CN108985662A - A kind of train operation optimization method based on parallel immunity particle cluster algorithm - Google Patents
A kind of train operation optimization method based on parallel immunity particle cluster algorithm Download PDFInfo
- Publication number
- CN108985662A CN108985662A CN201810982142.0A CN201810982142A CN108985662A CN 108985662 A CN108985662 A CN 108985662A CN 201810982142 A CN201810982142 A CN 201810982142A CN 108985662 A CN108985662 A CN 108985662A
- Authority
- CN
- China
- Prior art keywords
- train
- population
- time
- energy consumption
- particles
- 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
- 239000002245 particle Substances 0.000 title claims abstract description 136
- 238000005457 optimization Methods 0.000 title claims abstract description 86
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 76
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000036039 immunity Effects 0.000 title abstract description 5
- 238000005265 energy consumption Methods 0.000 claims abstract description 64
- 238000013486 operation strategy Methods 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 6
- 238000002255 vaccination Methods 0.000 claims description 6
- 229960005486 vaccine Drugs 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 9
- 230000001172 regenerating effect Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 230000002068 genetic effect Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- 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
- 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
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a kind of train operation optimization methods based on parallel immunity particle cluster algorithm, it first passes through parallel particle swarm algorithm the train operation route in each section is optimized to obtain the corresponding relationship of energy consumption and time, keep the general plan runing time of train constant again, according to the corresponding relationship of energy consumption and time, distribution again is carried out using runing time of the parallel particle swarm algorithm to each section, utilize such two steps optimization method, the energy consumption time relationship that train can more be effectively utilized reasonably distributes the runing time in each section, energy consumption in train journey is reduced in the case where not changing train general plan runing time, parallel immunity particle cluster algorithm has been selected during optimization simultaneously, it can effectively prevent falling into local optimum, keep this method optimal speed fast, reach good effect of optimization.
Description
Field of the method
The invention belongs to the field of train operation optimization, and mainly relates to a train operation optimization method based on a parallel immune particle swarm algorithm.
Background method
Urban rail transit is an important transportation mode in China, and due to the characteristics of frequent operation and large transportation volume, the train has very large power consumption, so that the urban rail transit train energy-saving optimization method has important significance for running energy-saving optimization of urban rail transit trains. In the prior art, certain research has been carried out on the energy-saving optimization problem of the train. Chang optimizes the lazy control strategy by using a genetic algorithm in an ATO mode for the first time. Firstly, establishing an energy-saving, punctual, safe and comfortable multi-target model, then determining the operating condition turning point of the train interval operation, and then performing multi-target optimization by using a genetic algorithm. The result shows that the genetic algorithm has good searching capability and can find the optimal solution of multiple targets. Vanderbei and the like establish a train energy-saving optimization model by taking train operation gears and idle points as optimization variables and taking train operation energy consumption and travel time accuracy as optimization targets, and optimize the model by using a multidimensional parallel genetic algorithm in combination with overtime penalty factors. Huang and the like comprehensively consider the change situation of the following train tracking interval based on the mobile block section, and optimize the energy-saving running problem of the tracked train by using a genetic algorithm. Yang develops a set of schedule optimization algorithm, so that the regenerative braking energy consumption generated by braking trains can be utilized by the traction locomotive to the maximum extent, and the total energy consumption is saved. However, when the prior art performs energy-saving optimization on train operation, due to the limitation of the algorithm, the optimization result is difficult to avoid the situation of local convergence.
Disclosure of Invention
The invention aims to provide a train operation optimization method based on a parallel immune particle swarm algorithm aiming at the defects of the existing method, so that the train operation energy consumption is effectively reduced under the condition of not changing the total planned operation time of the train, the optimization speed is high, and a good optimization effect is achieved.
In order to solve the technical problems, the invention is implemented by the following method scheme:
a train operation optimization method based on a parallel immune particle swarm algorithm is characterized by comprising the following steps:
s1, determining basic parameters of the train line interval to be optimized; the basic parameters comprise train parameters, line parameters and operation parameters;
s2, establishing a train multi-objective optimization model; the objective function of the train multi-objective optimization model is that the energy consumption and the time error of train operation are both minimized, the constraint condition is the limiting index of train operation and the basic parameter, and the solution is the train operation strategy; the time error is the error between the train running time and the interval planning running time;
s3, setting the section planning operation time of the train line section to be optimized, resolving the train multi-target optimization model by using a parallel immune particle swarm algorithm to obtain a plurality of optimal solutions, and fitting the optimal solutions to obtain the corresponding relation between the operation energy consumption and the operation time of the train in the train line section;
s4, optimizing a plurality of train line sections on the same train line to be optimized by using the steps S1-S3 to obtain corresponding relations between operation energy consumption and operation time corresponding to the train line sections;
s5, establishing a train operation strategy optimization model, wherein the objective function of the train line optimization model is that the total energy consumption of train operation is minimum, the constraint conditions comprise that the operation time of the train in the line interval is greater than the minimum operation time of the train, and the sum of the operation time of the train in each interval is less than the total planned operation time of the train in the train line to be optimized; the minimum running time of the train is the minimum running time which can be reached by the train running in the train line section under the condition that the train meets the limiting conditions of the line; the solution of the train line optimization model is the running time distribution of each interval of the train;
s6, setting the total planned running time of the train, using a parallel immune particle swarm algorithm, and combining the corresponding relation between the running energy consumption and the running time of the train; and calculating the train operation strategy optimization model to obtain the train operation time of each section.
Further, in the step S1, the train parameters include train quality, train tractive force, train braking force, train number marshalling; the line parameters comprise the speed limit, the length, the gradient and the curvature of the line; the operation parameter is interval operation time.
Further, in step S2, the limit index includes a safety index, a comfort index, a speed limit index, and an accurate stop index of train operation.
Further, in step S3, the objective function of the train multi-objective optimization model is:
minF{fE,fT}＝αfE+β·fT·γ
wherein α and β are respectively an on-time index weight and an energy consumption index weight, α + β is 1, gamma is a time penalty factor, fEEnergy consumption for train operation; f. ofTIs the time error of the train.
Further, the step S3 includes:
s301, setting initial parameters of a parallel immune particle swarm algorithm; the initial parameters comprise a population size M, a particle dimension D, acceleration factors c1 and c2 and an inertia factor;
s302, initializing a population P containing M particles0Each particle is a group of train operation strategies;
s303, calculating the operation energy consumption of the operation time allocation plan represented by each particle according to the correspondence between the operation energy consumption and the operation time corresponding to the plurality of train line sections obtained in the step S4, calculating the fitness of the operation energy consumption of each particle, and calculating the population P according to the fitness0The particle concentration of (a); calculating the optimal position gbest of the population and storing the gbest in a vaccine memory bank;
s304, arranging the M particles from large to small according to the fitness, dividing the former S particles into a high-quality population Y, and dividing the rest particles into a low-quality population N;
s305, evolving the particles in the population Y according to a particle swarm algorithm speed and position updating formula, performing vaccination on the particles in the population N by using the optimal population position gbest, and updating the particles in the population N;
s306, merging the population Y and the population N into a population C, evolving the particles in the population C according to a particle swarm algorithm speed and position updating formula, and calculating an individual extreme value and a population extreme value;
s307, judging whether the individual extreme value and the group extreme value both meet the constraint condition of the train multi-objective optimization model, if so, executing S308, otherwise, returning to S303;
s308, storing the operation strategy of the particles in the population C to an optimal operation strategy library;
s309, judging whether the iteration times of the algorithm meet the preset iteration times, if so, executing S310, otherwise, returning to S303;
and S310, fitting the running time and the running energy consumption of all the running strategies in the optimal running strategy library to obtain the corresponding relation between the energy consumption and the time of the train in the energy-saving running process.
Further, step S6 includes;
s601, setting initial parameters of a parallel immune particle swarm algorithm; the initial parameters comprise a population size M, a particle dimension D, acceleration factors c1 and c2 and an inertia factor;
s602, initializing a population P containing M particles0Each particle is a running time distribution scheme of each interval of a group of trains;
s603, calculating the time error and the operation energy consumption of the operation strategy represented by each particle, calculating the fitness of the time error and the operation energy consumption of each particle, and calculating the population P according to the fitness0The particle concentration of (a); calculating the optimal position gbest of the population and storing the gbest in a vaccine memory bank;
s604, arranging the M particles from large to small according to the fitness, dividing the former S particles into a high-quality population Y, and dividing the rest particles into a low-quality population N;
s605, evolving the particles in the population Y according to a particle swarm algorithm speed and position updating formula, performing vaccination on the particles in the population N by using the optimal population position gbest, and updating the particles in the population N;
s606, merging the population Y and the population N into a population C, evolving the particles in the population C according to a particle swarm algorithm speed and position updating formula, and calculating an individual extreme value and a population extreme value;
s607, judging whether the individual extreme value and the group extreme value both meet the constraint condition of the train operation strategy optimization model, if so, executing S608, otherwise, returning to S603;
s608, judging whether the iteration times of the algorithm meet preset iteration times, if so, executing S609, and otherwise, returning to S603;
and S609, finishing the optimization and outputting an optimization result.
Compared with the prior art, the method has the following beneficial effects:
the invention discloses a train operation optimization method based on a parallel immune particle swarm algorithm, which comprises the steps of optimizing train operation lines in each interval through the parallel particle swarm algorithm to obtain the corresponding relation between energy consumption and time, keeping the total planned operation time of a train unchanged, and redistributing the operation time of each interval by using the parallel particle swarm algorithm according to the corresponding relation between the energy consumption and the time, compared with the traditional method for directly optimizing the operation time of the train, the two-step optimization method can more effectively utilize the energy consumption time relation of the train to reasonably distribute the operation time of each interval, reduces the operation energy consumption of the train under the condition of not changing the total planned operation time of the train, simultaneously selects the parallel immune particle swarm algorithm in the optimization process, and compared with the traditional particle swarm algorithm and the serial immune particle swarm algorithm, the parallel immune particle swarm algorithm can reduce the effect of the artificial immune algorithm as much as possible in the initial stage of the algorithm, fully exert the algorithm efficiency and realize quick positioning; the later-period action on artificial immunity is increased, and the local optimum can be effectively prevented from being trapped, so that the method is high in optimization speed and achieves a good optimization effect.
Drawings
FIG. 1 is a schematic diagram illustrating the steps of a train operation optimization method based on a parallel immune particle swarm algorithm according to the present invention;
FIG. 2 shows the result of the "energy consumption-time" curve optimization of the train in the G-H interval described in example 1 of the present invention.
Detailed Description
In order to fully understand the objects, features and effects of the present invention, the concept, specific steps and effects of the method of the present invention will be further described with reference to the accompanying drawings and the detailed description.
As shown in figure 1, the invention discloses a train operation optimization method based on a parallel immune particle swarm algorithm, which is characterized by comprising the following steps of:
s1, determining basic parameters of the train line interval to be optimized; the basic parameters comprise train parameters, line parameters and operation parameters; specifically, the train parameters include train quality, train traction, train braking force, and train number marshalling; the line parameters comprise the speed limit, the length, the gradient and the curvature of the line; the operation parameter is interval operation time.
S2, establishing a train multi-objective optimization model; the objective function of the train multi-objective optimization model is that the energy consumption and the time error of train operation are both minimized, the constraint condition is the limiting index and the basic parameter of train operation, and the solution is the operation strategy of the train; the time error is the error between the train running time and the interval planning running time; specifically, the objective function of the train multi-objective optimization model is as follows:
minF{fE,fT}＝αfE+β·fT·γ
wherein, α and β are respectively an on-time index weight and an energy consumption index weight, and α + β1 is ═ 1; gamma is a time penalty factor; f. ofEEnergy consumption for train operation; f. ofTTime error of the train; specifically, the running energy consumption of the train is as follows:
fE＝Es＝Eall-Ereg
wherein E isregFor regenerating braking energy, the calculation step length is designed to be delta t, and t is usednIndicating the time of the nth calculation, i.e., the nth at time. Then tnTime t andn+1the train speeds at the times are v respectivelynAnd vn+1. The motion equation of the train is as follows:
wherein, C (t)n) Is tnResultant force of train; m (t)n) Is tnThe train quality; sn，sn-1Respectively at t for trainn+1And tnTrain displacement at the moment.
At t of trainnInstantaneous traction power P (t) at timen) Can be expressed as:
P(tn)＝F(tn)·v(tn)
wherein, F (t)n) Is tnThe tractive effort of the train.
The traction energy consumed during the Δ t time is then:
E(tn)＝F(tn)·v(tn)·Δt
therefore, the total line traction energy consumption EallThe integral sum of the traction energy consumption at each moment, namely:
the train running in the braking condition has 0 tractive force and can generate a part of regenerative braking energy feedback. When the train starts to brake, only basic resistance and braking force are applied, and the total energy is the kinetic energy of the train. During the period from the braking start to the stopping of the train, the kinetic energy of the train is totally three parts to go: running resistance consumption, braking force consumption and regenerative braking energy feedback.
If tmThe train starts to brake at the moment, and the running speed, the running resistance and the braking resistance of the train are respectively as follows: v (t)m)、W(tm) And B (t)m) (ii) a Resultant force is C (t)m) The running distance of the train is Sm。
The kinetic energy variation E in the time Δ tmechIs composed of
Overcoming the overcoming resistance E of the train during the time delta tWThe energy consumed was:
EW＝W(tm)·v(tm)·Δt
therefore, the regenerative braking energy EregCan be expressed as:
Ereg＝(Emech-EW)·ηreg
wherein eta isregThe regenerative braking energy feedback coefficient for the train is the ratio of the feedback network energy to the total braking energy. Since there is a portion of the energy to supply the train auxiliary power, it does not return completely to the grid, typically at 95%.
Specifically, the limit indexes include a safety index, a comfort index, a speed limit index and an accurate parking index of train operation; in order to ensure that passengers normally get on or off the subway, the train parking deviation must be within 25 cm. Therefore, the accurate parking index is taken as a constraint condition, and can be expressed as:
wherein:for the train consist ofnStation runs to An+1Distance under the traction condition;
for the train consist ofnStation runs to An+1Distance under cruise conditions;
for the train consist ofnStation runs to An+1Distance under the time-coasting condition;
for the train consist ofnStation runs to An+1Distance under the braking condition;
Snis AnStand to An+1Total distance of the interval of stations;
Δ S is typically 25 cm.
The constraints for analyzing the security indicators may be expressed as:
the constraints for analyzing the comfort index may be expressed as:
ga＝|a|≤amax
a′min≤ga'＝|a′|≤a′max
in addition, there are other constraints such as:
the initial and final speeds of the train section are all 0:
v(S0)＝v(ST)＝0
analyzing the speed limit condition of the line, wherein the running speed of the train can not exceed the speed limit value:
0≤v≤Vline
due to the structural limitation of the subway vehicle, in the running process of the train, the traction force and the braking force have certain limiting values, and the expression is as follows:
0≤F(t)≤Fmax(t)
0≤B(t)≤Bmax(t)
the train operates in a certain sequence of 'traction, cruise, coasting and braking'.
Therefore, the constraint conditions of the train multi-objective optimization model are as follows:
specifically, the train generally includes four working conditions of traction, cruising, inertia and braking during the interval operation, so the solution of the objective function, i.e. the turning point S of the train operation strategy including the traction working condition and the constant speed working conditiona-vTurning point S of constant speed working condition and idle working conditionv-cTurning point S of idle running working condition and braking working conditionc-b(ii) a Wherein S isc-bNeed to pass through Sa-v、Sv-cAnd (6) calculating. The solution of the objective function is therefore Sa-v、Sv-c、Sc-bThe composed ordered sequence of numbers;
s3, setting the section planning operation time of the train line section to be optimized, resolving the train multi-target optimization model by using a parallel immune particle swarm algorithm to obtain a plurality of optimal solutions, and fitting the optimal solutions to obtain the corresponding relation between the operation energy consumption and the operation time of the train in the train line section;
specifically, step S3 includes:
s301, setting initial parameters of a parallel immune particle swarm algorithm; the initial parameters comprise a population size M, a particle dimension D, acceleration factors c1 and c2 and an inertia factor;
s302, initializing a population P containing M particles0Each particle is a group of train operation strategies;
s303, calculating the operation energy consumption of the operation time distribution scheme represented by each particle according to the corresponding relation between the operation energy consumption and the operation time corresponding to the plurality of train line intervals obtained in the step S4, calculating the fitness of the operation energy consumption of each particle, and calculating the population P according to the fitness0The particle concentration of (a); calculating the optimal position gbest of the population and storing the gbest in a vaccine memory bank;
s304, arranging the M particles according to the fitness from big to small, dividing the first S particles into a high-quality population Y, and dividing the rest particles into a low-quality population N;
s305, evolving the particles in the population Y according to a particle swarm algorithm speed and position updating formula, performing vaccination on the particles in the population N by using a population optimal position gbest, and updating the particles in the population N;
s306, merging the population Y and the population N into a population C, evolving the particles in the population C according to a particle swarm algorithm speed and position updating formula, and calculating an individual extreme value and a population extreme value;
s307, judging whether the individual extreme value and the group extreme value both meet the constraint condition of the train multi-objective optimization model, if so, executing S308, otherwise, returning to S303;
s308, storing the operation strategy of the particles in the population C to an optimal operation strategy library;
s309, judging whether the iteration times of the algorithm meet the preset iteration times, if so, executing S310, otherwise, returning to S303;
s310, fitting the running time and the running energy consumption of all the running strategies in the optimal running strategy library to obtain the corresponding relation between the energy consumption and the time of the train in the energy-saving running process;
s4, optimizing a plurality of train line sections of the same train line to be optimized by using the steps S1-S3 to obtain corresponding relations between operation energy consumption and operation time corresponding to the plurality of train line sections;
s5, establishing a train operation strategy optimization model, wherein the objective function of the train line optimization model is that the total energy consumption of train operation is minimum, the constraint conditions comprise that the operation time of the train in line intervals is greater than the minimum operation time of the train, and the sum of the operation time of the train in each interval is less than the total planned operation time of the train; the solution of the train line optimization model is the running time distribution of each interval of the train; specifically, the objective function is expressed as:
wherein E isiIs the energy consumption of the ith interval.
The constraint is expressed as:
wherein,the actual running time of the ith interval;the minimum running time of the ith interval;planned runtime for the ith interval;
specifically, the minimum running time of the train is the minimum running time which can be reached by the train running in the train line section under the condition that the train meets the limiting condition of the line; specifically, the train is assumed to make an accelerating movement with the maximum traction force, and after reaching the speed-limiting section, the train keeps running at the maximum speed, and finally, the train is braked by the maximum braking force. The train operation time in this process is the minimum operation time Tmin.
S6, setting the total planned running time of the train, using a parallel immune particle swarm algorithm, and combining the corresponding relation between the running energy consumption and the running time of the train; and calculating the train operation strategy optimization model to obtain the train operation time of each section.
Specifically, step S6 includes;
s601, setting initial parameters of a parallel immune particle swarm algorithm; the initial parameters comprise a population size M, a particle dimension D, acceleration factors c1 and c2 and an inertia factor;
s602, initializing a population P containing M particles0Each particle is a running time distribution scheme of each interval of a group of trains;
s603, calculating the time error and the operation energy consumption of the operation strategy represented by each particle, calculating the fitness of the time error and the operation energy consumption of each particle, and calculating the population according to the fitnessP0The particle concentration of (a); calculating the optimal position gbest of the population and storing the gbest in a vaccine memory bank;
s604, arranging the M particles from large to small according to the fitness, dividing the former S particles into a high-quality population Y, and dividing the rest particles into a low-quality population N;
s605, evolving the particles in the population Y according to a particle swarm algorithm speed and position updating formula, performing vaccination on the particles in the population N by using a population optimal position gbest, and updating the particles in the population N;
s606, merging the population Y and the population N into a population C, evolving the particles in the population C according to a particle swarm algorithm speed and position updating formula, and calculating an individual extreme value and a population extreme value;
s607, judging whether the individual extreme value and the group extreme value both meet the constraint condition of the train operation strategy optimization model, if so, executing S608, otherwise, returning to S603;
s608, judging whether the iteration times of the algorithm meet preset iteration times, if so, executing S609, and otherwise, returning to S603;
and S609, finishing the optimization and outputting an optimization result.
Example 1
In this embodiment, the train energy-saving operation optimization method is used. And substituting the actual data of a certain subway line in China for optimization. The total 14 stations of the whole line of the line total 22.73km, including 8 aboveground stations measuring 13.81km and 6 underground stations measuring 8.92km of tunnels. Due to the complexity and typicality of the line, the line has become a classic case of train simulation operation. The interval "G-H" is selected for simulation, and the specific line data of the interval is shown in table 1 and table 2.
TABLE 1 Rongjing east street to Wanyuan street line gradient parameter
TABLE 2G-H LINE SPEED LIMIT PARAMETERS
Through the steps S1-S3 of implementing the train energy-saving optimization algorithm, simulation is carried out on the interval to obtain a plurality of operation strategies of the train in the G-H interval, energy consumption values corresponding to different operation times are fitted to obtain an energy consumption-time curve of train operation, and the energy consumption-time curve is shown in figure 2.
The minimum run time between stations was obtained by simulation, as shown in table 3.
TABLE 3 minimum run time scheme
And (5) operating the steps S4-S6 of the train energy-saving optimization algorithm again, wherein the optimization result is shown in the table 4.
TABLE 4
The simulation result table 4 shows that the total train running time is unchanged through the second stage running time optimization, the running energy consumption is 187.666 kW.h, and 11.235% of energy is saved compared with 208.751 kW.h before optimization. According to the simulation, the effects of train timing energy-saving optimization and running time optimization are better, and the method has certain reference significance for actual running scheduling of the subway.
By the train operation optimization method based on the parallel immune particle swarm algorithm, the train operation lines of each interval are optimized through the parallel particle swarm algorithm to obtain the corresponding relation between energy consumption and time, then the total planned operation time of the train is kept unchanged, the parallel particle swarm algorithm is used for redistributing the operation time of each interval according to the corresponding relation between the energy consumption and the time, compared with the traditional method for directly optimizing the train operation time, the two-step optimization method is utilized to more effectively utilize the energy consumption time relation of the train and reasonably distribute the operation time of each interval, the train operation energy consumption is reduced under the condition of not changing the total planned operation time of the train, meanwhile, the parallel immune particle swarm algorithm is selected in the optimization process, compared with the traditional particle swarm algorithm and the serial immune particle swarm algorithm, the parallel immune particle swarm algorithm can reduce the effect of the artificial immune algorithm as much as possible in the initial stage of the algorithm, fully exert the algorithm efficiency and realize quick positioning; the later-period action on artificial immunity is increased, and the local optimum can be effectively prevented from being trapped, so that the method is high in optimization speed and achieves a good optimization effect.
While the preferred embodiments of the present invention have been described in detail, it should be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings without inventive faculty. Therefore, any process solutions available to a person skilled in the art of the present process based on the present inventive concept through logical analysis, reasoning or based on limited experimentation, shall be considered within the scope of protection defined by the claims.
Claims (6)
1. A train operation optimization method based on a parallel immune particle swarm algorithm is characterized by comprising the following steps:
s1, determining basic parameters of the train line interval to be optimized; the basic parameters comprise train parameters, line parameters and operation parameters;
s2, establishing a train multi-objective optimization model; the objective function of the train multi-objective optimization model is that the energy consumption and the time error of train operation are both minimized, the constraint condition is the limiting index of train operation and the basic parameter, and the solution is the train operation strategy; the time error is the error between the train running time and the interval planning running time;
s3, setting the section planning operation time of the train line section to be optimized, resolving the train multi-target optimization model by using a parallel immune particle swarm algorithm to obtain a plurality of optimal solutions, and fitting the optimal solutions to obtain the corresponding relation between the operation energy consumption and the operation time of the train in the train line section;
s4, optimizing a plurality of train line sections on the same train line to be optimized by using the steps S1-S3 to obtain corresponding relations between operation energy consumption and operation time corresponding to the train line sections;
s5, establishing a train operation strategy optimization model, wherein the objective function of the train line optimization model is that the total energy consumption of train operation is minimum, the constraint conditions comprise that the operation time of the train in the line interval is greater than the minimum operation time of the train, and the sum of the operation time of the train in each interval is less than the total planned operation time of the train in the train line to be optimized; the minimum running time of the train is the minimum running time which can be reached by the train running in the train line section under the condition that the train meets the limiting conditions of the line; the solution of the train line optimization model is the running time distribution of each interval of the train;
s6, setting the total planned running time of the train, using a parallel immune particle swarm algorithm, and combining the corresponding relation between the running energy consumption and the running time of the train; and calculating the train operation strategy optimization model to obtain the train operation time of each section.
2. A parallel immune particle swarm algorithm-based train operation optimization method according to claim 1, wherein in step S1, the train parameters comprise train quality, train traction, train braking force, train number marshalling; the line parameters comprise the speed limit, the length, the gradient and the curvature of the line; the operation parameter is interval operation time.
3. The train operation optimization method based on the parallel immune particle swarm algorithm according to claim 1, wherein in the step S2, the limit indexes comprise a safety index, a comfort index, a speed limit index and an accurate stop index of train operation.
4. The method for optimizing train operation based on the parallel immune particle swarm algorithm according to claim 1, wherein in step S3, the objective function of the train multi-objective optimization model is:
minF{fE,fT}＝αfE+β·fT·γ
wherein α and β are respectively an on-time index weight and an energy consumption index weight, α + β is 1, gamma is a time penalty factor, fEEnergy consumption for train operation; f. ofTIs the time error of the train.
5. The train operation optimization method based on the parallel immune particle swarm algorithm according to claim 1, wherein the step S3 comprises:
s301, setting initial parameters of a parallel immune particle swarm algorithm; the initial parameters comprise a population size M, a particle dimension D, acceleration factors c1 and c2 and an inertia factor;
s302, initializing a population P containing M particles0Each particle is a group of train operation strategies;
s303, calculating the operation energy consumption of the operation time allocation plan represented by each particle according to the correspondence between the operation energy consumption and the operation time corresponding to the plurality of train line sections obtained in the step S4, calculating the fitness of the operation energy consumption of each particle, and calculating the population P according to the fitness0The particle concentration of (a); calculating the optimal position gbest of the population and storing the gbest in a vaccine memory bank;
s304, arranging the M particles from large to small according to the fitness, dividing the former S particles into a high-quality population Y, and dividing the rest particles into a low-quality population N;
s305, evolving the particles in the population Y according to a particle swarm algorithm speed and position updating formula, performing vaccination on the particles in the population N by using the optimal population position gbest, and updating the particles in the population N;
s306, merging the population Y and the population N into a population C, evolving the particles in the population C according to a particle swarm algorithm speed and position updating formula, and calculating an individual extreme value and a population extreme value;
s307, judging whether the individual extreme value and the group extreme value both meet the constraint condition of the train multi-objective optimization model, if so, executing S308, otherwise, returning to S303;
s308, storing the operation strategy of the particles in the population C to an optimal operation strategy library;
s309, judging whether the iteration times of the algorithm meet the preset iteration times, if so, executing S310, otherwise, returning to S303;
and S310, fitting the running time and the running energy consumption of all the running strategies in the optimal running strategy library to obtain the corresponding relation between the energy consumption and the time of the train in the energy-saving running process.
6. The train operation optimization method based on the parallel immune particle swarm algorithm according to claim 1, wherein the step S6 comprises;
s601, setting initial parameters of a parallel immune particle swarm algorithm; the initial parameters comprise a population size M, a particle dimension D, acceleration factors c1 and c2 and an inertia factor;
s602, initializing a population P containing M particles0Each particle is a running time distribution scheme of each interval of a group of trains;
s603, calculating the time error and the operation energy consumption of the operation strategy represented by each particle, calculating the fitness of the time error and the operation energy consumption of each particle, and calculating the population P according to the fitness0The particle concentration of (a); calculating the optimal position gbest of the population and storing the gbest in a vaccine memory bank;
s604, arranging the M particles from large to small according to the fitness, dividing the former S particles into a high-quality population Y, and dividing the rest particles into a low-quality population N;
s605, evolving the particles in the population Y according to a particle swarm algorithm speed and position updating formula, performing vaccination on the particles in the population N by using the optimal population position gbest, and updating the particles in the population N;
s606, merging the population Y and the population N into a population C, evolving the particles in the population C according to a particle swarm algorithm speed and position updating formula, and calculating an individual extreme value and a population extreme value;
s607, judging whether the individual extreme value and the group extreme value both meet the constraint condition of the train operation strategy optimization model, if so, executing S608, otherwise, returning to S603;
s608, judging whether the iteration times of the algorithm meet preset iteration times, if so, executing S609, and otherwise, returning to S603;
and S609, finishing the optimization and outputting an optimization result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810982142.0A CN108985662A (en) | 2018-08-27 | 2018-08-27 | A kind of train operation optimization method based on parallel immunity particle cluster algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810982142.0A CN108985662A (en) | 2018-08-27 | 2018-08-27 | A kind of train operation optimization method based on parallel immunity particle cluster algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108985662A true CN108985662A (en) | 2018-12-11 |
Family
ID=64546697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810982142.0A Pending CN108985662A (en) | 2018-08-27 | 2018-08-27 | A kind of train operation optimization method based on parallel immunity particle cluster algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108985662A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615135A (en) * | 2018-12-12 | 2019-04-12 | 广州地铁集团有限公司 | A kind of train operation optimization method based on dynamic programming algorithm |
| CN109977553A (en) * | 2019-03-28 | 2019-07-05 | 广西大学 | A kind of subway train energy conservation optimizing method based on improved adaptive GA-IAGA |
| CN111325462A (en) * | 2020-02-18 | 2020-06-23 | 中国铁道科学研究院集团有限公司 | Motor train unit auxiliary driving method and system |
| CN111460633A (en) * | 2020-03-19 | 2020-07-28 | 南京理工大学 | Train energy-saving operation method based on multi-target particle swarm algorithm |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014003151A2 (en) * | 2012-06-29 | 2014-01-03 | Mitsubishi Electric Corporation | Method for determining an optimal run-curve for a vehicle |
| CN105785795A (en) * | 2016-05-05 | 2016-07-20 | 北京交通大学 | Train operation speed curve energy saving optimization method based on particle swarm algorithm |
| CN106503804A (en) * | 2016-10-11 | 2017-03-15 | 南京理工大学 | A kind of train timing energy-saving operation method based on Pareto multi-objective genetic algorithms |
| CN106873366A (en) * | 2017-01-19 | 2017-06-20 | 重庆交通大学 | The method that ATO speed control commands are generated for municipal rail train |
| CN107458424A (en) * | 2017-07-31 | 2017-12-12 | 湖南福德电气有限公司 | A kind of more train energy-saving operating methods applied on multi-site |
-
2018
- 2018-08-27 CN CN201810982142.0A patent/CN108985662A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014003151A2 (en) * | 2012-06-29 | 2014-01-03 | Mitsubishi Electric Corporation | Method for determining an optimal run-curve for a vehicle |
| CN105785795A (en) * | 2016-05-05 | 2016-07-20 | 北京交通大学 | Train operation speed curve energy saving optimization method based on particle swarm algorithm |
| CN106503804A (en) * | 2016-10-11 | 2017-03-15 | 南京理工大学 | A kind of train timing energy-saving operation method based on Pareto multi-objective genetic algorithms |
| CN106873366A (en) * | 2017-01-19 | 2017-06-20 | 重庆交通大学 | The method that ATO speed control commands are generated for municipal rail train |
| CN107458424A (en) * | 2017-07-31 | 2017-12-12 | 湖南福德电气有限公司 | A kind of more train energy-saving operating methods applied on multi-site |
Non-Patent Citations (1)
| Title |
|---|
| 张宏艺等: "改进免疫粒子群算法在矿车调度优化中的应用", 《计 算 机 系 统 应 用》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615135A (en) * | 2018-12-12 | 2019-04-12 | 广州地铁集团有限公司 | A kind of train operation optimization method based on dynamic programming algorithm |
| CN109977553A (en) * | 2019-03-28 | 2019-07-05 | 广西大学 | A kind of subway train energy conservation optimizing method based on improved adaptive GA-IAGA |
| CN109977553B (en) * | 2019-03-28 | 2022-07-05 | 广西大学 | Subway train energy-saving optimization method based on improved genetic algorithm |
| CN111325462A (en) * | 2020-02-18 | 2020-06-23 | 中国铁道科学研究院集团有限公司 | Motor train unit auxiliary driving method and system |
| CN111460633A (en) * | 2020-03-19 | 2020-07-28 | 南京理工大学 | Train energy-saving operation method based on multi-target particle swarm algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105460048B (en) | Comprehensive energy-saving control method and method integrating optimized manipulation and traffic scheduling for urban rail transit | |
| CN111311017B (en) | Urban rail transit train operation schedule and speed operation curve optimization method | |
| CN105930937B (en) | A kind of train operation scheduling Integrated Optimization considering subway speed line | |
| CN108985662A (en) | A kind of train operation optimization method based on parallel immunity particle cluster algorithm | |
| Yang et al. | A survey on energy-efficient train operation for urban rail transit | |
| Li et al. | An energy-efficient scheduling and speed control approach for metro rail operations | |
| ShangGuan et al. | Multiobjective optimization for train speed trajectory in CTCS high-speed railway with hybrid evolutionary algorithm | |
| Xing et al. | Energy consumption optimization of tramway operation based on improved PSO algorithm | |
| CN109978350A (en) | A kind of subway train energy conservation optimizing method based on regime decomposition dynamic programming algorithm | |
| CN112307564A (en) | Method and device for optimizing ATO target running speed curve of train | |
| Zhou et al. | Smart train operation algorithms based on expert knowledge and reinforcement learning | |
| CN110239600B (en) | Train operation control method and system for regenerative braking energy utilization | |
| CN109858154A (en) | A kind of energy-saving train operation method based on multiple-objection optimization | |
| Fernández-Rodríguez et al. | Balancing energy consumption and risk of delay in high speed trains: A three-objective real-time eco-driving algorithm with fuzzy parameters | |
| CN106651009B (en) | Energy-saving optimization control method for any multi-vehicle cooperation of urban rail transit | |
| CN112633598A (en) | Comprehensive energy-saving optimization method for speed curve and timetable of urban rail transit train | |
| CN103963805A (en) | Energy-saving method of train operation of urban mass transit | |
| CN106672027B (en) | A kind of urban track traffic ATS energy conservation timetable preparation method | |
| CN104881527A (en) | ATO (Automatic Train Operation) speed command optimization method of urban rail transit train | |
| CN106056238B (en) | Planning method for train interval running track | |
| Li et al. | Dynamic trajectory optimization design for railway driver advisory system | |
| He et al. | Optimal control of metro energy conservation based on regenerative braking: A complex model study of trajectory and overlap time | |
| Zhang et al. | Data-driven train operation models based on data mining and driving experience for the diesel-electric locomotive | |
| CN112507464A (en) | Freight train operation curve optimization method based on improved multi-target wolf algorithm | |
| Bin et al. | Energy-efficient speed profile optimization for high-speed railway considering neutral sections |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |