A modular bus operation programming method based on variable rolling time windows

Abstract

A modular bus operation compilation method based on a variable rolling time window. The present invention belongs to the field of urban public transportation technology, and specifically relates to a modular bus operation compilation method based on a dynamic time window. The purpose of the present invention is to solve the problem that the existing modular bus system cannot handle the mismatch between the supply of bus system capacity resources and the temporal and spatial demand of passenger flow, resulting in long passenger travel time, low modular bus operation efficiency, and high bus company operating costs. The present invention decomposes the bus operation cycle into a number of continuous variable time windows, and completes two-stage interactive optimization control at the beginning of each time window. At the same time, the departure interval information of the compilation plan is passed to the rolling time domain, and the first departure interval length obtained by the plan is used as the variable window length to execute the car grouping plan. Until the total plan cycle is completed, the departure interval timetable of the departure station and the number of cars sent by each train at each station are obtained.

Description

Technical Field

[0001] The invention belongs to the technical field of urban public transportation, and in particular relates to a modular public transportation operation compilation method based on a dynamic time window. Background Art

[0002] Under the current background of urban transportation development in my country, the "public transport priority" policy aims to solve the contradiction between the growing travel demand of residents and limited transportation resources by giving priority to the development of public transportation systems, and to alleviate the traffic pressure caused by the continuous surge in the number of motor vehicles. The emergence of modular autonomous vehicles (MAV) has brought disruptive changes and huge development space to the traditional public transportation system. Based on the design concept of flexibility and customizability of public transportation vehicles, this type of vehicle can flexibly adjust public transportation capacity through the application of dissociation / coupling technology, give full play to the advantages of intensive transportation in high-flow sections, and provide personalized transportation services in low-density sections, thereby achieving accurate matching of capacity supply and passenger demand and optimizing resource allocation.

[0003] However, in the current research on modular bus operation systems, the operation plan of modular vehicles (carriage formation, departure interval) lacks dynamism and real-time performance, resulting in the failure of transport supply to fully adapt to passenger demand in terms of time and space, causing vehicles to be empty or crowded in certain periods or sections. In order to achieve effective allocation of transport resources and give full play to the functional advantages of small modular buses, the main problems that need to be solved are as follows:

[0004] 1) The existing modular bus system fails to achieve dynamic adjustment of departure intervals. The time intervals between vehicles leaving the departure station are mostly fixed values ​​or simply divided into peak and off-peak periods, which makes it impossible to dynamically adjust the departure time according to the passenger flow arrival rules. The system may experience passenger delays or redundant capacity during certain periods. At the same time, the determination of departure intervals requires interactive calculation of passenger flow data collection and vehicle operation status data. Fixed or static time interval adjustment methods are difficult to meet this requirement. The system urgently needs a flexible and variable rolling time domain control method to generate departure intervals of unequal time intervals based on the system operation status to adapt to the temporal volatility of passenger flow demand.

[0005] 2) In the current research on modular bus operation, the optimal configuration of the number of carriages is mostly focused on single lines, without considering the operation process of seamless passenger transfers between lines, ignoring the functional advantages of modular vehicles that can achieve barrier-free splitting / combination, and insufficient research on the cycle operation of modular vehicles at the road network level and the resulting transfer connection optimization problems. Therefore, the system needs to master the spatial law of passenger flow distribution on the line and incorporate the passenger transfer operation between lines into the construction process of the optimization model, so as to design a reasonable carriage grouping plan and send different numbers of carriages at different bus stops to adapt to the changing passenger needs of the section.

[0006] Therefore, in the face of the problem of mismatch between the supply of transport resources in the bus system and the temporal and spatial demand of passenger flow, on the one hand, it is necessary to use rolling time windows to design flexible and variable departure intervals to adapt to the temporal fluctuation characteristics of passenger demand; on the other hand, it is necessary to consider seamless transfer operations when designing car grouping schemes on the line network to dynamically adapt to the spatial distribution characteristics of passenger demand. In other words, it is crucial to explore a modular bus operation compilation method that can dynamically adapt to the temporal and spatial changes of passenger flow. Summary of the invention

[0007] The purpose of the present invention is to solve the problem that the existing modular bus system cannot handle the mismatch between the supply of bus system capacity resources and the temporal and spatial demand of passenger flow, resulting in long passenger travel time, low modular bus operation efficiency and high bus company operating costs, and propose a modular bus operation compilation method based on a variable rolling time window.

[0008] A modular bus operation compilation method based on a variable rolling time window. The specific process is as follows:

[0009] Step 1: Start the operation and get the current time As the start time of the operation cycle window and establish a time point set;

[0010] collection Time and operation cycle The passenger flow arrival information within the window is based on Time and operation cycle The passenger demand matrix is ​​obtained from the passenger flow arrival information within the window ;

[0011] Based on bus routes , site data , transfer station data , Passenger demand matrix , Ratio of people getting off at each station and the ratio of passengers transferring between lines , forming a database ;

[0012] Step 2: Database based on step 1 , calculate the vehicle running status :

[0013] Step 3: Based on the database and vehicle operating status , establish a modular bus operation optimization model, that is, construct the objective function of minimizing the total cost, and solve the objective function of minimizing the total cost under the constraints to obtain The best operation plan ;

[0014] Step 4: Set the total scheduling cycle length ,but ; The length of planning prepared for the operation of the bus system;

[0015] Start time , Step 1: Obtain passenger arrival information data , Step 2: Get the vehicle running status data , step 3 obtain ;

[0016] Get departure interval collection The first value is used as the length of the rolling window ,Pick As a carriage group ;

[0017] when When the bus performs carriage formation ;

[0018] when When the operation is completed, re-execute steps 1 to 3 to obtain a new round of operation compilation plan. , take a new round of operation plan Middle departure interval assembly The first value is used as the length of the rolling window ; Get a new round of operation plan middle As a carriage group ;

[0019] when , bus execution car marshaling ;

[0020] when When the operation is completed, re-execute steps 1 to 3 to obtain a new round of operation compilation plan. , take a new round of operation plan Middle departure interval assembly The first value is used as the length of the rolling window ; Get a new round of operation plan middle As a carriage group ;

[0021] until , that is, when the time window rolls to the end of the planned time, the rolling time domain terminates the response and the operation ends; obtain the departure interval table of the starting station within the planned time and the train marshaling plan for each train station .

[0022] The beneficial effects of the present invention are:

[0023] In view of the deficiencies in the prior art, the present invention provides a modular bus operation compilation method based on a variable rolling time window. The bus operation cycle is decomposed into a number of continuous variable time windows, and two-stage interactive optimization control is completed at the beginning of each time window, including the first stage of multi-source information real-time collection and the second stage of operation plan decision-making to obtain the operation compilation plan (carriage grouping, departure interval). At the same time, the departure interval information of the compilation plan is passed to the rolling time domain, and the first departure interval length obtained by the plan is used as the variable window length to execute the car grouping plan. Because the passenger flow demand distribution is different in different time periods, in order to adapt to the volatility of passenger flow, the departure interval solved for each cycle is also different, so the rolling time window is dynamically variable. The above process is executed cyclically until the total planning cycle is completed, and the system will obtain the departure interval timetable of the departure station and the number of cars dispatched by each train at each station. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Optimize the control block diagram for two-stage interaction;

[0025] Figure 2 It is a schematic diagram of the flexible disassembly site service mode;

[0026] Figure 3 This is a schematic diagram of cross transfer;

[0027] Figure 4 This is a schematic diagram of overlapping section transfer;

[0028] Figure 5 This is a schematic diagram of the departure interval and train car formation plan at the departure station. DETAILED DESCRIPTION

[0029] Specific implementation method 1: This implementation method is a modular bus operation compilation method based on a variable rolling time window. The specific process is as follows:

[0030] Step 1: Start the operation and get the current time As the start time of the operation cycle window and establish a time point set;

[0031] collection Time and operation cycle The passenger flow arrival information within the window is based on Time and operation cycle The passenger demand matrix is ​​obtained from the passenger flow arrival information within the window ;

[0032] Based on bus routes , site data , transfer station data , Passenger demand matrix , Ratio of people getting off at each station and the ratio of passengers transferring between lines , forming a database ;

[0033] Step 2: Database based on step 1 , calculate the vehicle running status :

[0034] Based on the functional characteristics of modular bus coupling / decoupling, a service model of flexible disassembly of single lines and seamless transfer between lines is designed;

[0035] Vehicle operating status data include: , , Time point The bus departing from the station Number of carriages 、Real-time number of passengers in the vehicle , the number of passengers stranded at the station , Number of passengers waiting at the station Number of passengers transferring between the lines ;

[0036] It is a 0-1 variable, and a value of 1 indicates a time interval. Selected as the departure interval, a value of 0 indicates Not selected as the departure interval; Indicates time point and time point the time interval formed;

[0037] It is a 0-1 variable. When the value is 1, it means it is at the transfer node. Number of trains Perform seamless transfer operation. A value of 0 indicates that the transfer node Number of trains No seamless transfer operation is performed;

[0038] Step 3: Based on the database and vehicle operating status , establish a modular bus operation optimization model, that is, construct the objective function of minimizing the total cost, and solve the objective function of minimizing the total cost under the constraints to obtain The best operation plan ;

[0039] Step 4: Set the total scheduling cycle length ,but ; The planned duration of the bus system operation is usually one morning or one afternoon of bus operation;

[0040] Start time , Step 1: Obtain passenger arrival information data , Step 2: Get the vehicle running status data , step 3 obtain ;

[0041] Get departure interval collection The first value is used as the length of the rolling window (The value obtained in step 3 ), take the first value As a carriage group ;

[0042] when When the bus performs carriage formation ;

[0043] As the time pointer rolls, the state changes of the time window are continuously monitored. When the operation is completed, re-execute steps 1 to 3 to obtain a new round of operation compilation plan. , take a new round of operation plan Middle departure interval assembly The first value is used as the length of the rolling window ; Get a new round of operation plan middle As a carriage group ;

[0044] when , bus execution car marshaling ;

[0045] when When the operation is completed, re-execute steps 1 to 3 to obtain a new round of operation compilation plan. , take a new round of operation plan Middle departure interval assembly The first value is used as the length of the rolling window (The newly obtained The first value in ); get the new round of operation compilation plan middle As a carriage group ;

[0046] until , that is, when the time window rolls to the end of the planned time, the rolling time domain terminates the response and the operation ends; obtain the departure interval table of the starting station within the planned time and the train marshaling plan for each train station .

