Scheduling optimization method for intelligent rail train connected with high-speed rail

Abstract

The invention relates to the technical field of train dispatching, and discloses a dispatching optimization method for an intelligent rail train connected with a high-speed rail. The method comprises the following steps: acquiring intelligent rail train operation data and passenger flow OD data in a set time period; based on the intelligent rail and high-speed rail connection scene, taking the passenger travel time cost and the intelligent rail running time cost as optimization targets, and taking the intersection constraint, the timetable constraint and the passenger flow constraint as constraint conditions, constructing an intelligent rail train dispatching multi-target optimization model considering high-speed rail station connection; solving the intelligent rail train dispatching multi-objective optimization model considering high-speed rail station connection to obtain a Pareto optimal solution set; and determining a train scheduling optimal solution according to the Pareto optimal solution set, and generating a corresponding intelligent rail train working diagram. According to the method, the train dispatching optimization problem of the intelligent rail line connected with the high-speed rail station can be better solved, and the travel experience of passengers and the operation service efficiency of intelligent rail enterprises are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of train dispatching, and in particular to a method for optimizing dispatching of a smart rail train connected to a high-speed railway. Background Art

[0002] Smart rail transportation has brought new options for solving urban transportation problems. For large cities, smart rail trains can serve as a supplement to large-city rail transit and improve the convenience of urban residents' travel; for small and medium-sized cities, smart rail trains can play the main role of urban public transportation, combined with the existing conventional bus system, to jointly serve the travel of urban residents. However, due to the short time from research and development to commissioning of smart rail trains, compared with urban rail transit and traditional buses, smart rail trains lack operating experience. From the perspective of operation and management, the preparation of the operation plan of smart rail trains still needs to be optimized, which directly affects the operating efficiency of smart rail trains, the operating costs of smart rail companies, and the travel satisfaction of passengers.

[0003] At present, most intelligent rail trains adopt a single route and stop-at-every-station operation organization mode. The compilation of their train timetables is mainly based on the significant characteristics of the section passenger flow during the morning and evening peak hours, and adopts a time-divided equal-interval compilation strategy. For ordinary stations on the intelligent rail line, the passenger flow entering the station is evenly distributed, so this train timetable compilation strategy can effectively cope with passenger flow fluctuations, ensure the stability and efficiency of operations, and has strong adaptability. However, for the intelligent rail connecting the high-speed rail station, its passenger flow mainly comes from the arrival of the high-speed rail. Due to the discrete and non-balanced arrival of railway trains, the conventional intelligent rail train timetable is difficult to adapt to the time-varying passenger flow demand, which will cause passengers to wait too long at the station, which is not conducive to the evacuation of passenger flow and increases safety hazards; with the uneven distribution of passenger flow and the continuous increase in passenger flow demand, this traditional operation organization model will also lead to problems such as waste of transportation capacity and reduced passenger travel experience. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for optimizing the scheduling of smart rail trains connected to high-speed railways, so as to better solve the train scheduling optimization problems of the smart rail lines connected to high-speed railway stations, and improve the travel experience of passengers and the operational service efficiency of smart rail enterprises.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] A method for optimizing the dispatching of a smart rail train connected to a high-speed railway comprises the following steps:

[0007] Obtain the operation data of smart rail trains and passenger flow OD data within a set period;

[0008] Based on the scenario of intelligent rail and high-speed rail connection, the multi-objective optimization model of intelligent rail train scheduling considering high-speed rail station connection is constructed with passenger travel time cost and intelligent rail running time cost as optimization objectives, and route constraints, timetable constraints and passenger flow constraints as constraints.

[0009] Solve the multi-objective optimization model of intelligent rail train scheduling considering high-speed rail station connection and obtain the Pareto optimal solution set;

[0010] The optimal solution for train scheduling is determined based on the Pareto optimal solution set, and the corresponding smart rail train operation diagram is generated.

[0011] Preferably, the smart rail train operation data includes:

[0012] Operating speed, interval operating time, stop time and maximum passenger capacity data.

[0013] As a preferred embodiment, the passenger travel time cost is taken as the optimization target, specifically:

[0014] minZ1=T wait +T ivt +T crowd

[0015]

[0016]

[0017] Among them, min is the minimization function, Z1 is the passenger travel time cost, T wait is the waiting time cost, T ivt is the time cost in transit, T crowd is the congestion time cost, K is the set of smart rail train numbers, T is the set of discretized set time ranges, N is the number of stations on the smart rail line, is the departure time of smart rail train k at station i, f k，i，j，t is the number of passengers arriving at station i at time t and heading to station j who take train k, i is the weight of station i, Q is a positive integer, T max is the maximum waiting time, Passengers arriving at station i at time t and heading to station j take the virtual smart rail train K v Number of people, is the arrival time of smart rail train k at station j, Additional cost for congestion, p k，i is the number of people in the smart rail train k when it departs from station i, r i is the running time of the smart rail train between stations i and i+1.

[0018] As a preferred method, the travel time cost of the smart rail is used as the optimization target, specifically:

[0019] minZ2=T operator

[0020]

[0021] Among them, min is the minimization function, Z2 is the travel time cost of the smart track, T operator is the time conversion cost, K is the number of smart rail trains, is a 0-1 variable indicating whether the smart rail train k chooses route h1 to run, N is the number of stations on the smart rail line, r i is the running time of the smart rail train between stations i and i+1, s i is the dwelling time of the smart rail train at station i, ε k，i is a 0-1 variable indicating whether the smart rail train k stops at station i, is a 0-1 variable indicating whether the smart rail train k chooses route h2 to run, m is the starting station number of route h2, and n is the terminal station number of route h2.

[0022] As a preferred embodiment, the route constraints include route constraints and train stop range constraints;

[0023] The route opening constraints are specifically:

[0024]

[0025] Among them, z k,h is a 0-1 variable indicating whether the smart rail train k chooses route h to run, H is the route set, A 0-1 variable indicating whether the smart rail train k chooses route h3 for operation;

[0026] The specific constraints of the train stop range are:

[0027]

[0028] in, is a 0-1 variable indicating whether the smart rail train k chooses route h1 to run. is a 0-1 variable indicating whether the smart rail train k chooses route h2 for operation. is a 0-1 variable indicating whether the smart rail train k chooses route h3 to run, ε k,i A 0-1 variable indicating whether the smart rail train k stops at station i.

[0029] Preferably, the timetable constraints include smart rail train operation time range constraints, safety interval constraints, interval operation time constraints and stop time constraints;

[0030] The specific operating time range constraints of the smart rail train are:

[0031]

[0032] Among them, T e This is the earliest time when the smart rail train will be put into operation. is a 0-1 variable indicating whether the smart rail train k chooses route h3 to run. is the departure time of smart rail train k at station i, T l The latest time for the operation of the Smart Rail train;

[0033] The safety interval constraints are specifically:

[0034]

[0035]

[0036] Among them, h min is the minimum safe interval of the smart rail train at the station, Q is a set positive integer, is a 0-1 variable indicating whether the smart rail train k chooses route h3 to run. is the departure time of smart rail train k at station i, h max The maximum safe interval for smart rail trains at stations. is a 0-1 variable indicating whether the smart rail train k chooses route h1 to run. is the arrival time of smart rail train k at station i, and M is the maximum number of smart rail trains;

[0037] The interval running time constraints are specifically:

[0038]

[0039] Among them, r i is the running time of the smart rail train between stations i and i+1;

[0040] The stop time constraint is specifically:

[0041]

[0042] Among them, ε k,i is a 0-1 variable indicating whether the smart rail train k stops at station i, s i is the dwelling time of the smart rail train at station i.

[0043] Preferably, the passenger flow constraints include passenger flow distribution constraints, passenger boarding and alighting number constraints, and train capacity constraints;

[0044] The passenger flow allocation constraints are specifically:

[0045] fk,i,j,t ≤δ·C·ε k,i ·ε k,j

[0046]

[0047] 0≤w k,i,j,t ≤T max ,iff k,i,j,t >0

[0048]

[0049] Among them, f k，i，j，t is the number of passengers arriving at station i at time t and going to station j who take the smart rail train k, δ is the capacity limit coefficient of the smart rail train, C is the number of passengers on the smart rail train, and ε k , i is a 0-1 variable indicating whether the smart rail train k stops at station i, ε k，j is a 0-1 variable indicating whether the smart rail train k stops at station j, W k,i，j，t is the waiting time for the smart rail train k for passengers arriving at station i and going to station j at time t, is the departure time of the kth smart rail train at station i, T max is the maximum waiting time, K is the number of smart rail trains, Passengers arriving at station i at time t and heading to station j take the virtual smart rail train K v The number of people, F i，j (t) is the passenger flow from station i to station j at time t;

[0050] The specific constraints on the number of passengers getting on and off the bus are:

[0051]

[0052] in, is the number of people boarding the smart rail train k at station i to station j, is the number of people boarding the smart rail train k at station i, N is the number of stations on the smart rail line, is the number of passengers getting off the smart rail train k at station i, is the number of people who boarded the smart rail train k at station j and went to station i;

[0053] The train capacity constraint is specifically:

[0054] p k,i ≤δ·C

[0055] Among them, p k，i is the number of people in the smart rail train k when it departs from station i, δ is the capacity limitation coefficient of the smart rail train, and C is the number of passengers on the smart rail train.

[0056] As a preferred method, solving the multi-objective optimization model for smart rail train scheduling considering high-speed rail station connection includes:

[0057] A non-dominated sorting genetic algorithm with an elite strategy is used to solve the multi-objective optimization model of smart rail train scheduling considering high-speed railway station connections.

[0058] As a preferred method, a non-dominated sorting genetic algorithm with an elite strategy is used to solve the multi-objective optimization model of smart rail train scheduling considering high-speed rail station connection.

[0059] The route selection and running interval of smart rail trains are taken as decision variables, and the chromosome encoding is carried out using integer coding.

[0060] As a preferred method, a non-dominated sorting genetic algorithm with an elite strategy is used to solve the multi-objective optimization model of smart rail train scheduling considering high-speed rail station connection.

[0061] The individual fitness value is calculated based on the passenger travel time cost and the smart rail travel time cost.

[0062] The present invention has the following beneficial effects:

[0063] 1. The present invention proposes a method for optimizing the dispatching of smart rail trains connected to high-speed rail. Considering the proportion of high-speed rail station connecting passenger flow in the overall passenger flow of the smart rail line at different time periods, a certain weight is given to it when the optimization model is established to indicate the importance of high-speed rail station connecting passenger flow at different time periods of the smart rail line, which is more in line with the actual situation.

[0064] 2. The present invention proposes an optimization method for smart rail trains that comprehensively considers the large and small route plans and stop plans of smart rail trains. Compared with existing studies, instead of optimizing the large and small route plans and stop plans separately, the two are combined, which is conducive to the optimization of the overall operation plan of the smart rail train.

[0065] 3. The present invention outputs the Pareto optimal solution set, which can provide enterprises with multiple optimal selection options. Enterprises can choose the corresponding smart rail train scheduling plan based on the investment cost. On the basis of the Pareto optimal solution set, they can further select the enterprise optimal solution, the trade-off optimal solution and the passenger optimal solution from different angles and compare them with the actual operation plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A flow chart of a method for optimizing the dispatching of smart rail trains connected to high-speed rails;

[0067] Figure 2 This is a schematic diagram of the Zhigui line station;

[0068] Figure 3Flowchart for solving the model based on the NSGA-II algorithm. DETAILED DESCRIPTION

[0069] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0070] like Figure 1 As shown, an embodiment of the present invention provides a method for optimizing the scheduling of a smart rail train connected to a high-speed railway, comprising the following steps S1 to S4:

[0071] S1. Obtain the operation data of the smart rail train and the passenger flow OD data within the set time period;

[0072] In an optional embodiment of the present invention, the smart rail train operation data acquired in step S1 includes:

[0073] Operating speed, interval operating time, stop time and maximum passenger capacity data.

[0074] This embodiment first collects and processes the required data. Specifically, the actual operation plan, operation speed, interval operation time, stop time, maximum passenger capacity and other data of the smart rail train are collected, and the entry and exit data of each station on the smart rail line are obtained from the relevant departments of the smart rail enterprise, and the passenger flow OD data during the study period is obtained after processing.

[0075] S2. Based on the scenario of intelligent rail and high-speed rail connection, the multi-objective optimization model of intelligent rail train scheduling considering high-speed rail station connection is constructed with passenger travel time cost and intelligent rail running time cost as optimization targets, and route constraints, timetable constraints and passenger flow constraints as constraints;

[0076] In an optional embodiment of the present invention, step S2 takes a smart rail line connecting a high-speed railway station as the research object, takes the connection with the high-speed railway station as a consideration, optimizes the comprehensive route plan and the stop plan, and constructs a multi-objective optimization model from two aspects: the passenger travel time cost and the smart rail travel time cost.

[0077] Specifically, a one-way intelligent rail line with N intelligent rail stations is constructed. Station g is the station where the high-speed rail and the intelligent rail line connect. Station g can be any station on the line, such as Figure 2 As shown in the figure, model-related assumptions are proposed, the comprehensive route plan and stop plan are optimized, and a multi-objective optimization model is constructed from two aspects: passenger travel time cost and smart rail travel time cost.

[0078] This embodiment makes assumptions about the passenger part and train operation related to the model; it is assumed that except for the passengers transferring to the smart rail at the high-speed railway station, the rest of the smart rail passengers arrive randomly and the arriving passenger flow is known; all passengers choose to take the direct smart rail train, and the situation of transferring on the main line in the middle is not considered; the platform capacity of the smart rail station is sufficient to accommodate stranded passengers; and it is also assumed that all smart rail trains use locomotives and vehicles of the same composition and performance; smart rail trains in the same direction arrive first and depart first, and only one smart rail train is allowed to stay at the station at the same time; all smart rail trains have the same running time between the same stations and the same stopping time at the same station; there is no situation of overtaking or avoiding trains on the smart rail trains; the starting and ending stations of large and small intersections on the line have the conditions for turning back.

[0079] In the multi-objective optimization model constructed in this embodiment, the passenger travel time cost includes waiting time cost, en route time cost and congestion cost. Therefore, the passenger travel time cost is taken as the optimization objective as follows:

[0080] minZ1=T wait +T ivt +T crowd

[0081] Among them, min is the minimization function, Z1 is the passenger travel time cost, T wait is the waiting time cost, T ivt is the time cost in transit, T crowd The cost of congestion time.

[0082] This embodiment takes into account the particularity of high-speed rail and smart rail transfer stations, and divides the calculation of the total waiting time of passengers into two parts: the total waiting time of passengers at high-speed rail and smart rail transfer stations and the total waiting time of passengers at other ordinary smart rail stations. Set the weight of the station to λ i , by adjusting the weight λ i The different values ​​of can set the importance and status of each station in the whole line of intelligent rail stations. When i = g, it represents the weight coefficient λ of the high-speed rail and intelligent rail connecting station g The total waiting time for passengers is as follows:

[0083]

[0084] Among them, T is the discretized set time range, N is the number of stations on the smart rail line, is the departure time of the kth smart rail train at station i, f k，i，j，t is the number of passengers arriving at station i at time t and heading to station j who take train k, i is the weight of station i.

[0085] In this embodiment, the maximum waiting time that passengers can tolerate is set to T maxIf the time exceeds this range, the passenger satisfaction will be greatly reduced, so passengers will choose to leave the station and take other means of transportation. v , its capacity is infinite, when the waiting time of passengers exceeds T max When the virtual train is launched, passengers will board the virtual train. The waiting time for some passengers on the virtual train is as follows:

[0086]

[0087] Among them, Q is a set positive integer, T max is the maximum waiting time, Passengers arriving at station i at time t and heading to station j take the virtual smart rail train K v Number of people.

[0088] Therefore, by weighting the total waiting time of passengers, the waiting time cost is obtained as follows:

[0089]

[0090] Among them, K is the set of smart rail train numbers.

[0091] In this embodiment, the transit time cost is specifically:

[0092]

[0093] in, is the arrival time of smart rail train k at station j.

[0094] In this embodiment, the congestion cost refers to the cost caused by congestion perceived by passengers when the smart rail train reaches a certain level of congestion. The cross-sectional passenger flow p of the kth smart rail train in the section between station i and station i+1 is k，i , that is, the number of passengers in the kth smart rail train when it departs from station i is expressed as:

[0095]

[0096] Among them, i′ is the boarding station of the passengers who are already on board when the kth smart rail train arrives at station i.

[0097] The additional charges due to congestion are calculated as follows:

[0098]

[0099] in, Additional cost for congestion, p k，iis the number of people in the smart rail train k when it departs from station i, α is the crowding correction parameter when the train capacity does not exceed the train capacity, β is the crowding correction parameter when the train capacity exceeds the train capacity, B is the number of seats in the smart rail train, C is the capacity of the smart rail train, and 6 is the capacity limitation coefficient of the smart rail train.

[0100] Crowding in the car will lead to a deviation in passengers' perception of the ride time. The time cost of crowding can be calculated by converting the degree of crowding into:

[0101]

[0102] Among them, r i is the running time of the smart rail train between stations i and i+1.

[0103] In the multi-objective optimization model constructed in this embodiment, the travel time cost of the smart rail is taken as the optimization target, specifically:

[0104] minZ2=T operator

[0105] Among them, min is the minimization function, Z2 is the travel time cost of the smart track, T operator Discount the cost for time.

[0106] The calculation method of time conversion cost is:

[0107]

[0108] Among them, T operator is the time conversion cost, K is the number of smart rail trains, is a 0-1 variable indicating whether the smart rail train k chooses route h1 to run, N is the number of stations on the smart rail line, r i is the running time of the smart rail train between stations i and i+1, s l is the dwelling time of the smart rail train at station i, ε k，i is a 0-1 variable indicating whether the smart rail train k stops at station i, is a 0-1 variable indicating whether the smart rail train k chooses route h2 to run, m is the starting station number of route h2, and n is the terminal station number of route h2. In the above formula, the first part represents the running time cost of the smart rail train running on the large route h1, and the second part represents the running time cost of the smart rail train running on the small route h2.

[0109] This embodiment constructs the constraint conditions of the optimization model from three aspects: route constraints, timetable constraints and passenger flow constraints. The route constraints are limited by the route and the range of train stops. The timetable constraints are limited by the operating time range, safety interval, section running time and stop time of the smart rail train. The passenger flow constraints are limited by passenger flow distribution, the number of passengers getting on and off the train and the train capacity.

[0110] In this embodiment, the route constraints include route operation constraints and train stop range constraints.

[0111] The route constraints limit each smart rail train to only one route, which are:

[0112]

[0113] Among them, z k,h is a 0-1 variable indicating whether the smart rail train k chooses route h for operation. H is a route set, H = {h1, h2, h3}, where h1 is a large route and h2 is a small route. If the smart rail train does not operate, route h3 is selected.

[0114] When the kth smart rail train selects the route h3, it means that the smart rail train will not run during the study period due to the timetable constraint, so the subsequent smart rail trains cannot run either, which is expressed as:

[0115]

[0116] in, A 0-1 variable indicating whether the smart rail train k chooses route h3 to run.

[0117] The train stop range constraint restricts each smart rail train to stop at stations within its selected route range. For smart rail trains running in small route sections, they do not stop at stations outside the small route sections, which can be expressed as:

[0118]

[0119] in, is a 0-1 variable indicating whether the smart rail train k chooses route h1 to run. is a 0-1 variable indicating whether the smart rail train k chooses route h2 for operation. is a 0-1 variable indicating whether the smart rail train k chooses route h3 to run, ε k,i It is a 0-1 variable indicating whether the smart rail train k stops at station i.

[0120] The vehicle must stop at the turnaround station (1,m,n,N) where the large and small roads meet, expressed as:

[0121]

[0122] In this embodiment, the timetable constraints include the operating time range constraints of the smart rail train, the safety interval constraints, the section operation time constraints and the stop time constraints.

[0123] The operating time range constraint of the smart rail train limits the departure time of the smart rail train to the earliest and latest operating time, which is expressed as:

[0124]

[0125] Among them, T e This is the earliest time when the smart rail train will be put into operation. is a 0-1 variable indicating whether the smart rail train k chooses route h3 to run. is the departure time of smart rail train k at station i, T l The latest time for the operation of the Smart Rail train;

[0126] The departure interval is the time difference between two adjacent smart rail trains departing from the station. In order to ensure the safe operation of smart rail trains, the adjacent departure interval should not be less than the minimum departure interval; in order to avoid passengers waiting too long, the adjacent departure interval should not be greater than the maximum departure interval. Similarly, the arrival time of smart rail trains also has a maximum and minimum arrival interval. This constraint can be uniformly expressed by the train safety interval. The safety interval between the k-1th smart rail train and the kth smart rail train at station i is expressed as:

[0127]

[0128] Among them, h min is the minimum safe interval of the smart rail train at the station, Q is a set positive integer, is a 0-1 variable indicating whether the smart rail train k chooses route h3 to run. is the departure time of smart rail train k at station i, h max The maximum safe interval for smart rail trains at stations. is a 0-1 variable indicating whether the smart rail train k chooses route h1 to run. is the arrival time of smart rail train k at station i, and M is the maximum number of smart rail trains;

[0129] The running time of all intelligent rail trains in the interval is known, so the interval running time constraint of the kth intelligent rail train is expressed as:

[0130]

[0131] Among them, r i is the running time of the smart rail train between stations i and i+1;

[0132] The stop time of all smart rail trains at the station is known, so the stop time constraint of the kth smart rail train is expressed as:

[0133]

[0134] Among them, ε k，i is a 0-1 variable indicating whether the smart rail train k stops at station i, s i is the dwelling time of the smart rail train at station i.

[0135] In this embodiment, the passenger flow constraints include passenger flow distribution constraints, passenger boarding and alighting number constraints, and train capacity constraints.

[0136] Passenger flow allocation constraints restrict that only when the kth smart rail train directly reaches station i and station j and stops at both stations, the passengers who arrive at station i at time t and whose destination is station j can take the kth smart rail train, which can be expressed as:

[0137] f k，i，j，t ≤δ·C·ε k，i ·ε k，j

[0138] Passengers' waiting time shall not exceed T max , otherwise take a virtual train, expressed as:

[0139]

[0140] 0≤w k，i，j，t ≤T max , iff k，i，j，t >0

[0141] The number of people arriving at station i and heading to station j at time t is equal to the sum of the number of people taking the smart rail train and the virtual train, expressed as:

[0142]

[0143] Among them, f k，i，j，t is the number of passengers arriving at station i at time t and going to station j who take the smart rail train k, δ is the capacity limit coefficient of the smart rail train, C is the number of passengers on the smart rail train, and ε k，i is a 0-1 variable indicating whether the smart rail train k stops at station i, ε k，j is a 0-1 variable indicating whether the smart rail train k stops at station j, w k，i，j，t is the waiting time for the smart rail train k for passengers arriving at station i and going to station j at time t, is the departure time of the kth smart rail train at station i, T max is the maximum waiting time, K is the number of smart rail trains, Passengers arriving at station i at time t and heading to station j take the virtual smart rail train K v The number of people, F i，j(t) is the passenger flow from station i to station j at time t;

[0144] The number of passengers who take the kth smart rail train at station i to station j is equal to the sum of the number of passengers who arrive at station i and take the kth smart rail train to station j during the study period, expressed as:

[0145]

[0146] It can be obtained that the number of people boarding the kth smart rail train at station i is expressed as:

[0147]

[0148] Similarly, the number of passengers getting off the kth smart rail train at station i is expressed as:

[0149]

[0150] in, is the number of people boarding the smart rail train k at station i to station j, is the number of people boarding the smart rail train k at station i, N is the number of stations on the smart rail line, is the number of passengers getting off the smart rail train k at station i, is the number of people who boarded the smart rail train k at station j and went to station i;

[0151] The train capacity constraint limits the capacity of the smart rail train to not exceed the maximum passenger capacity of the smart rail train, which can be expressed as:

[0152] p k ,i≤δ·C

[0153] Among them, p k，i is the number of people in the smart rail train k when it departs from station i, δ is the capacity limitation coefficient of the smart rail train, and C is the number of passengers on the smart rail train.

[0154] S3. Solve the multi-objective optimization model of intelligent rail train scheduling considering high-speed rail station connection and obtain the Pareto optimal solution set;

[0155] In an optional embodiment of the present invention, step S3 uses a non-dominated sorting genetic algorithm with an elite strategy to solve the multi-objective optimization model for smart rail train scheduling considering high-speed rail station connections.

[0156] This embodiment designs a solution algorithm for the intelligent rail train scheduling optimization model considering high-speed rail station connection based on the NSGA-II algorithm from four aspects: chromosome design, fitness function selection, genetic operator design, and algorithm termination conditions. The NSGA-II algorithm process can be summarized as the following steps:

[0157] Step 1: Generate a parent population PP0 with U individuals;

[0158] Step 2: Generate the offspring population Q0 through selection, crossover and mutation operations;

[0159] Step 3: Merge the parent population P0 and the child population Q0, and obtain a new parent population P1 with U individuals through elite selection strategy, fast non-dominated sorting and crowding calculation;

[0160] Step 4: Obtain a new offspring population through selection, crossover, and mutation operations, and repeat Step 3 and Step 4 until the algorithm reaches the maximum number of iterations.

[0161] The algorithm flow chart of solving the intelligent rail train scheduling optimization model considering high-speed rail station connection based on NSGA-II algorithm is as follows: Figure 3 shown.

[0162] The "chromosome design" method is as follows: The model established by the present invention includes four decision variables, namely, the route selection, stop plan, departure time and arrival time of the smart rail train. In order to improve the efficiency of algorithm writing and operation, the original four decision variables are integrated into two decision variables, namely, the route selection and driving interval of the smart rail train. The route selection and stop plan of the smart rail train can be calculated from the route selection plan of the smart rail train, and the departure time and arrival time of the smart rail train can be calculated based on the driving interval of the smart rail train. Among the two decision variables designed by the algorithm, the route selection of the smart rail train needs to set the route alternative plan and number it according to the specific problem. Each smart rail train randomly selects a plan, that is, each smart rail train is assigned an alternative route number; the driving interval is a random integer between the minimum safe interval and the maximum safe interval of the smart rail train. In order to more realistically and intuitively reflect the solution, this embodiment adopts integer encoding for all decision variables, and the length of the chromosome is the number of decision variables. If k smart rail trains are optimized, each train selects a route alternative, and there is a running interval between every two smart rail trains that meets the constraints, then the decision variable consists of two parts: the route selection plan of k smart rail trains, with a length of k, and the running interval between adjacent smart rail trains, with a length of k-1, so the chromosome length is 2k-1.

[0163] The "fitness function selection" method is as follows: The fitness function calculation includes two parts: passenger travel time cost and smart rail travel time cost.

[0164] The “genetic operator design” involved in step S4 specifically includes: (1) selection, (2) crossover, and (3) mutation.

[0165] (1) Select

[0166] The selection process of the genetic algorithm is to retain the individuals with strong fitness in the parent population and put them into the offspring. The present embodiment adopts the roulette method. The roulette method is to calculate the probability of each individual appearing in the offspring according to the fitness value of the individual, and randomly select individuals to form the offspring population according to this probability. The starting point of this selection strategy is that the better the fitness value, the greater the probability of the individual being selected. When solving the maximization problem, the fitness value can be directly used for selection, but when solving the minimization problem, the fitness function of the problem needs to be converted (such as using the inverse or opposite number) to convert it into a maximization problem. Therefore, for the dual-objective minimization problem of the present embodiment, the fitness function is first inverted, and then the passenger travel cost and the enterprise travel time cost are normalized and summed to convert the problem into a maximization problem.

[0167] (2) Cross

[0168] Crossover refers to exchanging gene fragments of chromosomes of two individuals in a specific form under certain probability conditions to generate new individuals, which is an important means for genetic algorithms to obtain excellent individuals. This embodiment adopts single-point crossover.

[0169] (3) Variation

[0170] Mutation refers to changing some genes on chromosomes to produce new individuals. Mutation can improve the diversity of the population and prevent the algorithm from falling into a local optimum. This embodiment uses multi-point mutation.

[0171] The "algorithm termination condition" method uses reaching the maximum number of iterations as the termination condition of the algorithm.

[0172] S4. Determine the optimal solution for train scheduling based on the Pareto optimal solution set and generate the corresponding smart rail train operation diagram.

[0173] In an optional embodiment of the present invention, step S4 determines the enterprise optimal solution, the trade-off optimal solution and the passenger optimal solution according to the Pareto optimal solution set, and compares the passenger travel time cost, the smart rail running time cost, the number of smart rail trains in operation and the number of small-circuit trains in operation of the enterprise optimal solution, the trade-off optimal solution and the passenger optimal solution with the actual operation plan data to obtain the optimized final output solution, and outputs the smart rail train operation diagram corresponding to the solution for reference and decision-making by smart rail enterprise personnel.

[0174] In this embodiment, the enterprise optimal solution is: in the output Pareto diagram, the point in the upper left position represents the plan when the smart rail travel time cost is the smallest. At this time, the maximum consideration is given to reducing the smart rail travel time cost, so the passenger travel time cost is the largest, which is called the enterprise optimal solution.

[0175] The optimal solution of the trade-off in this embodiment is: considering the balance between the travel time cost of the smart rail and the travel cost of passengers, a trade-off optimal solution is selected, that is, the process of simulating the enterprise to weigh a series of optimization plans (a set of Pareto optimal solutions) and then select one of the optimization plans. The selected optimization plan must meet one of the following two conditions: (1) Compared with the original plan, the travel time cost of the smart rail and the travel time cost of passengers are both reduced; (2) Compared with the original plan, the travel time cost of passengers is slightly increased, but the travel time cost of the smart rail is greatly reduced.

[0176] In this embodiment, the optimal solution for passengers is: in the output Pareto diagram, the point at the lower right position represents the solution when the travel time cost of the smart rail is the largest. At this time, the maximum consideration is given to reducing the travel time cost of passengers, so the travel time cost of passengers is the smallest, which is called the optimal solution for passengers.

[0177] In this embodiment, the multi-objective genetic algorithm NSGA-II is used to solve the smart rail train scheduling optimization model that takes into account the high-speed rail station connection needs. The Pareto chart, smart rail train operation diagram, flow chart, etc. are drawn using software tools such as Visio, Excel, and MATLAB.

[0178] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0179] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0180] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0181] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0182] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A method for optimizing the dispatching of intelligent rail trains connected to high-speed rail. It is characterized in that The following steps are involved: Obtain the operation data of smart rail trains and passenger flow OD data within a set period; Based on the scenario of intelligent rail and high-speed rail connection, the multi-objective optimization model of intelligent rail train scheduling considering high-speed rail station connection is constructed with passenger travel time cost and intelligent rail running time cost as optimization objectives, and route constraints, timetable constraints and passenger flow constraints as constraints. Solve the multi-objective optimization model of intelligent rail train scheduling considering high-speed rail station connection and obtain the Pareto optimal solution set; The optimal solution for train scheduling is determined based on the Pareto optimal solution set, and the corresponding smart rail train operation diagram is generated.

2. According to claim 1, a method for optimizing the dispatching of smart rail trains connected to high-speed rails, It is characterized in that The operation data of the smart rail train includes: Operating speed, interval operating time, stop time and maximum passenger capacity data.

3. According to claim 1, a method for optimizing the dispatching of smart rail trains connected to high-speed rails, It is characterized in that Taking the passenger travel time cost as the optimization target, the specific optimization target is: Among them, min is the minimization function, Z 1 is the passenger travel time cost, T wait is the waiting time cost, T ivt is the time cost in transit, T crowd is the congestion time cost, K is the set of smart rail train numbers, T is the set of discretized set time ranges, N is the number of stations on the smart rail line, is the departure time of smart rail train k at station i, f k,i,j,t is the number of passengers arriving at station i at time t and heading to station j who take train k, i is the weight of station i, Q is a positive integer, T max is the maximum waiting time, f kv,i,j,t Passengers arriving at station i at time t and heading to station j take the virtual smart rail train K v Number of people, is the arrival time of smart rail train k at station j, Additional cost for congestion, p k,i is the number of people in the smart rail train k when it departs from station i, r i is the running time of the smart rail train between stations i and i+1.

4. According to claim 1, a method for optimizing the dispatching of a smart rail train connected to a high-speed railway, It is characterized in that The specific optimization target is to take the travel time cost of the smart track as follows: Among them, min is the minimization function, Z 2 is the travel time cost of the smart rail, T operator is the time conversion cost, K is the number of smart rail trains, Whether the smart rail train k chooses route h 1 is a 0-1 variable for the number of lines opened, N is the number of stations on the smart rail line, r i is the running time of the smart rail train between stations i and i+1, s i is the dwelling time of the smart rail train at station i, ε k,i is a 0-1 variable indicating whether the smart rail train k stops at station i, Whether the smart rail train k chooses route h 2 0-1 variable for the line, m is the intersection h 2 The starting station number of the route, n is the intersection h 2 The terminal number.

5. According to claim 1, a method for optimizing the dispatching of smart rail trains connected to high-speed rails, It is characterized in that Route constraints include route constraints and train stop range constraints; The route opening constraints are specifically: Among them, z k,h is a 0-1 variable indicating whether the smart rail train k chooses route h to run, H is the route set, Whether the smart rail train k chooses route h 3 0-1 variable for opening a line; The specific constraints of the train stop range are: in, Whether the smart rail train k chooses route h 1 The 0-1 variable of the opening line, Whether the smart rail train k chooses route h 2 The 0-1 variable of the opening line, Whether the smart rail train k chooses route h 3 The 0-1 variable of the opening line, ε k,i A 0-1 variable indicating whether the smart rail train k stops at station i.

6. According to claim 1, a method for optimizing the dispatching of smart rail trains connected to high-speed rails, It is characterized in that The timetable constraints include the operating time range constraints, safety interval constraints, interval operation time constraints and stop time constraints of the smart rail trains; The specific operating time range constraints of the smart rail train are: Among them, T e This is the earliest time when the smart rail train will be put into operation. Whether the smart rail train k chooses route h 3 The 0-1 variable of the opening line, is the departure time of smart rail train k at station i, T l The latest time for the operation of the Smart Rail train; The safety interval constraints are specifically: Among them, h min is the minimum safe interval of the smart rail train at the station, Q is a set positive integer, Whether the smart rail train k chooses route h 3 The 0-1 variable of the opening line, is the departure time of smart rail train k at station i, h max The maximum safe interval for smart rail trains at stations. Whether the smart rail train k chooses route h 1 The 0-1 variable of the opening line, is the arrival time of smart rail train k at station i, and M is the maximum number of smart rail trains; The interval running time constraints are specifically: Among them, r i is the running time of the smart rail train between stations i and i+1; The stop time constraint is specifically: Among them, ε k,i is a 0-1 variable indicating whether the smart rail train k stops at station i, s i is the dwelling time of the smart rail train at station i.

7. According to claim 1, a method for optimizing the dispatching of a smart rail train connected to a high-speed railway, It is characterized in that Passenger flow constraints include passenger flow distribution constraints, passenger boarding and alighting constraints, and train capacity constraints; The passenger flow allocation constraints are specifically: f k,i,j,t ≤δ·C·e k,i ·e k,j 0≤w k,i,j,t ≤T max ,if f k,i,j,t >0 Among them, f k,i,j,t is the number of passengers arriving at station i at time t and going to station j who take the smart rail train k, δ is the capacity limit coefficient of the smart rail train, C is the number of passengers on the smart rail train, and ε k,i is a 0-1 variable indicating whether the smart rail train k stops at station i, ε k,j is a 0-1 variable indicating whether the smart rail train k stops at station j, w k,i,j,t is the waiting time for the smart rail train k for passengers arriving at station i and going to station j at time t, is the departure time of the kth smart rail train at station i, T max is the maximum waiting time, K is the number of smart rail trains, Passengers arriving at station i at time t and heading to station j take the virtual smart rail train K v The number of people, F i,j (t) is the passenger flow from station i to station j at time t; The specific constraints on the number of passengers getting on and off the bus are: in, is the number of people boarding the smart rail train k at station i to station j, is the number of people boarding the smart rail train k at station i, N is the number of stations on the smart rail line, is the number of passengers getting off the smart rail train k at station i, is the number of people who boarded the smart rail train k at station j and went to station i; The train capacity constraint is specifically: p k,i ≤δ·C Among them, p k,i is the number of people in the smart rail train k when it departs from station i, δ is the capacity limitation coefficient of the smart rail train, and C is the number of passengers on the smart rail train.

8. According to claim 1, a method for optimizing the dispatching of a smart rail train connected to a high-speed railway, It is characterized in that Solving the multi-objective optimization model for smart rail train scheduling considering high-speed rail station connection includes: A non-dominated sorting genetic algorithm with an elite strategy is used to solve the multi-objective optimization model of smart rail train scheduling considering high-speed railway station connections.

9. A method for optimizing the dispatching of smart rail trains connected to high-speed rails according to claim 8, It is characterized in that When the non-dominated sorting genetic algorithm with elite strategy is used to solve the multi-objective optimization model of smart rail train scheduling considering high-speed rail station connection, The route selection and running interval of smart rail trains are taken as decision variables, and the chromosome encoding is carried out using integer coding.

10. A method for optimizing the dispatching of smart rail trains connected to high-speed rails according to claim 8, It is characterized in that When the non-dominated sorting genetic algorithm with elite strategy is used to solve the multi-objective optimization model of smart rail train scheduling considering high-speed rail station connection, The individual fitness value is calculated based on the passenger travel time cost and the smart rail travel time cost.

Images