CN110990950B - Multi-target train energy-saving optimization method based on hybrid operation mode - Google Patents
Multi-target train energy-saving optimization method based on hybrid operation mode Download PDFInfo
- Publication number
- CN110990950B CN110990950B CN201911228280.0A CN201911228280A CN110990950B CN 110990950 B CN110990950 B CN 110990950B CN 201911228280 A CN201911228280 A CN 201911228280A CN 110990950 B CN110990950 B CN 110990950B
- Authority
- CN
- China
- Prior art keywords
- train
- particles
- solution
- value
- particle
- 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
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L27/00—Central railway traffic control systems; Trackside control; Communication systems specially adapted therefor
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The embodiment of the invention discloses a multi-target train energy-saving optimization method based on a mixed operation mode, which comprises the steps of inputting a basic scheme and related parameters of a train based on the mixed operation mode into a test platform, establishing a train operation strategy scheme variable set, obtaining a Pareto frontier solution of the train mixed operation mode by the platform through a multi-target particle swarm algorithm in the mixed operation mode, analyzing the train operation solution of each mixed operation mode by a train evaluation simulation module in the test platform, simulating the train operation according to the solution to obtain an evaluation value corresponding to a specific strategy, applying the obtained Pareto frontier solution to a train system, and selecting one of the Pareto frontier solutions as the train operation strategy by a train control system according to conditions. The safety, energy saving, punctuality, parking accuracy and comfort index of the train are considered, so that the ATO system can select the train operation tracking curve in due time according to different operation environments, thereby reducing the train operation energy consumption, ensuring the efficient operation of the train and improving the passenger satisfaction.
Description
Technical Field
The invention relates to the field of signal control of urban rail trains, in particular to a multi-target train energy-saving optimization method based on a hybrid operation mode.
Background
With the development of train automatic control systems (CBTC) and controller technologies in recent years, Automatic Train Operation (ATO) has been widely applied in the field of urban rail transit, and how to adopt a suitable train energy-saving driving strategy becomes a research hotspot in the field of rail transit.
The traditional ATO train energy-saving driving strategy is mainly characterized in that a parking precision index, an energy-saving index, a comfort level index and a train punctuality rate index of a train are considered according to existing train running information and line conditions to generate an optimal speed curve, and the ATO adopts inertia tracking to achieve the purpose of saving energy according to the generated optimal speed curve and the current position of the train. The traditional ATO operation focuses on multi-target train control research under a single operation mode, multiple train operation modes such as cruising, coasting, cruising plus coasting and the like are not considered at the same time, and the energy-saving strategy requirement under a specific target cannot be met under a specific condition.
In the implementation process of the invention, the inventor analyzes and improves the existing ATO system aiming at the defect that the existing method is single in a multi-target processing mode and a running mode, adopts a multi-target decision method, and comprehensively considers a plurality of indexes of safety, energy conservation, punctuality, parking accuracy and comfort of a train, so that the ATO system can select a train running tracking curve in time according to different running environments, thereby reducing the train running energy consumption, ensuring the efficient running of the train and improving the satisfaction degree of passengers.
Disclosure of Invention
Aiming at the problems in the background art, the invention provides a multi-target train energy-saving optimization method based on a hybrid operation mode, which mixes three common train operation modes of traction, cruising and coasting, establishes a multi-target train operation model and reflects the multi-target train operation model into a Pareto frontier solution, and provides a solution set with a larger range and more choices so as to better reduce the train operation energy consumption. A multi-target train energy-saving optimization method based on a hybrid operation mode comprises the following steps:
basic scheme for basing train on hybrid operation mode and related parameters thereofInputting the number into a test platform, and establishing a train operation strategy scheme variable set V ═ S, muT,μB,μRT,μD,VCR,VCO,VRT,SCOAnd the set comprises parameter values related to three common train operation modes and a train operation mode indicating value S specifically related to the three common train operation modes. Wherein muTIs the proportional value of traction force to maximum traction force under traction command, muBIs the proportional value of the braking force to the maximum braking force under a braking command, muRTFor the proportional value of tractive effort to maximum tractive effort under a re-traction command, μDIs the proportional value of braking force to maximum braking force under deceleration command, VCRAt cruising speed, VCOFor the coasting speed value, VRTFor the reacceleration speed value, SCOIs the coasting position point.
The invention relates to three common train operation modes, which comprise:
the sequence of commands in the cruise running mode is external traction, constant-speed cruise and braking in sequence. The maximum allowable cruising speed is the speed limit value of the line.
And starting the traction strategy in the idle running mode, starting to execute an idle command after the train reaches the idle speed, and executing acceleration operation when the train speed is lower than the re-traction speed, and repeating the operation until the train is braked.
And under the cruising coasting running mode, the train accelerates to the cruising speed value after running is started again, then the coasting command is implemented after the train reaches the position point where the coasting starts, and finally the train still executes the braking command to stop.
Other parameters of interest to the test platform include: train quality, a function of traction characteristic and braking characteristic, a train resistance calculation formula, a station position, a line slope value and a line speed limit value.
The test platform adopts a multi-target particle swarm algorithm in a mixed operation mode to obtain a Pareto front solution of the mixed operation mode of the train; a train evaluation simulation module in the test platform analyzes the train operation solution of each mixed operation mode, and simulates the train operation according to the train operation solution to obtain an evaluation value of a corresponding specific strategy; the multi-target particle swarm algorithm in the hybrid operation mode comprises the following steps:
step 1: initializing a particle group, setting the particle group speed to 0, and evaluating the initialized particle group;
step 2: initializing the optimal historical position of each particle in the population and the corresponding fitness value thereof;
and step 3: selecting non-dominant particles and storing them and their fitness values into the precipitated particles;
and 4, step 4: initializing a hypercube and corresponding particle distribution;
and 5: randomly selecting global optimal particles from the precipitated particles, and then updating the speed and the position of the particles according to an updating formula in the traditional particle swarm optimization;
step 6: performing cross operation on the particle swarm, and checking whether each particle is still in the boundary;
and 7: evaluating the particle swarm;
and 8: renewing the precipitating particles by combining nondominant with precipitating particles;
and step 9: updating the optimal particle swarm and the fitness value thereof;
step 10: repeating the steps 5 to 9 until the termination condition is satisfied;
after the algorithm is completed, we can obtain Pareto frontiers through the fitness value of the precipitated particles. In order to obtain the value of the evaluation index, each particle position is input into the train evaluation simulation system. The train evaluation simulation system calculates a fitness function value by using a train operation model established before, and returns the train fitness function value to a multi-target particle swarm algorithm in a hybrid operation mode.
Wherein the step of obtaining the particle evaluation index value comprises the following steps:
step 1: for a certain solution, firstly, obtaining the solution of the strategy of the hybrid operation mode and analyzing the solution to obtain a specific value of the solution;
step 2: judging an operation mode corresponding to the strategy solution according to the value of S in the strategy solution, wherein different operation modes execute different operation command judgment functions;
and step 3: repeatedly updating the train running state by using a time clock cycle, and recording the component of the train running evaluation index in each time clock until the train stops running at the destination;
and 4, step 4: obtaining a final train operation evaluation value according to the evaluation index component in each train operation step length;
applying the Pareto front solution obtained by the method to a train system, and designing a train control system to select one solution from the Pareto front solution as a train operation strategy according to conditions, wherein the method comprises the following steps:
step 1: acquiring a running section, a time requirement and a comfort degree allowable range of the train;
step 2: the train control system selects a corresponding train operation strategy solution from the Pareto front solution according to the set conditions, and converts the corresponding train operation strategy solution into a train control strategy and a corresponding operation speed curve;
and step 3: the train drives according to the operation strategy obtained in the hybrid operation mode, and the operation speed curve is compared in real time to adjust the control output quantity;
the method of the invention has the following beneficial effects:
the hybrid train operation mode adopted by the invention comprises but is not limited to a common single train operation mode, the solution set search domain is wider, the optimization is solved by adopting a multi-target particle swarm algorithm under the hybrid operation mode, the Pareto frontier solution distribution is good, and the final train operation energy-saving effect is obviously better than that of a single strategy. And the test platform has better portability and can be applied to different lines and train models.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly explained below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and therefore should not be considered as limiting the scope, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without inventive effort.
Fig. 1 is a flowchart of a train energy-saving optimization method based on a hybrid operation mode according to an embodiment of the present invention.
Fig. 2 is a flowchart of a train hybrid operation mode energy-saving optimization method based on a test platform according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of three modes included in a train hybrid mode according to an embodiment of the present invention.
Fig. 4 is a flowchart of a multi-target particle swarm algorithm in a hybrid operation mode according to an embodiment of the present invention.
Fig. 5 is a flowchart for selecting and tracking a final train operation strategy from the Pareto frontier solution according to an embodiment of the present invention.
Icon: 100: a test platform; 110: a parameter entering module; 120: a solving module; 130: a curve module; 140: a signal module; 150: other devices.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a user skilled in the art without inventive work based on the embodiments of the present invention, are within the scope of protection of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Fig. 1 is a flowchart of a train energy-saving optimization method based on a hybrid operation mode according to an embodiment of the present invention. Referring to fig. 1, the specific method includes:
the train energy-saving optimization method based on the hybrid operation mode is applied to the test platform (100), and the test platform (100) can be, but is not limited to, a personal computer, a vehicle-mounted computer, a tablet computer and the like. Referring to fig. 1, fig. 1 shows a block schematic diagram of an experimental platform (100) according to an embodiment of the present invention, where the experimental platform (100) includes an entry module 110, a solution module 120, a running module 130, a signal module 140, and other devices 150. The information transmission from the parameter module 110 to the solving module 120 to the operation module 130 is a one-way information transmission module, and the information transmission from the operation module 130 to the signal module 140 and other devices 150 is a two-way information transmission module.
The parameter module 110 mainly includes two input modules, namely a variable input module and a parameter input module. The dependent variable may be expressed by the set V, where V ═ S, μT,μB,μRT,μD,VCR,VCO,VRT,SCOAnd the set comprises parameter values related to three common train operation modes and a train operation mode indicating value S specifically related to the three common train operation modes. Wherein muTIs the proportional value of traction force to maximum traction force under traction command, muBIs the proportional value of the braking force to the maximum braking force under a braking command, muRTFor the proportional value of tractive effort to maximum tractive effort under a re-traction command, μDIs the proportional value of braking force to maximum braking force under deceleration command, VCRAt cruising speed, VCOFor the coasting speed value, VRTFor the reacceleration speed value, SCOIs the coasting position point. The relevant parameters comprise a line gradient parameter, a line speed limit parameter, a station position, train quality, train length and a train traction and braking characteristic curve. The variable value range provides an interface for a user to set, and other data are imported into the system in an excel table form.
The solving module 120 uses a multi-objective particle swarm algorithm in the hybrid operation mode to solve the Pareto front edge corresponding to the train operation curve set used in the operation module 130. The solving module 120 is modified to be a multi-target particle swarm algorithm suitable for a hybrid strategy based on a conventional multi-target particle swarm algorithm for reference. The solution module 120 performs a Train state calculation and update on the Train operation condition by using Train Performance evaluation Simulation (TPS for short) in combination with a Train operation model, where the Train state calculation and update is based on a time clock. The results of the algorithm output may be saved in memory in excel table form. The variables are usually described in a matrix form during solving, and the method not only can quickly describe a large amount of data, but also can improve the calculation efficiency.
The operation module 130 is a command controller for train operation. The operation module 130 mainly functions to select an appropriate operation solution (corresponding to a recommended speed curve) according to the train operation condition, and control the train operation using the selected solution. The curve selection logic before train operation is mainly based on the requirement of common operation speed curve time value and the storage capacity of the speed curve of the controller, and in addition, the curve with the shortest operation time is selected by the controller as the alternative curve of the emergency. When the train runs, reasonable control command types and numerical values are output by combining the gear corresponding to the current position of the selected curve and the real-time speed, so that the curve is tracked and controlled, and the aim of saving energy is finally fulfilled.
The signal module 140 is responsible for receiving and transmitting related signals in the test platform. The signal module 140 is in two-way communication with the operation module 130, and can exchange data in real time. That is, the signal module 140 is responsible for transmitting the signal data collected from other places to the train operation module 130, and the train operation module 130 sends out a train control instruction according to the relevant information and transmits the train control instruction back to the signal module 140. This direct communication mode depends mainly on the specific bearing platform of the two modules.
The other devices 150 are devices involved in the control signals during train operation. The device needs to have the signal processing function of the corresponding system and can send transmission related signals, and the bearer of the device can be real application equipment or electronic simulation equipment.
It should be understood that the configuration shown in fig. 1 is merely a schematic illustration of a structural application of the test platform (100), and that the test platform (100) may also include more or fewer components than shown in fig. 1, or have a different configuration than shown in fig. 1. The components shown in fig. 1 may be implemented in hardware, software, or a combination thereof.
Based on the above test platform (100), an implementation manner of a train energy-saving optimization method based on a hybrid operation mode is given below, and an execution subject of the process may be the above test platform (100), please refer to fig. 2. Fig. 2 shows a flowchart of a test platform-based energy-saving optimization method for a hybrid operation mode of a train according to an embodiment of the present invention, which may include the following steps:
and S101, inputting the basic scheme of the train based on the hybrid operation mode and related parameters of the basic scheme into a test platform.
The basic scheme of the train based on the hybrid operation mode comprises three common train operation modes: cruise mode, coasting plus cruise mode. The three modes have different characteristics in the aspects of operation sequences and energy consumption, and the mixed application of the three modes in the train operation logic value enables the final operation strategy Pareto front solution to have a cluster advantage.
S102, the test platform obtains a Pareto front solution of the train hybrid operation mode by adopting a multi-target particle swarm algorithm in the hybrid operation mode.
The invention designs a hybrid strategy multi-target particle swarm algorithm corresponding to a hybrid operation strategy to solve the problem. The invention combines simulation and algorithm, and a train evaluation simulation module in a test platform analyzes the train operation solution of each mixed operation mode to obtain the evaluation value of the corresponding specific strategy.
S103, taking a selected Pareto front edge de-embedded train system as a train operation strategy.
The Pareto front solution obtained in the S102 is applied to a train system, and before the train runs, the system selects a solution which is most suitable for the current running of the train from a solution set to serve as a train running strategy. And outputs and adjusts control instructions and values in combination with the actual conditions of the train in the actual operation.
Based on the above flow step S101, a schematic diagram of three modes included in a train hybrid mode is given below, an execution subject of the method may be the above test platform (100), and please refer to fig. 3-1, fig. 3-2, and fig. 3-3 for a train operation speed curve of the three modes. Fig. 3 shows three hybrid train operation modes provided by the embodiment of the invention.
FIG. 3-1 illustrates a typical train operating speed profile in a cruise mode of operation of the train;
under a simpler cruise strategy, the orders of external traction, constant-speed cruise and braking are performed in sequence. The maximum allowable cruising speed is the speed limit value of the line. Wherein the variable parameter of the strategy design comprises a ratio value mu of the traction force to the maximum traction force under the train traction commandTProportional value mu of braking force to maximum braking force under braking commandBAnd, cruise speed VCR. Fig. 3-1 shows a curve diagram for a simpler speed limit situation, and a more complex speed limit situation will also include acceleration and deceleration caused by a speed limit change, i.e. the parameter also includes the proportional value mu of the tractive force to the maximum tractive force under the re-traction commandRTWith the proportional value mu of the braking force to the maximum braking force on deceleration commandD。
3-2 illustrate a common train operating speed profile for a coasting mode of train operation;
the coasting strategy is still started with the traction strategy, the coasting command is started after the train reaches the coasting speed, the acceleration operation is executed when the train speed is lower than the retracing speed, and the steps are repeated until the train is braked. Wherein the variable parameter of the strategy design comprises a ratio value mu of the traction force to the maximum traction force under the train traction commandTProportional value mu of braking force to maximum braking force under braking commandBThe ratio of tractive effort to maximum tractive effort, mu, under a re-traction commandRTValue of coasting speed VCOValue of reacceleration velocity VRT. Fig. 3-2 shows a graph for a simpler speed limit situation, and a more complex speed limit situation will also include deceleration caused by a speed limit change, i.e. the parameters also include the proportional value μ of braking force to maximum braking force under a deceleration commandD。
3-3 illustrate a typical train operating speed profile for a cruise plus coast mode of operation of a train;
and as for the cruising coasting strategy, the train accelerates to the cruising speed value after running is started again, then the coasting command is implemented after the train reaches the position point where the coasting starts, and finally the train still executes the braking command to stop. Wherein the variable parameter of the strategy design comprises a ratio value mu of the traction force to the maximum traction force under the train traction commandTProportional value mu of braking force to maximum braking force under braking commandBCruise speed VCRAnd, a coasting position point SCO. Fig. 3-3 show a graph for a simpler speed limit situation, and a more complex speed limit situation will also include acceleration and deceleration caused by a speed limit change, i.e. the parameter also includes the proportional value mu of the tractive force to the maximum tractive force under the re-traction commandRTWith the proportional value mu of the braking force to the maximum braking force on deceleration commandD。
Based on the above step S102, a flowchart of a multi-target particle swarm algorithm in a hybrid operation mode is given below, and an execution subject of the method may be the above test platform (100), please refer to fig. 4. Similar to the particle swarm algorithm, the HS-MOPSO algorithm uses velocity and position to find the pareto surface. Fig. 4 shows a flowchart of a multi-target particle swarm algorithm in a hybrid operation mode according to an embodiment of the present invention.
S201, initializing a particle swarm.
27 particle position by the variable muT,μB,μRT,μD,VCR,VCO,VRTAnd SCOThe composition, the particles, in turn, make up the population,
indicating the location of the population containing information on the location of all particles, the invention uses the labeling of the relevant patterns in order to indicate which pattern is to be performed when calculating the fitness of the particles
Adding into
Thereby obtaining the position of the population containing the information of the operation mode
Namely, it is
Initializing a population of particle populations
And the group velocity of the particles
Is set to 0. Then evaluating fitness function of particle swarm population through TPS simulation
The method comprises the following steps of train operation energy consumption f (E), train operation time f (T) and train operation comfort f (C). .
S202, initializing the optimal historical position of each particle in the population and the corresponding fitness value of each particle.
And assigning the optimal historical position of the initialized particle in the population as the initialized particle position, and initializing the corresponding fitness value of the optimal historical position of the initialized particle as the fitness value of the initialized particle.
S203, selecting non-dominant particles and storing them and their fitness values, respectively.
32, sorting and screening the fitness value of the particle by adopting non-dominated quick sorting, and notably, the sorting process needs to ensure the correspondence between the position of the particle and the fitness value and store the particle to the precipitation particle
The corresponding fitness values of the particles are stored in
In (1).
S204, initializing the hypercube and the corresponding particle distribution.
The hypercube is a design model for recording particle distribution in the algorithm of the invention, cuts the range value of the particles and judges the number of the particles in each small spatial domain. When the particles are crossed in selection, the hypercube value is taken as the probability basis of selection and inheritance.
S205, randomly selecting from the precipitation particles, and updating the speed and the position of the particles.
36 random from
To select a globally optimal particle
The updates are then distributed according to the following two formulas
And
wherein
Is a weight equal to 0.4 and,
and
is a learning factor with a value of 0 to 2.
S206, performing intersection operation on the particle swarm, and checking whether each particle is still in the boundary.
The particle swarm cross operation is the basis of population continuation, and because the cross operation has certain randomness, the crossed particles may exceed the value running range of the solution, so the crossed particles need to check the boundary condition one by one, and if the particles go out of the boundary, numerical correction is needed.
S207, the particle group is evaluated.
40 train operation simulation is an important content for studying train operation. The designed TPS system of the train operation simulation platform can realize train operation simulation under a specified line and a specified vehicle type, and line and train signals can be changed by providing different parameter configuration files. And (3) continuously advancing the time clock, calculating the state of the train by the platform, obtaining the speed, position and acceleration parameters of the train at each moment, and finally outputting three evaluation indexes f (E), f (T) and f (C) as solution values corresponding to each operation strategy.
S208, renewing the precipitation particles by combining the nondominant with the precipitation particles.
The particles obtained in S207 are subjected to non-dominant solution judgment and generated together with the previous precipitation particles to obtain new precipitation particles, and the particles are updated and stored in
The corresponding fitness value of the particles is updated and stored in
In (1).
S209, updating the optimal particle swarm and the fitness value thereof
The optimal particle swarm
And its fitness value
Updating and storing.
S210, judging whether a termination condition is met.
46 the end condition is to solve the algebra generated by particle storage, when the band number of the particle swarm is larger than the algebra in the end condition, the operation is stopped and the final particle swarm is output
And its corresponding fitness value
Otherwise, the test platform (100) will continue to execute S205 to S210.
Based on the above step S103, a process of selecting a final train operation strategy from the Pareto frontier solution and tracking the strategy is provided below, and an execution subject of the method may be the above test platform (100), please refer to fig. 5. Fig. 5 shows a flowchart for selecting and tracking a final train operation strategy from the Pareto frontier solution according to an embodiment of the present invention.
S301, acquiring a Pareto front edge of the current operation interval.
After S102 is executed, the testing platform (100) may obtain a Pareto front edge of the station operating solution according to the currently operating station. And reading the solution set and the corresponding fitness value into the system.
And S302, acquiring the running time requirement and comfort degree allowable range of the train.
51, filtering a train operation solution set within a comfort degree allowable range, and sending operation requirements such as allowable operation time to trains through a station control room by a subway master control. The operating hours at peak and off-peak times may be slightly different to account for passenger demand at different times.
And S303, selecting a train operation recommendation curve according to the operation time.
After the test platform (100) finishes executing S302, the test platform (100) selects the closest solution in the Pareto frontier according to the requirements of running time and the like, and inputs the solution into the controller as the train running basis.
And S304, comparing the current speed, the recommended speed and the recommended gear of the train.
After the test platform (100) sends out a train operation instruction, the test platform operates according to the recommended gear. And in the running process of the train, the test platform (100) monitors the current running speed of the train in real time, and compares and records the rest recommended speeds and recommended gears in real time.
S305, adjusting the type and the value of the train control command.
On the basis of obtaining the current running speed, the recommended speed and the recommended gear of the train, the test platform (100) adjusts the type and the value of the train output instruction in real time according to the fuzzy logic so as to achieve the real-time tracking control of the recommended running strategy and the corresponding speed curve thereof and finally achieve the energy-saving purpose of the train.
And S306, judging whether the train reaches the terminal.
In general, the stop condition is that the train enters the parking range and the speed is 0, and when the condition is not satisfied, the test platform (100) continues to execute steps S304 to S306.
Claims (6)
1. A multi-target train energy-saving optimization method based on a hybrid operation mode is characterized by comprising the following steps:
inputting a basic scheme of the train based on a hybrid operation mode and related parameters thereof into a test platform, and establishing a train operation strategy scheme variable set V ═ S, muT,μB,μRT,μD,VCR,VCO,VRT,SCOThe set comprises parameter values related to three common train operation modes and a train operation mode indicated value S specifically related to the set;
wherein muTIs the proportional value of traction force to maximum traction force under traction command, muBIs the proportional value of the braking force to the maximum braking force under a braking command, muRTFor the proportional value of tractive effort to maximum tractive effort under a re-traction command, μDIs the proportional value of braking force to maximum braking force under deceleration command, VCRAt cruising speed, VCOFor the coasting speed value, VRTFor the reacceleration speed value, SCOIs the coasting position point;
the test platform adopts a multi-target particle swarm algorithm in a mixed operation mode to obtain a Pareto front solution of the mixed operation mode of the train; a train evaluation simulation module in the test platform analyzes the train operation solution of each mixed operation mode, and simulates the train operation according to the train operation solution to obtain an evaluation value of a corresponding specific strategy;
and applying the obtained Pareto front solution to a train system, and designing a train control system to select one solution from the Pareto front solution as a train operation strategy according to the time requirement and comfort degree allowable range of train interval operation.
2. The method according to claim 1, characterized in that the test platform obtains a Pareto frontier solution of the train hybrid operation mode by using a multi-target particle swarm algorithm in the hybrid operation mode; a train evaluation simulation module in the test platform analyzes the train operation solution of each mixed operation mode, and simulates the train operation according to the train operation solution to obtain an evaluation value of a corresponding specific strategy; the multi-target particle swarm algorithm in the hybrid operation mode comprises the following steps:
step 1: initializing a particle group, setting the particle group speed to 0, and evaluating the initialized particle group;
step 2: initializing the optimal historical position of each particle in the population and the corresponding fitness value thereof;
and step 3: selecting non-dominant particles and storing them and their fitness values into the precipitated particles;
and 4, step 4: initializing a hypercube and corresponding particle distribution;
and 5: randomly selecting global optimal particles from the precipitated particles, and then updating the speed and the position of the particles according to an updating formula in the traditional particle swarm optimization;
step 6: performing cross operation on the particle swarm, and checking whether each particle is still in the boundary;
and 7: evaluating the particle swarm;
and 8: renewing the precipitating particles by combining nondominant with precipitating particles;
and step 9: updating the optimal particle swarm and the fitness value thereof;
step 10: repeating the steps 5 to 9 until the termination condition is satisfied;
after the algorithm is completed, the Pareto frontier can be obtained through the fitness value of the sediment particles, in order to obtain the value of an evaluation index, the position of each particle is input into a train evaluation simulation system, the train evaluation simulation system calculates the fitness function value by using a train operation model established by the train evaluation simulation system, and the fitness function value is returned to the multi-target particle swarm algorithm in the hybrid operation mode.
3. The method as claimed in claim 2, characterized in that the evaluation values of the driving Strategy are obtained, the corresponding operating Strategy is executed according to the Strategy parameters, the system accumulatively calculates the energy consumption ei per unit time and the length Δ ti per unit time at each time clock, and the acceleration difference value under two adjacent time clocks is obtained, wherein the step of obtaining the evaluation index values is as follows:
step 1: for a certain solution, firstly, obtaining the solution of the strategy of the hybrid operation mode and analyzing the solution to obtain a specific value of the solution;
step 2: judging an operation mode corresponding to the strategy solution according to the value of S in the strategy solution, wherein different operation modes execute different operation command judgment functions;
and step 3: repeatedly updating the train running state by using a time clock cycle, and recording the component of the train running evaluation index in each time clock until the train stops running at the destination;
and 4, step 4: and obtaining a final train operation evaluation value according to the evaluation index component in each train operation step length.
4. The method of claim 2, wherein the Pareto frontier solution obtained in the previous method is applied to a train system, and a train control system is designed to select one solution from the Pareto frontier solutions as a train operation strategy according to circumstances, comprising the steps of:
step 1: acquiring a running section, a time requirement and a comfort degree allowable range of the train;
step 2: the train control system selects a corresponding train operation strategy solution from the Pareto front solution according to the set conditions, and converts the corresponding train operation strategy solution into a train control strategy and a corresponding operation speed curve;
and step 3: and the train drives according to the operation strategy obtained in the hybrid operation mode, and the operation speed curve is compared in real time to adjust the control output quantity.
5. The method according to claim 3, wherein the execution subject of the multi-target particle swarm algorithm in the hybrid operation mode is the test platform (100), and similar to the particle swarm algorithm, the hybrid mode multi-target particle swarm HS-MOPSO algorithm uses speed and position to find the pareto plane, and the specific process is as follows:
initializing a particle swarm population;
the position of the particles being determined by the variable μT,μB,μRT,μD,VCR,VCO,VRTAnd SCOThe composition, the particles, in turn, make up the population,
indicating the position of the population containing the position information of all the particles, and marking the relevant mode of the particles in order to indicate which operation mode is executed when the fitness is calculated by the particles
Adding into
Thereby obtaining the position of the population containing the information of the operation mode
Namely, it is
Initialization
And the group velocity of the particles
Setting the fitness function to be 0, and then carrying out simulation evaluation on the fitness function of the particle swarm population through a TPS (train operating simulation platform) of the train operation simulation platform
The method comprises the following steps of (1) train operation energy consumption f (E), train operation time f (T) and train operation comfort f (C);
initializing the optimal historical position and the corresponding fitness value of each particle in the population: assigning the optimal historical position of the initialized particle in the population as the initialized particle position, and initializing the corresponding fitness value of the optimal historical position of the initialized particle as the fitness value of the initialized particle;
selecting non-dominant particles and storing them and their fitness values, respectively: sorting and screening the fitness value of the particles by adopting non-dominated quick sorting, and is remarkable in that the sorting process needs to ensure the correspondence between the positions of the particles and the fitness value and store the particles into precipitated particles
The corresponding fitness values of the particles are stored in
Performing the following steps;
initializing hypercube and its corresponding particle distribution: the hypercube is a design model for recording particle distribution in the algorithm, cuts the particle range value, judges the particle quantity in each small space domain, and takes the hypercube value as the probability basis of selection and inheritance when the particles are selected and crossed;
randomly selecting from the sediment particles, updating the speed and position of the particles: random slave
To select a globally optimal particle
The updates are then distributed according to the following two formulas
And
wherein
Is a weight equal to 0.4 and,
and
is a learning factor with a value of 0 to 2;
the population of particles is interleaved and checked for whether each particle is still within the boundary: the particle swarm cross operation is the basis of population continuation, and because the cross operation has certain randomness, the crossed particles can possibly exceed the value running range of the solution, so that the crossed particles need to check the boundary condition one by one, and if the particles go out of the boundary, numerical correction is needed;
evaluating the particle swarm: the train operation simulation is an important content for researching train operation, a designed train operation simulation platform TPS system can realize train operation simulation under a specified line and a specified train type, different parameter configuration files are provided, line and train information can be changed, the platform calculates the state of a train through a time clock, the speed, position and acceleration parameters of the train at each moment are obtained, and finally three evaluation indexes f (E), f (T) and f (C) are output as solution values corresponding to each operation strategy;
renewing the precipitating particles by combining nondominant with precipitating particles: the obtained particles are subjected to non-dominant solution judgment and generated together with the previous precipitation particles to obtain new precipitation particles, and the particles are updated and stored in
The corresponding fitness value of the particles is updated and stored in
Performing the following steps;
updating the optimal particle swarm and the fitness value thereof to optimize the particle swarm
And its fitness value
Updating and storing;
judging whether a termination condition is met: the end condition is to solve the algebra generated by particle storage, when the band number of the particle swarm is larger than the algebra in the end condition, the operation is stopped and the final algebra is output
And corresponding thereto
Otherwise, the test platform (100) continuously executes the process of updating the particles to the final value condition judgment.
6. The method according to claim 3, wherein a process for selecting and tracking a final train operation strategy from the Pareto frontier solution is performed by the test platform (100), and the specific process includes:
the method comprises the steps that a Pareto front edge of a current operation interval is obtained, a test platform (100) can obtain the Pareto front edge of a solution set operated by a station according to the current operation station, and the solution set and a corresponding fitness value of the solution set are read into a system;
the method comprises the steps of obtaining train operation time requirements and a comfort degree allowable range, filtering train operation solution sets in the comfort degree allowable range, sending operation requirements such as allowable operation time to trains through a station control room by a subway master control, and slightly different operation time at peak and off-peak moments so as to meet passenger transport requirements at different moments;
selecting a train operation recommended curve according to the operation time, and after the test platform (100) is executed, inputting a solution which is closest to the Pareto front edge and is selected by the test platform (100) according to the requirements of the operation time and the like into a controller to serve as a train operation basis;
comparing the current speed, the recommended speed and the recommended gear of the train, operating the test platform (100) according to the recommended gear after sending a train operation instruction, monitoring the current operation speed of the train in real time by the test platform (100) in the train operation process, and comparing and recording the rest recommended speed and the recommended gear in real time;
the train control command type and value are adjusted, and the test platform (100) adjusts the train output command type and value in real time according to fuzzy logic on the basis of obtaining the current running speed, the recommended speed and the recommended gear of the train, so that the real-time tracking control of the recommended running strategy and the corresponding speed curve is achieved, and the energy-saving purpose of the train is finally achieved;
and judging whether the train reaches the end point, wherein the end condition is that the train enters a parking range and the speed is 0 under the general condition, and when the condition is not met, the test platform (100) continuously executes the step of comparing the current speed with the recommended gear.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911228280.0A CN110990950B (en) | 2019-12-04 | 2019-12-04 | Multi-target train energy-saving optimization method based on hybrid operation mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911228280.0A CN110990950B (en) | 2019-12-04 | 2019-12-04 | Multi-target train energy-saving optimization method based on hybrid operation mode |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110990950A CN110990950A (en) | 2020-04-10 |
| CN110990950B true CN110990950B (en) | 2021-12-21 |
Family
ID=70090041
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911228280.0A Active CN110990950B (en) | 2019-12-04 | 2019-12-04 | Multi-target train energy-saving optimization method based on hybrid operation mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110990950B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111898185B (en) * | 2020-07-17 | 2024-05-28 | 交控科技股份有限公司 | Rail transit design scheme evaluation method and device and readable storage medium |
| CN112896244B (en) * | 2021-02-22 | 2022-07-19 | 同济大学 | Inter-station operation control method and device based on target speed curve tracking |
| CN112948971B (en) * | 2021-03-04 | 2023-09-29 | 北京交通大学 | Energy-saving optimization method for speed curve of maglev train |
| CN113147841A (en) * | 2021-05-13 | 2021-07-23 | 中车长春轨道客车股份有限公司 | Rail vehicle capacity management and energy-saving auxiliary driving method and related device |
| CN113326574A (en) * | 2021-06-30 | 2021-08-31 | 南京信息工程大学 | ATO speed command energy-saving optimization method considering train accurate parking factor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN106777717A (en) * | 2016-12-23 | 2017-05-31 | 南京理工大学 | A kind of ATO speed command energy conservation optimizing methods for considering handling capacity of passengers change |
| CN108932356A (en) * | 2017-05-27 | 2018-12-04 | 南京理工大学 | A kind of ATO speed command energy conservation optimizing method considering Train delay factor |
| CN109376437A (en) * | 2018-10-30 | 2019-02-22 | 中车南京浦镇车辆有限公司 | A kind of vehicle-mounted energy storage optimal control method of new energy based on multi-objective genetic algorithm |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102360401B (en) * | 2011-10-14 | 2013-06-05 | 南京理工大学 | Method for designing urban rail transit energy-saving run chart based on genetic algorithm |
| US9299032B2 (en) * | 2012-11-12 | 2016-03-29 | International Business Machines Corporation | Multi objective design selection |
| CN104881527B (en) * | 2015-05-13 | 2018-06-12 | 南京理工大学 | Urban railway transit train ATO speed command optimization methods |
| CN105243430B (en) * | 2015-09-07 | 2018-10-09 | 北京交通大学 | The optimization method of the target velocity curve of energy-saving train operation |
| CN106126870B (en) * | 2016-08-31 | 2019-07-16 | 中车青岛四方机车车辆股份有限公司 | A kind of energy saving boundary evaluation method of train traction system |
| CN106842901B (en) * | 2017-01-19 | 2019-11-08 | 重庆交通大学 | For the method for train automated driving system formation speed control command |
| CN107368920B (en) * | 2017-07-01 | 2021-04-06 | 南京理工大学 | Energy-saving optimization method for multi-train operation in off-peak hours |
| KR102185222B1 (en) * | 2018-01-24 | 2020-12-01 | 현대로템 주식회사 | Control system for ato operation train and railway train having the same and control method thereof |
| CN109815536B (en) * | 2018-12-19 | 2020-10-30 | 西南交通大学 | Urban rail transit energy-saving timetable and operation curve optimization method |
| CN110497943B (en) * | 2019-09-03 | 2020-07-14 | 西南交通大学 | Online optimization method for energy-saving operation strategy of urban rail train based on reinforcement learning |
-
2019
- 2019-12-04 CN CN201911228280.0A patent/CN110990950B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN106777717A (en) * | 2016-12-23 | 2017-05-31 | 南京理工大学 | A kind of ATO speed command energy conservation optimizing methods for considering handling capacity of passengers change |
| CN108932356A (en) * | 2017-05-27 | 2018-12-04 | 南京理工大学 | A kind of ATO speed command energy conservation optimizing method considering Train delay factor |
| CN109376437A (en) * | 2018-10-30 | 2019-02-22 | 中车南京浦镇车辆有限公司 | A kind of vehicle-mounted energy storage optimal control method of new energy based on multi-objective genetic algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 基于Pareto多目标遗传算法的单列车定时节能研究;李婷等;《广西大学学报(自然科学版)》;20171025(第05期);全文 * |
| 多种群分层联合优化的城轨列车ATO研究;徐凯等;《铁道学报》;20180615(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110990950A (en) | 2020-04-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110990950B (en) | Multi-target train energy-saving optimization method based on hybrid operation mode | |
| CN108791367B (en) | Energy-saving operation method for train | |
| CN111439260B (en) | Network-connected commercial diesel vehicle cruise running optimization control system oriented to individual requirements | |
| CN111267830B (en) | Hybrid power bus energy management method, device and storage medium | |
| CN106427589A (en) | Electric car driving range estimation method based on prediction of working condition and fuzzy energy consumption | |
| CN113788021B (en) | Adaptive following cruise control method combined with preceding vehicle speed prediction | |
| CN105868787A (en) | Electric car driving range evaluation method based on working condition identification and fuzzy energy consumption | |
| CN111038478B (en) | Vehicle running speed determination method and device | |
| CN106777717A (en) | A kind of ATO speed command energy conservation optimizing methods for considering handling capacity of passengers change | |
| CN103869809A (en) | Railway locomotive running dynamic simulation test device and simulation method thereof | |
| CN106056238B (en) | Planning method for train interval running track | |
| CN114074680B (en) | Vehicle channel change behavior decision method and system based on deep reinforcement learning | |
| CN111680413B (en) | Tramcar timing energy-saving operation optimization method and system based on double-layer algorithm | |
| CN110427690A (en) | A kind of method and device generating ATO rate curve based on global particle swarm algorithm | |
| CN113562039A (en) | Multi-vehicle cooperation oriented automatic operation diagram adjusting method and device | |
| CN110116730A (en) | A kind of vehicle brake control method, system, equipment and medium | |
| Bin et al. | Energy-efficient speed profile optimization for high-speed railway considering neutral sections | |
| CN114047765B (en) | Virtual marshalling-oriented multi-train self-adaptive cruise control method and system | |
| CN110481331A (en) | A kind of electric car slides energy reclaiming method | |
| Kan et al. | A deep learning engine power model for estimating the fuel consumption of heavy-duty trucks | |
| CN115056825B (en) | Train acceleration method with self-adaptive performance | |
| CN116039715A (en) | Train virtual marshalling operation control method and device | |
| Škugor et al. | Dynamic programming-based design of shift scheduling map taking into account clutch energy losses during shift transients | |
| CN114547774B (en) | Heavy-load train arrival curve optimization method and system | |
| CN113635900B (en) | Channel switching decision control method based on energy management in predicted cruising process |
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 |