[0047] Specific implementation method 2: This implementation method is different from the specific implementation method 1 in that the operation starts in step 1 and the current time is taken As the start time of the operation cycle window and establish a time point set;

[0048] collection Time and operation cycle The passenger flow arrival information within the window is based on Time and operation cycle The passenger demand matrix is ​​obtained from the passenger flow arrival information within the window ;

[0049] Based on bus routes , site data , transfer station data , Passenger demand matrix , Ratio of people getting off at each station and the ratio of passengers transferring between lines , forming a database ;

[0050] The specific process is:

[0051] Step 11: Get the current time As the start time of the operation cycle window, ;

[0052] Collect bus topology information, including bus lines, stations and transfer station data and define their numbers;

[0053] The up-and-down process of a bus from the starting station to the terminal and then back to the starting station is regarded as a complete trip;

[0054] Bus Stops From 1 to To number, , Indicates bus stops, Indicates the total number of bus stops;

[0055] All bus routes in the road network From 1 to To number, , Indicates bus routes, Indicates the total number of bus routes;

[0056] Intersections of lines in the road network will generate transfer stations. , Indicates transfer stations, Represents all transfer stations in the road network;

[0057] It is the total departure number of all lines; Indicates The set of departure times of a bus line (for example, there are 10 buses on route 8, the first one leaves at 7:00, the second one leaves at 7:30, and so on);

[0058] Step 1 and 2: Establish a time point set, and obtain a passenger arrival demand matrix based on the time point set;

[0059] Step 13: Based on the total number of bus routes , Total number of bus stops , all transfer stations in the network , Passenger demand matrix , Ratio of people getting off at each station and the ratio of passengers transferring between lines , forming a database .

[0060] The other steps and parameters are the same as those in the first embodiment.

[0061] Specific implementation method three: This implementation method is different from specific implementation methods one or two in that a time point set is established in steps one and two, and a passenger arrival demand matrix is ​​obtained based on the time point set; the specific process is as follows:

[0062] Step 121: If the operating time required for a bus to complete a trip is , at fixed intervals Will Split into time points, forming a time point set ;

[0063]

[0064] Where: , are the vehicle operation start time and end time respectively;

[0065] Step 122: Collection Time and operation cycle The passenger flow arrival information within the window is based on Time and operation cycle The passenger flow in the window is obtained by operating time The passenger arrival situation at each station, that is, the passenger demand matrix ;

[0066]

[0067]

[0068] In the formula, From the first station to the station Journey time; Indicates the site exist Number of passengers at the time; Indicates the site In the kth fixed interval Number of passengers in the Indicates the site At the first fixed interval Number of passengers in the Indicates the site At the nth fixed interval Number of passengers in the For Site Passenger demand matrix; For Site Passenger demand matrix.

[0069] The other steps and parameters are the same as those in the first or second embodiment.

[0070] Specific implementation method 4: This implementation method is different from any one of the specific implementation methods 1 to 3 in that the database based on step 1 in step 2 , calculate the vehicle running status ; The specific process is:

[0071] Step 21: Design a service model for flexible disassembly of single lines and seamless transfer between lines; the specific process is as follows:

[0072] Flexible disassembly of single lines: In order to minimize the two goals of vehicle operating costs and passenger time costs and achieve a dynamic balance between "capacity supply-passenger demand", a flexible disassembly operation mode is adopted for single bus lines;

[0073] The up and down lines are considered as a complete trip, and the attraction weight coefficient The higher the station is, the higher the passenger travel demand is. The high travel demand stations are set as split stations.

[0074] Allow modular vehicles to be coupled or uncoupled at the disassembly site, and keep the number of car modules unchanged at general sites (sites other than the disassembly site);

[0075] The splitting and combining sites are the top 25% (including 25%) of all values ​​of the attraction weight coefficient;

[0076]

[0077] In the formula, for Site pair The attraction weight coefficient of the site, for The total number of passengers boarding the station. for The total number of people getting off at the station, from which high travel demand stations can be selected as split-and-combine stations;

[0078] Seamless transfer between lines: According to the structure of the bus line network, there are two main ways for passengers to transfer between lines: overlapping transfer and cross transfer;

[0079] In the cross transfer, the two lines intersect in the up and down directions to form four transfer stations. At the transfer stations, modular buses can realize seamless transfer operations at the same station, such as Figure 3 ;

[0080] Similarly, the overlapping transfer lines will form two transfer stations, one for separation and the other for convergence. The lines in the overlapping sections follow the principle of independent operation and non-interference with each other, and when they arrive at the transfer station, passengers can perform directional transfers or convergence transfers according to their transfer intentions. Figure 4 ;

[0081] Step 2: Define decision variables and calculate departure intervals and time point The bus departing from the station Number of carriages ; The specific process is:

[0082] Step 221: Define decision variables; the specific process is:

[0083] It is a 0-1 variable. When the value is 1, it means that the site s is a split site, and when the value is 0, it means that it is a general site.

[0084] It is a 0-1 variable, and a value of 1 indicates a time interval. Selected as the departure interval, a value of 0 indicates Not selected as the departure interval;

[0085] It is a 0-1 variable. When the value is 1, it means it is at the transfer node. Number of trains Perform seamless transfer operation. A value of 0 indicates that the transfer node Number of trains No seamless transfer operation is performed;

[0086] Step 222: Gather according to time point , establish a set of time intervals ; The specific process is:

[0087] remember For time point ,but For time point , For time point , Indicates time point and time point The time interval formed, ;

[0088] A time interval is a one-way arc connecting two consecutive points in time;

[0089]

[0090] In the formula, For time point , For time point ;

[0091] Step 223: Calculate the time point The bus departing from the station Number of carriages ; The specific process is:

[0092] The train can only be transformed at the disassembly station, and passengers can only be picked up and dropped off at the ordinary station. Therefore, the number of carriages sent out from each station is composed of the sum of the number of modules before entering the station and the number of changes in assembled and disassembled carriages.

[0093]

[0094] In the formula, For time point The bus departs on The number of modules assembled at the station, For time point The bus departs on The number of modules removed from the station;

[0095] For time point The bus departing from the station The number of carriages;

[0096] Step 23: Calculate the real-time passenger capacity of the vehicle ;

[0097] Step 24: Consider the situation where passengers are waiting at the station due to vehicle overload, and calculate the time point The bus arrives at the station Number of waiting passengers ;

[0098] Step 25: Calculate the number of transfer passengers between different transfer trains based on the seamless transfer service model of the line network and ;

[0099] Step 26: Build vehicle operation status ;

[0100] .

[0101] The other steps and parameters are the same as those in Specific Embodiments 1 to 3.

[0102] Specific implementation method 5: This implementation method is different from the specific implementation methods 1 to 4 in that the real-time number of passengers in the vehicle is calculated in steps 2 and 3. ; The specific process is:

[0103] Step 231: The number of passengers getting off at a station is the sum of the number of passengers planning to get off at all stations before the station. The number of passengers getting off the bus at station s is ;

[0104]

[0105] In the formula, For time point The bus departing from the station Number of passengers;

[0106] For the site You need to go to the station to get on the bus The proportion of passengers who get off the bus;

[0107] Step 232: The bus departs from the departure station, and passengers get on and off at each station. The real-time passenger capacity of the vehicle is calculated by the cumulative difference in the number of passengers getting on and off:

[0108]

[0109] In the formula, Indicates time point The bus arrives at the station The number of passengers on board at the time;

[0110] For time point The bus departing from the station The number of people getting off the bus.

[0111] The other steps and parameters are the same as those in Specific Embodiments 1 to 4.

[0112] Specific implementation method 6: This implementation method is different from the specific implementation methods 1 to 5 in that the step 24 takes into account the situation where the passenger is waiting at the station due to vehicle overload, and calculates the time point The bus arrives at the station Number of waiting passengers ; The specific process is:

[0113] Due to the limitation of module capacity and number of trains, the number of passengers may exceed the remaining capacity of the arriving vehicle. At this time, some passengers will not be able to board the bus, and the passengers stranded at the station will be forced to wait for the next bus.

[0114] Time The number of passengers on the bus dispatched at station s is expressed as:

[0115]

[0116] Based on time point The number of passengers on the bus at stop s , calculate the time point The bus was stranded at the station due to insufficient passenger capacity. Number of passengers ; expressed as:

[0117]

[0118] In the formula, Indicates time The remaining passenger capacity of the dispatched buses arriving at station s;

[0119] Indicates time The passenger capacity of the bus that arrives at station s;

[0120] For time point The bus arrives at the station Number of waiting passengers at the time;

[0121] Indicates time point The bus was stranded at the station due to insufficient passenger capacity. Number of passengers;

[0122] If the time interval is selected as the departure interval, then the time interval The number of passengers arriving at station s is :

[0123]

[0124] In the formula, For time point , For time point , For time point ;

[0125] Time The bus arrives at the station The number of waiting passengers at the previous time point The bus was stranded at the station due to insufficient passenger capacity. Number of passengers and Arrive at the station within the interval Number of passengers The sum of is expressed as:

[0126] .

[0127] The other steps and parameters are the same as those in Specific Implementation Methods 1 to 5.

[0128] Specific implementation method 7: This implementation method is different from the specific implementation methods 1 to 6 in that the number of transfer passengers between different transfer trains is calculated based on the seamless transfer service mode of the line network in step 25. and ; The specific process is:

[0129] Both lines have two transfer directions for passengers at each transfer station.

[0130] At the transfer station The number of people transferring is arrive or arrive The number of passengers in two transfer directions, that is, the number of transfer passengers between different transfer trains , ;

[0131] It is expressed as:

[0132]

[0133]

[0134] In the formula, Indicates the train number At the transfer station Number of people boarding at the previous stops; Indicates the train number At the transfer point Number of people boarding at the previous stops;

[0135] Indicates the number of passengers at each station from To train The transfer ratio; Indicates the number of passengers at each station from To train The transfer ratio;

[0136] express arrive Number of transfer passengers in the transfer direction, express arrive The number of transfer passengers in the transfer direction.

[0137] The other steps and parameters are the same as those in Specific Embodiments 1 to 6.

[0138] Specific implementation eight: This implementation differs from one of the specific implementations one to seven in that the step three is based on the database and vehicle operating status , establish a modular bus operation optimization model, that is, construct the objective function of minimizing the total cost, and solve the objective function of minimizing the total cost under the constraints to obtain The best operation plan ; The specific process is:

[0139] Considering the dynamic constraints of the number of carriages and passenger flow demand, the passenger waiting time cost of the system is calculated separately. , transportation time cost , Transfer time cost and vehicle operating costs , construct the objective function with the minimum system comprehensive cost, solve the modular bus operation optimization model, and obtain the optimal modular vehicle formation plan and the departure interval decision of the origin station ;

[0140] Step 31: Establish the objective function; the specific process is:

[0141] There are two main costs incurred by the operation of modular bus systems: system costs for vehicle operation and time costs related to passenger service quality. The passenger time cost is further divided into waiting time cost. , transportation time cost and the cost of transfer time , the direct cost of vehicle operation is .

[0142] Step 3: Calculate the cost of passenger waiting time ; The specific process is:

[0143] Passenger waiting time cost It consists of two parts. One is for the previous time that has not been boarded. The waiting time of the bus that was delayed at the station , and the second is the time interval The cumulative time of passengers arriving at station s within 1 hour;

[0144] Using the complex trapezoidal formula as an approximation, the interval between two consecutive departures is The cumulative passenger demand at station s is converted into a time function ;

[0145] Time interval Contains A set of time points, set the independent variable , function value Corresponding to The passenger demand at fixed intervals is , then the time interval The cumulative time of passengers arriving at station s for:

[0146]

[0147] In the formula, Indicates time interval The cumulative time of passengers arriving at station s within , express Time Passenger demand at fixed intervals, express Time Passenger demand at fixed intervals;

[0148] in,

[0149]

[0150]

[0151] In the formula, Indicates the site In the Fixed interval The number of passengers in Indicates the site In the Fixed interval The number of passengers in Indicates the site In the Fixed interval Number of passengers in the

[0152] Will , , Substitute the value of into the formula , sorted as follows:

[0153]

[0154] Therefore, the waiting time cost :

[0155]

[0156] Where: Indicates that the previous time was not logged in The bus was stranded at the station. of passengers’ waiting time, , Indicates the last time The bus was stranded at the station due to insufficient passenger capacity. Number of passengers;

[0157] Indicates the bus line number; represents a set of time arcs; represents the unit transportation time cost of passengers; represents the unit penalty cost of stranded waiting passengers;

[0158] Step 3.12: Calculate the time cost of transportation ; The specific process is:

[0159] The passenger's transportation time cost is affected by two main factors: the number of passengers and the travel time.

[0160]

[0161] In the formula, is the running time of buses between adjacent stops, , From the first station to the station The travel time, From the first station to the station Journey time;

[0162] Step 3: Calculate the cost of passenger transfer time ; The specific process is:

[0163] If the passenger fails to complete the seamless transfer operation at the transfer station, the time consumed during the transfer process mainly includes the additional loss time waiting for the connecting train and the inconvenience time of carrying luggage, which is recorded as ;

[0164]

[0165]

[0166]

[0167] Where: For train number of transfer passengers at transfer stations Time spent on transfers; For train number of transfer passengers at transfer stations Time spent on transfers;

[0168] Indicates the train number At the node The departure time of Indicates the train number At the node The departure time of Indicates the train number At the node The departure time of Indicates the train number At the node The departure time of

[0169] For train number of transfer passengers at transfer stations Time spent on transfers and train number of transfer passengers at transfer stations Time spent on transfers of and;

[0170] Combined with the transfer decisions of the vehicle marshaling plan at each station, the cost of passenger transfer time lost due to seamless transfer failure can be obtained. for:

[0171]

[0172] Where: is the unit time cost caused by the failure of seamless transfer;

[0173] Represents the set of all transfer stations in the road network;

[0174] It is a 0-1 variable. When the value is 1, it means it is at the transfer node. Number of trains Perform seamless transfer operation. A value of 0 indicates that the transfer node Number of trains No seamless transfer operation is performed;

[0175] is the total number of departures for all lines, For train number ;

[0176] Step 3-4: Calculate vehicle operating costs ;

[0177] The relationship between bus operators’ vehicle operating costs and their fleet size There is a direct linear correlation, and different numbers of module groups will produce corresponding unit operating costs and vehicle fixed usage costs:

[0178]

[0179] Where: represents the unit operating cost of a carriage with N modules;

[0180] represents the fixed cost of using a vehicle with N modules;

[0181] is the average speed of the module vehicle;

[0182] Step 315: Based on the cost of passenger waiting time , transportation time cost , Transfer time cost , Vehicle operating costs , the objective function of minimizing the comprehensive cost is constructed as ;

[0183] Step 32: Set constraints.

[0184] The other steps and parameters are the same as those in Specific Embodiments 1 to 7.

[0185] Specific implementation method 9: This implementation method is different from the specific implementation methods 1 to 8 in that the step 315 is based on the passenger waiting time cost. , transportation time cost , Transfer time cost , Vehicle operating costs , the objective function of minimizing the comprehensive cost is constructed as ; The specific process is:

[0186] .

[0187] The other steps and parameters are the same as those in Specific Embodiments 1 to 8.

[0188] Specific implementation method 10: This implementation method is different from specific implementation methods 1 to 9 in that the constraint conditions are set in step 32; the specific process is:

[0189] Consider the dynamic relationship between the number of module vehicles and passenger flow demand, transfer time threshold and other constraints.

[0190] ① The actual technical limitations and the impact of the length of the marshaling on the driving safety of the vehicle, the number of marshaling is not greater than the maximum suspension number threshold:

[0191]

[0192] In the formula, is the maximum hanging quantity threshold;

[0193] ② Modular vehicles only allow the decoupling / decoupling of the carriages at the disassembly and assembly sites, with the help of auxiliary variables Relevant constraints for changing carriage groups:

[0194]

[0195] In the formula, For time point Buses departing from the station The number of modules assembled; For time point Buses departing from the station Number of modules removed; , is a positive integer; is a very large integer;

[0196] It is a 0-1 variable. When the value is 1, it means that the site s is a split site, and when the value is 0, it means that it is a general site.

[0197] ③ In order to ensure the effective operation of the bus and the passengers' satisfaction with the modular bus, the number of modular seats removed from the disassembly station should not be less than the number of passengers getting off the bus, so as to ensure that passengers can get off the bus smoothly at the destination station:

[0198]

[0199] In the formula, is the passenger capacity of a module vehicle;

[0200] ④ To avoid disruption of the original operating schedule by transfers, and to ensure the stability of the operating system and the balance of bus arrival intervals, the difference in arrival time between two buses at the transfer node cannot exceed the time threshold interval. :

[0201]

[0202] In the formula, For train number At the transfer station The departure time, For train number At the transfer station The departure time of

[0203] ⑤ The seamless transfer operation between modular bus modules requires the modules to be disassembled in advance. Therefore, the number of modules in the previous station of the transfer station should be greater than the number of modules required to be disassembled due to the transfer reorganization, and the remaining car capacity should meet the passenger demand;

[0204] The number of modules that need to be transformed at the transfer nodes of each line is calculated based on the number of transfer passengers between trains. The number of modules is an integer and needs to be rounded:

[0205]

[0206]

[0207] In the formula, is the passenger capacity of a module vehicle;

[0208] For train number At the transfer station The number of modules that need to be transformed, For train number At the transfer station The number of modules that need to be transformed;

[0209] , It is to round up the calculation result;

[0210] express arrive Number of transfer passengers in the transfer direction, express arrive Number of transfer passengers in the transfer direction;

[0211]

[0212]

[0213] In the formula, For train number At the transfer station The number of departure modules of the previous station; For train number At the transfer station The number of departure modules of the previous station;

[0214] For train number Arrival site The number of passengers on board at the time; For train number Arrival site The number of passengers on the bus at that time.

[0215] The other steps and parameters are the same as those in Specific Embodiments 1 to 9.

[0216] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A modular bus operation compilation method based on a variable rolling time window, characterized in that: The specific process of the method is: Step 1: Start the operation and take the current time t curr As the start time of the operation cycle window, t0 = t curr and establish a time point set; Collect the time t0 and the operation cycle [t0, t0+t g ] window, based on the time t0 and the operation cycle [t0, t0+t g ] The passenger flow arrival information within the window is used to obtain the passenger demand matrix B; Based on bus line L and station data Transfer station data Passenger demand matrix B, the proportion of people getting off at each station ξ z,s and the ratio of passengers transferring between linesζ m,m′,s′ , forming database I curr ; Step 2: Database I based on step 1 curr , calculate the vehicle running status S curr : Step 3: Based on database I curr and vehicle operating status S curr , establish a modular bus operation optimization model, that is, construct the objective function of minimum total cost, solve the objective function of minimum total cost under the constraints and obtain t g The best operation plan G i (Δt i,j ,N i,s ); Step 4: Set the total scheduling cycle length t d , then t end =t star +t d ;t d The length of planning for bus system operations; Starting time t curr =t star , Step 1: Obtain passenger flow arrival information data I curr , Step 2: Obtain vehicle running status data S curr , step 3 obtain G i (Δt i,j ,N i,s ); Take the departure interval set Δt i,j The first value of the rolling window is taken as the time length Δt1, and N i,s As carriage set N1; When t star <t curr <t star At +Δt1, the bus executes carriage formation N1; When t curr =t star +Δt1, re-execute steps 1 to 3 to obtain a new round of operation compilation plan G i (Δt i,j ,N i,s ), take the new round of operation compilation plan G i (Δt i,j ,N i,s ) in the departure interval set Δt i,j The first value of is used as the time length of the rolling window Δt2; take the new round of operation compilation plan G i (Δt i,j ,N i,s ) i,s As carriage formation N2; When t star +Δt1<t curr <t star +Δt1+Δt2, the bus executes carriage formation N2; When t curr =t star +Δt1+Δt2, re-execute steps 1 to 3 to obtain a new round of operation compilation plan G i (Δt i,j ,N i,s ), take the new round of operation compilation plan G i (Δt i,j ,N i,s ) in the departure interval set Δt i,j The first value of is used as the time length of the rolling window Δt3; take the new round of operation compilation plan G i (Δt i,j ,N i,s ) i,s As carriage formation N3; Until t curr =t end , that is, when the time window rolls to the end of the planned time, the rolling time domain terminates the response and the operation ends; obtain the departure interval table of the starting station within the planned time and the train marshaling plan for each train station In step 1, the operation starts, and the current time t is taken curr As the start time of the operation cycle window, t0 = t curr and establish a time point set; Collect the time t0 and the operation cycle [t0, t0+t g ] window, based on the time t0 and the operation cycle [t0, t0+t g ] The passenger flow arrival information within the window is used to obtain the passenger demand matrix B; Based on bus line L and station data Transfer station data Passenger demand matrix B, the proportion of people getting off at each station ξ z,s and the ratio of passengers transferring between linesζ m,m′,s′ , forming database I curr ; The specific process is: Step 1. Take the current time t curr As the start time of the operation cycle window, t0 = t curr ; The up-and-down process of a bus from the starting station to the terminal and then back to the starting station is regarded as a complete trip; Bus stops from 1 to To number, s represents the sth bus stop, Indicates the total number of bus stops; Number all bus routes l in the road network from 1 to L, l = 1, 2, ..., L, l represents the lth bus route, and L represents the total number of bus routes; Intersections of lines in the road network will generate transfer stations. s′ represents the s′th transfer station, Represents all transfer stations in the road network; M=(M 1 ,M 2 ,...,M l ,...,M L ) is the total number of departures of all lines; M l Represents the set of departure times of the lth bus route; Step 1 and 2: Establish a time point set, and obtain a passenger arrival demand matrix based on the time point set; Step 13: Based on the total number of bus routes L and the total number of bus stops All transfer stations in the network Passenger demand matrix B, the proportion of people getting off at each station ξ z,s and the ratio of passengers transferring between linesζ m,m′,s′ , forming database I curr ; In the steps 1 and 2, a time point set is established, and a passenger arrival demand matrix is ​​obtained based on the time point set. The specific process is as follows: Step 121: If the operating time required for a bus to complete a trip is t g , and set t at fixed intervals δ g Divide into n time points to form a time point set T; T={t0,t0+δ,t0+2δ,...,t0+kδ,...,t0+nδ},k∈{0,1,...,n} Where: t0 and t0+nδ are the start and end time of vehicle operation respectively; Step 122: Collect the time t0 and the operation period [t0, t0+t g ] window, based on the time t0 and the operation cycle [t0, t0+t g ] The passenger flow in the window is obtained by operating time t g The passenger arrival situation at each station, that is, the passenger demand matrix B; In the formula, is the travel time from the first station to station s; represents the number of passengers at station s at time t0; represents the number of passengers at station s in the kth fixed interval δ; represents the number of passengers at station s in the first fixed interval δ; represents the number of passengers at station s in the nth fixed interval δ; B s is the passenger demand matrix of station s; For Site Passenger demand matrix; The database I in step 2 is based on step 1 curr , calculate the vehicle running status S curr ; The specific process is: Step 21: Design a service model for flexible disassembly of single lines and seamless transfer between lines; the specific process is as follows: Flexible disassembly on a single line: allows modular vehicles to be coupled or decoupled at disassembly stations, while keeping the number of carriage modules unchanged at general stations; The splitting and combining sites are the top 25% of all the values ​​of the attraction weight coefficient; In the formula, F a,b is the attraction weight coefficient of site b to site a, O a is the total number of passengers boarding at station a, D b is the total number of people getting off at station b; Seamless transfer between lines: In the cross transfer, the two lines intersect in the up and down directions to form four transfer stations, at which the modular buses can realize seamless transfer operations at the same station; Overlapping transfer lines will form two transfer stations, one separating and one merging; Step 2: Define decision variables and calculate the departure interval Δt i,j and the number of carriages N of the bus departing at time i at station s i,s ; The specific process is: Step 221: Define decision variables; the specific process is: α s It is a 0-1 variable. When the value is 1, it means that the site s is a split site, and when the value is 0, it means that it is a general site. X i,j It is a 0-1 variable, and when it takes the value 1, it represents the time interval Δt i,j Selected as the departure interval, a value of 0 indicates Δt i,j Not selected as the departure interval; β m,m′,s′ It is a 0-1 variable. When the value is 1, it means that the trains m and m' perform seamless transfer at the transfer node s'. When the value is 0, it means that the trains m and m' do not perform seamless transfer at the transfer node s'. Step 222: Based on the time point set T, establish the time interval set T t ; The specific process is: Let t0+kδ be the time point t(k), then t0+iδ is the time point t(i), t0+jδ is the time point t(j), Δt i,j represents the time interval between time point t(i) and time point t(j), Δt i,j ∈T t ; A time interval is a one-way arc connecting two consecutive points in time; Δt i,j =t(j)-t(i),t(0)≤t(i)<t(j)≤t(n) Where, t(n) is the time point t0+nδ, t(0) is the time point t0; Step 223: Calculate the number of carriages N of the bus departing at time point t(i) at station s i,s ; The specific process is: In the formula, is the number of modules assembled at station s for the bus departing at time point t(i), is the number of modules disassembled at station s by the bus departing at time t(i); N i,s-1 is the number of carriages of the bus departing at time t(i) at station s-1; Step 23: Calculate the real-time passenger capacity of the vehicle O i,s ; Step 24: Calculate the number of waiting passengers W when the bus departing at time point t(j) arrives at station s j,s ; Step 25: Calculate the number of transfer passengers g between different transfer trains m,m′,s′ and g m′,m,s′ ; Step 26: Construct vehicle operation status S curr ; S curr =(N i,s ,L i,s ,O i,s ,x i,j ,b m,m′,s′ ,W j,s ,g m,m′,s′ ) In the steps 2 and 3, the real-time number of passengers carried by the vehicle is calculated. i,s ; The specific process is: Step 231: The number of passengers getting off the bus at stop s at time point t(i) is V i,s ; Where U i,z is the number of passengers at station z for the bus departing at time point t(i); ξ z,s is the proportion of passengers who board at station z and need to get off at station s; Step 232: Calculate the real-time number of passengers in the vehicle: In the formula, O i,s represents the number of passengers on the bus when the bus departs at time t(i) arrives at station s; V i,z is the number of passengers who get off the bus at station z at time point t(i); In step 24, the number of waiting passengers W is calculated when the bus departing at time point t(j) arrives at station s. j,s ; The specific process is: The number of passengers on the bus departing at time point t(i) at station s is expressed as: U i,s =min(W i,s ,CN i,s -(THE i,s -V i,s )) The number of passengers U at the bus stop s based on the time point t(i) i,s , calculate the number of passengers L that are stranded at station s due to insufficient passenger capacity of the bus departing at time point t(i) i,s ; expressed as: In the formula, CN i,s -(O i,s -V i,s ) represents the remaining passenger capacity of the bus departing at time i and arriving at station s; CN i,s represents the passenger capacity of the bus departing at time i and arriving at station s; W i,s is the number of waiting passengers when the bus departing at time t(i) arrives at station s; L i,s represents the number of passengers stranded at station s due to insufficient passenger capacity of the bus departing at time t(i); If the time interval Δt i,j is selected as the departure interval, then the time interval Δt i,j The number of passengers arriving at station s is d i,j,s : Where t(k) is the time point t0+kδ ， t(i) is the time point t0+iδ, t(j) is the time point t0+jδ; The number of waiting passengers when the bus departing at time point t(j) arrives at station s is the number of passengers L who were stranded at station s due to insufficient passenger capacity of the bus departing at the previous time point t(i) i,s and Δt i,j The number of passengers arriving at station s during the interval d i,j,s The sum of is expressed as: W j,s =d i,j,s +L i,s In step 25, the number of transfer passengers g between different transfer trains is calculated. m,m′,s′ and g m′,m,s′ ; The specific process is: The number of people transferring at the transfer station s′ is composed of the number of passengers transferring from m to m′ or from m′ to m, that is, the number of transfer passengers between different transfer trains g m,m′,s′ , g m′,m,s′ ; It is expressed as: Where U z,m,s′ represents the number of passengers boarding the train m at each station before the transfer station s′; U z,m′,s′ represents the number of passengers boarding the train m′ at each station before the transfer node s′; ζ m,m′,s′ represents the transfer ratio of passengers from train number m to train number m′ at each station; ζ m′,m,s′ represents the transfer ratio of passengers from train number m′ to train number m at each station; g m,m′,s′ represents the number of transfer passengers from m to m′, g m′,m,s′ represents the number of transfer passengers from m′ to m; In the step 3, based on database I curr and vehicle operating status S curr , establish a modular bus operation optimization model, that is, construct the objective function of minimum total cost, solve the objective function of minimum total cost under the constraints and obtain t g The best operation plan G i (Δt i,j ,N i,s ); the specific process is: Step 31: Establish the objective function; the specific process is: Step 3: Calculate the passenger waiting time cost Z D ; The specific process is: Time interval Δt i,j Contains A set of time points, set the independent variable Function value f(X p ) corresponds to the passenger demand of the (i+p)th fixed interval, that is, Then the time interval Δt i,j The cumulative time of passengers arriving at station s is: Where D i,j,s Represents the time interval Δt i,j The cumulative time of passengers arriving at station s within 10 seconds, f(i) represents the passenger demand at the ith fixed interval when p = 0, and f(j) represents The passenger demand at the jth fixed interval at time ; in, In the formula, represents the number of passengers at station s within the i-th fixed interval δ, represents the number of passengers at station s in the jth fixed interval δ, represents the number of passengers at station s in the (i+p)th fixed interval δ; f(X p ), f(i), and f(j) are substituted into the formula Arranged as: Therefore, the waiting time cost Z D : Where: E i,j,s represents the waiting time of passengers who are stranded at station s but fail to board the bus that left at the previous time t(i), E i,j,s =L i,s ×pδ,L i,s represents the number of passengers stranded at station s due to insufficient passenger capacity of the bus that departed at the previous time t(i); L represents the bus line number; A t represents the set of time arcs; φ represents the unit transportation time cost of passengers; represents the unit penalty cost of stranded waiting passengers; Step 3.12: Calculate the travel time cost Z O ; The specific process is: In the formula, r i,s is the running time of buses between adjacent stops, is the travel time from the first station to station s, is the travel time from the first station to station s-1; Step 3: Calculate the passenger transfer time cost Z g ; The specific process is: T m,m′,s′ =T m,s′ +T m′,s′ Where: T m,s′ is the transfer time spent by the transfer passengers of train number m at the transfer station s′; T m′,s′ is the transfer time spent by the transfer passengers of train number m′ at the transfer station s′; t m,s′ represents the departure time of train number m at node s′; t m′,s′ represents the departure time of train number m′ at node s′; t m+1,s′ represents the departure time of train number m+1 at node s′; t m′+1,s′ represents the departure time of train number m′+1 at node s′; T m,m′,s′ is the transfer time T spent by the transfer passengers of train number m at the transfer station s′ m,s′ The transfer time T spent by the transfer passengers of train number m′ at the transfer station s′ m′,s′ of and; Passenger transfer time cost Z g for: Where: η is the unit time loss cost caused by the failure of seamless transfer; Represents the set of all transfer stations in the road network; β m,m′,s′ It is a 0-1 variable. When the value is 1, it means that the trains m and m' perform seamless transfer at the transfer node s'. When the value is 0, it means that the trains m and m' do not perform seamless transfer at the transfer node s'. M is the total number of departures of all lines, and m is the train number m; Step 3-4: Calculate vehicle operating cost Z V ; Where: C N represents the unit operating cost of a carriage with N modules; F N represents the fixed cost of using a vehicle with N modules; v is the average speed of the module vehicle; Step 315: Based on the passenger waiting time cost Z D , transportation time cost Z O , transfer time cost Z g , Vehicle operating cost Z V , the objective function of minimizing the comprehensive cost is constructed as minZ; Step 32: Set constraints; In step 315, based on the passenger waiting time cost Z D , transportation time cost Z O , transfer time cost Z g , Vehicle operating cost Z V , the objective function of minimizing the comprehensive cost is constructed as minZ; the specific process is: In step 32, constraints are set; the specific process is: ① The number of groups is not greater than the maximum suspension number threshold: Where N max is the maximum hanging quantity threshold; ② With the help of auxiliary variables Relevant constraints for changing carriage groups: In the formula, is the number of modules assembled at station s for the bus departing at time t(i); is the number of modules disassembled at station s by the bus departing at time t(i); is a very large integer; α s It is a 0-1 variable. When the value is 1, it means that the site s is a split site, and when the value is 0, it means that it is a general site. ③The number of modular seats removed at the disassembly station should not be less than the number of people getting off the bus: In the formula, C is the passenger capacity of a modular vehicle; ④ The difference between the arrival time of two trains at the transfer node cannot exceed the time threshold interval [-t0, t0]: -t0≤t m,s′ -t m′,s′ ≤t0 In the formula, t m,s′ is the departure time of train number m at transfer station s′, t m′,s′ is the departure time of train number m′ at transfer station s′; ⑤ The number of module groups at the previous station of the transfer station should be greater than the number of modules required to be disassembled due to the transfer reorganization, and the remaining carriage capacity should meet the passenger demand; The number of modules that need to be transformed at the transfer nodes of each line is calculated based on the number of transfer passengers between trains. The number of modules is an integer and needs to be rounded: In the formula, C is the passenger capacity of a modular vehicle; B m,s′ is the number of modules that train number m needs to change at transfer station s′, B m′,s′ is the number of modules that train m′ needs to transform at transfer station s′; It is to round up the calculation result; g m,m′,s′ represents the number of transfer passengers from m to m′, g m′,m,s′ represents the number of transfer passengers from m′ to m; (n m,s′-1 -B m,s′ )×C≥O m,s′ (n m′,s′-1 -B m′,s′ )×C≥O m′,s′ Where n m,s′-1 is the number of departure modules of train number m at the previous station of transfer station s′; n m′,s′-1 is the number of departure modules of train number m′ at the station before the transfer station s′; O m,s′ is the number of passengers on the bus when bus number m arrives at station s′; O m′,s′ is the number of passengers on the bus when bus number m′ arrives at station s′.

Images