CN116424397A - Method and system for generating driving adjustment strategy under urban rail transit operation fault - Google Patents
Method and system for generating driving adjustment strategy under urban rail transit operation fault Download PDFInfo
- Publication number
- CN116424397A CN116424397A CN202310293799.7A CN202310293799A CN116424397A CN 116424397 A CN116424397 A CN 116424397A CN 202310293799 A CN202310293799 A CN 202310293799A CN 116424397 A CN116424397 A CN 116424397A
- Authority
- CN
- China
- Prior art keywords
- driving adjustment
- operation fault
- adjustment strategy
- rail transit
- driving
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 230000006870 function Effects 0.000 claims description 24
- 238000013527 convolutional neural network Methods 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 10
- 238000012546 transfer Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 8
- 238000012512 characterization method Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 2
- 230000003750 conditioning effect Effects 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 16
- 239000013598 vector Substances 0.000 description 9
- 210000002569 neuron Anatomy 0.000 description 7
- 238000011176 pooling Methods 0.000 description 7
- 238000011160 research Methods 0.000 description 5
- 210000004027 cell Anatomy 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- PCTMTFRHKVHKIS-BMFZQQSSSA-N (1s,3r,4e,6e,8e,10e,12e,14e,16e,18s,19r,20r,21s,25r,27r,30r,31r,33s,35r,37s,38r)-3-[(2r,3s,4s,5s,6r)-4-amino-3,5-dihydroxy-6-methyloxan-2-yl]oxy-19,25,27,30,31,33,35,37-octahydroxy-18,20,21-trimethyl-23-oxo-22,39-dioxabicyclo[33.3.1]nonatriaconta-4,6,8,10 Chemical compound C1C=C2C[C@@H](OS(O)(=O)=O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2.O[C@H]1[C@@H](N)[C@H](O)[C@@H](C)O[C@H]1O[C@H]1/C=C/C=C/C=C/C=C/C=C/C=C/C=C/[C@H](C)[C@@H](O)[C@@H](C)[C@H](C)OC(=O)C[C@H](O)C[C@H](O)CC[C@@H](O)[C@H](O)C[C@H](O)C[C@](O)(C[C@H](O)[C@H]2C(O)=O)O[C@H]2C1 PCTMTFRHKVHKIS-BMFZQQSSSA-N 0.000 description 1
- CIWBSHSKHKDKBQ-JLAZNSOCSA-N Ascorbic acid Chemical compound OC[C@H](O)[C@H]1OC(=O)C(O)=C1O CIWBSHSKHKDKBQ-JLAZNSOCSA-N 0.000 description 1
- 206010017472 Fumbling Diseases 0.000 description 1
- 206010033799 Paralysis Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 238000011272 standard treatment Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L27/00—Central railway traffic control systems; Trackside control; Communication systems specially adapted therefor
- B61L27/10—Operations, e.g. scheduling or time tables
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L27/00—Central railway traffic control systems; Trackside control; Communication systems specially adapted therefor
- B61L27/40—Handling position reports or trackside vehicle data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention belongs to the technical field of urban rail transit fault processing, and provides a method and a system for generating a driving adjustment strategy under urban rail transit operation faults. Acquiring a downlink vehicle adjustment multidimensional feature of urban rail transit operation faults, searching whether the downlink vehicle adjustment multidimensional feature is in an operation fault driving adjustment strategy decision-making library, and if so, calling an operation fault downlink driving adjustment strategy corresponding to the downlink vehicle adjustment multidimensional feature; otherwise, generating a model by adopting a trained driving adjustment strategy according to the driving adjustment multidimensional feature, predicting the driving adjustment strategy under the operation fault, and storing the corresponding relation between the driving adjustment strategy under the operation fault and the driving adjustment multidimensional feature obtained by prediction into an operation fault driving adjustment strategy decision-making library. The method can achieve the purpose of intelligently and rapidly acquiring the driving adjustment strategy suitable for different urban rail transit operation fault scenes.
Description
Technical Field
The invention belongs to the technical field of urban rail transit fault treatment, and particularly relates to a method and a system for generating a driving adjustment strategy under urban rail transit operation faults.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The urban rail transit construction scale is continuously increased, the passenger flow demand is increased year by year, the full rate of the train carriages is gradually increased, the contradiction between overload operation and construction scale is increasingly prominent, the frequency of operation faults is increased, and the passenger flow emergency evacuation difficulty is also increased. Urban rail traffic is usually arranged underground, belongs to a system with relatively closed environment, narrow internal space and denser people flow, and once operation faults occur, the normal operation order of an accident area can be disturbed, the influence on adjacent stations, intervals or lines can be caused, chain reaction is initiated, if the effective treatment is not completed, the network transportation capacity is reduced, paralysis even occurs, and the ground traffic system can be influenced under serious conditions.
Under urban rail transit operation faults, the original train operation plan cannot fully meet the travel demands of passengers. How to reasonably adjust urban rail transit driving plans, effectively evacuate sudden large passenger flows, recover normal operation order, and ensure passenger travel safety and efficiency is a great difficulty to be solved urgently.
After operation faults occur, large passenger flow evacuation becomes a primary problem, emergency measures should be timely taken to induce evacuation of passenger flow, and influence on passengers and driving is reduced. At present, the regular driving adjustment of urban rail transit has more researches, and the driving adjustment is an effective mode adopted by operation enterprises except for measures such as station sealing and current limiting. Under operation faults, the driving emergency adjustment difficulty mainly comprises the following two aspects:
firstly, driving adjustment influence feature extraction has certain difficulty under operation fault, and operation fault information data acquisition is relatively difficult. Different from a normal train operation adjustment method, the train emergency operation adjustment needs to fully consider the influence of factors such as emergency characteristics, passenger flow space-time characteristics, driving influence, network operation and the like, and the current related research is relatively few. Therefore, the multi-source heterogeneous big data are required to be combined, and the influence factors influencing the train operation under the operation faults are extracted by analyzing the operation fault attribute of the urban rail transit, the influence effect of the operation faults on passengers and traveling, and the like, so that a rapid generation method of the train emergency operation adjustment strategy is explored.
Secondly, in the field of urban rail transit emergency study, most studies tend to the theoretical exploration of sudden large passenger flow evacuation, focus on sudden large passenger flow propagation mechanism or driving adjustment optimization modeling, and relatively less studies on train operation adjustment under different influence effects; after an operation fault occurs, rail transportation enterprises generally relieve and evacuate passenger flow by improving the rail transportation capacity or adopting a method of limiting current. In the practical application technology, rail transit operation enterprises often adjust the traveling crane by adopting a manual mode based on historical experience, and the emergency adjustment scheme formed by manually judging the proper traveling crane adjustment mode has a certain reference, but the judgment of the traveling crane adjustment mode under manual intervention is long in time consumption, the traveling crane adjustment is relatively lagged, the efficiency is low, the feedback is slow, and the interference is many. After the emergency occurs, the emergency adjustment requirement is urgent, and if the emergency treatment is not performed, important influences can be generated on the traveling of passengers and the running of trains. Therefore, the research on the train operation adjustment strategy under the urban rail transit operation fault is worth further discussion and fumbling.
Currently, emergency evacuation of sudden passenger flows of a rail transit system is in an empirical operation stage, and no standard treatment method exists. After the operation fault occurs, the emergency adjustment requirement is urgent, and if the emergency adjustment requirement is not urgent, important influences can be generated on the traveling of passengers and the running of trains. Therefore, an intelligent generation method for establishing a train operation adjustment strategy under operation faults is needed to provide technical support for efficient evacuation of sudden large passenger flows.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a method and a system for generating a driving adjustment strategy under urban rail transit operation faults, which can realize the purpose of intelligently and rapidly acquiring driving adjustment strategies suitable for being adopted under different operation fault scenes.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
the first aspect of the invention provides a driving adjustment strategy generation method under urban rail transit operation faults.
A method for generating a driving adjustment strategy under urban rail transit operation faults comprises the following steps:
acquiring downlink vehicle adjustment multidimensional features of urban rail transit operation faults, searching whether the driving adjustment multidimensional features are in an operation fault driving adjustment strategy decision-making library, and if so, calling an operation fault driving adjustment strategy corresponding to the driving adjustment multidimensional features;
otherwise, generating a model by adopting a trained driving adjustment strategy according to the driving adjustment multidimensional feature, predicting the driving adjustment strategy under the operation fault, and storing the corresponding relation between the driving adjustment strategy under the operation fault and the driving adjustment multidimensional feature obtained by prediction into an operation fault driving adjustment strategy decision-making library.
Further, the driving adjustment strategy generation model training process comprises the following steps:
based on the driving adjustment multidimensional feature, capturing a spatial relationship of the driving adjustment multidimensional feature by adopting a convolutional neural network (ConvolutionalNeural Network, CNN) model, and capturing time dependence in data processed by the CNN model by adopting a long short-time memory network (LongShort-TermMemory, LSTM) model to obtain driving adjustment feature data;
based on driving adjustment characteristic data, a fully connected neural network (Dense) is adopted for fusion, probability distribution is output by means of a normalized exponential function (softmax function), a cross entropy function is used as a loss function of a classification part, the error descending direction is searched in gradient counter propagation through an adam optimizer, and when the set iteration times are reached, a global optimal solution is obtained.
Further, the multi-dimensional characteristic of the down-going vehicle adjustment of the urban rail transit operation fault comprises an operation fault event attribute, an influence representation factor of the operation fault on driving and an influence representation factor of the operation fault on passenger traveling.
Still further, the operation fault event attribute includes an issue date, an issue time, an issue duration, an issue position and an issue reason, the operation fault impact characterization factors on driving include an issue line departure interval, an issue line length and an issue period plan issue train number, and the operation fault impact characterization factors on passenger travel include an issue station inbound passenger flow rate, an issue station outbound passenger flow rate, an OD passenger flow rate and an issue station transfer passenger flow rate.
Wherein, O (origin) represents an initial station of the passenger travel, D (Destination) represents a destination station, OD travel is the travel between two stations, and OD passenger flow represents the number of passengers traveling between the two stations.
Further, when an operation failure occurs in a section or zone, the section or zone is identified as an event station.
Further, the operation fault running adjustment strategy comprises 15 operation fault running adjustment strategies which are randomly combined in four modes of only stopping trains, only stopping trains or stopping trains in stations, only going beyond stations and only clearing people and turning back.
Further, the method further comprises the step of preprocessing the operation fault driving adjustment information characteristic data before searching whether the driving adjustment multidimensional characteristic is in the driving adjustment strategy decision-making library under the operation fault.
The second aspect of the invention provides a driving adjustment strategy generation system under urban rail transit operation faults.
A driving adjustment strategy generation system under urban rail transit operation faults comprises:
a search module configured to: acquiring downlink vehicle adjustment multidimensional features of urban rail transit operation faults, searching whether the driving adjustment multidimensional features are in an operation fault driving adjustment strategy decision-making library, and if so, calling an operation fault driving adjustment strategy corresponding to the driving adjustment multidimensional features; otherwise, the method proceeds to a prediction module;
a prediction module configured to: and generating a model by adopting a trained driving adjustment strategy according to the driving adjustment multidimensional characteristic, predicting the driving adjustment strategy under the operation fault, and storing the corresponding relation between the driving adjustment strategy under the operation fault and the driving adjustment multidimensional characteristic obtained by prediction into an operation fault driving adjustment strategy decision library.
A third aspect of the present invention provides a computer-readable storage medium.
A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps in the urban rail transit operation failure driving adjustment policy generation method according to the first aspect described above.
A fourth aspect of the invention provides a computer device.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the urban rail transit operation failure driving adjustment policy generation method according to the first aspect described above when the program is executed.
Compared with the prior art, the invention has the beneficial effects that:
the invention can intelligently and pre-judge the driving adjustment strategy adopted for evacuating the sudden passenger flow by inputting the multidimensional influence characteristics of the urban rail transit operation faults, can realize the purpose of intelligently and rapidly acquiring the driving adjustment strategy adopted under different operation fault scenes, reduces the manual interference, shortens the decision time consumption of emergency treatment, provides references for the formulation of the driving adjustment scheme, effectively shortens the decision time consumption of emergency treatment, and can provide references and support for the formulation of the driving emergency adjustment method and the adjustment of the operation diagram under the sudden operation faults of the urban rail transit.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
FIG. 1 is a flow chart of a driving adjustment strategy generation method under urban rail transit operation faults, which is shown in the invention;
FIG. 2 is a block diagram of a driving adjustment strategy generation method under urban rail transit operation faults shown in the invention;
FIG. 3 is a schematic diagram of an urban rail transit operation failure in a station and section shown in the present invention;
FIG. 4 is a schematic diagram of the spatial distance relationship under the operation fault of the urban rail transit shown in the invention;
fig. 5 is a diagram of a decision network structure of a driving adjustment strategy under the operation fault of urban rail transit.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
It is noted that the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present disclosure. It should be noted that 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 logical functions specified in the various embodiments. 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 flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by special purpose hardware-based systems which perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.
Example 1
As shown in fig. 1, the present embodiment provides a method for generating a driving adjustment policy under an urban rail transit operation fault, and the present embodiment is applied to a server for illustration by using the method, and it can be understood that the method may also be applied to a terminal, may also be applied to a system and a server, and is implemented through interaction between the terminal and the server. The server can be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, and can also be a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network servers, cloud communication, middleware services, domain name services, security services CDNs, basic cloud computing services such as big data and artificial intelligent platforms and the like. The terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited herein. In this embodiment, the method includes the steps of:
(1) Data preprocessing: including normalization, deletion of missing values, etc.
(2) Model training: and inputting data of the training set into a model for training, and primarily extracting features by the CNN layer and learning time sequence features by the LSTM. The CNN module consists of 3 layers of one-dimensional convolutional neural networks, and the number and the convolution size of each layer are respectively 64, 32, 32 and 5 multiplied by 1,3 multiplied by 1 and 3 multiplied by 1. The relation between driving adjustment characteristic data under the operation fault of urban rail transit is discovered through the convolution kernel depths of multiple layers of different shapes, and hidden characteristics in the data are learned. And then, inputting the primary features learned by the CNN module into the LSTM layer to perform deeper feature extraction. The LSTM module consists of two layers of LSTM neural networks, the number of each layer of neurons is 64 and 32 respectively, irrelevant variables in shallow characteristics are ignored through a unique forgetting mechanism of the LSTM neurons, and the most critical data characteristics for regulating the descending vehicles with operation faults are reserved.
(3) And (3) outputting a prediction result: and finally learning the advanced features learned by the LSTM layer by using the Dense layer. The Dense layer connects the neurons of each layer with all neurons of the upper layer, establishes global connection with the high-dimensional features extracted by the LSTM, integrates global information of the whole input, outputs probability distribution by means of softmax, and finally uses cross entropy functions as a loss function of a classification part, and searches the direction of error reduction in gradient back propagation by an adam optimizer. Through a large number of iterative operations, the global optimal solution can be found.
And establishing a driving adjustment strategy decision base under operation faults. The driving adjustment strategy multidimensional feature under the operation fault comprises 13 influencing factors, different operation fault scenes can be built by random combination of the factors, the specific rule is obtained by training, testing and training based on the operation fault multidimensional feature and the driving adjustment method, and the specific rule is fed back into the driving adjustment strategy decision library so as to achieve feature extraction and matching of the driving adjustment method and the multidimensional feature under different operation fault scenes. The decision flow of the driving adjustment strategy under the operation fault is shown in figure 1.
Adjusting driving of operation faults to multidimensional features
Inputting the feature library X, judging whether the feature library X can be searched for +.>
If there is corresponding characteristic information and +.>
If the operation fault driving adjustment strategy library is matched, searching +.>Corresponding driving adjustment strategies; if there is no->
Based on->
Predicting driving adjustment strategy under operation fault, outputting corresponding driving adjustment strategy, and simultaneously adding +.>
And feeding back the corresponding driving adjustment strategy and storing the driving adjustment strategy into an operation fault driving adjustment strategy library X.
The embodiment provides a method for generating a train operation adjustment strategy by considering the attribute of an operation fault event of urban rail transit and the influence effect of operation faults on passengers and traveling, and the method is as shown in fig. 2.
The basic data of the research method provided by the embodiment comprises the following steps: urban rail transit operator interview data, passenger travel willingness investigation data, AFC data, urban rail transit network geographic information data and driving adjustment record data under operation faults. The driving adjustment record data is used for recording the occurrence time, place, influence result and actually adopted emergency driving adjustment method of the urban rail transit operation fault in detail.
The common driving adjustment method under the operation fault determination mainly comprises four types of train stop (A), station buckling or interval temporary stopping (B), train station crossing driving (C) and train clear turning (D).
Specifically, according to four types of modes of train operation adjustment under operation faults, the driving adjustment strategies can be generated by combining the four types of driving adjustment modes, and 15 combination modes are provided, wherein A, B, C, D, AB, AC, AD, BC, BD, CD, ABC, ABD, ACD, BCD, ABCD is included, and the driving adjustment strategies under operation faults and the descriptions are shown in table 1.
TABLE 1 traffic adjustment strategy and description under operation failure
Wherein, determining driving adjustment influencing factors under operation faults, and can be used
Indicating (I)>
The driving adjustment influencing factors under the operation faults comprise: the operation fault event attribute, the influence of the operation fault on the traveling of passengers and the influence of the operation fault on the traveling are three 13 influence factors.
The operational failure event attributes include: date of occurrence x 1 (day of the week in which operation failure occurred), time of occurrence x 2 (specific time-division of operation faults occurring in one day), estimated time of occurrence x 3 The incident position (distance x between the incident position and the initial station 4 Distance x between the incident location and the nearest reentrant station 5 ) Cause of occurrence x 6 (vehicle signal fault, ground signal fault, vehicle fault, screen door fault or other fault);
the impact of an operational fault on passengers is quantitatively expressed in terms of passenger traffic, including: incoming passenger flow x of bus stop 7 Outbound passenger flow x of bus stop 8 OD passenger flow volume x 9 Transfer passenger flow x with departure station 10 ；
The influence of operation faults on driving comprises the following steps: hairlineDistance x for road departure 11 Length of incident line x 12 Number x of planned released trains in event period 13 。
The operation fault occurrence position comprises three parts of an accident station, an accident interval and an accident section. Schematic diagrams of the occurrence of the station and the occurrence of the interval are shown in fig. 3.
N stations are arranged on the operation fault occurrence line, and the station serial numbers are 1,2, … and n in sequence according to the uplink direction. When the operation faults occur in the section, the stations at the two ends of the section are always affected incidentally, and the stations at the two ends of the section are considered to be in an operation fault occurrence state when the operation faults occur in the section. Similarly, when an operation failure occurs in a section, stations at both ends of the section can be considered to be in an emergency state. Therefore, when the operation fault occurs in the section or the zone, the whole section or the zone is in the accident state, and the section or the zone can be kneaded into an accident station e, that is, whether the operation fault occurs in the station, the section or the zone, the running adjustment strategy research can consider the station e as the accident position, and the distance relation is shown in fig. 4.
In FIG. 4, z is a return station, e is an incident station, l oe For the shortest distance between the accident station and the originating station, l ze For the shortest distance between the accident station and the turn-back station, l od To deal with the length of the line, l z Is the return distance.
The respective positional relationships are expressed as follows:
1≤z≤e≤n(3)
wherein d i,i+1 Indicating the distance between station i and station i+1.
The calculation method of the passenger flow of the departure station in-and-out station is as follows:
wherein t is e Representing an operational failure duration; f (F) 1 Representing the arrival passenger flow of the incident station within the duration of the operation fault; f (F) 2 Indicating the outbound passenger flow of the incident station within the duration of the operation fault; x is x 1,j The arrival passenger flow of the station is sent out on the j th day before and after the fault occurs within the duration of the operation fault; x is x 2,j The outbound passenger flow of the accident train station at the j th day before and after the occurrence of the fault is represented within the duration of the operation fault; g represents the number of days of arrival and departure of passenger flow in the same period of history of the departure station.
The change in OD passenger flow also reflects a change in passenger travel traffic patterns, which is directly related to driving adjustment. OD passenger flow can be obtained by calculation of clear model of urban rail transit system.
Specifically, if the departure station is a transfer station, the departure station has a great influence on driving adjustment and also includes passenger flow transferred from the transfer line to the departure line. When an operation fault occurs in a transfer station, the driving adjustment and passenger flow evacuation pressure are relatively large.
Wherein F is h Representing passenger flow from other lines to the incident line at the incident station within the duration of the operation fault; q χ Representing the transfer passenger flow of the line χ to the accident station within the duration of the operation fault; w represents the number of lines having a transfer function with the incident line, and w=0 represents that the station is a normal station at the incident station and transfer cannot be performed.
Influence of operation faults on driving by accident line length l od Distance of return l z Four indexes of train departure intervals and planned departure quantity in an operation fault influence period are quantified. The train departure interval is represented by the departure interval Δt of the departure line in normal state. For the planned departure number n in the occurrence period of operation failure e The method comprises the following steps:
based on a pytorch framework, a CNN model and an LSTM model are fused to construct a train operation adjustment strategy generation model under operation faults, and a method for constructing a train operation adjustment strategy library under operation faults is provided.
The LSTM can well complete time feature extraction and is used for processing time series data, but when an input sequence is overlong, important data information is lost, and CNN is needed to process original data and filter out a part of unimportant information. Meanwhile, the CNN model has better spatial feature information learning capability. The down going vehicle of the urban rail transit operation fault adjusts multidimensional characteristic information to have time characteristics (such as time of occurrence and duration of occurrence) and space characteristics (such as distance between the occurrence position and the starting station, distance between the occurrence position and the turning-back station and the like). The planned departure quantity in the departure time period, the arrival and departure passenger flow of the departure station and the like are all related to the departure time length. Therefore, in order to enable the proposed method to have the characteristic expression capability of time and space, the spatial relationship of the characteristic data is captured by adopting CNN, LSTM is combined after CNN, and the time dependence in the data is captured by adopting LSTM. The advantages of the CNN model and the LSTM model are combined, the characteristic value training is carried out, and a generation method of a driving adjustment strategy under proper operation faults is established. The network architecture of the design is shown in fig. 5.
Firstly, carrying out data preprocessing on characteristic data of driving adjustment information of operation faults, including data normalization, deletion of missing values and the like, then inputting the processed data into one-dimensional CNN and LSTM models, extracting driving adjustment effective information, and filtering to invalid characteristic information.
The CNN network has the main function of extracting the characteristics of the data and consists of a convolution layer and a pooling layer. The convolution layer is the key to extracting features, and deeper implicit features are extracted through convolution kernels of a plurality of different shapes. The operation fault driving adjustment related data has stronger nonlinearity, the characteristics of the original data can not well reflect the change of driving adjustment strategies, and the CNN model is required to be adopted to extract effective information in the original characteristics. The structure of a one-dimensional CNN includes 3 convolutional layers, 2 max pooling layers, and 1 average pooling layer.
Feature vector pair using convolutional layers
The convolution operation is carried out, and the output of the convolution layer is represented as follows:
in the method, in the process of the invention,
from the output vector of the upper layer +.>
Calculating to obtain; />
A bias representing a jth feature map; />
Representing the weight of the convolution kernel.
After the operation fault driving adjustment characteristic data is subjected to convolution operation, the operation fault driving adjustment characteristic data is transmitted into a pooling layer to further reduce the parameter quantity, compress the data dimension and reduce the overfitting. And screening out characteristic information with the greatest influence on the driving adjustment method under the operation fault by adopting a maximum pooling method, and predicting a proper driving adjustment method. The maximum pooling method is expressed as follows:
where r represents the step size of the pooled region.
Specifically, LSTM receives the CNN extracted significant feature vector sequence. LSTM neural network is implemented by extracting feature vector x t State memory cell c t-1 Intermediate output h t-1 Input to forget door f t Thereby determining the part of the state memory unit which needs to be forgotten; input gate i t X in (2) t After respectively passing through sigmoid and tanh activation functions, the vectors needing to be reserved in the state memory unit are determined together; intermediate output h t From updated state memory cell C t And an output gate O t And (5) jointly determining.
f t For forgetting the output information of the gate, the expression is as follows:
f t ＝sigmoid(W f ·[h t-1 ,x t ]+b f )(10)
in which W is f Connection weight representing forget gate; b f A bias vector representing a forgetting gate.
Further, the hidden state information h outputted by the previous LSTM unit t-1 And the information x currently entered t Respectively inputting into sigmoid function and tanh function to obtain i t And
the formula is as follows:
i t ＝sigmoid(W i ·[h t-1 ,x t ]+b i )(11)
in which W is i 、W c Representing connection weights associated with the input gates; b i 、b c Representing the bias vector of the input gate.
At i t And
on the basis of (a), an update of the cell state can be obtained, the formula is as follows:
in the method, in the process of the invention,
representation->
The information updating of the LSTM unit can be completed through the above formula.
H for output result of output gate t Representing, first, the hidden state information h outputted from the previous LSTM unit t-1 And the information x currently entered t Input into sigma function to obtain O t As shown in equation (14).
O t ＝sigmoid(W o ·[h t-1 ,x t ]+b o )(14)
Wherein W is o And b o Representing the connection weight and offset vector associated with the output gate, respectively.
State C of the cell at the present time t Inputting into a tanh function to obtain tanh (C t ) Then tan h (C t ) And O t Multiplying to obtain information h to be carried in hidden state t As shown in equation (15).
h t ＝O t ·tanh(C t )(15)
Finally, new cell state C t And new hidden state information h t To the LSTM element at the next instant.
Specifically, the driving adjustment characteristic data output by the CNN and LSTM technology are fused by using a Dense network. The neurons after pooling are unfolded into a one-dimensional vector form by the Dense layer, so that the data are more conveniently processed, the neurons of each layer are connected with all neurons of the upper layer by the Dense layer, the whole input global information is integrated, and after each piece of data is processed, the connection weight is changed according to the error between the output and the expected result. Finally, the classification prediction of the driving adjustment strategy is carried out by utilizing Softmax.
Example two
The embodiment provides a driving adjustment strategy generation system under urban rail transit operation faults.
A driving adjustment strategy generation system under urban rail transit operation faults comprises:
a search module configured to: acquiring downlink vehicle adjustment multidimensional features of urban rail transit operation faults, searching whether the driving adjustment multidimensional features are in an operation fault driving adjustment strategy decision-making library, and if so, calling an operation fault driving adjustment strategy corresponding to the driving adjustment multidimensional features; otherwise, the method proceeds to a prediction module;
a prediction module configured to: and generating a model by adopting a trained driving adjustment strategy according to the driving adjustment multidimensional characteristic, predicting the driving adjustment strategy under the operation fault, and storing the corresponding relation between the driving adjustment strategy under the operation fault and the driving adjustment multidimensional characteristic obtained by prediction into an operation fault driving adjustment strategy decision library.
It should be noted that the search module and the prediction module are the same as the examples and application scenarios implemented by the steps in the first embodiment, but are not limited to the disclosure of the first embodiment. It should be noted that the modules described above may be implemented as part of a system in a computer system, such as a set of computer-executable instructions.
Example III
The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps in the urban rail transit operation failure driving adjustment policy generation method according to the above embodiment.
Example IV
The embodiment provides a computer device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the steps in the method for generating the driving adjustment strategy under the urban rail transit operation fault according to the embodiment when executing the program.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, magnetic disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, are implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-only memory (ROM), a random access memory (RandomAccessMemory, RAM), or the like.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The method for generating the driving adjustment strategy under the urban rail transit operation fault is characterized by comprising the following steps:
acquiring downlink vehicle adjustment multidimensional features of urban rail transit operation faults, searching whether the driving adjustment multidimensional features are in an operation fault driving adjustment strategy decision-making library, and if so, calling an operation fault driving adjustment strategy corresponding to the driving adjustment multidimensional features;
otherwise, generating a model by adopting a trained driving adjustment strategy according to the driving adjustment multidimensional feature, predicting the driving adjustment strategy under the operation fault, and storing the corresponding relation between the driving adjustment strategy under the operation fault and the driving adjustment multidimensional feature obtained by prediction into an operation fault driving adjustment strategy decision-making library.
2. The method for generating a driving adjustment strategy under urban rail transit operation failure according to claim 1, wherein the process of training the driving adjustment strategy generation model comprises the following steps:
capturing the spatial relationship of the driving adjustment multi-dimensional characteristics by adopting a convolutional neural network model based on the driving adjustment multi-dimensional characteristics, and capturing the time dependence in the data processed by the convolutional neural network model by adopting a long-short-term memory network model to obtain driving adjustment characteristic data;
based on driving adjustment characteristic data, a fully connected neural network is adopted for fusion, probability distribution is output by means of a normalized exponential function, a cross entropy function is used as a loss function of a classification part, the error descending direction is searched in gradient back propagation through an adam optimizer, and when the set iteration times are reached, a global optimal solution is obtained.
3. The method for generating the driving adjustment strategy under the urban rail transit operation fault according to claim 1, wherein the urban rail transit operation fault downlink adjustment multidimensional feature comprises an operation fault event attribute, an operation fault influence characterization factor on driving and an operation fault influence characterization factor on passenger traveling.
4. The method for generating a driving adjustment strategy under urban rail transit operation fault according to claim 3, wherein the operation fault event attribute comprises an occurrence date, an occurrence time, an occurrence duration, an occurrence position and an occurrence reason, the operation fault influence characterization factors comprise an occurrence line departure interval, an occurrence line length and an occurrence period plan release train quantity, and the operation fault influence characterization factors comprise an occurrence station arrival passenger flow volume, an occurrence station departure passenger flow volume, an OD passenger flow volume and an occurrence station transfer passenger flow volume.
5. The method for generating a traffic adjustment strategy under urban rail transit operation failure according to claim 4, wherein when the operation failure occurs in a section or zone, the section or zone is identified as an accident station.
6. The method for generating the running adjustment strategy under the urban rail transit operation fault according to claim 1, wherein the running adjustment strategy under the operation fault comprises 15 running adjustment strategies under the operation fault, which are randomly combined in four modes of only stopping a train, only stopping a train or a section, only crossing the train and only turning back a clear person.
7. The method for generating a traffic adjustment strategy under urban rail transit operation failure according to claim 1, wherein the step of preprocessing the operation failure traffic adjustment information feature data before searching whether the traffic adjustment multidimensional feature is in the traffic adjustment strategy decision library under operation failure is further included.
8. The system for generating the driving adjustment strategy under the urban rail transit operation fault is characterized by comprising the following components:
a search module configured to: acquiring downlink vehicle adjustment multidimensional features of urban rail transit operation faults, searching whether the driving adjustment multidimensional features are in an operation fault driving adjustment strategy decision-making library, and if so, calling an operation fault driving adjustment strategy corresponding to the driving adjustment multidimensional features; otherwise, the method proceeds to a prediction module;
a prediction module configured to: and generating a model by adopting a trained driving adjustment strategy according to the driving adjustment multidimensional characteristic, predicting the driving adjustment strategy under the operation fault, and storing the corresponding relation between the driving adjustment strategy under the operation fault and the driving adjustment multidimensional characteristic obtained by prediction into an operation fault driving adjustment strategy decision library.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the steps in the urban rail transit operation failure driving adjustment policy generation method according to any one of claims 1-7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps in the method for generating a traffic conditioning policy under urban rail transit operation failure according to any one of claims 1-7 when the program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310293799.7A CN116424397A (en) | 2023-03-21 | 2023-03-21 | Method and system for generating driving adjustment strategy under urban rail transit operation fault |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310293799.7A CN116424397A (en) | 2023-03-21 | 2023-03-21 | Method and system for generating driving adjustment strategy under urban rail transit operation fault |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116424397A true CN116424397A (en) | 2023-07-14 |
Family
ID=87082502
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310293799.7A Pending CN116424397A (en) | 2023-03-21 | 2023-03-21 | Method and system for generating driving adjustment strategy under urban rail transit operation fault |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116424397A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116739392A (en) * | 2023-08-14 | 2023-09-12 | 北京全路通信信号研究设计院集团有限公司 | Multi-system rail transit emergency collaborative decision-making method and device based on physical elements |
-
2023
- 2023-03-21 CN CN202310293799.7A patent/CN116424397A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116739392A (en) * | 2023-08-14 | 2023-09-12 | 北京全路通信信号研究设计院集团有限公司 | Multi-system rail transit emergency collaborative decision-making method and device based on physical elements |
| CN116739392B (en) * | 2023-08-14 | 2023-10-27 | 北京全路通信信号研究设计院集团有限公司 | Multi-system rail transit emergency collaborative decision-making method and device based on physical elements |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | Short‐term traffic flow prediction with linear conditional Gaussian Bayesian network | |
| CN107103758B (en) | A kind of city area-traffic method for predicting based on deep learning | |
| CN108831153A (en) | A kind of traffic flow forecasting method and device using spatial and temporal distributions characteristic | |
| DE112018007119T5 (en) | OPTIMIZE AND PREDICT THE AVAILABILITY OF RESOURCES IN A BICYCLE SHARED ENVIRONMENT | |
| CN109635246B (en) | Multi-attribute data modeling method based on deep learning | |
| CN115440032B (en) | Long-short-period public traffic flow prediction method | |
| CN113326974B (en) | Multi-source traffic flow prediction method based on multi-task hypergraph convolutional network | |
| Zhao et al. | Research on traffic accident prediction model based on convolutional neural networks in VANET | |
| CN108629503A (en) | A kind of taxi based on deep learning is got on the bus the prediction technique of demand | |
| CN111667092A (en) | Method and system for predicting short-term passenger flow of rail transit based on graph convolution neural network | |
| CN110570035B (en) | People flow prediction system for simultaneously modeling space-time dependency and daily flow dependency | |
| CN116424397A (en) | Method and system for generating driving adjustment strategy under urban rail transit operation fault | |
| CN114626585B (en) | Urban rail transit short-time passenger flow prediction method based on generation countermeasure network | |
| WO2024098956A1 (en) | Method for fusing social media data and moving track data | |
| Kong et al. | The method and application of big data mining for mobile trajectory of taxi based on MapReduce | |
| CN113554060B (en) | LSTM neural network track prediction method integrating DTW | |
| Hsueh et al. | A short-term traffic speed prediction model based on LSTM networks | |
| CN113761020B (en) | Urban rail transit abnormal large passenger flow real-time prediction method | |
| Hlaing et al. | Big traffic data analytics for smart urban intelligent traffic system using machine learning techniques | |
| CN115762147B (en) | Traffic flow prediction method based on self-adaptive graph meaning neural network | |
| Yu et al. | Estimating left behind patterns in congested metro systems: a Bayesian model | |
| Ayman et al. | Neural Architecture and Feature Search for Predicting the Ridership of Public Transportation Routes | |
| CN115409276A (en) | Traffic prediction transfer learning method based on space-time diagram self-attention model | |
| Jia et al. | Traffic flow prediction based on self-attention mechanism and deep packet residual network | |
| CN113535978B (en) | Construction method of rail transit system operation risk ontology |
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 |