CN111291856B - Multi-objective optimization method and system for operation and control of subway train - Google Patents
Multi-objective optimization method and system for operation and control of subway train Download PDFInfo
- Publication number
- CN111291856B CN111291856B CN202010072449.4A CN202010072449A CN111291856B CN 111291856 B CN111291856 B CN 111291856B CN 202010072449 A CN202010072449 A CN 202010072449A CN 111291856 B CN111291856 B CN 111291856B
- Authority
- CN
- China
- Prior art keywords
- optimization
- train
- objective
- subway
- subway train
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 155
- 238000000034 method Methods 0.000 title claims abstract description 59
- 239000002245 particle Substances 0.000 claims abstract description 78
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 31
- 230000004927 fusion Effects 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 21
- 238000004088 simulation Methods 0.000 claims abstract description 20
- 230000002068 genetic effect Effects 0.000 claims abstract description 15
- 238000005265 energy consumption Methods 0.000 claims abstract description 13
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 230000007246 mechanism Effects 0.000 claims abstract description 11
- 230000006872 improvement Effects 0.000 claims abstract description 9
- 230000001133 acceleration Effects 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 10
- 230000003750 conditioning effect Effects 0.000 claims description 9
- 238000013500 data storage Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000004134 energy conservation Methods 0.000 abstract description 7
- 238000012360 testing method Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012803 optimization experiment Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Primary Health Care (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention provides a multi-objective optimization method and system for subway train operation control. In order to solve the optimization problem that the train operation process needs to meet a plurality of performance indexes such as energy conservation, punctuality, comfort and the like at the same time, the invention takes the energy consumption, comfort and punctuality of the train as optimization targets, and establishes a train operation control multi-target optimization model; the particle swarm improvement optimization algorithm introducing the genetic evolution mechanism adopts a multi-target included angle cosine as an evaluation standard of the solution quality, and simultaneously adopts the fusion distance as a judgment condition for judging that the particle swarm is gathered on the extremum particles so as to obtain the comprehensive performance index of the optimal subway train operation process. According to the optimization method, a corresponding optimization system is designed, and simulation tests are carried out under two different subway train automatic driving scenes. The test result shows that the technical scheme of the invention has better optimizing performance and can obtain better optimizing result under the condition that the subway train and the running line thereof are appointed and the planned running time is the same.
Description
Technical Field
The invention relates to the technical field of subway train operation, in particular to a subway train operation control multi-objective optimization method and system.
Background
The operation control optimizing system in the automatic train driving system has good operation control optimizing performance so as to be capable of giving a subway train target speed track which enables all optimizing indexes such as comfort, energy saving, punctual and accurate stopping in the automatic train driving operation process to be optimized as much as possible. The method is an important premise for ensuring that the subway train operation process has operation control comprehensive performance indexes with good comfortableness, energy conservation and punctuality.
With the vigorous development of subway traffic industry, subway traffic is becoming popular with more and more people due to the characteristics of rapidness, convenience and comfort. However, the train running process is an optimization problem that needs to meet multiple performance indexes such as energy saving, punctual time, comfort and the like at the same time, and more than one or even an infinite number of Pareto solutions which are mutually non-inferior exist in the optimized solution set. Therefore, the optimization solution obtained by means of single optimization target optimization cannot simultaneously achieve optimization of the comprehensive performance indexes of multiple performance indexes such as energy conservation, punctual performance and comfort performance. In order to achieve the aim of optimizing all optimization indexes such as comfort, energy conservation, punctual time, accurate stopping and the like in the automatic driving operation process of the subway train as much as possible, a multi-objective operation and control optimizing method and system for the subway train with good operation and control optimizing performance are required to be designed.
Disclosure of Invention
According to the technical problems, the method and the system for optimizing the operation and control of the subway train are provided. The invention establishes a multi-target optimizing model for train operation control by taking train energy consumption, comfort and punctuality as optimizing targets, and provides a multi-target genetic particle swarm algorithm based on an included angle cosine and a fusion distance. Compared with a linear weighting method, the method adopts the cosine of the included angle between the solution target vector and the target demand vector as the evaluation index of the solution quality, and is more objective and reasonable, so that the problem of blind selection of subjective parameters is avoided. Compared with Euclidean distance, the fusion distance can be adopted to give consideration to the relativity and independence of multiple characteristic variables, and whether individual aggregation phenomena exist at the end of iteration of particles can be accurately detected, so that local convergence of the particles is restrained. Under the condition that the subway train and the running line thereof are designated and the planned running time is the same, the multi-objective optimizing method and system for the running operation of the subway train designed by the invention have better optimizing performance and can obtain better optimizing results.
The invention adopts the following technical means:
a subway train operation control multi-objective optimization method comprises the following steps:
s1, taking train energy consumption, comfort and punctuality as optimization targets, and establishing a train operation control multi-target optimization model;
s2, a particle swarm improvement optimization algorithm of a genetic evolution mechanism is introduced, and a group of optimal train control sequences are selected to serve as an optimal solution;
s3, on the basis of the step S2, taking the cosine of the multi-target included angle as an evaluation standard for resolving the quality, and taking the fusion distance as a judgment condition for judging that the particle swarm is gathered on the extremum particles to obtain the best comprehensive performance index of the subway train in the running process.
Further, the train operation manipulation multi-objective optimization model established in the step S1 is as follows:
wherein K is E Represents the energy consumption measurement index, and the energy consumption measurement index,
K A indicating comfort metrics->
K T Indicating punctual metrics>
Wherein a is i Acceleration representing the ith operating condition, si representing displacement of the ith operating condition, R i Resistance representing the i-th operating condition; min represents the minimum value of the function F (u), if the objective vector of the multi-objective optimization problem is F (x) = (F) 1 (x),f 2 (x),…,f k (x) A specific optimization model thereof can be expressed as:
where k is the number of optimization objectives, x is the decision variable, g (x) is the equality constraint and inequality constraint of x, x "is the absolute optimal solution (Pareto optimal), and F (x") is superior to F (x') if and only if any of D is.
Further, the step S3 includes the steps of:
s31, initializing:
initializing the speed and the position of a population, setting the maximum iteration number d, giving an acceleration constant c1, c2 and a weight coefficient omega, initializing the optimal position of a particle individual, and initializing the optimal position of the particle population;
s32, calculating a cosine value of the multi-target included angle of each particle according to the following formula;
where γ represents the cosine of the included angle, (T, C) represents the dot product of the solution target vector T and the target demand vector C, the term A represents the modular length of vector A, represents a numerical multiplication, T i And c i Normalized values representing the ith optimization objective of the solution objective vector T and the objective demand vector C;
s33, obtaining an individual extremum and a global extremum of the particle swarm by using sequencing;
s34, updating the speed and the position of each particle by adopting an updating formula of the speed and the position of the particle; the update formula for the velocity and position of the particles is as follows:
wherein i e [1,2, ], N]The serial number of the particle, t e [1, 2., d.]The t-th dimension of the particles is represented, d represents the iteration number, c1 and c2 are acceleration constants, rand is a random real number in the interval (0, 1), ω is a weight coefficient, and the position vector is represented as
The velocity vector is denoted +.>
The optimal position of the individual particles is marked as->
The optimal position of the population is marked as->
S35, calculating a fusion distance, wherein a calculation formula of the fusion distance is as follows:
d Mix ＝ω×MD(X,Y)+(1-ω)×ED(X,Y)
wherein d Mix Represents fusion distance, MD represents Mahalanobis distance, C Y A correlation coefficient matrix representing a sample set Y, n representing the number of samples in the sample set Y, Y i (i=1, …, n) represents the samples in the sample set YThe book, ρ, represents the correlation coefficient;
s36, judging whether the particle swarm is aggregated to the individual polar value by adopting the fusion distance calculated in the step S35, and if the particle swarm is not aggregated, turning to the step S37; if the particles are aggregated, adopting the operations of selection, crossing and mutation of a genetic algorithm on the corresponding particles, and turning to step S37;
s37, judging whether the maximum iteration number d is reached, if not, adding 1 to the iteration number, and turning back to the execution step S33; otherwise, outputting the comprehensive performance index of the running process of the optimal subway train.
Further, the judgment formula for judging whether the particle swarm is aggregated in the extremum of the individual in the step S36 is as follows:
in the method, in the process of the invention,
the sample space formed by the individual extremum of the particles is represented, dis|A-B| represents the distance from the sample A to the sample space B, epsilon represents the threshold value, and Dis adopts the fusion distance.
The invention also provides a subway train operation control multi-objective optimization system, which comprises: a lower control loop and an upper optimization loop;
the lower control loop comprises a sensor, a conditioning circuit, a controller, a traction transmission link, a main circuit breaker, a train line data storage unit and an intermediate link circuit which are electrically connected;
the upper layer optimization loop comprises a network connecting wire, a transmitter, a signal processor, an optimizer, a main circuit breaker, a train line data storage unit and an intermediate link simulation circuit which are electrically connected;
the system has the functions of off-line optimization and on-line optimization; the subway train operation control multi-objective optimization method is applied to an offline optimization function of the system, is embedded in an offline optimization function module area of an optimizer core chip, does not need actual train tracking control during offline optimization, and is replaced by a simulation circuit by isolating a lower control loop from an intermediate link circuit; during on-line optimization, the lower control loop and intermediate link circuit are enabled to implement the actual train tracking control.
Further, the off-line optimization is used for optimizing the control sequence to the greatest extent based on the comprehensive performance index of the train operation process obtained by off-line data optimization when the train stops running; the on-line optimization is used for adjusting the control sequence obtained by off-line optimization by combining the current control condition when the train runs on line so as to obtain the optimal running optimization effect of actual tracking control.
Further, the optimizer is used for providing the corresponding target speed track for the controller;
the controller is used for controlling the traction transmission link to track and control the target speed track;
the train line data storage unit is used for providing necessary related data, conditions and parameters of lines and trains for subway train operation optimization and tracking control;
the sensor, the network connecting wire and the transmitter respectively acquire corresponding data for the controller and the optimizer to implement tracking control and operation optimization calculation;
the conditioning circuit and the signal processor are respectively used for conditioning and processing the electric signals acquired by the sensor, the network connection line and the network signals acquired by the transmitter into signals acceptable by the core chips of the controller and the optimizer;
the traction transmission link is driven by a traction motor to apply power to the subway train, so that the subway train is braked or in traction operation, and the electric energy of the subway train is transmitted from a power grid by an intermediate link circuit;
the main breaker is used for switching on and off a circuit and protecting a system.
Compared with the prior art, the invention has the following advantages:
1. the invention adopts the multi-objective included angle cosine as the evaluation index, and can effectively avoid the problem that part of parameters need to be subjectively determined in the evaluation index.
2. The fusion distance is adopted, and the correlation and the independence of decision variables can be considered, so that whether the phenomenon of particle aggregation exists can be detected more accurately, and the local convergence of a particle swarm algorithm is better inhibited.
Based on the reasons, the invention can be widely popularized in the fields of subway train operation and the like.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort to a person skilled in the art.
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a schematic diagram showing cosine of an included angle between a solution target vector and a target demand vector in the method of the present invention.
FIG. 3 is a system design of the present invention.
Fig. 4 is a schematic diagram of a ramp and speed limiting curve (large-connection subway 12 line) provided by an embodiment of the invention.
Fig. 5 is a schematic diagram of a ramp and speed limit curve (Jin Pu subway line 1) according to another embodiment of the present invention.
Fig. 6 is a track curve of an automatic driving target speed of a train (large subway 12).
Fig. 7 is a distance curve (large subway 12) of an automatic train driving control mode according to an embodiment of the present invention.
Fig. 8 is an iteration convergence curve (large-connection subway line 12) of each optimization index of different optimization algorithms provided by the embodiment of the invention.
Fig. 9 is an iteration convergence curve (large-connection subway line 12) of different optimization algorithm evaluation indexes provided by the embodiment of the invention.
Fig. 10 is a track curve of an automatic train driving target speed (Jin Pu subway line 1) according to another embodiment of the present invention.
Fig. 11 is a distance curve (Jin Pu subway line 1) of an automatic train driving control mode according to another embodiment of the present invention.
Fig. 12 is an iterative convergence curve (Jin Pu subway line 1) of each optimization index of a different optimization algorithm according to another embodiment of the present invention.
Fig. 13 is an iteration convergence curve (Jin Pu subway line 1) of an evaluation index of a different optimization algorithm according to another embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution of an embodiment of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiment of the present invention, and it is apparent that the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
The invention provides a multi-objective optimization method for subway train operation control, which comprises the following steps:
s1, taking train energy consumption, comfort and punctuality as optimization targets, and establishing a train operation control multi-target optimization model; the train operation control multi-objective optimization model established in the step S1 is as follows:
wherein K is E Represents the energy consumption measurement index, and the energy consumption measurement index,
K A indicating comfort metrics->
K T Indicating punctual metrics>
Wherein a is i Acceleration, s, representing the ith operating mode i Representing displacement of the ith operating mode, R i Resistance representing the i-th operating condition; min represents the minimum value of the function F (u), if the objective vector of the multi-objective optimization problem is F (x) = (F) 1 (x),f 2 (x),…,f k (x) A specific optimization model thereof can be expressed as:
where k is the number of optimization objectives, x is the decision variable, g (x) is the equality constraint and the inequality constraint of x, x "is the absolute optimal solution (Pareto optimal), and F (x") is superior to F (x') if and only if any of D is.
S2, a particle swarm improvement optimization algorithm of a genetic evolution mechanism is introduced, and a group of optimal train operation sequences are selected to serve as an optimal solution; in specific implementation, the train operation multi-objective optimization is to set any given train parameters, line conditions, constraint conditions and optimization objectives thereof, and a particle swarm improvement optimization algorithm adopting a genetic evolution mechanism finds a group of train operation sequences { o ] which are as optimal as possible 1 ,o 2 ,…,o k O, where o i ＝(u i ,p i ) Represents the ith manipulation mode, u i For its train steering control mode, p i I e [1, 2..k for which the switch point position is manipulated]. Employing the set of train operating sequences { o } 1 ,o 2 ,…,o k Control train operation so that various optimization metrics such as energy conservation, punctual, comfort, etc. are optimized to the greatest extent. Train control sequence { o } 1 ,o 2 ,…,o k I.e. an optimization solution of the train operation manipulation multi-objective optimization problem.
S3, on the basis of the step S2, taking the cosine of the multi-target included angle as an evaluation standard for resolving the quality, and taking the fusion distance as a judgment condition for judging that the particle swarm is gathered on the extremum particles to obtain the best comprehensive performance index of the subway train in the running process. As shown in fig. 1, the step S3 includes the following steps:
s31, initializing:
initializing the speed and the position of a population, setting the maximum iteration number d, giving an acceleration constant c1, c2 and a weight coefficient omega, initializing the optimal position of a particle individual, and initializing the optimal position of the particle population;
s32, calculating a cosine value of the multi-target included angle of each particle according to the following formula;
where γ represents the cosine of the included angle, (T, C) represents the dot product of the solution target vector T and the target demand vector C, the term A represents the modular length of vector A, represents a numerical multiplication, T i And c i Normalized values representing the ith optimization objective of the solution objective vector T and the objective demand vector C; the cosine of the included angle between the specific solution target vector and the target demand vector is shown in fig. 2.
S33, obtaining an individual extremum and a global extremum of the particle swarm by using sequencing;
s34, updating the speed and the position of each particle by adopting an updating formula of the speed and the position of the particle; the update formula for the velocity and position of the particles is as follows:
wherein i e [1,2, ], N]The serial number of the particle, t e [1, 2., d.]The t-th dimension of the particles is represented, d represents the iteration number, c1 and c2 are acceleration constants, rand is a random real number in the interval (0, 1), ω is a weight coefficient, and the position vector is represented as
The velocity vector is denoted +.>
The optimal position of the individual particles is marked as->
The optimal position of the population is marked as->
S35, calculating a fusion distance, wherein a calculation formula of the fusion distance is as follows:
d Mix ＝ω×MD(X,Y)+(1-ω)×ED(X,Y)
wherein d Mix Represents fusion distance, MD represents Mahalanobis distance, C Y A correlation coefficient matrix representing a sample set Y, n representing the number of samples in the sample set Y, Y i (i=1, …, n) represents samples in the sample set Y, ρ represents a correlation coefficient;
s36, judging whether the particle swarm is aggregated to the individual polar value by adopting the fusion distance calculated in the step S35, and if the particle swarm is not aggregated, turning to the step S37; if the particles are aggregated, adopting the operations of selection, crossing and mutation of a genetic algorithm on the corresponding particles, and turning to step S37;
the judgment formula for judging whether the particle swarm is gathered at the individual extremum is as follows:
in the method, in the process of the invention,
the sample space formed by the individual extremum of the particles is represented, dis|A-B| represents the distance from the sample A to the sample space B, epsilon represents the threshold value, and Dis adopts the fusion distance.
S37, judging whether the maximum iteration number d is reached, if not, adding 1 to the iteration number, and turning back to the execution step S33; otherwise, outputting the comprehensive performance index of the running process of the optimal subway train.
The invention also provides a multi-objective optimizing system for subway train operation control, as shown in fig. 3, comprising: a lower control loop and an upper optimization loop; the lower control loop comprises a sensor, a conditioning circuit, a controller, a traction transmission link, a main breaker, a train line data storage unit and an intermediate link circuit which are electrically connected; the upper layer optimizing loop comprises a network connecting wire, a transmitter, a signal processor, an optimizer, a main breaker, a train line data storage unit and an intermediate link simulation circuit which are electrically connected; the specific implementation method comprises the following steps:
the optimizer is used for providing the corresponding target speed track for the controller;
the controller is used for controlling the traction transmission link to track and control the target speed track;
the train line data storage unit is used for providing necessary related data, conditions and parameters of lines and trains for subway train operation optimization and tracking control;
the sensor, the network connecting wire and the transmitter respectively acquire corresponding data for the controller and the optimizer to implement tracking control and operation optimization calculation;
the conditioning circuit and the signal processor are respectively used for conditioning and processing the electric signals acquired by the sensor, the network connecting wires and the network signals acquired by the transmitter into signals acceptable by the core chips of the controller and the optimizer;
the traction transmission link is driven by a traction motor to apply power to the subway train so as to brake or carry out traction operation, and the electric energy of the traction transmission link is transmitted from a power grid by an intermediate link circuit;
when the fault occurs, the main circuit breaker is used for breaking the circuit and protecting the system, and when the corresponding fault is removed, the circuit can be switched on to enable the circuit to normally operate.
The system has the functions of off-line optimization and on-line optimization; the subway train operation control multi-objective optimization method is applied to an offline optimization function of the system, is embedded in an offline optimization function module area of an optimizer core chip, does not need actual train tracking control during offline optimization, and is replaced by a simulation circuit by isolating a lower control loop from an intermediate link circuit; the off-line optimization is used for optimizing an operating sequence to the greatest extent based on the comprehensive performance index of the train operation process obtained by off-line data optimization when the train stops running; during on-line optimization, the lower control loop and the intermediate link circuit are enabled to implement actual train tracking control. The on-line optimization is used for combining the current control situation when the train runs on line, and adjusting the control sequence obtained by off-line optimization to obtain the optimal running optimization effect of actual tracking control.
Examples
In the embodiment, a train with a large subway line 12 and a large subway line Lian Jinpu and a large subway line 1 (also called a large subway line 13 and Jin Pu intercity railways) and a running line thereof are selected as research objects. The simulation running line of the large-connection subway No. 12 line is from the Qing Shun Xingang to the Tien mountain town, the interval length is 2.94 km, and the simulation running line comprises two long downhill slopes and one long uphill slope. The large continuous subway No. 12 line is a rail traffic line which is running and extends from a river mouth station in a high new garden area of the large continuous city to a station in a station of the large continuous city station, which is a new port station of the large continuous city station, and 8 stations are arranged. The simulation running line of the large-connection Jin Pu No. 1 line is from Jiuyi to nineteen bureaus, and the interval length is 2.74km. The Jin Pu line 1 is an intercity railway line which is being constructed from Jiuli of Jinzhou district to Pulan shop district, the total length of which is 46.76km, and 11 stations are initially set. The ramp and speed limit curves of the subway simulation running lines of the large-connection subway No. 12 line and the subway simulation running line of the large-connection subway No. Jin Pu line are shown in fig. 4 and 5, and the specific basic train properties are shown in tables 1-2:
TABLE 1 major parameters of large-connection subway No. 12 subway train
Table 2 Jin Pu number 1 subway train main parameters
In the above embodiment, on the basis of considering both the convergence rate and the optimizing effect, the following initialization parameters are given by combining the related scientific literature, field experience and simulation effect of multiple tests. The simulation optimization result must satisfy the following conditions: the instantaneous speed of the train cannot exceed the speed limit; the whole process must be run; the parking error is less than 0.2 meter. The parameters in the examples are specifically as follows: particle swarm scale 30, weight coefficient 0.9, acceleration coefficient c 1 And c 2 The maximum iteration number is 150, the selection probability is 40%, the crossover probability is 85%, and the variation probability is 5%. The multi-objective optimization parameters of the large-connection subway No. 12 train operation control optimization scene are specifically as follows: when the planned running time is 180s, the basic requirement of the optimization target is K E ∈[80000，130000]、K A ∈[5,10]、K T ∈[0,02]Intrinsic weight ω' 1 、ω′ 2 And omega' 3 Taking 0.5, 0.3 and 0.2 respectively, and the target demand vector is [90000,5.2,0.01 ]]. The multi-objective optimization parameters of Jin Pu No. 1 train operation control optimization scene are specifically as follows: when the planned running time is 177s, the basic requirement of the optimization target is K E ∈[95000,150000]、K A ∈[6,10]、K T ∈[0,0.2]Intrinsic weight ω' 1 、ω′ 2 And omega' 3 Taking 0.5, 0.3 and 0.2 respectively, and the target demand vector is [98000,6.2,0.01 ]]。
Example 1
In the embodiment, a subway operation control optimization experiment verification platform based on a Matlab/GUI simulation system is adopted, and the subway operation control optimization experiment verification platform is mainly configured as follows:
the simulation software is Matlab/GUI software (Matlab Gui 2016 b); the systems of the display computer are configured as "cpu core i7" and "windows 10". In order to verify the effectiveness of the improvement of the invention, based on the large-connection subway No. 12 line train and the selected simulation running line (from the Qingshun to the Fengshan town), in the embodiment, the operation control multi-objective optimization method (marked as I-GA-PSO) of the subway train operation control, the particle swarm improvement optimization algorithm (marked as GA-PSO) introducing a genetic evolution mechanism and the particle swarm optimization algorithm (marked as PSO) are respectively adopted to carry out operation control multi-objective optimization so as to obtain the corresponding operation control optimization result of the subway train. The speed track curve of the specific automatic train driving target, the distance curve of the control mode, the iterative convergence curve of each optimization target of different optimization algorithms and the iterative convergence curve of the evaluation index are shown in fig. 6, 7, 8 and 9, and the final result of the automatic train driving operation optimization is shown in tables 3 and 4:
TABLE 3 final optimization results of various optimization indexes of different optimization algorithms (Dalian subway No. 12 line)
TABLE 4 final optimization results of various evaluation indexes of different optimization algorithms (Dalian subway No. 12 line)
As can be seen from the simulation results of tables 2 and 3, the optimized solution obtained by the subway train operation multi-objective optimization method I-GA-PSO provided by the invention is superior to the optimized solution obtained by the particle swarm optimization algorithm GA-PSO and the particle swarm optimization algorithm PSO which are introduced into the genetic evolution mechanism, and the three indexes of energy conservation, punctual and comfort are improved to a certain extent. In this embodiment, a hilly area of the travel gate area is selected at the operating line, and a plurality of hills are typical landforms of the travel gate area. In such terrain-controlled train operation, it is desirable that the control sequence be compact and that the long downhill section be utilized to accelerate and the long uphill section be utilized to decelerate as much as possible. Obviously, such manipulation is advantageous for saving energy consumption and avoiding jolting. As can be seen from fig. 6 and fig. 7, the multi-objective optimization method I-GA-PSO for subway train operation provided by the present invention can obtain extremely concise operation sequences and can maximally utilize long uphill sections and long downhill sections. As can be seen from the iterative convergence curves of FIGS. 8 and 9, compared with the particle swarm optimization algorithm GA-PSO and the particle swarm optimization algorithm PSO which introduce the genetic evolution mechanism, the subway train operation control multi-objective optimization method I-GA-PSO provided by the invention has higher convergence speed and stronger global optimizing capability even in the middle and late stages of iteration.
Example 2
In this embodiment, the subway operation control optimization experiment verification platform adopting the hardware-in-loop simulation system based on dsace is mainly configured as follows:
the simulation software is dSPACE software (Control Desk 7.0) and Matlab/Simulink software (Matlab/Simulink 2016 b); the system of the display computer is configured as 'cpu core i 7' and 'windows 10'; the embedded chip of the optimizer and controller is "TMS320F28335". In order to verify the effectiveness of the improvement of the invention, the model Yu Jinpu subway No. 1 line train and the selected simulation running line thereof (from Murray to nineteen bureau) are adopted in the embodiment, and the running operation multi-objective optimization method (marked as I-GA-PSO) of the subway train running operation is adopted, the particle swarm improvement optimization algorithm (marked as GA-PSO) introducing a genetic evolution mechanism and the existing particle swarm optimization algorithm (marked as PSO) are adopted to perform running operation multi-objective optimization so as to obtain corresponding subway train running operation optimization results. The specific speed track curve, tracking control speed track curve, control mode distance curve, iterative convergence curves of various optimization indexes and evaluation indexes of different optimization algorithms of the automatic train driving target speed track curve, tracking control speed track curve and the automatic train driving target speed track curve are shown in fig. 10, 11, 12 and 13, and the final results of automatic train driving control optimization and tracking control are shown in tables 5, 6 and 7:
TABLE 5 final optimization results for various optimization metrics for different optimization algorithms (Jin Pu Metro No. 1 line)
TABLE 6 final tracking control results for various optimization metrics of different optimization algorithms (Jin Pu Metro No. 1 line)
TABLE 7 final optimization results of various evaluation indexes of different optimization algorithms (Jin Pu Metro No. 1 line)
As can be seen from the simulation results of tables 5 and 6 and 7, the optimized solution obtained by the I-GA-PSO (matrix-particle swarm optimization) method for operating and controlling the subway train provided by the invention is superior to the optimized solution obtained by the GA-PSO and PSO (particle swarm optimization) methods for introducing a genetic evolution mechanism, and the three indexes of energy conservation, time alignment and comfort are improved to a certain extent. In the embodiment, the undulating region of the Dalianjinzhou area at the operation line is selected, the train is operated on the terrain, the operation sequence is simple, and the undulating slope section is utilized to accelerate and the long ascending slope section to decelerate as much as possible, so that the energy consumption is saved and the jolt is avoided. As can be seen from fig. 10 and fig. 12, the multi-objective optimization method for operating and controlling the subway train I-GA-PSO according to the present invention can obtain an extremely concise operating sequence and can utilize the undulating slope to the greatest extent, thereby obtaining a better objective speed trajectory. Further, as shown in fig. 11 (particularly obvious in 2 amplifying regions), the target speed track obtained by the I-GA-PSO based on the multi-target optimizing method for operating and controlling the subway train according to the present invention is tracked and controlled, so that a better tracking and controlling speed track curve can be obtained, because the target speed track obtained by the I-GA-PSO is smoother, less severe operating and switching conditions exist, and thus, the tracking and controlling are easier. As can be seen from the iterative convergence curve of FIG. 13, compared with the particle swarm optimization algorithm GA-PSO and the particle swarm optimization algorithm PSO which introduce the genetic evolution mechanism, the subway train operation control multi-objective optimization method I-GA-PSO provided by the invention has higher convergence speed and stronger global optimizing capability even in the middle and later stages of iteration.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (6)
1. The multi-objective optimizing method for subway train operation control is characterized by comprising the following steps:
s1, taking train energy consumption, comfort and punctuality as optimization targets, and establishing a train operation control multi-target optimization model;
s2, a particle swarm improvement optimization algorithm of a genetic evolution mechanism is introduced, and a group of optimal train control sequences are selected to serve as an optimal solution;
s3, taking cosine of a multi-target included angle as an evaluation standard of the solution quality on the basis of the step S2, and taking the fusion distance as a judgment condition for judging that the particle swarm is gathered on the extremum particles to obtain an optimal comprehensive performance index in the subway train running process; the step S3 includes the steps of:
s31, initializing:
initializing the speed and the position of a population, setting the maximum iteration number d, giving an acceleration constant c1, c2 and a weight coefficient omega, initializing the optimal position of a particle individual, and initializing the optimal position of the particle population;
s32, calculating a cosine value of the multi-target included angle of each particle according to the following formula;
where γ represents the cosine of the included angle, (T, C) represents the dot product of the solution target vector T and the target demand vector C, the term A represents the modular length of vector A, represents a numerical multiplication, T i And c i Normalized values representing the ith optimization objective of the solution objective vector T and the objective demand vector C;
s33, sorting by utilizing cosine values of multiple target included angles to obtain an individual extremum and a global extremum of the particle swarm;
s34, updating the speed and the position of each particle by adopting an updating formula of the speed and the position of the particle; the update formula for the velocity and position of the particles is as follows:
wherein i e [1,2, ], N]The serial number of the particle, t e [1, 2., d.]The t-th dimension of the particles is represented, d represents the iteration number, c1 and c2 are acceleration constants, rand is a random real number in the interval (0, 1), ω is a weight coefficient, and the position vector is represented as
The velocity vector is denoted +.>
The optimal position of the individual particles is marked as->
The optimal position of the population is marked as->
S35, calculating a fusion distance, wherein a calculation formula of the fusion distance is as follows:
d Mix ＝ω×MD(X,Y)+(1-ω)×ED(X,Y)
wherein d Mix Represents fusion distance, MD represents Mahalanobis distance, C Y A correlation coefficient matrix representing a sample set Y, n representing the number of samples in the sample set Y, Y i (i=1, …, n) represents samples in the sample set Y, ρ represents a correlation coefficient;
s36, judging whether the particle swarm is gathered in an individual extremum by adopting the fusion distance calculated in the step S35, and if the particle swarm is not gathered, turning to the step S37; if the particles are aggregated, adopting the selection, crossing and mutation operations of a genetic algorithm on the corresponding particles, and turning to step S37;
s37, judging whether the maximum iteration number d is reached, if not, adding 1 to the iteration number, and turning back to the execution step S33; otherwise, outputting the comprehensive performance index of the running process of the optimal subway train.
2. The subway train operation manipulation multi-objective optimization method according to claim 1, wherein the train operation manipulation multi-objective optimization model established in step S1 is as follows:
wherein K is E Represents the energy consumption measurement index, and the energy consumption measurement index,
K A represents a measure of the comfort level,
K T indicating punctual metrics>
Wherein a is i Acceleration, s, representing the ith operating mode i Representing displacement of the ith operating mode, R i Resistance representing the i-th operating condition; min represents the minimum of the function F (u), if the objective vector of the multi-objective optimization problem is F (x) = (F) 1 (x),f 2 (x),…,f k (x) A specific optimization model thereof can be expressed as:
where k is the number of optimization objectives, x is the decision variable, g (x) is the equality and inequality constraints of x, x "is the absolute optimal solution if and only if D * F (x ") is superior to F (x').
3. The method for optimizing operation and control of a subway train according to claim 1, wherein the judging formula for judging whether the particle swarm is gathered in the individual extremum in the step S36 is as follows:
in the method, in the process of the invention,
representing individual extremum structures of particlesThe resulting sample space, dis|A-B|, represents the distance of sample A to sample space B, ε represents the threshold, dis employs the fusion distance.
4. A subway train operation manipulation multi-objective optimization system based on the subway train operation manipulation multi-objective optimization method according to any one of claims 1 to 3, comprising: a lower control loop and an upper optimization loop;
the lower control loop comprises a sensor, a conditioning circuit, a controller, a traction transmission link, a main circuit breaker, a train line data storage unit and an intermediate link circuit which are electrically connected;
the upper layer optimization loop comprises a network connecting wire, a transmitter, a signal processor, an optimizer, a main circuit breaker, a train line data storage unit and an intermediate link simulation circuit which are electrically connected;
the system has the functions of off-line optimization and on-line optimization; the subway train operation control multi-objective optimization method is applied to an offline optimization function of the system, is embedded in an offline optimization function module area of an optimizer core chip, does not need actual train tracking control during offline optimization, and is replaced by a simulation circuit after a lower control loop and an intermediate link circuit are isolated; during on-line optimization, the lower control loop and the intermediate link circuit are enabled to implement actual train tracking control.
5. The subway train operation control multi-objective optimization system according to claim 4, wherein the off-line optimization is used for controlling a sequence which is optimized to the greatest extent based on the comprehensive performance index of the train operation process obtained by off-line data optimization when the train stops running; the on-line optimization is used for adjusting the control sequence obtained by off-line optimization by combining the current control condition when the train runs on line so as to obtain the optimal running optimization effect of actual tracking control.
6. The subway train operation manipulation multi-objective optimization system according to claim 4, wherein,
the optimizer is used for providing the corresponding target speed track for the controller;
the controller is used for controlling the traction transmission link to track and control the target speed track;
the train line data storage unit is used for providing necessary related data, conditions and parameters of lines and trains for subway train operation optimization and tracking control;
the sensor, the network connecting wire and the transmitter respectively acquire corresponding data for the controller and the optimizer to implement tracking control and operation optimization calculation;
the conditioning circuit and the signal processor are respectively used for conditioning and processing the electric signals acquired by the sensor, the network connection wires and the network signals acquired by the transmitter into signals acceptable by the core chips of the controller and the optimizer;
the traction transmission link is driven by a traction motor to apply power to the subway train, so that the subway train is braked or in traction operation, and the electric energy of the subway train is transmitted from a power grid by an intermediate link circuit;
the main breaker is used for switching on and off a circuit and protecting a system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010072449.4A CN111291856B (en) | 2020-01-21 | 2020-01-21 | Multi-objective optimization method and system for operation and control of subway train |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010072449.4A CN111291856B (en) | 2020-01-21 | 2020-01-21 | Multi-objective optimization method and system for operation and control of subway train |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111291856A CN111291856A (en) | 2020-06-16 |
| CN111291856B true CN111291856B (en) | 2023-05-23 |
Family
ID=71023500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010072449.4A Active CN111291856B (en) | 2020-01-21 | 2020-01-21 | Multi-objective optimization method and system for operation and control of subway train |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111291856B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112070275A (en) * | 2020-08-14 | 2020-12-11 | 青岛海信网络科技股份有限公司 | Method, device and equipment for cooperatively limiting current of rail transit road network |
| CN112307564B (en) * | 2020-11-10 | 2024-05-10 | 交控科技股份有限公司 | Train ATO target running speed curve optimization method and device |
| CN112487695B (en) * | 2020-11-30 | 2021-11-26 | 中南大学 | Multi-target intelligent comprehensive line selection method for railway in complex environment |
| CN112989699B (en) * | 2021-03-12 | 2022-05-24 | 重庆交通大学 | New energy automobile performance evaluation method based on deep reinforcement learning |
| CN113268879A (en) * | 2021-05-31 | 2021-08-17 | 深圳市大数据研究院 | Train operation optimization method, device, equipment and storage medium |
| CN113807599B (en) * | 2021-09-26 | 2022-05-24 | 中建电子信息技术有限公司 | Intelligent construction informatization integrated management cloud platform |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992051A (en) * | 2017-12-26 | 2018-05-04 | 江南大学 | Unmanned vehicle paths planning method based on improved multi-objective particle swarm algorithm |
| US20180183650A1 (en) * | 2012-12-05 | 2018-06-28 | Origin Wireless, Inc. | Method, apparatus, and system for object tracking and navigation |
| CN110427690A (en) * | 2019-07-29 | 2019-11-08 | 交控科技股份有限公司 | A kind of method and device generating ATO rate curve based on global particle swarm algorithm |
-
2020
- 2020-01-21 CN CN202010072449.4A patent/CN111291856B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180183650A1 (en) * | 2012-12-05 | 2018-06-28 | Origin Wireless, Inc. | Method, apparatus, and system for object tracking and navigation |
| CN107992051A (en) * | 2017-12-26 | 2018-05-04 | 江南大学 | Unmanned vehicle paths planning method based on improved multi-objective particle swarm algorithm |
| CN110427690A (en) * | 2019-07-29 | 2019-11-08 | 交控科技股份有限公司 | A kind of method and device generating ATO rate curve based on global particle swarm algorithm |
Non-Patent Citations (6)
| Title |
|---|
| "Improvement of multi-objective differential evolutionary algorithm and its application for Hybrid electric vehicles";Mou Liu 等;《The 31st Chinese Control and Decision Conference (2019 CCDC)》;全文 * |
| "基于人工蜂群算法的推荐速度曲线节能优化";刘海娜;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;全文 * |
| "基于粒子群算法的城市轨道交通列车节能优化研究";李玲玉;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;全文 * |
| "基于自适应学习的多目标粒子群优化算法";尹呈 等;《计算机应用研究》;第29卷(第9期);全文 * |
| "融入偏好信息的列车运行过程多目标优化算法";王龙达 等;《交通运输系统工程与信息》;第17卷(第6期);全文 * |
| "融合集群度与距离均衡优化的K-均值聚类算法";王日宏 等;《计算机应用》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111291856A (en) | 2020-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111291856B (en) | Multi-objective optimization method and system for operation and control of subway train | |
| Cao et al. | Bio-inspired speed curve optimization and sliding mode tracking control for subway trains | |
| CN104881527B (en) | Urban railway transit train ATO speed command optimization methods | |
| CN110610260B (en) | Driving energy consumption prediction system, method, storage medium and equipment | |
| CN109840641B (en) | Method for quickly optimizing train multi-section operation curve | |
| Hwang | Control strategy for optimal compromise between trip time and energy consumption in a high-speed railway | |
| Ke et al. | Optimization of train-speed trajectory and control for mass rapid transit systems | |
| CN110533242B (en) | Energy-saving optimization method under train interconnection and intercommunication overline operation | |
| CN111582750B (en) | Method and system for identifying reverse passengers and determining riding class of rail transit | |
| Brenna et al. | Application of genetic algorithms for driverless subway train energy optimization | |
| CN112116207A (en) | Multi-constraint-condition train operation adjustment calculation method and device | |
| CN110990950B (en) | Multi-target train energy-saving optimization method based on hybrid operation mode | |
| CN109229160A (en) | Cope with boisterous automatic train control method, device and mobile unit | |
| CN107757584B (en) | A kind of brake control method, two-shipper multi-locomotive braking system and braking method | |
| CN103955135A (en) | Method for calculating locomotive optimized operation sequence based on double-layer mode curve | |
| CN112530177A (en) | Kalman filtering-based vehicle queuing length estimation method in Internet of vehicles environment | |
| Chen et al. | Towards human-like speed control in autonomous vehicles: A mountainous freeway case | |
| Yang et al. | Multi-objective operation optimization for electric multiple unit-based on speed restriction mutation | |
| CN109739245A (en) | One kind being based on unpiloted end to end model appraisal procedure and device | |
| CN109017884A (en) | Automatic train operation control method based on study | |
| CN110895631B (en) | Identification method and system for wheel-rail adhesion characteristics | |
| Yin et al. | Balise arrangement optimization for train station parking via expert knowledge and genetic algorithm | |
| Wang et al. | Energy-efficient speed profile optimization for urban rail transit with considerations on train length | |
| CN112329340B (en) | Auxiliary parking area arrangement optimization method for normally-conductive high-speed magnetic levitation unidirectional running line | |
| CN112446086B (en) | Simulation method and device for magnetic suspension train |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |