiment of the invention takes the three gorges-Ge Zhou dam hub as an example, and provides a maintenance period ship passing gate scheduling optimization method based on a balk algorithm, and particularly as shown in fig. 1, the optimization method comprises the following steps:
step S1: acquiring ship information, navigation requirements and navigation characteristics of a water transportation junction in a maintenance period;
in an embodiment of the present invention, the ship information includes: ship-to-anchor time, ship length, ship width, ship name, ship type, ship displacement;
the navigation requirements include: when a ship passes through the three gorges ship lock, in order to ensure the passing safety of two locks, the lock interval between adjacent lock chambers needs to be ensured. For example, typically, the shortest interval for one pass of the three isthmus wire operation (five-stage operation) is 90 minutes, and the shortest interval required for one pass of the three isthmus wire operation is 90 minutes. In the ship lock, the maximum ship speed of the ship lock is 1m/s, and the maximum ship speed between two adjacent ship locks is 0.6m/s. The ship lift can only bear one ship at a time, the time interval for transporting the ship in the same direction is 30 minutes, and the speed of the ship entering and exiting the ship lift can not exceed 1m/s.
The navigation features of the overhaul period comprise: when one line of the three gorges ship lock is shut down for overhauling, the other line implements a strategy of unidirectional operation and timing reversing.
Step S2: constructing a ship passing-gate scheduling model in a maintenance period by taking the minimum ship waiting time as an optimization target; establishing constraint conditions met by ship passing-gate scheduling in a maintenance period according to the scheduling model;
the method for constructing the ship gate-crossing scheduling model in the overhaul period by taking the minimum ship gate-waiting time as an optimization target comprises the following steps of:
step S21: calculating the average time to be turned on of the ship; the calculation formula is as follows:
in the above, deltat is the average time of waiting for the ship, deltat pq The waiting time of the ship p in the q stage is represented, and N represents the total number of ships; p, q represent the vessel p in q phase; s represents a group of scheduling units; s= { (p, q) |1. Ltoreq.p.ltoreq.N, 1. Ltoreq.q.ltoreq.q (p);
step S22: constructing constraint conditions met by the ship passing-gate scheduling in the overhaul period according to the calculated average ship waiting time; the constraint conditions are as follows:
constraint 1:
constraint 1 shows the minimum time interval constraint for two adjacent lock runs of a ship lock. In the reversing period, the Ge Zhouba single-stage ship lock runs in the same direction, and when reversing is carried out, the three gorges five-stage ship lock runs in the opposite direction, and a reverse lock operation is needed to be added.
Wherein r is ij Is a judgment variable r ij E {0,1}, if the transport service (i, j) is running, r ij =1, otherwise r ij =0;t ij Is the planned transit time, t, of the transport service (i, j) ij E r+. i and j respectively represent the jth gate of the ith gate chamber; j-1 represents the last time; t is t i(j-1) Representing a planned crossing time for an ith gate room, a jth-1 gate transport service (i, j);
representing the time interval of use of the lock i; />Representing the switching time of the ship lock i;
d ij representing the transport service direction of the jth gate of the ith gate chamber; d, d i(j-1) Representing the transport service direction of the jth-1 gate of the ith gate chamber; d, d ij E { -1,1} represents the transport service direction of (i, j), "1" represents up, "-1" represents down.
Constraint 2: z ijpq ·t b ≤z ijpq ·t ij ≤z ijpq ·t e ;
Constraint 2 represents a constraint on the opening time that the planned ending time should not be earlier than the earliest starting time of a given period. Wherein Z is ijpq To determine the variables, indicating whether the dispatch unit (p, q) is transported by the transport service (i, j), z ijpq E {0,1}, if the dispatch unit (p, q) is transmitted by the transport service (i, j), Z ijpq =1; otherwise Z ijpq =0;t b ,t e Indicating the beginning and end of a planning period;
constraint 3: z ijpq ·r ij =z ijpq ;
Constraint 3 indicates that each programmed gate must run. Since the Ge Zhou dam has three locks, which are usually operated in the same mode, the distribution of the ship among the three locks should be balanced;
Constraint 4:
constraint 4 represents the lock balance of Ge Zhou dam # 1, # 2, and # 3 locks. Lambda (lambda) 1 、λ 2 、λ 3 Representing the optimal workload balance of the ship locks 1, 2 and 3; n is n i The opening times of the lock chamber i are represented; epsilon represents the imbalance variance, n of the three chambers of the Ge Zhou dam 3 、n 4 、n 5 Ge Zhou dam # 1, # 2, # 3 ship locks are shown.
Constraint 5:
constraint 5 ensures that each dispatch unit is transferred by only one lock chamber. L (L) (p,q) Representing a set of available chambers of transfer vessel p at q-stage; s represents a group of scheduling units; s= { (p, q) |1. Ltoreq.p.ltoreq.N, 1. Ltoreq.q.ltoreq.q (p);
constraint 6:
constraint 6 ensures that the dispatch unit can only be transferred by the available ship lock; l represents a ship lock service set, L = { (i, j) |1 is less than or equal to i and less than or equal to 5, and 1 is less than or equal to j and less than or equal to n i -wherein (i, j) represents the transport service of the j-th lock of lock i; i=1, 2, 3, 4, 5, 6 represent three gorges north line locks, ship lifts, ge Zhou dams 1, 2, 3 locks, three gorges south line locks, respectively;
constraint 7:
constraint 7 indicates that the time of approach of the vessel cannot be earlier than the vessel arrivalFor anchor time.Representing the arrival time of the scheduling units (p, q); t is t ij Representing a planned crossing time of (i, j); />Representing the arrival time of the scheduling units (p, q);
constraint 8: d, d ij ·z ijpq =v ij ·z ijpq (i,j)∈L,(p,q)∈S;
Constraint 8 ensures that the ship heading and the ship lock travel direction are consistent. v ij Representing the shipping direction of the dispatch units (p, q), "1" representing up, "-1" representing down;
constraint 9:
constraint 9 indicates that phase q can only be entered after the previous phase q-1 is completed;
constraint 10:
the restraint 10 ensures that for all vessels entering the lock chamber, their hulls must not exceed the lock chamber boundaries. w (w) pq 、l pq Respectively representing the width and length of the ship p at the q-stage; w (W) ij ,L ij Representing the available width and length of the j-th lock sub-ship lock i; x is x pq E R represents the abscissa value of the ship p placed in the upper right corner of the lock chamber at stage q thereof; y is pq E R represents the vertical coordinate value of the ship p placed in the upper right corner of the lock chamber at stage q thereof;
constraint 11:
constraint 11 ensures that two vessels laid out in the same lock do not overlap, where Z ijmn E {0,1}, if the scheduling unit (m, n) is transmitted by the transport service (i, j)Z is then ijmn =1; otherwise Z ijmn =0;
Z ijmn Indicating whether the dispatch unit (m, n) is transported by the transport service (i, j), Z if the dispatch unit (m, n) is transported by the transport service (i, j) ijmn =1; otherwise Z ijmn =0、Is a step function, w mn 、l mn Representing the width and length of the vessel m at stage n; x is x mn E R represents the abscissa value of the ship m placed in the upper right corner of the lock chamber at stage n thereof; y is mn E R represents the vertical coordinate value of the ship m placed in the upper right corner of the lock chamber at stage n thereof; x is x pq E R represents the abscissa value of the ship m placed in the upper right corner of the lock chamber at stage n thereof; y is pq E R represents the vertical coordinate value of the ship m placed in the upper right corner of the lock chamber at stage n thereof;
step S23: deriving whether dispatch unit is transported by transportation service based on constraint condition Z ijpq And a planned transit time t for the transport service ij 。
Step S3: respectively solving three sub-problems of a ship lock distribution problem, a lock chamber arrangement problem and a ship lock planning problem of ship lock scheduling in the overhaul period to obtain an initial ship lock scheduling scheme in the overhaul period;
further, the method for solving the three sub-problems of the ship lock distribution problem, the lock room arrangement problem and the ship lock planning problem of the ship lock dispatch in the overhaul period respectively comprises the following steps:
step S31: determining the total number of times of opening each ship lock and corresponding scheduling of the times by using the ship lock state and the generated scheduling unit and considering the corresponding requirements and constraints of ship load balance among the three ship locks of the Ge Zhou dam;
specifically, in terms of lock chamber area utilization, the system sets a range of values for lock chamber area utilization in consideration of ship lock demand and lock throughput. If the three gorges lock is more than or equal to 75%, the first lock of the Ge Zhou dam is more than or equal to 65%, and the second lock of the Ge Zhou dam is more than or equal to 75%. In terms of the equalization of the number of passes, The balance proportion of the number of locks reflects the balance of the workload of the locks, and the lock management department hopes that the workload of each lock can be balanced as much as possible. For example, when Ge Zhouba three-line ship locks are operated normally, the ratio of the number of locks should be as close to lambda as possible 1 :λ 2 :λ 3 =16:20:42。
The number of service times for each lock is calculated based on the declared ship type and the total lock area utilization. During the overhaul of the ship lock, the total transit times of the three gorges dam north line ship lock and the three gorges ship lift can be calculated as follows:
wherein,,for a three isthmus north line lock to serve a total number of times in the d direction, d=1, -1, "1" indicating up, "-1" indicating down; />The three gorges ship lift is served for the total times in the direction d; />Applying for the total area of the ship passing through the three gorges dam in the d direction; η (eta) 1 Is an estimated value of the area utilization rate of the ship lock of the north line of three gorges, 75% is taken here; l (L) 1 、W 1 Respectively representing the length and the width of a three gorges north line lock chamber; l (L) 2 、W 2 Respectively representing the length and width of the ship lift.
Considering the work load balance of the Ge Zhou dam 3 locks, the total number of locks for the 3 locks can be calculated as follows:
wherein,,the total number of times the Ge Zhou dam 1#, 2#, 3# locks are serviced in the d direction,/->Applying for the total area of the ship passing through the three gorges dam in the d direction; η (eta) 3 ,η 4 Is the estimated value of the area utilization rate of the No. 1 ship lock and the No. 2 ship lock of the Ge Zhou dam, which are respectively 70 percent and 75 percent, L 3 、W 3 Respectively representing the length and the width of a Ge Zhou dam 1# lock chamber; l (L) 4 、W 4 Respectively representing the length and the width of a Ge Zhou dam No. 2 lock chamber; l (L) 5 、W 5 The length and width of the Ge Zhou dam 3# chamber are shown respectively.
Step S32: performing brake chamber programming solution based on a Bottom-Left algorithm;
it should be noted that the layout problem of the lock chamber can be described by a two-dimensional Packing model, and is an NP-complete problem. The lock chambers are a rigid rectangular space, i.e. the ships must not exceed the lock chamber boundary during ship displacement, and a safe distance should be maintained between the internal ships and between the ships and the inner wall of the lock chamber. In order to simplify the ship arrangement problem, the ship and the ship lock are abstracted into rectangles with different lengths and widths, so that the process of ship arrangement can be regarded as a process of filling a large rectangle with a small rectangle. The specific steps are as follows:
step S321: determining a ship passing sequence;
step S322: determining a ship gate-entering dischargeable point set; the set of points where the lock chambers are available for the ship to lay out is ap (i, j, a),a represents a number of dischargeable points in the gate chamber, and when a=1, ap (i, j, a) = (0, 0) is set;
step S323: selecting a v-th ship (v < N, ship serial number is smaller than available waiting ship) from the ship lock sequence to be arranged;
step S324: v=1 is set.
Step S325: the auxiliary variable a, a=1 is set.
Step S326: based on the ship-discharging constraints in the constraints (10) and (11), it is checked whether v ships can be discharged into the available point ap (i, j, a) of the lock i. If yes, go to step S327; otherwise, step S328 is performed.
Step S327: setting upz ijpq =1. Deletion of the Point +.>And adds two new points ((x) in the set of available points pq ,y pq +w pq ),(x pq +l pq ,y pq ) Let a=a+2). And then goes to S329.
Step S328: if a is<A, a=a+1 is set, and step S326 is returned; otherwise, set z ijpq =0, and then proceeds to step S329.
Step S329: if v < N, v=v+1 is set and step S326 is returned; otherwise, stopping.
Step S33: a ship lock schedule is generated based on the ship lock allocation and the lock room arrangement.
It is to be noted that, it is assumed that a set of scheduling units (p, q) waiting for a gate run (i, j) is sp (i, j); the generation of the ship lock schedule based on the ship lock allocation and lock room arrangement comprises the following steps:
step S331: set d=1, d ij E { -1,1} represents the transport service direction of (i, j), "1" represents upward, "-1" represents downward;
step S332: setting i=1, i representing the i-th chamber;
step S333: setting j=1, j representing a j-th chamber;
step S334: if r ij =1, the corresponding transport service (i, j) is deleted from the current pending queue.
Step S335: if i=1, t ij =t b ,r ij =1, i.e. the planned transit time of the transport service (i, j) is equal to the planned period start time, otherwise, proceeding to S337;
step S336: finding the vessel that finally arrives at the anchor from sp (i, j), assuming that the arrival time of this vessel isr ij =1;
Step S337: if j<n i Go to step S338; otherwise, go to step S33X;
step S338: setting upr ij =1,j=j+1;
Step S339: if t ij -t b >T, d= -d, return to step S332;
step S33X: if i <4, let i=i+1, return to step S333; otherwise, stopping the algorithm and generating a ship passing schedule.
Step S4: and carrying out iterative solution on the ship brake scheduling problem in the overhaul period ship brake scheduling model based on a balying searching algorithm with elite group guiding, population memory crossing and self-adapting mechanisms, so as to obtain an optimized scheduling scheme.
It should be noted that, the elite group guiding strategy can accelerate the convergence rate by the optimal individual guiding update, but is guided by only a single individual, which easily results in excessive learning strength and loss of population diversity. If the optimal solution found by the algorithm in the optimizing process is a local extremum, the population is easy to fall into a local optimal neighborhood, and the problems of premature convergence and the like are caused. For the problem of single information source, the random selection of top-level individuals to replace a single optimal solution is a solution, and in the first stage of position update of the adaptive bald eagle searching algorithm, a learning mode based on elite group guidance is proposed. Firstly, a certain proportion of elite individuals are selected to construct an elite group, then, when each dimension of a target individual is updated, corresponding dimension information of the elite individuals is randomly selected from the elite group to conduct guiding, so that a novel collective guiding mode of updating the individuals by the whole elite group is formed, overhigh learning intensity of a single source is avoided, and inter-population communication is enhanced. The update method is as follows:
Wherein,,representing the d-dimensional condition of the ith bald eagle after updating; />Representing elite individuals randomly extracted from elite groups; />The corresponding dimension of the average position of the bald eagle group distribution in the searching process; />Indicating the current position of the ith bald eagle d dimension; r is a random number with a value of (0, 1); m represents a parameter for controlling the magnitude of the position change, and the range thereof is (1.5, 2).
P re Are elite individuals randomly recombined from elite groups ranked in the top 5% from the current population fitness value. This update mode retains the dominant population, allowing the algorithm to converge quickly while eliminating single-entity high-intensity guidance, overcoming the limitations of the first search stage.
The population memory crossover strategy comprises: (1) population memory construction: when the position of the population of the bald eagle searching optimization algorithm is updated, new individual information is temporarily stored, and if the newly generated bald eagle individual is an invalid search, the new individual information is not stored. The manner of invalidating the rejection, while reducing the cost of running the algorithm, may miss potentially valid information of the individual's cognition. The method comprises the steps of providing a position updating strategy for establishing a population memory to participate in searching, forming a population memory library by reserving a certain scale of individual searching memories, and randomly extracting individuals from the population memory library to participate in current individual position updating when the individuals perform position updating. The population memory library fully reserves the results of the search space of the individuals in the search process, so that each individual has the opportunity to reserve the potential effective information of the individual, and provides additional search information through the difference between the population memory set and the current population set, thereby providing more search possibilities for the individual and relieving the problems of premature convergence and the like. The individual performs position updating under the guidance of the population memory library and the position mean value information in the following mode:
P i,new =P i +x(i)*(P i -P mean )+y(i)*(P r1 -P r2 )
Wherein P is r1 ,P r2 Is a different individual randomly selected from the union of the current species of cluster and the population memory set, respectively. In order to avoid overlarge population memory storage space, the population memory size is set to be the same as the population scale. At each iteration, if the population memory size exceeds a threshold N, individuals are randomly removed to maintain the population memory size unchanged.
(2) Cross reorganization strategy based on population memory; the bald hawk searching optimization algorithm gradually slows down and even stagnates in the later searching period, and is difficult to jump out of a local extremum, and the main reason is that new effective information is lacking, so that the population is in local dilemma. Cross-recombination strategies are fused based on population memory to maintain population diversity. And selecting each target individual and the corresponding individual in the memory set to perform cross operation dimension by dimension, so as to construct a new individual. The operation is defined as follows:
wherein CR is a uniform random number of (0, 1), CR E (0, 1) is the crossing rate, representing the new individualThe proportion of elements is replicated from the population memory set. drand is a random natural number between 1-D. />Representing the value of the ith individual after the d-th dimension crossing,/->And->The values representing the ith individual's d-th dimension selected from the population memory set or the current population, respectively. The PM-C strategy can improve population diversity by memorizing and reserving 'invalid solutions' and introducing difference information, and effectively solves the second limitation of the bald eagle search optimization algorithm. In addition, the third limitation is also resolved because the dive stage search redundancy deletes the third stage.
In addition, the self-adaptive strategy can fully utilize effective information in the evolution process to carry out parameter self-adaptive adjustment, endow individuals with the required capacities in different stages, and improve the searching performance of the algorithm. M, a, R and CR in the improved bald hawk search optimization algorithm have important influence on the group search behavior. In the selection stage, m controls the searching step length in the first stage, when the value of m is larger, the flying step length of the bald hawk is larger, the space can be explored in a large scale, and when m is smaller, the deep development of the local area can be carried out. M in the bald hawk searching optimization algorithm is a fixed value, so that the requirement of the searching process is not matched with parameters, and the searching capability is affected. a and R determine the intensity of the spiral trajectory and the distribution of the individual, respectively. Under the condition of a certain population scale, the larger a is, the more the spiral circulation times are, the larger the searching range is, so that the weaker the depth searching around an individual is; the larger R the more widely distributed the individual, the greater the search range.
In order to effectively improve the searching behavior of the bald eagle group, the embodiment of the invention adopts a self-adaptive adjustment strategy of parameters, so that the parameter adjustment meets the different requirements of searching in different periods of the bald eagle. The parameters m, a and R are perturbed by the following formulas respectively, so that the bald eagle can find the optimal value more quickly. Individuals are converted in dispersion and aggregation in the population space, so that the randomness of searching is ensured, the diversity of the population is maintained, and the algorithm is prevented from falling into a local extremum.
Wherein T represents the current iteration number, T represents the maximum iteration number, m 0 、a 0 、R 0 Representing the initial values of the parameters. In the present embodiment, m is recommended 0 =3,a 0 =10,R 0 =1。
In addition, CR determines individual crossover probabilities, a larger crossover rate does help particles escape the current local dilemma, but at the same time it is possible to increase the difficulty and complexity of the algorithm. Therefore, qin et al propose an adaptive differential algorithm to adaptively take the CR. Herein, based on the adaptive strategy, the CR is updated to makeA CR value is generated for each individual in the current population, and the CR that successfully facilitates location updating in each generation is denoted SCR. And calculating the average value of the SCR according to the recorded result of a period of time. The above steps were repeated with a newly generated normal distribution mean (SCR) and a variance of 0.1, where oCR =0.5. The process can self-learn the proper CR value range, thereby adapting to different problems.
In the embodiment of the invention, the bald eagle searching algorithm is essentially an optimizing algorithm aiming at a real space, and the feasible region where each individual is located is a continuous space, so that the information of the individual still belongs to the continuous space in the whole population optimizing process. The objective of the above schedule optimization problem is to obtain a reasonable scheduling plan, i.e. a scheduling matrix of the brake and the ship, which is a non-numerical optimization problem belonging to discrete space. Therefore, in order to realize the effective mapping of the individual information and the ship arrangement scheme, the invention designs a coding and decoding scheme for realizing the individual information based on the combination of the individual position information and the arrangement scheme on the basis of continuous space coding.
Further, the iterative solution to the ship brake scheduling problem in the overhaul period ship brake scheduling model based on the balying search algorithm with elite group guidance, population memory intersection and self-adaption mechanism comprises the following steps:
step S41: and initializing parameters. Includes determining an individual dimension D, search space range [ lb, ub ]]And the initial values of parameters for controlling the position change of the flying bald eagle and the spiral track in the algorithm, namely the initial values m of m, a and R 0 ,a 0 And R is 0 ;
Step S42: determining the coding and decoding modes of the bald eagle individuals, constructing an initial population according to the number of the ships and the ship gate sequence, and calculating the fitness value of the bald eagle population individuals with the average time to gate of the ship as a target;
further, the coding and decoding modes of the bald eagle individual are as follows:
the location information of any one individual may be expressed as p= (P) 1 ,p 2 ,…,p n ) Wherein n is the number of waiting for the ship to pass through the gate,i is a chamber selected by ship passing. The elements in P are ordered according to the increment sequence, so that a sequence which is equal to P can be obtained i And a corresponding array S. Based on the array S, p i The corresponding original index number constructs another corresponding sequence O, and the sequence O is continuously adjusted along with the updating of the individual position in the algorithm iteration process. And (3) associating the sequence O with a corresponding ship passing sequence in a ship scheduling problem, so as to construct a mapping model between the solution space of the optimization target and the individual position expression.
Based on the above ideas, the individual location information and the order information contained therein are conveniently represented and marked in the form of a two-dimensional array (see table 1 below). Wherein the first dimension of the array is used for recording the position information of an individual in a continuous space (a gate chamber selected by a ship gate), and the second dimension is used for recording the sequence of elements in the position information (the gate time of the selected gate chamber operation). After each time the position of the individual is moved, the positions are reordered to obtain a new order O, so that the adjustment of the ship passing brake times is realized.
TABLE 1 two-dimensional individual design
Individual position (p) i ) | p i1 | p i2 | p i3 | …… | p iN |
Position order (S) | S 1 | S 2 | S 3 | …… | S j |
According to the above coding rule, as shown in fig. 2, an array S based on the element arrangement in P can be obtained. And taking the index number of the position where each element in P is positioned as the number of the ship to be braked, and determining the ship brake entering sequence O according to the array S. Based on the determined loading sequence, a ship passing scheduling scheme A can be determined according to a gate scheduling algorithm and a gate room allocation algorithm.
Step S43: the elite population is determined and elite individuals are selected. According to the fitness of the individuals in the populationSelecting balying with the current population fitness value ranked at the top 5% as elite population, randomly extracting one elite individual from the balying population, and marking as P re ;
Step S44: selecting a search space, and calculating P according to the average position of bald eagle group distribution in the search process mean Updating the position of the bald eagle population, and reserving the global optimal solution at the current stage;
wherein, the formula for updating the position in the selected search space is as follows:
in the above-mentioned method, the step of,representing the d-dimensional condition of the ith bald eagle after updating; />Representing elite individuals randomly extracted from elite groups; />The corresponding dimension of the average position of the bald eagle group distribution in the searching process; />Indicating the current position of the ith bald eagle d dimension; r is a random number with a value of (0, 1); m represents a parameter for controlling the magnitude of the position change, and the range thereof is (1.5, 2).
Step S45: constructing a population memory bank, randomly reserving optimizing records of individuals in the searching process, and forming the maximum-scale N population memory bank;
further, the method for constructing the population memory library comprises the following steps:
step S451: the method comprises the steps of reserving a certain scale of individual search memory to form a population memory library, and randomly extracting individuals from the population memory library to participate in current individual position updating when the individuals perform position updating.
Step S452: in order to avoid overlarge population memory storage space, the population memory size is set to be the same as the population scale. At each iteration, if the population memory size exceeds a threshold N, individuals are randomly removed to maintain the population memory size unchanged.
Step S46: searching space prey, randomly selecting different individuals P from the union of the current species cluster and the group memory cluster r1 And P r2 Updating the population position, and reserving the global optimal solution at the current stage;
further, the search space prey update formula is as follows:
P i,new =P i +x(i)*(P i -P mean )+y(i)*(P r1 -P r2 )
wherein, x (i) and y (i) are in a spiral relation to guide the balk to fly, the values of the balk are (-1, 1), and the polar coordinate model is expressed as follows:
xr(i)=r(i)*sin[θ(i)],yr(i)=r(i)*cos[θ(i)]
θ(i)=a*π*rand
r(i)=θ(i)+R*rand
wherein θ (i) and r (i) are the polar angle and the polar diameter of the spiral equation, respectively; a and R are parameters for controlling the spiral track, and the values are (5, 10) and (0.5, 2) respectively.
Step S47: cross reorganization of population memories, selecting a target individual and corresponding individuals in the memory set to perform cross operation dimension by dimension, constructing a new individual, calculating a fitness value, comparing the fitness value with a global optimal individual, and reserving the individual with a better fitness value;
further, the cross-reorganization operation of the population memory in the step S47 is defined as follows:
each target individual and the corresponding individual in the memory set are selected to carry out cross operation dimension by dimension, and a new individual is constructed:
wherein CR is a uniform random number of (0, 1), CR is the crossing rate, and represents a new individualThe proportion of elements is replicated from the population memory set. d, d rand Is a random natural number between 1-D. />Representing the value of the ith individual after the d-th dimension crossing,/->Andthe values representing the ith individual's d-th dimension selected from the population memory set or the current population, respectively.
Step S48: judging whether the algorithm reaches the maximum iteration number T, if so, terminating the algorithm to obtain final ship brake passing chamber arrangement, brake order arrangement and ship brake passing schedule information; otherwise, the process returns to step S43.
To further illustrate, embodiments of the present invention build cases with three gorges-Ge Zhou dam ladder rungs to the actual vessel navigation data. When congestion is not formed at the upstream and downstream of the hub, the ship passing scheduling is dynamic scheduling, at the moment, the aim of minimizing the ship waiting time is preferably achieved, and the scheduling scheme of the lock chamber is formulated by considering the ship arrival time. The dispatch time range is selected from 20 days of 2020 to 21 days of 2020, 24 hours of ship data to be braked (as shown in table 2).
Table 2 dynamic scheduling case (uplink)
FIG. 3 is a diagram of a comparison table of solving a dynamic scheduling problem according to an embodiment of the present invention; from the results shown in fig. 3, the solution result of the adaptive balying search algorithm is superior to that of the balying search optimization algorithm, the fitness value is rapidly reduced along with the increase of the iteration times, and the average time to be on the ship can be reduced by about 3 hours; from the iteration process, the bald hawk searching optimization algorithm falls into a local extremum before and after 60 generations, because parameters in a position updating formula are fixed, the population cannot be adaptively adjusted along with the optimizing condition, so that the diversity of the population is seriously lost in the later stage, a new solution cannot be searched, and the algorithm is premature. And an adaptive parameter adjustment strategy is added, so that the population searching range is larger in the early iteration stage, the global searching capability of the population is enhanced, and the population can be quickly gathered to the optimal neighborhood in the initial stage. Along with the change of iteration times, the parameter is adaptively adjusted and has random disturbance, so that the algorithm strengthens the local searching capability, maintains population diversity, avoids sinking into a local optimal value, and improves the convergence accuracy of the algorithm. From this, it is known that the adaptive bald hawk search algorithm is effective in solving the problem of ship scheduling.
Taking the result of the dynamic scheduling experiment on the three gorges north line ship lock as an example, a lock chamber arrangement diagram of part of the locks is selected, and the safety distance between ships is included in the ship size, as shown in fig. 4. The three gorges north line uplink dynamic scheduling single-lock plan is shown in table 3, the average lock chamber area utilization rate can reach more than 75%, and the lock chamber resources of the ship are effectively utilized; the average time for waiting for the ship is effectively improved, and the requirement of limiting the time for waiting for the ship is met. The result generated by the whole dispatching algorithm is a dispatching plan of all ship locks in operation in one overhaul period, and fig. 5 is a dispatching plan diagram of the dynamic dispatching locks on the ship in one overhaul period provided by the embodiment of the invention. Since the Ge Zhou dam 3# locks and the three gorges locks are more ordered, both locks only show the entire cyclic arrangement.
Table 3 dynamic scheduling plan profile for three gorges north line ship over-dam
In addition, the embodiment of the invention also provides a maintenance period ship passing gate scheduling optimization system based on a bald eagle algorithm, which comprises the following steps:
the acquisition module is used for acquiring ship information, navigation requirements and navigation characteristics of the water transportation junction in a maintenance period;
the construction module is used for constructing a ship passing-gate scheduling model in a maintenance period by taking the minimum ship waiting time as an optimization target; establishing constraint conditions met by ship passing-gate scheduling in a maintenance period according to the scheduling model;
The initial scheme establishing module is used for respectively solving three sub-problems of a ship lock distribution problem, a lock chamber arrangement problem and a time schedule optimization problem of the ship lock scheduling in the overhaul period based on constraint conditions met by the ship lock scheduling in the overhaul period to obtain an initial scheme of the ship lock scheduling in the overhaul period;
and the optimization iteration module is used for carrying out iteration solution on the overhaul period ship pass scheduling model based on a balying search algorithm with elite group guidance, population memory intersection and a self-adaption mechanism to obtain an optimized scheduling scheme.
It can be understood that the maintenance period ship passing-gate dispatching optimization system based on the balk algorithm provided by the invention corresponds to the maintenance period ship passing-gate dispatching optimization method based on the balk algorithm provided by the foregoing embodiment, and the relevant technical characteristics of the maintenance period ship passing-gate dispatching optimization system based on the balk algorithm can refer to the relevant technical characteristics of the maintenance period ship passing-gate dispatching optimization method based on the balk algorithm, and are not repeated herein.
In summary, the embodiment of the invention has the following technical effects:
(1) A mathematical model of the three gorges Ge Zhou dam step junction ship passing-gate scheduling in the overhaul period is constructed. And (3) constructing a mathematical model of the ship passing-gate scheduling in the overhaul period on the basis of the characteristics of the ship passing-gate scheduling in the overhaul period by analyzing. The model sets the optimization objective to minimize the ship time to lock, and sets several related constraints for ship lock operation and lock room arrangement.
(2) The solution algorithm improves the study. In order to solve the problem of ship passing gate scheduling in the overhaul period, the method is used for analyzing and improving the defects that a basic bald eagle searching algorithm is easy to fall into a local extremum, population diversity is rapidly reduced, searching behavior is redundant and the like. Introducing elite group guidance, group memory crossing and other mechanisms to perform algorithm optimization, realizing algorithm self-adaption by improving core parameters, eliminating redundant searching mechanisms, and providing a self-adaption bald hawk searching algorithm (Adaptive Bald Eagle Search algorithm, ABES) with elite group guidance and group memory crossing.
(3) Model solving and instance verification. The problem of ship lock passing scheduling during the ship lock overhaul period is divided into three sub-problems: the ship lock distribution problem, the lock room arrangement problem and the schedule optimization problem are solved by adopting a self-adaptive balying search algorithm, and experimental results are given. In the experiment, the two dams pass through capacity of one maintenance period, average time to be shut down of the ship, average utilization rate of the area of the lock chamber and ship throughput are compared, and the effectiveness of the model and algorithm is verified.
The improved self-adaptive balying search algorithm and the rest 7 intelligent optimization algorithms are compared with each other in the CEC2013 function test set for optimizing performance, and both experimental results and Wilcoxon symbol rank test results show that the improved algorithm has stronger comprehensive optimizing performance, and the robustness of the algorithm is remarkably improved. In model solving, experimental verification is carried out by using actual historical data of three gorges-Ge Zhou dam step junction navigation, the optimization results of the bald hawk search algorithm before and after improvement are compared, the average time to lock of the ship can be reduced by about 3 hours in one scheduling period, and the average utilization rate of the lock chamber area can reach more than 75 percent.
Finally, it should be noted that: the foregoing description is only illustrative of the preferred embodiments of the present invention, and although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described, or equivalents may be substituted for elements thereof, and any modifications, equivalents, improvements or changes may be made without departing from the spirit and principles of the present invention.
Claims (10)
1. The ship passing gate scheduling optimization method based on the bald eagle algorithm in the overhaul period is characterized by comprising the following steps of:
step S1: acquiring ship information, navigation requirements and navigation characteristics of a water transportation junction in a maintenance period;
step S2: constructing a maintenance period ship passing-gate scheduling model by taking the minimum ship waiting time as an optimization target, and establishing constraint conditions met by maintenance period ship passing-gate scheduling according to the scheduling model;
step S3: respectively solving three sub-problems of a ship lock distribution problem, a lock chamber arrangement problem and a time table optimization problem of the ship lock scheduling in the overhaul period based on constraint conditions met by the ship lock scheduling in the overhaul period, so as to obtain an initial scheme of the ship lock scheduling in the overhaul period;
Step S4: and performing iterative solution on the initial scheme of the ship brake scheduling in the overhaul period based on a balying search algorithm with elite group guidance, population memory intersection and a self-adaptive mechanism to obtain an optimized scheduling scheme.
2. The method for optimizing the dispatch of a ship lock during a maintenance period based on a balk algorithm according to claim 1, wherein the ship information comprises: ship-to-anchor time, ship length, ship width, ship name, ship type, ship displacement;
the navigation requirements include: when a ship passes through a lock, in order to ensure the passing safety of two locks, the lock interval between adjacent lock chambers is required to be ensured;
the navigation features of the overhaul period comprise: when one line of the ship lock is shut down for overhauling, the other line implements a strategy of unidirectional operation and timing reversing.
3. The method for optimizing the ship passing schedule in the overhaul period based on the balk algorithm according to claim 1, wherein in step S2, the constructed ship passing schedule model in the overhaul period with the minimum ship waiting time as the optimization target, and the constraint condition satisfied by the ship passing schedule in the overhaul period is established according to the schedule model, comprising:
step S21: calculating the average time to be turned on of the ship;
Step S22: and constructing constraint conditions met by the ship passing-gate scheduling in the overhaul period according to the calculated average ship waiting time.
4. A method for optimizing ship lock-out schedule in a maintenance period based on a balk algorithm according to claim 3, wherein in step S21, the formula for calculating the average time to lock-out of the ship is as follows:
wherein Deltat is the average time of waiting for the ship, deltat pq The waiting time of the ship p in the q stage is represented, N represents the total number of ships, and p and q represent the ship p in the q stage; s denotes a set of scheduling units.
5. The method for optimizing ship lock dispatch in a maintenance period based on bald eagle algorithm according to claim 1, wherein in step S3, the solving of three sub-problems of a ship lock allocation problem, a lock room arrangement problem and a ship lock plan problem of ship lock dispatch in the maintenance period comprises the following steps:
step S31: determining the total number of times of opening the ship locks and corresponding arrangement of the locks;
step S32: performing brake chamber programming solution based on a Bottom-Left algorithm;
step S33: a ship lock schedule is generated based on the ship lock allocation and the lock room arrangement.
6. The method for optimizing ship lock scheduling during a maintenance period based on the balk algorithm according to claim 5, wherein in step S33, the algorithm for generating a ship lock schedule based on lock allocation and lock room arrangement is as follows:
Step S331: set d=1, d ij E { -1,1} represents the transport service direction of (i, j), "1" represents upward, "-1" represents downward;
step S332: setting i=1, i representing the i-th chamber;
step S333: setting j=1, j representing a j-th chamber;
step S334: if r ij =1, deleting the corresponding transport service (i, j) from the current pending queue;
step S335: if i=1, t ij =t b ,r ij =1, i.e. the planned transit time of the transport service (i, j) is equal to the planned period start time, otherwise, proceeding to S337;
step S336: finding the vessel that finally arrives at the anchor from sp (i, j), assuming that the arrival time of this vessel isr ij =1;
Step S337: if j<n i Go to step S338; otherwise, go to step S340;
step S338: setting up
Step S339: if t ij -t b >T, d= -d, return to step S332;
step S340: if i <4, let i=i+1, return to step S333; otherwise, stopping the algorithm and generating a ship passing schedule.
7. The method for optimizing ship passing scheduling in a overhaul period based on a balk algorithm according to claim 1, wherein in step S4, the iterative solution of the ship passing scheduling problem in the overhaul period ship passing scheduling model based on the balk search algorithm with elite group guidance, population memory crossover and adaptive mechanism comprises:
Step S41: determining an individual dimension D, search space range [ lb, ub ]]And the initial value of the parameter for controlling the position change of the flying bald eagle and the spiral track in the algorithm, namely the initial value m 0 ,a 0 And R is 0 ;
Step S42: constructing an initial population according to the number of the ships and the ship gate sequence, and calculating individual fitness values of the bald eagle population by taking the average time to be gate of the ships as a target;
step S43: calculating the fitness value of each bald eagle according to the fitness function, randomly extracting elite individuals from elite groups ranked in the top 5% of the current population fitness value, and marking as Pre;
step S44: calculating corresponding dimension P according to average position of bald eagle group distribution in searching process mean Updating the position of the bald eagle population, and reserving the current global optimal solution;
step S45: randomly preserving the results of the individual searching space in the searching process to form a population memory bank with the maximum scale of N;
step S46: randomly selecting different individuals P from the union set of the current species cluster and the population memory set respectively r1 And P r2 Updating the population position, and reserving the optimal solution at the current stage;
step S47: selecting a target individual and the corresponding memory set individuals to perform cross operation dimension by dimension, constructing a new individual, calculating the fitness value of the new individual, comparing the fitness value with the global optimal individual, and reserving the individual with better fitness value;
Step S48: judging whether the algorithm reaches the maximum iteration number T, if so, terminating the algorithm to obtain final ship brake passing chamber arrangement, brake order arrangement and ship brake passing schedule information; otherwise, the process goes to step S43.
8. The method for optimizing the ship lock-out schedule in a maintenance period based on the balding algorithm as claimed in claim 7, wherein in said step S44, the corresponding dimension P is calculated based on the average position of the balding group distribution during the search mean The formula of (2) is as follows:
in the above-mentioned method, the step of,representing the d-dimensional condition of the ith bald eagle after updating; />Representing elite individuals randomly extracted from elite groups; />The corresponding dimension of the average position of the bald eagle group distribution in the searching process; />Indicating the current position of the ith bald eagle d dimension; r is a random number with a value of (0, 1); m represents a parameter controlling the magnitude of the position change.
9. The method for optimizing the dispatch of the ship lock during the overhaul period based on the balk algorithm according to claim 8, wherein the step S45 of constructing the population memory library comprises the following steps:
step S351: forming a population memory library by reserving a certain scale of individual search memories, and randomly extracting individuals from the population memory library to participate in the current individual position updating when the individuals perform position updating;
Step S352: setting the size of the population memory to be the same as the size of the population, and randomly removing individuals to maintain the size of the population memory library unchanged if the size of the population memory exceeds a threshold value N during each iteration.
10. The utility model provides a maintenance period boats and ships gap dispatch optimizing system based on balk algorithm which characterized in that includes:
the acquisition module is used for acquiring ship information, navigation requirements and navigation characteristics of the water transportation junction in a maintenance period;
the construction module is used for constructing a ship passing-gate scheduling model in a maintenance period by taking the minimum ship waiting time as an optimization target; establishing constraint conditions met by ship passing-gate scheduling in a maintenance period according to the scheduling model;
the initial scheme establishing module is used for respectively solving three sub-problems of a ship lock distribution problem, a lock chamber arrangement problem and a time schedule optimization problem of the ship lock scheduling in the overhaul period based on constraint conditions met by the ship lock scheduling in the overhaul period to obtain an initial scheme of the ship lock scheduling in the overhaul period;
and the optimization iteration module is used for carrying out iteration solution on the initial dispatching scheme of the ship pass in the overhaul period based on a balying search algorithm with elite group guidance, population memory intersection and self-adaption mechanism, so as to obtain an optimized dispatching scheme.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310384235.4A CN116523210A (en) | 2023-04-11 | 2023-04-11 | Maintenance period ship passing gate scheduling optimization method and system based on bald eagle algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310384235.4A CN116523210A (en) | 2023-04-11 | 2023-04-11 | Maintenance period ship passing gate scheduling optimization method and system based on bald eagle algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116523210A true CN116523210A (en) | 2023-08-01 |
Family
ID=87395079
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310384235.4A Pending CN116523210A (en) | 2023-04-11 | 2023-04-11 | Maintenance period ship passing gate scheduling optimization method and system based on bald eagle algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116523210A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117151285A (en) * | 2023-08-29 | 2023-12-01 | 淮阴工学院 | Runoff forecasting method based on multi-element attention space-time diagram convolutional network |
-
2023
- 2023-04-11 CN CN202310384235.4A patent/CN116523210A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117151285A (en) * | 2023-08-29 | 2023-12-01 | 淮阴工学院 | Runoff forecasting method based on multi-element attention space-time diagram convolutional network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhong et al. | Multi-AGV scheduling for conflict-free path planning in automated container terminals | |
CN105787595B (en) | Emergency logistics transfer station site selection method based on improved ant colony algorithm | |
CN110222938B (en) | Short-term peak-load regulation scheduling collaborative optimization method and system for cascade hydropower station group | |
CN116523210A (en) | Maintenance period ship passing gate scheduling optimization method and system based on bald eagle algorithm | |
CN112836974B (en) | Dynamic scheduling method for multiple field bridges between boxes based on DQN and MCTS | |
CN112488452B (en) | Energy system management multi-time scale optimal decision method based on deep reinforcement learning | |
CN109710404A (en) | Method for scheduling task in distributed system | |
CN112149990B (en) | Fuzzy supply and demand matching method based on prediction | |
CN104077634A (en) | Active-reactive type dynamic project scheduling method based on multi-objective optimization | |
CN108537370B (en) | Super-large basin hydropower station group optimal scheduling method based on hybrid intelligent dimensionality reduction algorithm | |
CN114648165A (en) | Multi-agent game-based optimal scheduling method for multi-heat-source heat supply system | |
CN115271140A (en) | Dynamic programming based container ship stowage box position reservation method | |
CN114819622A (en) | Carrier-based aircraft tripping operation scheduling method, device, equipment and medium | |
CN109002878A (en) | A kind of GA Optimized BP Neural Network | |
CN117595392A (en) | Power distribution network joint optimization method and system considering light Fu Xiaona and light storage and charge configuration | |
CN116300755A (en) | Double-layer optimal scheduling method and device for heat storage-containing heating system based on MPC | |
Dong et al. | QCs scheduling scheme of genetic algorithm (GA) and improved firefly algorithm (FA) | |
CN115755801A (en) | SQP-CS-based ship building workshop process optimization method and system | |
CN104268639A (en) | Ship piloting and scheduling method based on forest model and improved genetic algorithm | |
CN115271254A (en) | Short-term wind power prediction method for optimizing extreme learning machine based on gull algorithm | |
CN108053084A (en) | A kind of short-term micro-grid load Forecasting Methodology based on ABC algorithms | |
CN112785022A (en) | Method and system for excavating electric energy substitution potential | |
CN117111620B (en) | Autonomous decision-making method for task allocation of heterogeneous unmanned system | |
TWI847733B (en) | Method and system for optimizing power generation unit scheduling in an electrical power system using a deep q-network | |
Sun et al. | Multi-Agent Deep Deterministic Policy Gradient Algorithm Based on Classification Experience Replay |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |