CN113450032A - Subway departure strategy method and system based on genetic algorithm - Google Patents
Subway departure strategy method and system based on genetic algorithm Download PDFInfo
- Publication number
- CN113450032A CN113450032A CN202111013857.3A CN202111013857A CN113450032A CN 113450032 A CN113450032 A CN 113450032A CN 202111013857 A CN202111013857 A CN 202111013857A CN 113450032 A CN113450032 A CN 113450032A
- Authority
- CN
- China
- Prior art keywords
- departure
- subway
- decision
- departure decision
- station
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002068 genetic effect Effects 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 37
- 239000011159 matrix material Substances 0.000 claims abstract description 50
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 36
- 230000007613 environmental effect Effects 0.000 claims abstract description 27
- 238000012546 transfer Methods 0.000 claims description 16
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000009395 breeding Methods 0.000 claims description 7
- 230000001488 breeding effect Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 11
- 210000000349 chromosome Anatomy 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 230000004083 survival effect Effects 0.000 description 4
- 206010064571 Gene mutation Diseases 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000036961 partial effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Evolutionary Computation (AREA)
- Operations Research (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a subway departure strategy method and system based on a genetic algorithm. The method comprises the following steps: acquiring subway line information, subway line passenger information and subway station waiting passenger information, establishing a subway environment state matrix, establishing a subway departure decision, and acquiring an environment reward value corresponding to the departure decision of each time point according to the subway environment state matrix; coding the departure decision at each time point into a departure gene of a genetic algorithm, establishing departure decision individuals comprising a plurality of departure genes of the genetic algorithm, and establishing a departure decision population comprising a plurality of departure decision individuals; randomly selecting one departure decision individual in the departure decision population as a departure decision parent, taking the rest departure decision individuals as departure decision parents, and randomly combining and reproducing through corresponding positions of genes to generate offspring departure decision individuals; and taking the departure decision individual with the highest environmental reward value as a final departure basis.
Description
Technical Field
The invention relates to the field of subway regulation and control, in particular to a subway departure strategy method and system based on a genetic algorithm.
Background
The subway is an important component of the public transportation of the big city, the passenger flow of the subway changes with the change of the living and working needs of people every day, and the travel rules of working days and resting days have obvious difference. And subway passenger flow has the characteristics of dynamics and time variation, and the density of dispatching a car is required to be adjusted according to the actual condition.
At present, departure time and departure interval of a subway are fixed, and only a few adjustments are made in a rush hour or a rest day. The so-called peak time is judged by the staff of the subway station according to the real-time passenger flow, has subjectivity and hysteresis, and brings bad experience to passengers in the peak time or the rest day. The effective solution is to predict the passenger flow in advance and adjust the departure density. The traditional method only counts the inbound and outbound information of all passengers to predict, but does not predict the complete travel of the passengers, so that the departure density prediction needs to be carried out according to the subway operation environment in a new mode.
Disclosure of Invention
The invention aims to provide a subway departure strategy method and system based on a genetic algorithm, aiming at the defects of the existing process, the method and the system are based on the genetic algorithm, departure decisions of each time point are taken as genes of the genetic algorithm, departure decision sets of all time points in a fixed time period are taken as departure decision individuals, different gene departure decision individuals in different time periods or the same time period are combined to generate a departure decision population for random combination and reproduction of the genes, offspring of the departure decisions are generated, and the optimal departure decision individuals are obtained according to calculation of an environment reward value.
A subway departure strategy method and system based on genetic algorithm adopts the following technical scheme:
the invention provides a subway departure strategy method and system based on genetic algorithm, the environmental data of the method and system comprises subway line information, passenger information on a subway line and passenger information waiting at a subway station, the three parts jointly form the environment state of the subway of the whole system, so that the passenger flow of the whole subway environment can be considered at the same time or the local passenger flow of the subway line and the subway station can be considered respectively, and the departure decision of filial generations can be made to better meet the scheduling requirements of the overall passenger flow and the local passenger flow by combining the genetic algorithm, thereby avoiding the situation that the departure strategy of the subway is judged based on the overall passenger flow while the rapid increase of the local passenger flow is realized, improving the satisfaction degree of passengers and reducing the overall operation cost;
the invention provides a subway departure strategy method and system based on a genetic algorithm, wherein the method and the system acquire information of a passenger station entering station and a target station from a subway environment state, judge whether the passenger can reach the target station directly or needs to transfer, and set different matrix parameters aiming at different types of passengers, so that the overall flow and the local flow can be comprehensively considered to design a dispatching scheme of the subway;
the invention provides a subway departure strategy method and system based on a genetic algorithm, wherein the method and the system select most of descendants with high environmental reward values to construct a descendant departure decision population by taking the environmental reward values as the basis of descendant genetic adaptability, select a small part of descendants with low environmental reward values to be added into the descendant departure decision population to serve as possible descendants of 'genetic mutation', wherein the descendants of 'genetic mutation' may have better genetic adaptability or poorer genetic adaptability, and reject the descendants with poorer genetic adaptability through a selection function;
further, acquiring subway line information, passenger information on a subway line and passenger information waiting at a subway station, and establishing a subway environment state matrix and a passenger waiting environment state matrix;
further, a subway departure decision is established, and an environment reward value corresponding to the departure decision of each time point is obtained according to the subway environment state matrix;
furthermore, coding the departure decision at each time point into a departure gene of a genetic algorithm, establishing departure decision individuals comprising a plurality of departure genes of the genetic algorithm, and establishing a departure decision population comprising a plurality of departure decision individuals;
further, randomly selecting one departure decision individual in the departure decision population as a departure decision parent, and taking the rest departure decision individuals as departure decision parents, wherein the departure decision parent traverses all the departure decision parents to randomly combine and propagate corresponding positions of the departure decision parents and the departure decision parent genes, so as to generate offspring departure decision individuals;
further, an environment reward value corresponding to each child departure decision individual is calculated, the departure decision individuals corresponding to the environment reward threshold value higher than the set environment reward threshold value are selected through a selection function to carry out gene random combination reproduction, and the departure decision individual with the highest environment reward value in the departure decision population is obtained and serves as a final departure decision individual.
Preferably, the selection probability of each child departure decision individual is calculated according to the environmental reward and through the selection function, and the calculation method of the selection probability is as follows: and each child departure decision individual environment reward value/the sum of all child departure decision individual environment reward values in the child departure decision population, and reserving a part of child departure decision individuals lower than the environment reward threshold value to carry out gene random combination reproduction.
Preferably, after the departure decision parent and all departure decision parents multiply, a child departure decision population is generated, one individual is randomly selected from the child departure decision population as a new departure decision parent and all the remaining new departure decision parents in the child departure decision population are subjected to gene random combination multiplication to generate a next child departure decision population, after the population is multiplied by multiple generations, the environmental reward value of each individual in each generation of population is calculated, and the individual with the highest environmental reward value is taken as a final departure decision.
Preferably, the environment bonus value is calculated by: (number of passengers on subway cars/total number of passengers in subway environment) -number of subway trains running-average waiting time per passenger, the higher the environmental reward value, the higher the individual genetic suitability.
Preferably, a minimum distance threshold value between two adjacent subway trains is set, if the distance between the two adjacent subway trains is smaller than the minimum distance threshold value in the running process of the subway, the current subway environment state is judged to be dead, the subway trains are prohibited from dispatching, and the subway trains are dispatched after the current subway environment state matrix is obtained again until the distance between the two subway trains is larger than the minimum distance threshold value.
Preferably, the method for updating the subway environment state matrix at each moment comprises the following steps: according to the station of passenger information, the station time of entering and the station of leaving, confirm the passenger quantity of other stations on the subway line of passenger waiting at the present subway station, when the subway train arrives at the station at a certain moment, the passenger waiting in the subway station gets on the bus partly or totally, and remove the passenger information of getting on the bus in the passenger information of waiting in subway station, add corresponding data in the passenger information on the subway line at the same time, produce the new subway environmental state matrix.
Preferably, if there are passengers waiting for the subway train to be transferred at the transfer station, the passengers getting off the subway train after arriving at the transfer station, removing the passenger information on the subway line, and entering the passengers getting off into the passenger information waiting for the current subway station by taking the current transfer station as a starting station and the next transfer station or destination station as a destination station to generate a new subway environment state matrix.
Preferably, a subway passenger flow prediction model is established through an LSTM + CNN neural network, and the subway environment state matrixes of all stations at the current moment are used as input and input into the model to predict the passenger flow at the next time point.
Drawings
Fig. 1 shows a subway departure strategy method and a system flow diagram based on a genetic algorithm.
FIG. 2 shows a schematic flow chart of the genetic algorithm of the present invention.
Fig. 3 is a schematic diagram showing the processing of the present invention in a subway environment state matrix of a non-transferable site.
Fig. 4 is a schematic processing diagram of the invention in a subway environment state matrix of a transfer station.
Detailed Description
The following description is presented to disclose the invention so as to enable any person skilled in the art to practice the invention. The preferred embodiments in the following description are given by way of example only, and other obvious variations will occur to those skilled in the art. The basic principles of the invention, as defined in the following description, may be applied to other embodiments, variations, modifications, equivalents, and other technical solutions without departing from the spirit and scope of the invention.
It is understood that the terms "a" and "an" should be interpreted as meaning that a number of one element or element is one in one embodiment, while a number of other elements is one in another embodiment, and the terms "a" and "an" should not be interpreted as limiting the number.
The invention discloses a subway departure strategy method and a system based on a genetic algorithm, which are combined with figures 1-4, and realize the optimized dispatching of subway departure based on the genetic algorithm, wherein a subway environment state matrix is established by acquiring the taking information of each passenger, the waiting information of passengers in a subway station and the information of passengers on a subway line, an environment reward value corresponding to a departure decision at each moment is calculated according to the subway environment state matrix, and the departure strategy of the subway can give consideration to the subway departure cost and the passenger taking experience by comprehensively considering local and overall passenger flow and combining the genetic algorithm.
Specifically, passengers can obtain the passenger's station-entering data by swiping the card to enter the station, the passenger's station-entering data comprises card swiping occurrence time, subway line ID, subway station ID, station-entering and station-exiting states and user identity ID, and all subway line information and each station data of each line can be obtained in a subway official network. And acquiring the state of the subway train, wherein the state of the train is divided into a survival state and a death state, the survival state is that the train is in operation, and the death state is that the train does not depart from the starting station or arrives at the terminal station. And acquiring the maximum passenger capacity of the subway train, wherein the passenger capacity of the train in operation cannot exceed the maximum passenger capacity allowed by the train.
After the taking information of all passengers is acquired, a subway environment state matrix is initialized, the subway environment state matrix is divided into a subway station waiting passenger state matrix and a subway line passenger state matrix, the subway environment state matrix is a two-dimensional matrix, the horizontal columns represent station-entering stations where the passengers take the bus, the vertical columns represent station-destination stations where the passengers leave the bus, and data in the matrix represent the number of the passengers from the station-entering stations to the station-destination stations. The subway environment state matrix comprises passenger waiting data of all subway stations at the current moment and passenger data on all subway lines in operation.
Specifically, when a passenger enters a station by swiping a card, corresponding passenger riding data is added to a passenger waiting state matrix of the subway station; and after passengers get on the train, removing the passenger state matrix of the passenger waiting at the subway station, and adding the passenger data into the passenger state matrix on the subway line. If the passenger is a direct passenger, removing the passenger state matrix of the passenger data on the subway line after the passenger arrives at the destination station; if the passenger is a transfer passenger, the passenger gets off the train after arriving at the transfer station, the data of the passenger state matrix on the subway line of the passenger is removed, and meanwhile, the passenger data is added into the passenger state matrix waiting at the subway station by taking the current transfer station as an initial station. Compared with the traditional scheme that the passenger flow can only be monitored from the passenger entering to the passenger leaving, the invention can make up the monitoring deficiency of passenger flow for passenger transfer, thereby improving the actual subway dispatching effect. It should be noted that, in the present invention, the case where the passenger stops at the station or the passenger sits down and stands will not be considered.
After the subway environment state matrix is set, the invention further combines the neural network to predict the subway passenger flow, and adjusts the subway departure strategy according to the prediction result. Specifically, the method comprises the following steps: acquiring a subway environment state matrix at each moment, taking every minute as a time interval, combining time information with the subway environment state matrix, and constructing a time passenger flow matrix as follows: (T, X)m,n) Wherein T represents a time node, Xm,nRepresenting a matrix of passenger numbers from m stations to n stations at time T, each time T having its own matrix of passenger numbers. For the invention, the passenger flow matrix X at each moment is usedm,nAs an input, data is collected in the passenger number matrix as a training set by setting the size of the sliding window, and trained in an LSTM + CNN neural network. Sliding deviceThe window size should be set to be less than or equal to the passenger traffic matrix Xm,nThe horizontal and vertical values of (2). The final prediction model is obtained by combining the calculation of the loss function of the neural network and the adjustment of the weight of the neural network. Considering that a neural network includes a plurality of neurons, the output of the entire neural network can be adjusted by setting the output weight of each neuron.
In the subway departure decision module, the part is used for judging whether a train should be departed at the current moment, the fixed earliest departure time of the train every day is used as the time for initializing departure, and the result trained according to a subway environment state matrix and a neural network is used as the judgment standard for judging whether the train should be departed at the later moment. Specifically, the survival state of the train at a certain time is firstly judged, wherein 1 represents survival and can perform departure operation, and 0 represents death and does not allow departure, so the departure decision at a certain time can be represented as (T, Vs), T represents a time point, and Vs represents a departure decision matrix.
It is worth mentioning that the genetic algorithm adopted by the invention is used for constructing departure decision individuals, and offspring with higher genetic adaptability is selected as a population for reproduction through genetic adaptability calculation of individual reproduction, so that the genetic reproduction of each offspring is basically ensured to have better genetic adaptability compared with that of a parent generation. Specifically, the method comprises the following steps: constructing a coding gene of a genetic algorithm, wherein n time points are set, obtaining departure actions of all lines in a subway environment at each time point, recording the departure actions as 1, and recording the departure actions as 0, so that the departure decision of 8 lines at the current time point can be expressed as: vt = (1, 1, 0, 0, 1, 1, 0, 1), t represents the current time point, the value in the departure decision of the code corresponds to the base in the gene, the base needs to be exchanged in the process of multiplication, and the departure decision of a single time point corresponds to the chromosome. And further calculating departure decisions corresponding to all time points in one day as departure decision individuals in the genetic algorithm. It should be noted that, by changing the genes in the departure decision individuals, different departure decision individuals are generated, and the changing manner of the genes includes: random substitution of base pairs in the gene:i.e. interchanging 0 and 1 in the partial departure decision; or increasing or decreasing the number of chromosomes: namely, different departure decisions at the same time point or the number of departure decisions at different time points are increased. After the establishment of a plurality of departure decision individuals, generating a departure decision population containing a plurality of departure decision individuals, wherein in a preferred embodiment of the invention, assuming that there is one departure decision every minute from 6 am to 24 pm, the departure behavior has 1080 departure decisions at time points, the invention establishes an initial population of individuals, wherein the population includes all possible genes, and the genes of each initialization individual are the same, for example: there is 2 when the above 8 lines are sent out8=256 possibilities, i.e. there are 256 different chromosomes, the present invention sets the individuals so that the chromosomes in each individual are the same, for example, all chromosomes of the first individual are: v1= (0, 0, 0, 0, 0, 0, 0, 0), meaning that no delivery was made at all time points, and all genes of the second individual are: v2= (1, 0, 0, 0, 0, 0, 0, 0), which means that only line 1 is issued at regular time intervals until 256 different individuals are set, thereby ensuring that all genes are present in the population, thereby enabling a better effect and speed of reproduction.
After the 256 initial population of departure decision individuals is established, randomly taking one individual in the initial population as a departure decision parent, wherein the departure decision parent traverses the remaining 255 departure decision individuals in the initial population, wherein the remaining 255 departure decision individuals serve as departure decision parents and the departure decision parent to perform random combined breeding of corresponding positions of genes, and the random gene exchange breeding mode includes but is not limited to: base pair selection interchange (0, 1 interchange), chromosome local selection interchange (departure decision part interchange), and chromosome whole selection interchange (departure decision whole interchange), wherein the interchange is random combination of corresponding gene positions of a father and a mother, namely, the control of corresponding base positions for free combination is equivalent to the simulation of human genetic reproduction function. And generating offspring departure decision individuals after the gene random exchange and multiplication are completed. Because each departure decision parent and each departure decision parent carry out gene random exchange and reproduction once, 255 child departure decision individuals can be reproduced by one departure decision parent. In another preferred embodiment of the present invention, all the decision-making individuals for departure are used as the decision-making parents for departure and the rest decision-making individuals for departure, so that a plurality of different child decision-making individuals for departure can be generated. It is worth mentioning that the invention takes the calculation of the environmental reward value of each departure decision-making individual as the genetic adaptability of the departure decision-making individual, especially the departure decision-making individuals of the offspring. Wherein the environmental reward value is calculated in a manner that includes: the method comprises the steps of obtaining a subway environment state matrix corresponding to each departure decision in a current departure decision individual, wherein the subway environment state matrix comprises passenger data of a transfer station, and therefore when passengers get on or off the train are recorded, waiting time of the passengers and time of the passengers in subway carriages can be obtained simultaneously. That is, the number of passengers in the corresponding subway car can be obtained according to the subway environment state matrix, the waiting time of passengers at the station can be obtained, and the environment reward value is further calculated, wherein the environment reward value = (the number of passengers at the subway car/the total number of passengers at the subway environment) -the number of running subway buses-the average waiting time of each passenger. The environment reward value is used as a parameter of departure decision genetic adaptability, and when the environment reward value is larger, the genetic adaptability is stronger, and the departure requirement is better met.
In a preferred embodiment of the invention, whether the environmental reward value of the departure decision individual is greater than the environmental reward threshold value or not is calculated, and if so, the environmental reward value is added into the population of the departure decision individual. The method is more suitable for the departure decision individuals of the filial generations, so that the departure decision population of the filial generations has better genetic adaptability on the whole. In another preferred embodiment of the present invention, a probability function is used to calculate the departure probability of the departure decision entity in the whole population, wherein the departure probability includes the environmental reward value. The specific method comprises the following steps: calculating the environmental reward value of each departure decision individual in the population, and calculating the sum of the environmental reward values of all the departure decision individuals in the population as the environmental reward value of the population, wherein the departure probability can be the ratio of the environmental reward value of each departure decision individual in the corresponding population. And taking the departure decision individual with the maximum ratio as the final departure operation behavior.
It is worth mentioning that after the genetic algorithm is executed to obtain the departure decision individuals of the offspring, the departure decision population of the offspring is established, one departure decision individual in the departure decision population of the offspring can be further randomly selected to serve as a new departure decision parent, the remaining departure decision individuals in the departure decision population of the corresponding offspring are taken as a new departure decision parent to carry out gene random switching and breeding, a second offspring departure decision individual in the second breeding is generated, and the second offspring departure decision population is established according to the environment reward threshold. After repeated reproduction according to the reproduction mode, the departure decision individual with the highest environmental reward value is obtained to serve as a final departure decision execution basis.
Since the genetic algorithm needs to consider the possibility of the "gene mutation", in another preferred embodiment of the present invention, it is necessary to add the departure decision individuals with partial environmental reward values below the environmental reward threshold value to the established departure decision initialization population or the child departure decision population, so that better child departure decision individuals of the "gene mutation" type can occur in the process of performing the gene random combination reproduction, and of course, worse child departure decision individuals of the "gene mutation" type can also occur, and due to the constraint of the environmental reward threshold value and the constraint of the selection function, the selection and reproduction of the child departure decision individuals are basically forward.
In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication section, and/or installed from a removable medium. The computer program, when executed by a Central Processing Unit (CPU), performs the above-described functions defined in the method of the present application. It should be noted that the computer readable medium mentioned above in the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wire segments, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless section, wire section, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (10)
1. A subway departure strategy method based on a genetic algorithm is characterized in that the method and the system comprise the following steps:
acquiring subway line information, passenger information on a subway line and passenger information waiting at a subway station, and establishing a subway environment state matrix which consists of the subway line information, the running time information between the subway stations and passenger riding information;
establishing a subway departure decision, and acquiring an environment reward value corresponding to the departure decision of each time point according to the subway environment state matrix;
coding the departure decision at each time point into a departure gene of a genetic algorithm, establishing departure decision individuals comprising a plurality of departure genes of the genetic algorithm, and establishing a departure decision population comprising a plurality of departure decision individuals;
randomly selecting one departure decision individual in the departure decision population as a departure decision parent, taking the rest departure decision individuals as departure decision parents, and traversing all the departure decision parents by the departure decision parents to randomly combine and propagate corresponding positions of the departure decision parents and departure decision parent genes to generate offspring departure decision individuals;
and calculating the environment reward value corresponding to each child departure decision individual, selecting the departure decision individual corresponding to the environment reward threshold value through a selection function to carry out gene random combination reproduction, and acquiring the departure decision individual with the highest environment reward value in the departure decision population as the final departure decision individual.
2. A subway departure strategy method based on genetic algorithm as claimed in claim 1, wherein the selection probability of each child departure decision individual is calculated according to said environment reward value and through said selection function, said selection probability is calculated by: and each child departure decision individual environment reward value/the sum of all child departure decision individual environment reward values in the child departure decision population, and reserving a part of child departure decision individuals lower than the environment reward threshold value to carry out gene random combination reproduction.
3. The subway departure strategy method according to claim 1, wherein after the departure decision parent and all departure decision parents are subjected to gene random combination breeding, a child departure decision population is generated, an individual is randomly selected from the child departure decision population as a new departure decision parent and all the remaining new departure decision parents in the child departure decision population are subjected to gene random combination breeding to generate a next child departure decision population, after the child departure decision populations are subjected to gene random combination breeding, the environmental reward value of each departure decision individual in each generation of departure decision population is calculated, and the departure decision individual with the highest environmental reward value is taken as a final departure decision.
4. A subway departure strategy method based on genetic algorithm as claimed in claim 2, wherein said environment reward value is calculated by: (number of passengers on subway cars/total number of passengers in subway environment) -number of subway trains running-average waiting time per passenger, the higher the environmental reward value, the higher the individual genetic suitability.
5. A subway departure strategy method based on genetic algorithm as claimed in claim 2, wherein said calculation method of subway environment state matrix comprises: judging the station entering, station entering and station exiting of passengers according to the riding information of the current passengers, judging the number of passengers from each current station to each target station in the subway at the next time point, and if the subway is arrived at the station in a shift, clearing the number of all the passengers taking the arriving station as the target station in the waiting passenger information of the subway station to generate a new subway environment state matrix.
6. The subway departure strategy method based on genetic algorithm as claimed in claim 2, wherein if it is found that there is a transfer station in the waiting passenger information of the subway station, after the subway train arrives at the transfer station, the number of passengers in the subway environment state matrix of the destination station where the transfer station is the train of passengers is cleared, and the number of passengers in the cleared number of passengers is added to the corresponding subway environment state matrix point by taking the current transfer station as the start station and the next transfer station or destination station as the destination station, so as to form a new subway environment state matrix.
7. A subway departure strategy method based on genetic algorithm as claimed in claim 1, wherein said subway departure strategy method and system further comprises: acquiring the subway environment state matrixes of all stations at the current moment, establishing a subway pedestrian flow prediction model through an LSTM + CNN neural network, inputting the subway environment state matrixes of all stations at the current moment into the pedestrian flow prediction model, and predicting the pedestrian flow prediction model at a later time point.
8. The subway departure strategy method based on the genetic algorithm as claimed in claim 1, wherein the distance between two adjacent trains of subway buses is obtained, a minimum distance threshold is set, if the distance between two adjacent trains of subway buses is smaller than the minimum distance threshold, it is determined that the current subway environment state is dead, the subway buses are prohibited from departure, the subway environment state is initialized until the distance between two adjacent trains of subway buses is greater than the minimum distance threshold, and the current subway environment state matrix is obtained again.
9. A subway departure strategy system, characterized in that it implements a subway departure strategy method based on genetic algorithm as claimed in any one of the preceding claims 1-8.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to execute a subway departure strategy method based on a genetic algorithm according to any one of the preceding claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111013857.3A CN113450032A (en) | 2021-08-31 | 2021-08-31 | Subway departure strategy method and system based on genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111013857.3A CN113450032A (en) | 2021-08-31 | 2021-08-31 | Subway departure strategy method and system based on genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113450032A true CN113450032A (en) | 2021-09-28 |
Family
ID=77819153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111013857.3A Pending CN113450032A (en) | 2021-08-31 | 2021-08-31 | Subway departure strategy method and system based on genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113450032A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399468A (en) * | 2018-01-31 | 2018-08-14 | 华南理工大学 | It is a kind of based on vehicle when cost optimization operation Time segments division method |
| CN110245779A (en) * | 2019-05-10 | 2019-09-17 | 杭州电子科技大学 | A kind of public transport dynamic based on genetic algorithm is dispatched a car method for optimizing scheduling |
| CN111340289A (en) * | 2020-02-25 | 2020-06-26 | 杭州电子科技大学 | Genetic algorithm-based bus departure and speed adjustment optimization method and system |
| CN113077086A (en) * | 2021-03-30 | 2021-07-06 | 东南大学 | Method for designing bus synchronous transfer timetable for connecting subway hubs |
| CN113222387A (en) * | 2021-04-30 | 2021-08-06 | 北京理工新源信息科技有限公司 | Multi-objective scheduling and collaborative optimization method for hydrogen fuel vehicle |
| CN113276915A (en) * | 2021-07-06 | 2021-08-20 | 浙江非线数联科技股份有限公司 | Subway departure scheduling method and system |
-
2021
- 2021-08-31 CN CN202111013857.3A patent/CN113450032A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399468A (en) * | 2018-01-31 | 2018-08-14 | 华南理工大学 | It is a kind of based on vehicle when cost optimization operation Time segments division method |
| CN110245779A (en) * | 2019-05-10 | 2019-09-17 | 杭州电子科技大学 | A kind of public transport dynamic based on genetic algorithm is dispatched a car method for optimizing scheduling |
| CN111340289A (en) * | 2020-02-25 | 2020-06-26 | 杭州电子科技大学 | Genetic algorithm-based bus departure and speed adjustment optimization method and system |
| CN113077086A (en) * | 2021-03-30 | 2021-07-06 | 东南大学 | Method for designing bus synchronous transfer timetable for connecting subway hubs |
| CN113222387A (en) * | 2021-04-30 | 2021-08-06 | 北京理工新源信息科技有限公司 | Multi-objective scheduling and collaborative optimization method for hydrogen fuel vehicle |
| CN113276915A (en) * | 2021-07-06 | 2021-08-20 | 浙江非线数联科技股份有限公司 | Subway departure scheduling method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | A bi-level model for single-line rail timetable design with consideration of demand and capacity | |
| Zhang et al. | Comprehensive optimization of urban rail transit timetable by minimizing total travel times under time-dependent passenger demand and congested conditions | |
| Al Islam et al. | A real-time network-level traffic signal control methodology with partial connected vehicle information | |
| CN109409560B (en) | Urban rail transit passenger flow induction method based on multi-agent simulation | |
| CN113276915B (en) | Subway departure scheduling method and system | |
| Ghanim et al. | Real-time dynamic transit signal priority optimization for coordinated traffic networks using genetic algorithms and artificial neural networks | |
| Hassannayebi et al. | Variable and adaptive neighbourhood search algorithms for rail rapid transit timetabling problem | |
| Sáez et al. | Hybrid predictive control strategy for a public transport system with uncertain demand | |
| Wu et al. | Distributed agent-based deep reinforcement learning for large scale traffic signal control | |
| Li et al. | Expected value model for optimizing the multiple bus headways | |
| Tang et al. | Scheduling local and express trains in suburban rail transit lines: Mixed–integer nonlinear programming and adaptive genetic algorithm | |
| Ke et al. | Optimizing online matching for ride-sourcing services with multi-agent deep reinforcement learning | |
| Han et al. | Train timetabling in rail transit network under uncertain and dynamic demand using advanced and adaptive NSGA-II | |
| Núñez et al. | Multiobjective model predictive control for dynamic pickup and delivery problems | |
| Li et al. | Train timetabling with the general learning environment and multi-agent deep reinforcement learning | |
| Aragon-Gómez et al. | Traffic-signal control reinforcement learning approach for continuous-time Markov games | |
| CN113935595B (en) | Urban rail transit road network peak large passenger flow dredging system | |
| CN113362618B (en) | Multi-mode traffic adaptive signal control method and device based on strategy gradient | |
| CN113033928A (en) | Design method, device and system of bus shift scheduling model based on deep reinforcement learning | |
| CN113420926A (en) | Method for compiling multi-traffic mode transportation capacity collaborative optimization scheme between urban hubs | |
| CN109118412A (en) | Urban rail transit network passenger flow on-line control system | |
| CN113450032A (en) | Subway departure strategy method and system based on genetic algorithm | |
| Zhen et al. | Train rescheduling model with train delay and passenger impatience time in urban subway network | |
| CN115208892B (en) | Vehicle-road collaborative online task scheduling method and system based on dynamic resource demand | |
| CN115481777A (en) | Multi-line bus dynamic schedule oriented collaborative simulation optimization method, device and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210928 |
|
| RJ01 | Rejection of invention patent application after publication |