CN111325409B - Method and system for site selection of battery replacement station and route planning of hybrid fleet - Google Patents
Method and system for site selection of battery replacement station and route planning of hybrid fleet Download PDFInfo
- Publication number
- CN111325409B CN111325409B CN202010158192.4A CN202010158192A CN111325409B CN 111325409 B CN111325409 B CN 111325409B CN 202010158192 A CN202010158192 A CN 202010158192A CN 111325409 B CN111325409 B CN 111325409B
- Authority
- CN
- China
- Prior art keywords
- operator
- node
- vehicle
- electric vehicle
- cost
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 239000000446 fuel Substances 0.000 claims abstract description 57
- 230000008859 change Effects 0.000 claims description 24
- 230000009977 dual effect Effects 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 16
- 230000009467 reduction Effects 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000007689 inspection Methods 0.000 claims description 8
- 238000013138 pruning Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 239000000295 fuel oil Substances 0.000 claims description 3
- 230000002040 relaxant effect Effects 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 abstract description 6
- 238000013461 design Methods 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 description 5
- 238000000354 decomposition reaction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3407—Route searching; Route guidance specially adapted for specific applications
- G01C21/3423—Multimodal routing, i.e. combining two or more modes of transportation, where the modes can be any of, e.g. driving, walking, cycling, public transport
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3407—Route searching; Route guidance specially adapted for specific applications
- G01C21/343—Calculating itineraries, i.e. routes leading from a starting point to a series of categorical destinations using a global route restraint, round trips, touristic trips
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
- G01C21/3469—Fuel consumption; Energy use; Emission aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0835—Relationships between shipper or supplier and carriers
- G06Q10/08355—Routing methods
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Automation & Control Theory (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Medical Informatics (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of logistics transportation, in particular to a method and a system for site selection of a power station and route planning of a mixed fleet. The method comprises the following steps: acquiring alternative data information; constructing an integer planning model for combined decision of power station selection and electric vehicle and fuel vehicle mixed fleet delivery; reconstructing the integer programming model into a main problem model and a sub problem model; the invention designs an accurate branch pricing algorithm with a self-adaptive selection mechanism aiming at the model, and realizes self-adaptive adjustment of operators aiming at different enterprise distribution scenes by designing seven accelerated solving operators so as to accelerate the solving speed. And simultaneously, expanding a complete search solution space by adopting an accurate tag to obtain an optimal solution. The invention can quickly provide the optimal delivery scheme and the optimal station building scheme for the large-scale mixed fleet delivery system of different enterprises under the condition of self-building of the power station, thereby realizing the real-time optimal scheduling of the enterprises.
Description
Technical Field
The invention relates to the technical field of logistics transportation, in particular to a method and a system for site selection of a battery replacement station and route planning of a hybrid fleet.
Background
With the development of new energy technology and the coming of relevant national guide policies, compared with the traditional fuel vehicles, the electric vehicle has greater advantages in aspects of financial subsidies, license plate restriction, energy conservation, environmental protection, operation cost and the like, so that more and more logistics enterprises begin to popularize the electric logistics vehicles in urban distribution. However, the electric logistics vehicles need to be charged during distribution due to the limitation of the driving range. However, the construction of the charging infrastructure in China is still in a starting stage at present, the network of the charging infrastructure is not perfect, and the charging time of the battery is generally long (the common slow charging technology generally takes 5-8 hours for full charging, and the quick charging technology generally takes 1-2 hours for full charging). Therefore, in actual operation of enterprises, pure electric vehicles do not completely replace fuel vehicles, and a mixed fleet of electric vehicles and fuel vehicles is built to complete distribution. Meanwhile, the distribution efficiency of the mixed fleet is improved by building a power exchanging station (the power exchanging time is extremely short and is generally equivalent to the fuel replenishing time of a fuel vehicle).
The electric vehicle and fuel vehicle mixed distribution system under the condition of self-building a power station is completely different from the traditional fuel vehicle distribution system. Under the condition of considering the electricity changing, the addition of the electric vehicle and the electricity changing station changes the input parameters of the original fuel vehicle optimal distribution system, if the path planning is not proper, the economic advantages of the electric vehicle on the operation cost cannot be fully exerted, and the distribution service can be delayed or interrupted due to the limitation of the endurance mileage of the electric vehicle.
The combined decision problem of site selection and path combination of the electric vehicle and the fuel vehicle hybrid fleet based on the battery replacement mode is a complex combined optimization problem, integrates and considers the facility site selection of an enterprise strategic level and the path planning of an operation level, and is the combination of the site selection problem of the battery replacement facility and the distribution path planning problem of the electric vehicle and the fuel vehicle hybrid fleet. Compared with a single optimization site selection model or a path model, the joint decision can be used for comprehensively and integrally optimizing the whole distribution system, so that the distribution cost is optimized to the maximum extent, and meanwhile, the whole distribution system is more complex and more difficult to solve.
The hybrid fleet path planning problem takes into account two distinct vehicles-electric vehicles and fuel vehicles. The electric vehicle has low running cost, but has the limitation conditions of short driving range, battery charging and replacement in the distribution process and the like. The fuel-oil vehicle has longer driving distance, but the driving cost is higher. Meanwhile, the configuration parameters of the two vehicle types in the aspects of load and the like are different. Therefore, the distribution system needs to synchronously coordinate the path distribution of the two vehicles, so that the distribution system has more limiting factors, and the modeling and the solving are more complicated compared with a pure fuel oil fleet or a pure electric vehicle fleet which is independently configured. At present, methods for solving route planning of a hybrid fleet are few, most methods are solved based on traditional heuristic algorithms such as genetic algorithm and neighborhood search, and solving quality and solving stability are not ideal.
The electric vehicle charging and battery replacing facility site selection problem aims to decide the optimal energy supplement position and the optimal number of the station building in a distribution network, so that the electric vehicle with the least station building cost can serve the most, and the bypassing cost of the electric vehicle can be reduced. Most of the existing methods are independent decisions of charging station site selection decisions and fleet delivery path planning, namely, path planning is not considered, and site selection positions are decided only from factors such as service distance, station building cost, station building quantity and the like. And after the facility station building position is determined, independently planning the fleet delivery path. The method breaks the complete distribution system and is difficult to obtain a high-quality decision scheme.
The existing hybrid fleet distribution method under the battery replacement mode mainly has two defects: 1. the site selection of the power change station and the route planning of the hybrid fleet are not integrated for decision making, so that the obtained decision making scheme has the defects of improper site selection of the power change station and non-optimized delivery route, thereby causing high delivery cost of enterprises and reducing the profit of the enterprises. 2. In the aspect of algorithm design, although the currently designed precise algorithm can obtain an optimal solution, the solving efficiency is low, and particularly when a large-scale complex distribution system such as a site selection-path integration consideration is solved, a feasible solution scheme cannot be provided even within an acceptable time range, so that timely distribution of enterprises is influenced, and the customer satisfaction is reduced.
Disclosure of Invention
The invention aims to provide a method and a system for site selection of a power conversion station and route planning of a hybrid fleet, so as to solve the problems. In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
on one hand, the application provides a method for site selection of a power station and route planning of a hybrid fleet, and the method comprises the following steps:
acquiring alternative data information; constructing an integer planning model for combined decision of power station selection and electric vehicle and fuel vehicle mixed fleet delivery; reconstructing the integer programming model into a main problem model and a sub problem model by adopting a Dantzig-Wolfe decomposition method; and solving the main problem model and the sub problem model to obtain a distribution scheme with the lowest total cost and a power station site selection scheme.
Optionally, the alternative data information includes: the number and address of distribution centers; the number of the electric vehicles, the distribution cost of the electric vehicles per kilometer and the maximum driving mileage of the electric vehicles; the number of fuel vehicles, the delivery cost per kilometer of fuel vehicles; the location, demand and service time window of each waiting delivery customer; a location of each standby battery swap station.
Optionally, the integer programming model comprises the following formula:
the formula (4-1) in the model is an objective function and consists of three items of cost, wherein the first item is the construction cost of the power exchanging station, the second item is the transportation cost of the electric vehicle, and the third item is the transportation cost of the fuel vehicle; the formula (4-2) shows that each customer point and the power change station can only be accessed by one vehicle at the same time; the formula (4-3) restricts the available number of the delivery vehicles not to exceed the scale of the fleet; the formula (4-4) indicates that the started power change station can provide the power change service; equations (4-5) ensure conservation of each point stream; formulas (4-6) and (4-7) show that the total distribution task amount of the electric vehicle and the fuel vehicle does not exceed the maximum loading capacity; the formula (4-8) represents the time logical relation of the vehicle from the point i to the point j; the formula (4-9) represents the time logical relation of the electric vehicle from the power changing station f to the point j; equations (4-10) satisfy the service time window constraints for all points; the formula (4-11) satisfies the relation of electric quantity consumption when the electric vehicle runs from the point i to the point j; the formulas (4-12) represent that the electric vehicle has constant electric quantity when arriving at and leaving a customer point; formula (4-13) shows that the electric vehicle starts from a distribution center full-charge state and changes a full-charge battery after arriving at a battery replacement station; the formula (4-14) restricts the electric quantity of the electric vehicle at any point to be larger than zero, and ensures that the electric vehicle can return to a distribution center; equations (4-15) are 0-1 decision variable constraints.
Optionally, the main problem model comprises the following formula:
in equations (4-16) to (4-20), p is the vehicle travel path, Ω is the set of all feasible paths that satisfy the constraint,p belongs to omega; the vehicle route combination (p, k) indicates that the route p is delivered by the vehicle k;a variable of 0-1, indicating whether a vehicle path combination (p, k) is used in the final solution;a variable 0-1 indicating whether the vehicle path combination (p, k) passes through the arc (i, j);a variable of 0-1, indicating whether the path p passes through the point i;representing the cost of a vehicle path combination (p, k), including the cost of path transportation and the cost of station construction, λ k Taking λ according to type of vehicle k e Or λ c SaidCan be calculated by the following formula:
further, the subproblem model comprises the following formula:
in the formulas (4 to 21) and (4 to 22), pi = { pi = { [ pi ]) i ,π e ,π c The dual variables are respectively corresponding to formulas (4-17), (4-18) and (4-19) in the main problem model;representing the number of tests for each feasible path of Ω in the master problem model.
Further, the solving of the main problem model and the sub problem model includes the following steps:
s41, initializing node information, setting the global upper bound as positive infinity, setting the value of an active node set to be the same as that of a root node, and initializing operator pool information;
s42, judging the termination of calculation, namely judging whether the active node set is empty, if so, determining that the current upper bound is the optimal solution, and terminating the calculation, otherwise, turning to the step S43;
s43, selecting a node from the active node set according to the node selection strategy, and deleting the selected node from the active node set;
s44, solving the node, and if the node is not solved, turning to the step S42; otherwise, the linear relaxation optimal solution of the node is recorded as the local lower bound of the node;
s45, pruning, and if the local lower bound is greater than or equal to the global upper bound, turning to the step S42; otherwise, further judging whether the optimal relaxation solution obtained by the node is a score, if so, turning to the step S46; if the number of the nodes is an integer, updating the global upper bound into a local lower bound, deleting the nodes which are not smaller than the current global upper bound in the active node set, and turning to the step S42;
and S46, branching, selecting a branch variable from the relaxation optimal solution of the current node according to a branch strategy to divide a solution space to obtain sub-nodes, adding the new sub-nodes into an active node set, and turning to the step S42.
Further, the step S44 includes the following steps:
s4401, constructing a main problem model;
s4402, solving a main problem model;
s4403, transferring dual variables;
s4404, resetting the iteration times recorded in the counter to zero, and updating operator information;
s4405, randomly selecting operators from the operator pool, and solving a subproblem model;
s4406, judging whether a column with a negative check number is generated, if so, entering step S4407, and if not, entering step S4409;
s4407, adding 1 to the iteration number recorded in the counter, adding a score to the operator score, adding a column with a negative inspection number to the main problem model, and simultaneously restoring the operator pool;
s4408, judging the relationship between the iteration times and the small iteration cycles, and if the iteration times are smaller than the small iteration cycles, entering the step S4405; otherwise, go to step S4404;
s4409, adding 1 to the iteration number, and deleting the operator selected in the step S4405 in an operator pool;
s4410, judging whether the operator pool is empty, and if the operator pool is empty, entering the step S4411; otherwise, go to step S4405;
s4411, calling an accurate label extension method to solve a sub-problem model;
s4412, judging whether a column with a negative detection number is generated; if a column with a negative check number is generated, adding the column with the negative check number into the main problem model, and entering the step S4402; otherwise, the column generation iteration is terminated.
Further, the operator pool comprises the following 7 operators:
the first operator is used for expanding the current state of the solution to 2 stages in a greedy algorithm to obtain a second-order greedy operator;
and the second operator, namely the current node of the label to be expanded is expanded to the node of the negative cost arc only, and meanwhile, the accurate governing rule is kept unchanged.
And a third operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the accurate governing rule is kept unchanged.
And the fourth operator is used for expanding the current node of the label to be expanded only to the node of the negative cost arc, and relaxing the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule.
And a fifth operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule are relaxed.
And a sixth operator, wherein the current node of the label to be expanded is only expanded to the node of the negative cost arc, and simultaneously, the constraint of the load W of the electric vehicle and the fuel vehicle in the governing rule is relaxed.
And a seventh operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and simultaneously, the constraint of the electric vehicle and the fuel vehicle load W in the governing rule is relaxed.
Further, in step S44, the weight of each operator is updated every time Ct iteration calculation is performed, where Ct is the number of operators, and the operator weight may be calculated by the following formula:
in the formula (4-24), theta is a weight parameter and is a value theta epsilon [0,1 ]];S i Scoring the operator according to S n And updating, when the selected operator fails to solve the current subproblem S n =0, otherwise the operator is scored according to the following case:
when the operator selected for the 1 st time in the small period can solve the subproblem model, S n =10;
When the operator selected at the front cannot solve the sub-problem model and the operator selected at this time can solve the sub-problem model and is not the last operator, S n =20;
When the operator selected at the front cannot solve the sub-problem model until the operator selected at the last time can solve the problem, S n =30。
In another aspect, the present invention provides a system for site selection of a swapping station and path planning of a hybrid fleet, comprising:
the data acquisition module is used for acquiring alternative data information;
the model construction module is used for constructing an integer planning model for combined decision-making of power station selection and electric vehicle and fuel vehicle mixed fleet delivery;
the model reconstruction module is used for reconstructing the integer programming model into a main problem model and a sub problem model by adopting a Dantzig-Wolfe decomposition method;
and the calculation module is used for solving the main problem model and the sub problem model to obtain a distribution and power station site selection scheme with the lowest total cost.
The invention has the beneficial effects that:
the invention constructs an integer planning model for combined decision of the site selection of the power station and the path of the hybrid fleet, and can autonomously evaluate the solving performance of each operator through seven accelerating operators and a self-adaptive selection mechanism design, thereby flexibly calling the operators according to different enterprise delivery scenes, accelerating the solving speed, and particularly for a large-scale complex delivery system, rapidly obtaining a delivery scheme and realizing the real-time scheduling of enterprises. The invention can simultaneously decide the station building position of the power exchanging station and the path distribution of the electric vehicle and the fuel vehicle in the distribution system, and simultaneously can obtain the optimal path scheme and the station building scheme of the mixed distribution fleet under the condition of self-building the power exchanging station of an enterprise due to the properties of designed branch delimitation and accurate label expansion for completely searching the solution space.
The invention improves the solution of the sub-problem from the traditional label extension method into the solution of a plurality of acceleration operator combinations; the self-adaptive selection mechanism designed for the calling of the multiple solving operators can enable the performance of the operators to be automatically judged through the score condition of the operators in the algorithm, so that the optimal operators are preferentially selected for solving.
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 the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a method for site selection of a swapping station and route planning of a hybrid fleet according to an embodiment of the present invention;
FIG. 2 is a table of corresponding meanings of parameters or variables according to the embodiment of the present invention;
FIG. 3 is a flowchart illustrating the step S4 according to the embodiment of the present invention;
FIG. 4 is a flowchart illustrating the step S44 according to the embodiment of the present invention;
FIG. 5 is a schematic diagram of a second-order greedy algorithm according to an embodiment of the present invention;
fig. 6 is a schematic diagram of positions of a warehouse, a customer and a swap station according to an embodiment of the present invention;
FIG. 7 is a table of distances between points as described in the example of the present invention;
FIG. 8 is a first sub-problem model solution set list according to an embodiment of the present invention;
FIG. 9 is a second sub-problem model solution set list according to an embodiment of the present invention;
FIG. 10 is a third problem sub-model solution set list according to an embodiment of the present invention;
FIG. 11 is a fourth sub-problem model solution set list according to an embodiment of the present invention;
FIG. 12 is a schematic illustration of a pre-optimization delivery schedule in an embodiment of the present invention;
FIG. 13 is a schematic diagram of an optimized delivery scheme according to an embodiment of the present invention;
fig. 14 is a block diagram of a site selection and hybrid fleet path planning system for a power swapping station according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not construed as indicating or implying relative importance.
On one hand, as shown in fig. 1, the embodiment provides a method for site selection of a power conversion station and planning of a mixed fleet path, where the method includes step S1, step S2, step S3, and step S4.
S1, acquiring alternative data information;
s2, constructing an integer planning model for a combined decision of power station location selection and electric vehicle and fuel vehicle hybrid fleet delivery;
s3, reconstructing the integer programming model into a main problem model and a sub problem model by adopting a Dantzig-Wolfe decomposition method;
and S4, solving the main problem model and the sub problem model to obtain a distribution scheme and a power station site selection scheme with the lowest total cost.
Optionally, the alternative data information includes: the number and address of distribution centers; the number of the electric vehicles, the distribution cost of the electric vehicles per kilometer and the maximum driving mileage of the electric vehicles; the number of fuel vehicles, the delivery cost per kilometer of fuel vehicles; the location, demand and service time window for each customer waiting to be delivered; and the position of each alternative power conversion station.
Suppose that a distribution system of a logistics enterprise has a distribution center and a mixed fleet of electric vehicles and fuel vehicles, n customers with known positions and demands wait for distribution, and each customer has a service time window constraint that the vehicle needs to service the customers within a specified time period, needs to wait early and does not allow late arrival. Each vehicle is required to return to the starting point after the delivery is completed, each vehicle can only serve one path, and each customer can only be served by one vehicle. Because the electric vehicle has short driving mileage, part of positions can be selected from a given electric power station candidate set to build the electric power stations, and the station building cost is distributed to the calculation of the single delivery cost from the operation perspective. The site selection of the power station and the distribution path planning of the hybrid fleet need to be jointly decided, so that the distribution system meets the distribution requirements of all customers within the specified time, and the goal of lowest power station construction cost and hybrid fleet distribution cost is achieved.
Optionally, the integer programming model comprises the following formula:
the formula (4-1) in the model is an objective function and consists of three items of cost, wherein the first item is the construction cost of the power exchanging station, the second item is the transportation cost of the electric vehicle, and the third item is the transportation cost of the fuel vehicle; the formula (4-2) shows that each customer point and the power change station can only be accessed by one vehicle at the same time; the formula (4-3) restricts the available number of the delivery vehicles not to exceed the scale of the fleet; the formula (4-4) indicates that the started power change station can provide the power change service; equations (4-5) ensure that each point stream is conserved; formulas (4-6) and (4-7) show that the total distribution task amount of the electric vehicle and the fuel vehicle does not exceed the maximum loading capacity; the formula (4-8) represents the time logical relation of the vehicle from the point i to the point j; the formula (4-9) represents the time logical relation of the electric vehicle from the power change station f to the point j; equations (4-10) satisfy the service time window constraints for all points; the formula (4-11) satisfies the relation of electric quantity consumption when the electric vehicle runs from the point i to the point j; the formulas (4-12) represent that the electric vehicle has constant electric quantity when arriving at and leaving a customer point; formula (4-13) shows that the electric vehicle starts from the full-charge state of the distribution center and changes the full-charge battery after reaching the power change station; the formula (4-14) restricts the electric quantity of the electric vehicle at any point to be larger than zero, and ensures that the electric vehicle can return to a distribution center; equations (4-15) are 0-1 decision variable constraints.
The meanings of the parameters or variables in the formulas (4-1) to (4-15) are shown in the comparison table in FIG. 2.
Optionally, the main problem model comprises the following formula:
in the equations (4-16) to (4-20), p is a vehicle travel pathOmega is a set of all feasible paths meeting the constraint, and p belongs to omega; the vehicle route combination (p, k) indicates that the route p is delivered by the vehicle k;a variable of 0-1, indicating whether a vehicle path combination (p, k) is employed in the final solution;a variable 0-1 indicating whether the vehicle path combination (p, k) passes through the arc (i, j);a variable of 0-1, indicating whether the path p passes through the point i;representing the cost of a vehicle path combination (p, k), including the cost of path transportation and the cost of station construction, λ k Taking λ according to the type of vehicle k e Or λ c SaidCan be calculated by the following formula:
according to the principle of the simplex method, when the integer programming model solves the minimum target, a column with negative inspection number in the solution space needs to be found, and the column is added into the main problem model for continuous iteration until all the inspection numbers are positive.
Further, the subproblem model comprises the following formula:
in the formulas (4 to 21) and (4 to 22), pi = { pi = { [ pi ]) i ,π e ,π c The dual variables are respectively corresponding to formulas (4-17), (4-18) and (4-19) in the main problem model;representing the number of tests for each feasible path of Ω in the master problem model.
Further, as shown in fig. 3, the step S4 may include the following steps:
s41, initializing node information, setting the global upper bound as positive infinity, setting the value of an active node set to be the same as that of a root node, and initializing operator pool information;
s42, judging the termination of calculation, namely judging whether the active node set is empty, if so, determining that the current upper bound is the optimal solution, and terminating the calculation, otherwise, turning to the step S43;
s43, selecting a node from the active node set according to the node selection strategy, and deleting the selected node from the active node set;
s44, solving the node, and if the node is not solved, turning to the step S42; otherwise, recording the linear relaxation optimal solution of the node as a local lower bound of the node;
s45, pruning, and if the local lower bound is greater than or equal to the global upper bound, turning to the step S42; otherwise, further judging whether the optimal relaxation solution is a score, and if the optimal relaxation solution is a score, turning to the step S46; if the number of the nodes is an integer, updating the global upper bound into a local lower bound, deleting the nodes which are not smaller than the current global upper bound in the active node set, and turning to the step S42;
s46, branching, selecting a branch variable from the relaxation optimal solution of the current node according to a branch strategy to divide a solution space to obtain child nodes, adding the new child nodes into an active node set, and turning to the step S42.
In fig. 3, GUB is the global upper bound, that is, the current optimal integer solution objective function value of the original problem, and the smaller GUB is, the more effective pruning and fast operation can be promoted; GLB is the global lower bound, i.e. root node linear relaxation solutionThe objective function value of (1); LLB i The local lower bound is the objective function value of the linear relaxation solution of the current child node i; i is an active node set, namely a node set to be solved; the node is a partial solution space subproblem of the original integer programming problem and only comprises a partial solution space of a feasible domain of the original problem.
Further, as shown in fig. 4, the step S44 includes the following steps:
s4401, constructing a main problem model;
s4402, solving a main problem model;
s4403, transferring dual variables;
s4404, resetting the iteration times recorded in the counter to zero, and updating operator information;
s4405, randomly selecting operators from the operator pool, and solving a subproblem model;
s4406, judging whether a column with a negative check number is generated, if so, entering step S4407, and if not, entering step S4409;
s4407, adding 1 to the iteration number recorded in the counter, adding a score to the operator score, adding a column with a negative inspection number to the main problem model, and simultaneously restoring the operator pool;
s4408, judging the relationship between the iteration times and the small iteration cycles, and if the iteration times are smaller than the small iteration cycles, entering the step S4405; otherwise, go to step S4404;
s4409, adding 1 to the iteration number, and deleting the operator selected in the step S4405 in an operator pool;
s4410, judging whether the operator pool is empty, and if the operator pool is empty, entering the step S4411; otherwise, go to step S4405;
s4411, calling an accurate label extension method to solve a sub-problem model;
s4412, judging whether a column with a negative detection number is generated; if a column with a negative check number is generated, adding the column with the negative check number to the main problem model, and proceeding to step S4402; otherwise, the column generation iteration is terminated.
In step S41 initialization, the operator number H is given i Initial weight W i And the like and stored in the operator pool. When entering a column generation process with a self-adaptive selection mechanism, the Count is used as a counter to record the iteration times, the iteration times are recorded once each time a subproblem is solved, and the selected times N of operators are recorded at the same time i Setting a small iteration cycle Ct, and updating the weight W of each operator every time the Ct is iterated and calculated i And making each operator selected for a number of times N i And resetting the counter to 0 again, and restoring the number of the operators in the operator pool to the initial number of the operators. Initial weight W of given operator 0 And =10, in the subsequent algorithm execution, the weight update of each operator is influenced by the selection times of the operators and whether the current subproblem can be solved during the selection.
In FIG. 4, RMP is the main problem model; count is the number of iterations recorded by the counter; hi is the selected operator; s i Scoring the operator; ct is the number of operators; SP is a sub-problem model.
Further, the operator pool comprises the following 7 operators:
and the first operator H1 expands the current state of the solution to 2 stages in the greedy algorithm to obtain a second-order greedy operator, as shown in FIG. 5, the current state of the solution can be expanded to k stages in the greedy algorithm by using the strategy that the current solution state in the k-regret value algorithm is influenced by the multiple stages for reference, and the second-order greedy operator is obtained when the value of k is 2. In the second-order greedy operator, the current state of the solution needs to be updated in two stages according to the current node after expansion. Taking fig. 3 as an example, when the starting point o expands backward, only the cost of (o, i)/(o, j) cannot be considered, but a state that o expands to i and j and then expands from i and j is further considered, that is, the cost of (o, i, j)/(o, i, k)/(o, j, i)/(o, j, k)/(o, j, o ') needs to be considered, and the current solution is updated to be (o, j, o') with the minimum cost.
And a second operator H2, wherein the current node of the label to be expanded is only expanded to the node of the negative cost arc, and meanwhile, the accurate governing rule is kept unchanged.
And in the third operator H3, the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the accurate governing rule is kept unchanged.
And a fourth operator H4, wherein the current node of the label to be expanded is only expanded to the node of the negative cost arc, and meanwhile, the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule are relaxed.
And a fifth operator H5, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule are relaxed.
And a sixth operator H6, wherein the current node of the label to be expanded is only expanded to the node of the negative cost arc, and meanwhile, the constraint of the electric vehicle and fuel vehicle load W in the governing rule is relaxed.
And a seventh operator H7, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and simultaneously, the constraint of the load W of the electric vehicle and the fuel vehicle in the governing rule is relaxed.
Each operator can solve the subproblem independently, the solving speed of the method is higher than that of the precise label extension method, however, the subproblems of different data are different in solving space, meanwhile, the column generation algorithm is a process of continuously iterating and solving limited main problems and subproblems, the limited main problems can continuously add new columns along with the solving of the subproblems, the subproblems can be updated along with the change of dual variables of the limited main problems, the solving space of the subproblems changes in each iteration, and therefore the solving performance of each heuristic operator under different data cannot be determined artificially, and therefore a fixed operator solving and calling sequence cannot be given directly. Based on the self-adaptive large-scale neighborhood search concept, the self-adaptive selection mechanism is designed for the selection of the subproblem solving operators, the performance of the operators is scored through the self-adaptive selection mechanism, and the operators are automatically selected and inspired in the column generation algorithm to be solved.
Further, in step S44, the weight of each operator is updated every time Ct iteration calculation is performed, where Ct is the number of operators, and the operator weight may be calculated by the following formula:
in the formula (4-24), theta is a weight parameter and takes the value of theta from the [0,1 ]];S i Scoring the operator according to S n And updating, when the selected operator fails to solve the current subproblem S n =0, otherwise score the operator according to the following case:
when the operator selected for the 1 st time in the small period can solve the subproblem model, S n =10;
When the operator selected at the current time cannot solve the subproblem model and the operator selected at the current time can solve the subproblem model, the operator selected at the current time is not the last operator, S n =20;
When the operator selected at the front cannot solve the sub-problem model until the operator selected at the last time can solve, S n =30。
As shown in fig. 6, it is assumed that the logistics enterprise has a distribution center and 3 electric logistics vehicles (vehicle numbers 1-3) and 3 fuel logistics vehicles (vehicle numbers 4-6), and the distribution cost per kilometer of the electric vehicles is λ e =1, fuel vehicle distribution cost per kilometer is lambda c =1.57, there are 5 customers with known locations and demand waiting for delivery, and the vehicle can only serve the customer within each customer-specified time period. Each vehicle is required to return to the distribution center after the distribution is completed, each vehicle can only serve one path, and each customer can only be served by one vehicle. Because the endurance mileage of the electric vehicle is short (the maximum driving mileage is 77.75), an enterprise can select part of positions in a given swapping station candidate set to build a swapping station, and the cost lambda generated by building the swapping station is f =38.875, the electric vehicle can change the full charge battery in the established power station, and how the enterprise should make the station establishment scheme and the delivery scheme to minimize the total delivery cost of the system. Distance d between points ij As shown in fig. 7.
And constructing a main problem model and a sub problem model.
First, a feasible initial delivery scheme is given:
The total cost of the distribution scheme = station construction cost 0+ travel cost 356.39=356.39;
secondly, initializing an operator pool, wherein 7 operators such as H1, H2, … H7 and the like are shared, and each operator has an initial weight W i All are set to 10, then the probability of each operator being selected is the same at the beginning, and the small iteration cycle Ct =3 is set.
Column generates the first iteration, count =1:
(1) Constructing an RMP, and solving the current RMP:
Solving the current dual variable as pi 1 =218.23;π 2 =138.16;π 3 =π 4 =π 5 =0;π c =0。
(2) The arc costs are re-priced according to the dual variables, resulting in a solution set list of the first sub-problem model, as shown in FIG. 8.
(3) Roulette selects operators from the operator pool, and using H2 to solve for SP, assuming H2 is selected, solves for a path with a negative cost of-70.16 as follows:
The recording operator H2 is selected for 1 time, and the score is added by 10;
the column generates the second iteration, count =2:
(4) Constructing RMP, and solving the current RMP:
Solving the current dual variable as pi 1 =218.23;π 2 =68;π 3 =π 4 =0;π 5 =70.16;π e =π c =0
(5) The arc costs are re-priced according to the dual variables, resulting in a solution set list for the second sub-problem model, as shown in FIG. 9.
(6) The roulette selects an operator from the operator pool, H1 is used for solving SP (service provider) on the assumption that H1 is selected, and a negative cost path is not solved;
h1 is deleted from the operator pool, roulette selects from the operators left in the operator pool, H2 is supposed to be selected, SP is solved by H2, and a negative cost path is not solved;
removing H1 from the operator pool, selecting the roulette from the remaining operators in the operator pool, and solving for SP with H6, assuming H6 is selected, to find a path with a negative cost of-44.355 as follows:
The recording operator H6 is selected for 1 time, and the score is added for 20;
column generates the third iteration, count =3:
(7) Constructing RMP, and solving the current RMP:
Solving the current dual variable as:
π 1 =173.875;π 2 =68;π 3 =0;π 4 =44.355;π 5 =70.16;π e =π c =0
(8) The arc costs are re-priced according to the dual variables, resulting in a solution set list for the third sub-problem model, as shown in FIG. 10.
(9) The roulette selects an operator from the operator pool, H2 is used for solving SP (service provider) on the assumption that H2 is selected, and a negative cost path is not solved;
h2 is deleted from the operator pool, roulette selects from the operators left in the operator pool, H5 is selected, SP is solved by H5, and a negative cost path is not solved;
h5 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H3 is selected, the SP is solved by using the H3, and a negative cost path is not solved;
h3 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H4 is selected, the SP is solved by the H4, and a negative cost path is not solved;
h4 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H7 is selected, the SP is solved by using the H7, and a negative cost path is not solved;
h7 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H1 is supposed to be selected, SP is solved by H1, and a negative cost path is not solved;
h1 is deleted from the operator pool, roulette selects from the remaining operators in the operator pool, H6 is selected, and H6 is used to solve SP, solving two paths with negative cost of-25.515 as follows:
The recording operator H6 is selected for 2 times in total, and the score is added by 30;
at this time, count ≧ Ct =3, the operator information is updated:
h1: selecting 0 times, scoring as 0, and keeping the weight unchanged,W 1 =10;
h2: 1 hit, score 10, weight increase, W 2 =10+10=20;
H3: selecting 0 times, scoring 0, weight unchanged, W 3 =10;
H4: selecting 0 times, scoring 0, weight unchanged, W 4 =10;
H5: select 0 times, score 0, weight unchanged, W 5 =10;
H6: selecting 2 times, scoring 20+30=50, increasing weight, W 6 =10+50=60;
H7: selecting 0 times, scoring 0, weight unchanged, W 7 =10;
Reset count =1.
Column generates the fourth iteration, count =1:
(10) Constructing an RMP, and solving the current RMP:
Solving the current dual variable as:
π 1 =214.35;π 2 =108.47;π 3 =0;π 4 =-3.89;π 5 =111.47;π e =-29.69;π c =0
(11) The arc costs are re-priced according to the dual variables, resulting in a fourth sub-problem model solution set list, as shown in FIG. 11.
(12) The roulette selects an operator from the operator pool, and if H6 is selected (the probability of selecting H6 is the maximum), the SP is solved by using H6, and a negative cost path is not solved;
removing H6 from the operator pool, selecting the roulette from the remaining operators in the operator pool, and solving for SP by using H2 without solving for a negative cost path on the assumption that H2 is selected (the probability of selecting H2 is the second);
h2 is deleted from the operator pool, roulette selects from the operators left in the operator pool, H3 is selected, SP is solved by H3, and a negative cost path is not solved;
h3 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H4 is selected, the SP is solved by the H4, and a negative cost path is not solved;
h4 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H7 is selected, the SP is solved by using the H7, and a negative cost path is not solved;
h7 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H1 is selected, the SP is solved by using the H1, and a negative cost path is not solved;
h1 is deleted from the operator pool, the roulette selects from the operators left in the operator pool, H5 is selected, the SP is solved by the H5, and a negative cost path is not solved;
and (4) judging that the algorithm pool is empty (namely all operators are deleted), further calling the precision DP to solve, and if the precision DP does not solve the negative cost path, ending the column generation process.
Further judging the solution obtained by the current RMPThe integer solution is obtained, and branch pricing is finished without branch delimitation.
The optimal distribution scheme of the distribution system is finally obtained as follows:
1. the station building scheme comprises the following steps: establishing a battery replacement station at a candidate battery replacement station S6;
2. path scheme: 3 electric vehicles are selected to serve 5 customer points;
electric vehicle k 1 A driving path: 0-2-0;
electric vehicle k 2 A driving path: 0-3-6-1-0;
electric vehicle k 3 A driving path: 0-4-5-0;
3. total delivery cost = station cost 38.875+ travel cost 274=312.875.
The distribution scheme before optimization is shown in fig. 12, and the total distribution cost before optimization is 356.39;
the optimized delivery scheme is shown in fig. 13, and the optimized total delivery cost is 312.875.
On the other hand, the embodiment provides a system for site selection of a swapping station and path planning of a hybrid fleet, which includes a data acquisition module 51, a model construction module 52, a model reconstruction module 53, and a calculation module 54.
A data obtaining module 51, configured to obtain alternative data information;
the model building module 52 is used for building an integer planning model for combined decision making of power station selection and electric vehicle and fuel vehicle mixed fleet delivery;
a model reconstruction module 53, configured to reconstruct the integer programming model into a main problem model and a sub problem model by using a Dantzig-Wolfe decomposition method;
and the calculation module 54 is used for solving the main problem model and the sub problem model to obtain a distribution scheme and a power station location selection scheme with the lowest total cost.
The site selection of the swapping station and the route planning system of the hybrid fleet provided by the embodiment of the invention have the same implementation principle and technical effect as the embodiment of the site selection of the swapping station and the route planning method of the hybrid fleet, and for brief description, the corresponding contents in the embodiment of the site selection of the swapping station and the route planning method of the hybrid fleet can be referred to.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of the changes or substitutions within the technical scope of the present invention, and shall cover the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (3)
1. A method for power station selection and mixed fleet path planning is characterized by comprising the following steps:
acquiring alternative data information;
constructing an integer planning model for combined decision of power station selection and electric vehicle and fuel vehicle mixed fleet delivery;
reconstructing the integer programming model into a main problem model and a sub problem model;
solving the main problem model and the sub problem model to obtain a distribution scheme with the lowest total cost and a power station site selection scheme;
the integer programming model includes the following formula:
the formula (4-1) in the model is an objective function and consists of three items of cost, wherein the first item is the construction cost of the power exchanging station, the second item is the transportation cost of the electric vehicle, and the third item is the transportation cost of the fuel vehicle; the formula (4-2) shows that each customer point and the power change station can only be accessed by one vehicle at the same time; the formula (4-3) restricts the available number of the delivery vehicles not to exceed the scale of the fleet; the formula (4-4) indicates that the started power change station can provide the power change service; equations (4-5) ensure conservation of each point stream; formulas (4-6) and (4-7) show that the total distribution task amount of the electric vehicle and the fuel vehicle does not exceed the maximum loading capacity; the formula (4-8) represents the time logical relation of the vehicle from the point i to the point j; the formula (4-9) represents the time logical relation of the electric vehicle from the power change station f to the point j; equations (4-10) satisfy the service time window constraints for all points; the formula (4-11) satisfies the relation of electric quantity consumption when the electric vehicle runs from the point i to the point j; the formulas (4-12) represent that the electric vehicle has constant electric quantity when arriving at and leaving a customer point; formula (4-13) shows that the electric vehicle starts from the full-charge state of the distribution center and changes the full-charge battery after reaching the power change station; the formula (4-14) restricts the electric quantity of the electric vehicle at any point to be larger than zero, and ensures that the electric vehicle can return to a distribution center; the formula (4-15) is 0-1 decision variable constraint condition;
the main problem model comprises the following formula:
in the formulas (4-16) to (4-20), p is a vehicle driving path, omega is a set of all feasible paths meeting the constraint, and p belongs to omega; the vehicle route combination (p, k) indicates that the route p is delivered by the vehicle k;a variable of 0-1, indicating whether a vehicle path combination (p, k) is employed in the final solution;a variable 0-1 indicating whether the vehicle path combination (p, k) passes through the arc (i, j);a variable of 0-1, representing whether path p passes through point i;representing the cost of a vehicle path combination (p, k), including the cost of path transportation and the cost of station construction, λ k Taking λ according to the type of vehicle k e Or λ c SaidCalculated from the following equation:
the subproblem model comprises the following formula:
in the formulas (4 to 21) and (4 to 22), pi = { pi = { [ pi ]) i ,π e ,π c The dual variables are respectively corresponding to the formula (4-17), the formula (4-18) and the formula (4-19) in the main problem model;representing the inspection number of each feasible path of omega in the main problem model;
the method for solving the main problem model and the sub problem model comprises the following steps:
s41, initializing node information, setting the global upper bound as positive infinity, setting the value of an active node set to be the same as that of a root node, and initializing operator pool information;
s42, judging the termination of calculation, judging whether the active node set is empty, if so, determining that the current upper bound is the optimal solution, and terminating calculation, otherwise, turning to the step S43;
s43, selecting a node from the active node set according to the node selection strategy, and deleting the selected node from the active node set;
s44, solving the node, and if the node is not solved, turning to the step S42; otherwise, recording the linear relaxation optimal solution of the node as a local lower bound of the node;
s45, pruning, and if the local lower bound is greater than or equal to the global upper bound, turning to the step S42; otherwise, further judging whether the optimal relaxation solution obtained by the node is a score, if so, turning to the step S46; if the number of the nodes is an integer, the global upper bound is updated to be a local lower bound, the nodes which are not smaller than the current global upper bound in the active node set are deleted, and the step S42 is switched to;
s46, branching, namely selecting a branch variable from the relaxation optimal solution of the current node according to a branch strategy to divide a solution space to obtain sub-nodes, adding the new sub-nodes into an active node set, and turning to the step S42;
the step S44 includes the steps of:
s4401, constructing a main problem model;
s4402, solving a main problem model;
s4403, transferring a dual variable;
s4404, resetting the iteration times recorded in the counter to zero, and updating operator information;
s4405, randomly selecting operators from an operator pool, and solving a subproblem model;
s4406, judging whether a column with a negative check number is generated, if so, entering step S4407, and if not, entering step S4409;
s4407, adding 1 to the iteration number recorded in the counter, adding a score to the operator score, adding a column with a negative inspection number to the main problem model, and simultaneously restoring the operator pool;
s4408, judging the relationship between the iteration times and the small iteration cycles, and if the iteration times are smaller than the small iteration cycles, entering the step S4405; otherwise, go to step S4404;
s4409, adding 1 to the iteration times, and deleting the operator selected in the step S4405 in an operator pool;
s4410, judging whether the operator pool is empty, and if the operator pool is empty, entering the step S4411; otherwise, go to step S4405;
s4411, calling an accurate label extension method to solve a subproblem model;
s4412, judging whether a column with a negative detection number is generated; if a column with a negative check number is generated, adding the column with the negative check number into the main problem model, and entering the step S4402; otherwise, terminating the column generation iteration;
the operator pool comprises the following 7 operators:
the first operator is used for expanding the current state of the solution to 2 stages in a greedy algorithm to obtain a second-order greedy operator;
the second operator, the current node of the label to be expanded is expanded to the node of the negative cost arc only, and meanwhile, the accurate governing rule is kept unchanged;
a third operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the accurate governing rule is kept unchanged;
the fourth operator is used for expanding the current node of the label to be expanded only to the node of the negative cost arc, and simultaneously relaxing the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule;
a fifth operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule are relaxed;
the sixth operator, the current node of the label to be expanded is only expanded to the node of the negative cost arc, and simultaneously, the constraint of the electric vehicle and the fuel vehicle load W in the governing rule is relaxed;
a seventh operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and simultaneously, the constraint of the electric vehicle and the fuel vehicle load W in the governing rule is relaxed;
in the step S44, the weight of each operator is updated every time Ct iterative computation is performed, where Ct is the number of operators, and the operator weight is computed by the following formula:
in the formula (4-24), theta is a weight parameter and takes the value of theta from the [0,1 ]];S i Scoring the operator according to S n And updating, when the selected operator fails to solve the current subproblem S n =0, otherwise score the operator according to the following case:
when the operator selected for the 1 st time in the small period can solve the subproblem model, S n =10;
When the operator selected at the current time cannot solve the subproblem model and the operator selected at the current time can solve the subproblem model, the operator selected at the current time is not the last operator, S n =20;
When the operator selected at the front cannot solve the sub-problem model until the operator selected at the last time can solve the problem, S n =30;
In the above formula, I represents a customer set, I belongs to I, I = {1,2.. N };0 denotes a distribution center as a starting point; n +1 represents a distribution center as an end point; f denotes a charging station set, F = {1,2,.. F }; v denotes a set of all points, V = ibute {0} { n +1}; v 1 Representing a set of customers and origins, V 1 =I∪{0};V 2 Represents a set of customers, origins and charging stations, V 2 =I∪F∪{0};V 3 Representing a set of customers, terminals and charging stations, V 3 = I utouf utou { n +1}; k represents the set of all vehicles, K = {1,2 e +m c };K e Indicating a set of electric vehicles, K e ={1,2,...m e };K c Indicating fuel vehicle set, K c ={m e +1,...,m e +m c };W e Representing the maximum load capacity of the electric vehicle; w c Representing the maximum load capacity of the fuel vehicle; q represents the electric vehicle battery capacity; g represents the unit mileage power consumption rate of the electric vehicle; q. q.s i Representing clients iA demand amount; d ij Represents the distance between arcs (i, j); lambda e Representing the driving cost of the electric vehicle per mileage; lambda [ alpha ] c Representing the driving cost of the fuel vehicle per mileage; lambda f The construction cost of the power change station is shown; s i Represents the service time required by the client i; s. the f Indicating a battery replacement time; [ e ] a i ,l i ]A service time window representing customer i; t is t ij Represents the time required for the arc (i, j) to travel; t is t i Represents the time of arrival at point i;representing the residual capacity of the electric vehicle k when the electric vehicle k reaches the point i;representing the remaining capacity of the electric vehicle k when the electric vehicle k leaves the point i; x is the number of ijk Represents the variable 0-1, whether vehicle k passes arc (i, j) or not, and if so, x ijk =1, if not x ijk =0;y f Representing a variable of 0-1, selecting f points to build a power conversion station, and if the power conversion station is built, y points f =1, if not constructed y f =0。
2. The method for power station location selection and hybrid fleet path planning as set forth in claim 1, wherein said alternative data information comprises: the number and address of the distribution centers; the number of the electric vehicles, the distribution cost of the electric vehicles per kilometer and the maximum driving mileage of the electric vehicles; the number of fuel vehicles, the delivery cost per kilometer of fuel vehicles; the location, demand and service time window for each customer waiting to be delivered; and the position of each alternative power conversion station.
3. A system for site selection of a power conversion station and route planning of a hybrid fleet, the system comprising:
the data acquisition module is used for acquiring alternative data information;
the model construction module is used for constructing an integer planning model for combined decision-making of power station selection and electric vehicle and fuel vehicle mixed fleet delivery;
the model reconstruction module is used for reconstructing the integer programming model into a main problem model and a sub problem model;
the calculation module is used for solving the main problem model and the sub problem model to obtain a distribution scheme and a power station site selection scheme with the lowest total cost;
the integer programming model includes the following formula:
the formula (4-1) in the model is an objective function and consists of three terms of cost, wherein the first term is the construction cost of the power change station, the second term is the transportation cost of the electric vehicle, and the third term is the transportation cost of the fuel vehicle; the formula (4-2) shows that each customer point and the power change station can only be accessed by one vehicle at the same time; the formula (4-3) restricts the available number of the delivery vehicles not to exceed the scale of the fleet; the formula (4-4) indicates that the started power change station can provide the power change service; equations (4-5) ensure that each point stream is conserved; formulas (4-6) and (4-7) show that the total distribution task amount of the electric vehicle and the fuel vehicle does not exceed the maximum loading capacity; the formula (4-8) represents the time logical relationship of the vehicle from the point i to the point j; the formula (4-9) represents the time logical relation of the electric vehicle from the power change station f to the point j; equations (4-10) satisfy the service time window constraints for all points; the formula (4-11) satisfies the relation of electric quantity consumption when the electric vehicle runs from the point i to the point j; the formulas (4-12) represent that the electric vehicle has constant electric quantity when arriving at and leaving a customer point; formula (4-13) shows that the electric vehicle starts from a distribution center full-charge state and changes a full-charge battery after arriving at a battery replacement station; the formula (4-14) restricts the electric quantity of the electric vehicle at any point to be larger than zero, and ensures that the electric vehicle can return to a distribution center; the formula (4-15) is 0-1 decision variable constraint condition;
the main problem model comprises the following formula:
in the formulas (4-16) to (4-20), p is a vehicle driving path, omega is a set of all feasible paths meeting the constraint, and p belongs to omega; the vehicle route combination (p, k) indicates that the route p is delivered by the vehicle k;is a variable from 0 to 1, indicating whether a vehicle path combination (p, k) is employed in the final solution scheme);A variable 0-1 indicating whether the vehicle path combination (p, k) passes through the arc (i, j);a variable of 0-1, indicating whether the path p passes through the point i;representing the cost of a vehicle path combination (p, k), including the cost of path transportation and the cost of station construction, λ k Taking λ according to the type of vehicle k e Or λ c SaidCalculated from the following equation:
the subproblem model comprises the following formula:
in formulas (4 to 21) and (4 to 22), pi = { pi = i ,π e ,π c The dual variables are respectively corresponding to formulas (4-17), (4-18) and (4-19) in the main problem model;representing the inspection number of each feasible path of omega in the main problem model;
the method for solving the main problem model and the sub problem model comprises the following steps:
s41, initializing node information, setting the global upper bound as positive infinity, setting the value of an active node set to be the same as that of a root node, and initializing operator pool information;
s42, judging the termination of calculation, judging whether the active node set is empty, if so, determining that the current upper bound is the optimal solution, and terminating calculation, otherwise, turning to the step S43;
s43, selecting a node from the active node set according to the node selection strategy, and deleting the selected node from the active node set;
s44, solving the node, and if the node is not solved, turning to the step S42; otherwise, the linear relaxation optimal solution of the node is recorded as the local lower bound of the node;
s45, pruning, and if the local lower bound is greater than or equal to the global upper bound, turning to the step S42; otherwise, further judging whether the optimal relaxation solution is a score, and if the optimal relaxation solution is a score, turning to the step S46; if the number of the nodes is an integer, the global upper bound is updated to be a local lower bound, the nodes which are not smaller than the current global upper bound in the active node set are deleted, and the step S42 is switched to;
s46, branching, selecting a branch variable from the relaxation optimal solution of the current node according to a branch strategy to divide a solution space to obtain sub-nodes, adding the new sub-nodes into an active node set, and turning to the step S42;
the step S44 includes the steps of:
s4401, constructing a main problem model;
s4402, solving a main problem model;
s4403, transferring a dual variable;
s4404, resetting the iteration times recorded in the counter to zero, and updating operator information;
s4405, randomly selecting operators from the operator pool, and solving a subproblem model;
s4406, judging whether a column with a negative check number is generated, if so, entering step S4407, and if not, entering step S4409;
s4407, adding 1 to the iteration number recorded in the counter, adding a score to the operator score, adding a column with a negative inspection number to the main problem model, and simultaneously restoring the operator pool;
s4408, judging the relationship between the iteration times and the small iteration cycles, and if the iteration times are smaller than the small iteration cycles, entering the step S4405; otherwise, go to step S4404;
s4409, adding 1 to the iteration number, and deleting the operator selected in the step S4405 in an operator pool;
s4410, judging whether the operator pool is empty, and if the operator pool is empty, entering the step S4411; otherwise, go to step S4405;
s4411, calling an accurate label extension method to solve a subproblem model;
s4412, judging whether a column with a negative detection number is generated; if a column with a negative check number is generated, adding the column with the negative check number into the main problem model, and entering the step S4402; otherwise, terminating the column generation iteration;
the operator pool comprises the following 7 operators:
the first operator is used for expanding the current state of the solution to 2 stages in a greedy algorithm to obtain a second-order greedy operator;
the second operator, the current node of the label to be expanded is expanded to the node of the negative cost arc only, and meanwhile, the accurate governing rule is kept unchanged;
a third operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the accurate governing rule is kept unchanged;
the fourth operator, the current node of the label to be expanded is only expanded to the node of the negative cost arc, and meanwhile, the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule are relaxed;
a fifth operator, wherein the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the constraints of the electric quantity Q of the electric vehicle and the arrival time T of the fuel vehicle in the governing rule are relaxed;
the sixth operator, the current node of the label to be expanded is only expanded to the node of the negative cost arc, and simultaneously, the constraint of the electric vehicle and the fuel vehicle load W in the governing rule is relaxed;
the seventh operator, the current node of the label to be expanded is only expanded to the node of the cost reduction arc, and meanwhile, the constraint of the electric vehicle and the fuel vehicle load W in the governing rule is released;
in step S44, the weight of each operator is updated every time Ct iteration calculation is performed, where Ct is the number of operators and the operator weight is calculated by the following formula:
in the formula (4-24), theta is a weight parameter and is a value theta epsilon [0,1 ]];S i Scoring an operator according to S n And updating, when the selected operator fails to solve the current subproblem S n =0, otherwise the operator is scored according to the following case:
when the operator selected for the 1 st time in the small period can solve the sub-problem model, S n =10;
When the operator selected at the current time cannot solve the subproblem model and the operator selected at the current time can solve the subproblem model, the operator selected at the current time is not the last operator, S n =20;
When the operator selected at the front cannot solve the sub-problem model until the operator selected at the last time can solve the problem, S n =30;
In the above formula, I represents a customer set, I belongs to I, I = {1,2.. N };0 denotes a distribution center as a starting point; n +1 represents a distribution center as an end point; f denotes a charging station set, F = {1,2, · F }; v denotes a set of all points, V = ibute {0} { n +1}; v 1 Representing a set of customers and origins, V 1 =I∪{0};V 2 Represents a set of customers, origins and charging stations, V 2 =I∪F∪{0};V 3 Representing a set of customers, terminals and charging stations, V 3 = ibute { n +1}; k represents the set of all vehicles, K = {1,2 e +m c };K e Denotes the set of electric vehicles, K e ={1,2,...m e };K c Indicating fuel-oil vehicle set, K c ={m e +1,...,m e +m c };W e Representing the maximum load capacity of the electric vehicle; w c Representing the maximum load capacity of the fuel vehicle; q represents the electric vehicle battery capacity; g represents the unit mileage power consumption rate of the electric vehicle; q. q.s i Representing the demand of customer i; d ij Represents the distance between arcs (i, j); lambda [ alpha ] e Representing the driving cost of the electric vehicle per mileage; lambda [ alpha ] c Representing the driving cost of the fuel vehicle per mileage; lambda [ alpha ] f The construction cost of the power change station is shown; s i Indicating the service time required by the client i; s f Indicating the time for replacing the battery; [ e ] a i ,l i ]A service time window representing customer i; t is t ij Represents the time required for the arc (i, j) to travel; t is t i Represents the time of arrival at point i;representing the residual capacity of the electric vehicle k when the electric vehicle k reaches the point i;representing the remaining capacity of the electric vehicle k when the electric vehicle k leaves the point i; x is the number of ijk Represents the variable 0-1, whether vehicle k passes through arc (i, j), if so, x ijk =1, if not x ijk =0;y f Representing a variable of 0-1, selecting f points to build a power conversion station, and if the power conversion station is built, y points f =1, if not constructed y f =0。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010158192.4A CN111325409B (en) | 2020-03-09 | 2020-03-09 | Method and system for site selection of battery replacement station and route planning of hybrid fleet |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010158192.4A CN111325409B (en) | 2020-03-09 | 2020-03-09 | Method and system for site selection of battery replacement station and route planning of hybrid fleet |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111325409A CN111325409A (en) | 2020-06-23 |
CN111325409B true CN111325409B (en) | 2022-11-22 |
Family
ID=71173198
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010158192.4A Active CN111325409B (en) | 2020-03-09 | 2020-03-09 | Method and system for site selection of battery replacement station and route planning of hybrid fleet |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111325409B (en) |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111845426B (en) * | 2020-07-01 | 2021-09-24 | 大连理工大学 | Pure electric bus charging power distribution and optimization method based on column generation framework |
CN112488358B (en) * | 2020-10-31 | 2023-04-07 | 海南电网有限责任公司 | Electric vehicle charging path planning method and storage medium |
CN112347054A (en) * | 2020-11-10 | 2021-02-09 | 北京明略软件系统有限公司 | Personnel information comparison method, system, electronic equipment and computer readable storage medium |
CN113762855B (en) * | 2020-11-20 | 2023-12-05 | 北京京东振世信息技术有限公司 | Resource allocation method and device |
CN112762952B (en) * | 2020-12-24 | 2021-09-28 | 西南交通大学 | Method, device and equipment for optimizing driving track and readable storage medium |
CN112750063B (en) * | 2021-01-04 | 2023-12-05 | 李璐 | Random planning-based public bus team facility site selection-path planning-scheduling method |
CN112686458B (en) * | 2021-01-05 | 2023-03-07 | 昆明理工大学 | Optimized dispatching method for multi-vehicle fleet delivery process |
CN112862162A (en) * | 2021-01-20 | 2021-05-28 | 华中科技大学 | Vehicle scheduling method considering step freight rate subsidy |
CN113029144B (en) * | 2021-03-01 | 2023-03-17 | 汇链通产业供应链数字科技(厦门)有限公司 | Sub-heuristic algorithm path planning method for collaborative transportation |
CN114037180B (en) * | 2021-11-25 | 2023-02-03 | 电子科技大学 | Collaborative distribution path optimization method based on branch pricing and cutting algorithm |
CN114462693B (en) * | 2022-01-26 | 2023-02-10 | 电子科技大学 | Distribution route optimization method based on vehicle unmanned aerial vehicle cooperation |
CN116579511A (en) * | 2023-03-20 | 2023-08-11 | 广东工业大学 | Two-stage path planning optimization method based on SP model |
CN116362523B (en) * | 2023-06-01 | 2023-09-26 | 吉林大学 | Coordinated optimization method for site selection and operation strategy of power exchange station considering temperature adaptability |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107798867A (en) * | 2017-10-12 | 2018-03-13 | 上海交通大学 | A kind of congested traffic network equalization methods based on electric automobile and diesel locomotive mixed flow |
WO2018136687A1 (en) * | 2017-01-19 | 2018-07-26 | Massachusetts Institute Of Technology | A data-driven system for optimal vehicle fleet dimensioning, ride-sharing, and real-time dispatching based on shareability networks |
CN108764777A (en) * | 2018-04-26 | 2018-11-06 | 浙江工商大学 | Electronic logistic car dispatching method with time window and system |
CN109492791A (en) * | 2018-09-27 | 2019-03-19 | 西南交通大学 | Intercity highway network light based on charging guidance stores up charging station constant volume planing method |
CN109583650A (en) * | 2018-11-30 | 2019-04-05 | 浙江工商大学 | A kind of method of the addressing of electric vehicle electrical changing station and logistics distribution combined dispatching |
CN109886468A (en) * | 2019-01-22 | 2019-06-14 | 河海大学 | Charging station planing method based on improved self-adapted genetic algorithm |
CN109919369A (en) * | 2019-02-26 | 2019-06-21 | 浙江财经大学 | A kind of battery-exchange station addressing and electric car paths planning method |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10288439B2 (en) * | 2017-02-22 | 2019-05-14 | Robert D. Pedersen | Systems and methods using artificial intelligence for routing electric vehicles |
CN110059934A (en) * | 2019-03-27 | 2019-07-26 | 浙江工商大学 | The method of fuel vehicle and the scheduling of new energy vehicle coperating distribution |
-
2020
- 2020-03-09 CN CN202010158192.4A patent/CN111325409B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018136687A1 (en) * | 2017-01-19 | 2018-07-26 | Massachusetts Institute Of Technology | A data-driven system for optimal vehicle fleet dimensioning, ride-sharing, and real-time dispatching based on shareability networks |
CN107798867A (en) * | 2017-10-12 | 2018-03-13 | 上海交通大学 | A kind of congested traffic network equalization methods based on electric automobile and diesel locomotive mixed flow |
CN108764777A (en) * | 2018-04-26 | 2018-11-06 | 浙江工商大学 | Electronic logistic car dispatching method with time window and system |
CN109492791A (en) * | 2018-09-27 | 2019-03-19 | 西南交通大学 | Intercity highway network light based on charging guidance stores up charging station constant volume planing method |
CN109583650A (en) * | 2018-11-30 | 2019-04-05 | 浙江工商大学 | A kind of method of the addressing of electric vehicle electrical changing station and logistics distribution combined dispatching |
CN109886468A (en) * | 2019-01-22 | 2019-06-14 | 河海大学 | Charging station planing method based on improved self-adapted genetic algorithm |
CN109919369A (en) * | 2019-02-26 | 2019-06-21 | 浙江财经大学 | A kind of battery-exchange station addressing and electric car paths planning method |
Non-Patent Citations (3)
Title |
---|
An energy-efficient green-vehicle routing problem with mixed vehicle fleet, partial battery recharging and time windows;Giusy Macrina,etc;《European Journal of Operational Research》;20190801;第276卷(第3期);第971-982页 * |
不同车型车辆路径问题模型及混合算法;王琦等;《安徽工业大学学报(自然科学版)》;20170630;第34卷(第2期);第200-207页 * |
多车型电动汽车车辆路径问题的分支定价算法研究;揭婉晨;《系统工程理论与实践》;20160731;第36卷(第7期);第1795-1805页 * |
Also Published As
Publication number | Publication date |
---|---|
CN111325409A (en) | 2020-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111325409B (en) | Method and system for site selection of battery replacement station and route planning of hybrid fleet | |
Koç et al. | The electric vehicle routing problem with shared charging stations | |
Küçükoğlu et al. | Hybrid simulated annealing and tabu search method for the electric travelling salesman problem with time windows and mixed charging rates | |
Olsen et al. | A study on flow decomposition methods for scheduling of electric buses in public transport based on aggregated time–space network models | |
CN109919365B (en) | Electric vehicle path planning method and system based on double-strategy search | |
Li-ying et al. | Multiple charging station location-routing problem with time window of electric vehicle. | |
CN112001560B (en) | Two-stage bus scheduling algorithm based on iterative neighborhood search | |
Li et al. | Efficient and effective express via contextual cooperative reinforcement learning | |
CN110309436B (en) | Automobile service network collaborative site selection method and system considering owner selection behavior | |
Raeesi et al. | Coordinated routing of electric commercial vehicles with intra-route recharging and en-route battery swapping | |
CN113848970B (en) | Multi-target cooperative path planning method for vehicle-unmanned aerial vehicle | |
CN112613659B (en) | New energy freight charging station site selection-path optimization method based on three-dimensional space-time network | |
Wang et al. | A scheduling strategy of mobile parcel lockers for the last mile delivery problem | |
CN112801347B (en) | Multi-target city two-stage distribution planning method based on mobile transfer station and crowdsourcing | |
CN111860957B (en) | Multi-vehicle-type vehicle path planning method considering secondary distribution and balancing | |
CN114048924A (en) | Multi-distribution center site selection-distribution path planning method based on hybrid genetic algorithm | |
CN113762886A (en) | Oil product secondary logistics distribution optimization method and system based on variable neighborhood search algorithm | |
Wang et al. | A branch-and-price algorithm for a green location routing problem with multi-type charging infrastructure | |
Farvaresh et al. | A coalition structure algorithm for large-scale collaborative pickup and delivery problem | |
CN117035598A (en) | Vehicle-machine collaborative path planning method and system under regional limitation | |
Hua et al. | Optimality-guaranteed algorithms on the dynamic shared-taxi problem | |
CN112488358B (en) | Electric vehicle charging path planning method and storage medium | |
Lai et al. | A rolling-horizon decision framework for integrating relocation and user flexibility in one-way electric carsharing systems | |
Cortés-Murcia et al. | Multi-period profitable tour problem with electric vehicles and mandatory stops | |
Narayanan et al. | A Reinforcement Learning Approach for Electric Vehicle Routing Problem with Vehicle-to-Grid Supply |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |