WO2022237115A1 - 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置 - Google Patents

一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置 Download PDF

Info

Publication number
WO2022237115A1
WO2022237115A1 PCT/CN2021/132068 CN2021132068W WO2022237115A1 WO 2022237115 A1 WO2022237115 A1 WO 2022237115A1 CN 2021132068 W CN2021132068 W CN 2021132068W WO 2022237115 A1 WO2022237115 A1 WO 2022237115A1
Authority
WO
WIPO (PCT)
Prior art keywords
rail vehicle
train
speed limit
parameters
line
Prior art date
Application number
PCT/CN2021/132068
Other languages
English (en)
French (fr)
Inventor
马凯
李玲玉
王雷
高登科
Original Assignee
中车长春轨道客车股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中车长春轨道客车股份有限公司 filed Critical 中车长春轨道客车股份有限公司
Publication of WO2022237115A1 publication Critical patent/WO2022237115A1/zh

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B61RAILWAYS
    • B61LGUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L27/00Central railway traffic control systems; Trackside control; Communication systems specially adapted therefor
    • B61L27/20Trackside control of safe travel of vehicle or train, e.g. braking curve calculation

Definitions

  • the present application relates to the technical field of rail vehicles, and more specifically, relates to a method for capacity management and energy-saving assisted driving of rail vehicles and related devices.
  • this application provides a method of capacity management and energy-saving assisted driving of rail vehicles and related devices, so as to realize the planning function of automatic driving curve based on the multi-objective particle swarm algorithm, and improve the automation and control of rail vehicle operation control. Intelligent level.
  • a method for capacity management and energy-saving assisted driving of a rail vehicle comprising:
  • the automatic driving curve of the rail vehicle is solved using the basic parameters and the decision variables of each interval.
  • said dividing the line to be operated of said rail vehicle into a plurality of intervals includes:
  • the line to be operated of the rail vehicle is discretized into a plurality of small sections of equal length, and the static speed limit value of each small section is recorded to obtain the speed limit Discretized collections.
  • the initializing the decision variables of each of the intervals includes:
  • the distance constraints include: 0 ⁇ s ⁇ S, wherein s represents the running distance of the rail vehicle at any time during the operation of the small section, and S represents the total length of the small section;
  • the speed limit constraints include: 0 ⁇ v i ⁇ v lim , wherein, v i represents the speed of the rail vehicle at any position in the small section, and v lim represents the static speed limit value of the small section;
  • the acceleration constraints include: a min ⁇ a i ⁇ a max , where a i represents the acceleration of the rail vehicle, and a min and a max represent the minimum braking acceleration and maximum traction acceleration of the train, respectively.
  • the basic initialization parameters include:
  • the train parameters include at least: train model, marshalling situation, train quality, train traction and braking coefficient;
  • the line parameters at least include: line length, slope, curvature and speed limit;
  • the operating parameters at least include: train section running time;
  • the algorithm-related parameters at least include: particle dimension, population number and inertia factor.
  • the multi-objective particle swarm optimization algorithm using the basic parameters and the decision variables of each interval to solve the automatic driving curve of the rail vehicle includes:
  • train operation control model according to the fitness function of energy consumption, the fitness function of running time, the fitness function of parking accuracy and the fitness evaluation of speed limit, select non-inferior solutions from the set of train control sequences;
  • the train operation control model includes:
  • F( ) is the optimization objective function
  • c i represents the decision variable
  • f e (c i ) and f t (c i ) represent the energy consumption and time target variables of the optimization model, respectively
  • st represents the constraints to be followed
  • v 1 represents the speed of the rail vehicle at the initial moment
  • v n represents the speed of the rail vehicle at the stop moment.
  • the energy consumption fitness function includes:
  • E i represents the traction energy consumption of the rail vehicle in the i-th small section
  • E represents the traction energy consumption of the rail vehicle on the entire line to be operated
  • the fitness function of the running time includes:
  • T i represents the running time of the rail vehicle in the i-th small section
  • T represents the running time of the rail vehicle on the entire line to be operated.
  • the fitness function of the parking accuracy includes:
  • S a SS i ; where, S a represents the parking accuracy, S i represents the actual parking position of the rail vehicle, and S represents the target parking point;
  • the fitness evaluation of the speed limit includes: screening the set of train maneuvering sequences, and eliminating the train maneuvering sequences that do not meet the requirements of the static speed limit value.
  • a capacity management and energy-saving driving assistance system for rail vehicles comprising:
  • Section dividing module is used for dividing the line to be operated of described rail vehicle into a plurality of sections
  • a variable initialization module configured to initialize the decision variables of each interval
  • the curve solving module is used to solve the automatic driving curve of the rail vehicle by using the basic parameters and the decision variables of each interval based on the multi-objective particle swarm optimization algorithm.
  • the section division module is specifically used to discretize the track to be operated by the rail vehicle into a plurality of small sections of equal length, and record the slope of each small section to obtain a gradient discretization set;
  • the line to be operated of the rail vehicle is discretized into a plurality of small sections of equal length, and the static speed limit value of each small section is recorded to obtain the speed limit Discretized collections.
  • variable initialization module is specifically configured to, according to the slope discretization set and the speed limit discretization set, establish a distance constraint condition, a speed limit constraint condition, and an acceleration constraint condition;
  • the distance constraints include: 0 ⁇ s ⁇ S, wherein s represents the running distance of the rail vehicle at any time during the operation of the small section, and S represents the total length of the small section;
  • the speed limit constraints include: 0 ⁇ v i ⁇ v lim , wherein, v i represents the speed of the rail vehicle at any position in the small section, and v lim represents the static speed limit value of the small section;
  • the acceleration constraints include: a min ⁇ a i ⁇ a max , where a i represents the acceleration of the rail vehicle, and a min and a max represent the minimum braking acceleration and maximum traction acceleration of the train, respectively.
  • the parameter initialization module is specifically used to initialize train parameters, line parameters, operating parameters and algorithm-related parameters;
  • the train parameters include at least: train model, marshalling situation, train quality, train traction and braking coefficient;
  • the line parameters at least include: line length, slope, curvature and speed limit;
  • the operating parameters at least include: train section running time;
  • the algorithm-related parameters at least include: particle dimension, population number and inertia factor.
  • the curve solving module is specifically used to initialize the population of the multi-objective particle swarm optimization algorithm, so as to obtain a set of train maneuvering sequences that meet the requirements of the train maneuvering sequence;
  • train operation control model according to the fitness function of energy consumption, the fitness function of running time, the fitness function of parking accuracy and the fitness evaluation of speed limit, select non-inferior solutions from the set of train control sequences;
  • the train operation control model includes:
  • F( ⁇ ) is the optimization objective function
  • c i represents the decision variable
  • f e (c i ) and f t (c i ) represent the energy consumption and time target variables of the optimization model, respectively
  • st represents the constraints to be followed
  • v 1 represents the speed of the rail vehicle at the initial moment
  • v n represents the speed of the rail vehicle at the stop moment.
  • the energy consumption fitness function includes:
  • E i represents the traction energy consumption of the rail vehicle in the i-th small section
  • E represents the traction energy consumption of the rail vehicle on the entire line to be operated
  • the fitness function of the running time includes:
  • T i represents the running time of the rail vehicle in the i-th small section
  • T represents the running time of the rail vehicle on the entire line to be operated.
  • the fitness function of the parking accuracy includes:
  • S a SS i ; where, S a represents the parking accuracy, S i represents the actual parking position of the rail vehicle, and S represents the target parking point;
  • the fitness evaluation of the speed limit includes: screening the set of train maneuvering sequences, and eliminating the train maneuvering sequences that do not meet the requirements of the static speed limit value.
  • a capacity management and energy-saving driving assistance system for a rail vehicle comprising: a memory and a processor;
  • the memory is used to store program codes
  • the processor is used to call the program codes
  • the program codes are used to execute the method for capacity management and energy-saving assisted driving of rail vehicles described in any one of the above.
  • a storage medium on which a program code is stored, and when the program code is executed, the method for capacity management and energy-saving assisted driving of a rail vehicle described in any one of the above-mentioned methods is realized.
  • the embodiment of the present application provides a rail vehicle capacity management and energy-saving assisted driving method and related devices, wherein the rail vehicle capacity management and energy-saving assisted driving method initializes basic parameters and After dividing the to-be-operated line of the rail vehicle into multiple intervals and initializing the decision variables of each interval, based on the multi-objective particle swarm optimization algorithm, using the basic parameters and the decision variables of each interval to solve the automatic
  • the driving curve realizes the automatic generation of the automatic driving curve of the rail vehicle, improves the automation and intelligence level of the operation control of the rail vehicle, and reduces the labor intensity of the driver of the rail vehicle.
  • FIG. 1 is a schematic flow diagram of a method for capacity management and energy-saving driving assistance of a rail vehicle provided by an embodiment of the present application;
  • FIG. 2 is a schematic diagram of the overall distribution of non-dominated solutions and dominant solutions in the target space provided by an embodiment of the present application;
  • Fig. 3 is a structural schematic diagram of a capacity management and energy-saving driving assistance system of a rail vehicle provided by an embodiment of the present application;
  • FIG. 4 is a schematic structural diagram of a specific application system provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of the appearance of the system shown in FIG. 4 .
  • the embodiment of the present application provides a method for capacity management and energy-saving assisted driving of a rail vehicle, as shown in FIG. 1 , including:
  • S104 Based on the multi-objective particle swarm optimization algorithm, use the basic parameters and the decision variables in each interval to solve the automatic driving curve of the rail vehicle.
  • Multi-objective Optimization Problem can also be called multi-attribute optimization problem or multi-criteria optimization problem.
  • a multi-objective optimization problem includes: n decision variables, m objective functions and k constraints.
  • the mathematical description of MOP is as follows:
  • x represents the decision vector
  • represents the n-dimensional decision space
  • f(x) represents the target vector
  • h(x) represents the constraints.
  • the multi-objective problem description has the following important definitions:
  • h(x) (h 1 (x), h 2 (x), . . . , h k (x)) ⁇ 0 (3);
  • Feasible Solution Set (Feasible Solution Set): In the decision space ⁇ , the set composed of all feasible solutions x is called the feasible solution set, denoted by X ⁇ , X ⁇ X.
  • Equation (6) represents the mapping of Pareto optimal solution set in the target space and the Pareto frontier.
  • the solid line represents the Pareto front
  • the solid circles A and B are on the Pareto front, so they are both optimal solutions and non-dominated solutions
  • the hollow points C, D, and E are in the search space range, but not on the Pareto front, so they are not optimal solutions, and are in the dominated relationship, which is inferior to the solutions on the Pareto front.
  • multi-objective optimization problems are much more complicated, because the latter requires simultaneous optimization of multiple objectives.
  • a multi-objective optimization problem does not have an absolute optimal solution like a single-objective optimization problem, and the solution result of a multi-objective optimization problem is generally a set of Pareto optimal solutions.
  • some appropriate solutions should be selected from the optimal solution set as the optimal solution to the problem in combination with the actual problem and the choice preference of the decision maker. Therefore, for a multi-objective optimization problem, the most important thing in the solution process is to solve as many Pareto optimal solutions with relatively uniform distribution as possible.
  • the multi-objective particle swarm optimization algorithm can not only inherit the advantages of simple and fast convergence of the basic particle swarm optimization algorithm, but also solve the problem that the multi-objective evolutionary algorithm has a slow convergence speed and is easy to fall into a local optimal solution. , so in this embodiment, based on the multi-objective particle swarm optimization algorithm, the automatic driving curve of the rail vehicle is solved by using the basic parameters and the decision variables in each of the intervals.
  • the basic initialization parameters include:
  • the train parameters include at least: train model, marshalling situation, train quality, train traction and braking coefficient;
  • the line parameters at least include: line length, slope, curvature and speed limit;
  • the operating parameters at least include: train section running time;
  • the algorithm-related parameters at least include: particle dimension, population number and inertia factor.
  • Train parameters include many aspects, among which the parameters related to the research object of this paper mainly include: train marshalling mode, car body length, vehicle weight, passenger capacity under capacity and overcrowding, traction and braking characteristics of trains and other parameters. Some parameters are given below.
  • Train models include: trailer TC with driver's cab, trailer TP with pantograph, motor car M at both ends, and motor car M1 in the middle.
  • the weight of these four types of models are: 1M
  • TC vehicle about 33 tons
  • TP car about 33 tons
  • M car about 35 tons
  • M1 car weighs about 35 tons.
  • said dividing the line to be operated of said rail vehicle into a plurality of intervals includes:
  • the line to be operated of the rail vehicle is discretized into a plurality of small sections of equal length, and the static speed limit value of each small section is recorded to obtain the speed limit Discretized collections.
  • the initializing the decision variables of each of the intervals includes:
  • the distance constraints include: 0 ⁇ s ⁇ S, wherein s represents the running distance of the rail vehicle at any time during the operation of the small section, and S represents the total length of the small section;
  • the speed limit constraints include: 0 ⁇ v i ⁇ v lim , wherein, v i represents the speed of the rail vehicle at any position in the small section, and v lim represents the static speed limit value of the small section;
  • the acceleration constraints include: a min ⁇ a i ⁇ a max , where a i represents the acceleration of the rail vehicle, and a min and a max represent the minimum braking acceleration and maximum traction acceleration of the train, respectively.
  • the multi-objective particle swarm optimization algorithm using the basic parameters and the decision variables of each interval to solve the automatic driving curve of the rail vehicle includes:
  • train operation control model according to the fitness function of energy consumption, the fitness function of running time, the fitness function of parking accuracy and the fitness evaluation of speed limit, select non-inferior solutions from the set of train control sequences;
  • the train operation control model includes:
  • F( ⁇ ) is the optimization objective function
  • c i represents the decision variable
  • f e (c i ) and f t (c i ) represent the energy consumption and time target variables of the optimization model, respectively
  • st represents the constraints to be followed
  • v 1 represents the speed of the rail vehicle at the initial moment
  • v n represents the speed of the rail vehicle at the stop moment.
  • the energy consumption fitness function includes:
  • E i represents the traction energy consumption of the rail vehicle in the i-th small section
  • E represents the traction energy consumption of the rail vehicle on the entire line to be operated
  • the fitness function of the running time includes:
  • T i represents the running time of the rail vehicle in the i-th small section
  • T represents the running time of the rail vehicle on the entire line to be operated.
  • the fitness function of the parking accuracy includes:
  • S a SS i ; where, S a represents the parking accuracy, S i represents the actual parking position of the rail vehicle, and S represents the target parking point;
  • the fitness evaluation of the speed limit includes: screening the set of train maneuvering sequences, and eliminating the train maneuvering sequences that do not meet the requirements of the static speed limit value.
  • the following describes the capacity management and energy-saving assisted driving system of rail vehicles provided by the embodiments of the present application.
  • the capacity management and energy-saving assisted driving system of rail vehicles described below can interact with the capacity management and energy-saving assisted driving methods of rail vehicles described above. Corresponding reference.
  • the embodiment of the present application also provides a rail vehicle capacity management and energy-saving assisted driving system.
  • the rail vehicle capacity management and energy-saving assisted driving system includes:
  • Parameter initialization module 100 for initializing basic parameters
  • Section division module 200 for dividing the line to be operated of the rail vehicle into a plurality of sections
  • a variable initialization module 300 configured to initialize the decision variables of each interval
  • the curve solving module 400 is configured to solve the automatic driving curve of the rail vehicle by using the basic parameters and the decision variables in each interval based on the multi-objective particle swarm optimization algorithm.
  • FIG. 4 shows a schematic diagram of a scene when the rail vehicle capacity management and energy-saving driving assistance system is applied to a rail vehicle.
  • HMI Human Machine Interface
  • WTD Wireless Transmission Device
  • CCU Center Control Unit
  • EDAS Electronicgy Drive Advice System
  • HMI, WTD, CCU and EDAS establish a communication connection through an Ethernet switch, and communicate with each other through MVB (Multifunction Vehicle Bus, multifunctional vehicle bus) The communication link is established with other equipment of the rail vehicle.
  • MVB Multifunction Vehicle Bus, multifunctional vehicle bus
  • FIG. 5 shows a schematic diagram of the appearance of the system shown in FIG. 4 .
  • the section division module 200 is specifically configured to discretize the track to be operated by the rail vehicle into a plurality of small sections of equal length, and record the slope of each small section to obtain a gradient discretization set;
  • the line to be operated of the rail vehicle is discretized into a plurality of small sections of equal length, and the static speed limit value of each small section is recorded to obtain the speed limit Discretized collections.
  • variable initialization module 300 is specifically configured to, according to the gradient discretization set and the speed limit discretization set, establish a distance constraint condition, a speed limit constraint condition, and an acceleration constraint condition;
  • the distance constraints include: 0 ⁇ s ⁇ S, wherein s represents the running distance of the rail vehicle at any time during the operation of the small section, and S represents the total length of the small section;
  • the speed limit constraints include: 0 ⁇ v i ⁇ v lim , wherein, v i represents the speed of the rail vehicle at any position in the small section, and v lim represents the static speed limit value of the small section;
  • the acceleration constraints include: a min ⁇ a i ⁇ a max , where a i represents the acceleration of the rail vehicle, and a min and a max represent the minimum braking acceleration and maximum traction acceleration of the train, respectively.
  • the parameter initialization module 100 is specifically used to initialize train parameters, line parameters, operating parameters and algorithm-related parameters;
  • the train parameters include at least: train model, marshalling situation, train quality, train traction and braking coefficient;
  • the line parameters at least include: line length, slope, curvature and speed limit;
  • the operating parameters at least include: train section running time;
  • the algorithm-related parameters at least include: particle dimension, population number and inertia factor.
  • the curve solving module 400 is specifically configured to initialize the population of the multi-objective particle swarm optimization algorithm, so as to obtain a set of train maneuvering sequences that meet the requirements of the train maneuvering sequence;
  • train operation control model according to the fitness function of energy consumption, the fitness function of running time, the fitness function of parking accuracy and the fitness evaluation of speed limit, select non-inferior solutions from the set of train control sequences;
  • the train operation control model includes:
  • F( ) is the optimization objective function
  • c i represents the decision variable
  • f e (c i ) and f t (c i ) represent the energy consumption and time target variables of the optimization model, respectively
  • st represents the constraints to be followed
  • v 1 represents the speed of the rail vehicle at the initial moment
  • v n represents the speed of the rail vehicle at the stop moment.
  • the energy consumption fitness function includes:
  • E i represents the traction energy consumption of the rail vehicle in the i-th small section
  • E represents the traction energy consumption of the rail vehicle on the entire line to be operated
  • the fitness function of the running time includes:
  • T i represents the running time of the rail vehicle in the i-th small section
  • T represents the running time of the rail vehicle on the entire line to be operated.
  • the fitness function of the parking accuracy includes:
  • S a SS i ; where, S a represents the parking accuracy, S i represents the actual parking position of the rail vehicle, and S represents the target parking point;
  • the fitness evaluation of the speed limit includes: screening the set of train maneuvering sequences, and eliminating the train maneuvering sequences that do not meet the requirements of the static speed limit value.
  • the embodiment of the present application also provides a capacity management and energy-saving driving assistance system for rail vehicles, including: a memory and a processor;
  • the memory is used to store program codes
  • the processor is used to call the program codes
  • the program codes are used to execute the method for capacity management and energy-saving assisted driving of rail vehicles described in any one of the above embodiments.
  • the embodiment of the present application also provides a storage medium, on which a program code is stored, and when the program code is executed, the capacity management and energy-saving assisted driving of the rail vehicle described in any of the above-mentioned embodiments are realized. method.
  • the embodiment of the present application provides a rail vehicle capacity management and energy-saving assisted driving method and related devices, wherein the rail vehicle capacity management and energy-saving assisted driving method initializes basic parameters and sets the After the line to be operated of the rail vehicle is divided into multiple intervals and the decision variables of each interval are initialized, based on the multi-objective particle swarm optimization algorithm, the automatic driving curve of the rail vehicle is solved by using the basic parameters and the decision variables of each interval, The automatic generation of rail vehicle automatic driving curves is realized, the automation and intelligence level of rail vehicle operation control is improved, and the labor intensity of rail vehicle drivers is reduced.

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)
  • Train Traffic Observation, Control, And Security (AREA)

Abstract

一种轨道车辆的能力管理及节能辅助驾驶方法,包括初始化基本参数(S101),将轨道车辆的待运行线路划分为多个区间(S102),初始化各个区间的决策变量(S103),基于多目标粒子群算法,利用基本参数和各个区间的决策变量求解轨道车辆的自动驾驶曲线(S104)。实现了轨道车辆自动驾驶曲线的自动生成,提高了对轨道车辆运行控制的自动化和智能化水平,减轻轨道车辆驾驶员劳动强度。同时在多目标粒子群算法中,可将最低牵引能耗和轨道车辆运行时间等多个目标作为求解目标进行求解,可实现提升车辆经济性指标的目的。还公开一种轨道车辆的能力管理及节能辅助驾驶系统、一种存储介质。

Description

一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置
本申请要求于2021年05月13日提交中国专利局、申请号为202110523564.3、发明名称为“一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及轨道车辆技术领域,更具体地说,涉及一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置。
背景技术
目前我国铁路系统中高速和普速列车大多采用由列车驾驶员在车载安全设备监督和防护下驾驶列车的人工驾驶控制模式,随着路网规模扩大、运行间隔缩短、运行速度提高、铁路运能增大,列车能耗问题逐渐凸显,列车驾驶员工作强度日益增加。现行人工驾驶模式难以满足对列车运行控制系统自动化、智能化水平的需求。
发明内容
为解决上述技术问题,本申请提供了一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置,以基于多目标粒子群算法实现自动驾驶曲线的规划功能,提高对轨道车辆运行控制的自动化、智能化水平。
为实现上述技术目的,本申请实施例提供了如下技术方案:
一种轨道车辆的能力管理及节能辅助驾驶方法,包括:
初始化基本参数;
将所述轨道车辆的待运行线路划分为多个区间;
初始化各个所述区间的决策变量;
基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求 解所述轨道车辆的自动驾驶曲线。
可选的,所述将所述轨道车辆的待运行线路划分为多个区间包括:
将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的坡度,以获得坡度离散化合集;
根据所述轨道车辆的待运行线路的静态限速,将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的静态限速值,以获得限速离散化合集。
可选的,所述初始化各个所述区间的决策变量包括:
根据所述坡度离散化合集和所述限速离散化合集,建立距离约束条件、限制速度约束条件和加速度约束条件;
所述距离约束条件包括:0<s<S,其中,s表示所述轨道车辆在小区段中运行过程中任意时刻的运行距离,S表示小区段的总长度;
所述限制速度约束条件包括:0≤v i<v lim,其中,v i表示轨道车辆在小区段中任意一个位置的速度,v lim表示小区段的静态限速值;
所述加速度约束条件包括:a min≤a i≤a max,其中,a i表示轨道车辆在加速度,a min和a max分别表示列车的最小制动加速度和最大牵引加速度。
可选的,所述初始化基本参数包括:
初始化列车参数、线路参数、运行参数和算法相关参数;
其中,所述列车参数至少包括:列车型号、编组情况、列车质量、列车牵引和制动系数;
所述线路参数至少包括:线路长度、坡度、曲度和限速情况;
所述运行参数至少包括:列车区间运行时间;
所述算法相关参数至少包括:粒子维数、种群个数和惯性因子。
可选的,所述基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线包括:
对多目标粒子群算法的种群进行初始化,以获得满足列车操纵序列要求的列车操纵序列集合;
根据列车运行控制模型,依据能耗适应度函数、运行时间的适应度函数、停车精确度的适应度函数以及限速的适应度评价,从所述列车操控序列集合中筛选非劣解;
对筛选获得的非劣解进行个体极值更新和群体极值更新,以获得帕累托最优解;
依据所述帕累托最优解,生成所述轨道车辆的自动驾驶曲线。
可选的,所述列车运行控制模型包括:
Figure PCTCN2021132068-appb-000001
其中,F(·)为优化目标函数,c i表示所述决策变量,f e(c i)和f t(c i)分别表示优化模型的能耗和时间目标变量,s.t.表示遵从的约束条件,v 1表示初始时刻的轨道车辆的速度,v n表示停止时刻的轨道车辆的速度。
可选的,所述能耗适应度函数包括:
Figure PCTCN2021132068-appb-000002
其中,E i表示轨道车辆在第i个小区段的牵引能耗,E表示轨道车辆在整个待运行线路的牵引能耗;
所述运行时间的适应度函数包括:
Figure PCTCN2021132068-appb-000003
其中,T i表示轨道车辆在第i个小区段的运行时间,T表示轨道车辆在整个待运行线路的运行时间。
所述停车精确度的适应度函数包括:
S a=S-S i;其中,S a表示停车精度,S i表示轨道车辆实际停车位置,S表示目标停车点;
所述限速的适应度评价包括:对所述列车操纵序列集合进行筛选,剔除不满足静态限速值要求的列车操纵序列。
一种轨道车辆的能力管理及节能辅助驾驶系统,包括:
参数初始化模块,用于初始化基本参数;
区间划分模块,用于将所述轨道车辆的待运行线路划分为多个区间;
变量初始化模块,用于初始化各个所述区间的决策变量;
曲线求解模块,用于基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线。
可选的,所述区间划分模块具体用于将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的坡度,以获得坡度离散化合集;
根据所述轨道车辆的待运行线路的静态限速,将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的静态限速值,以获得限速离散化合集。
可选的,所述变量初始化模块具体用于,根据所述坡度离散化合集和所述限速离散化合集,建立距离约束条件、限制速度约束条件和加速度约束条件;
所述距离约束条件包括:0<s<S,其中,s表示所述轨道车辆在小区段中运行过程中任意时刻的运行距离,S表示小区段的总长度;
所述限制速度约束条件包括:0≤v i<v lim,其中,v i表示轨道车辆在小区段中任意一个位置的速度,v lim表示小区段的静态限速值;
所述加速度约束条件包括:a min≤a i≤a max,其中,a i表示轨道车辆在加速度,a min和a max分别表示列车的最小制动加速度和最大牵引加速度。
可选的,所述参数初始化模块具体用于,初始化列车参数、线路参数、运行参数和算法相关参数;
其中,所述列车参数至少包括:列车型号、编组情况、列车质量、列车牵引和制动系数;
所述线路参数至少包括:线路长度、坡度、曲度和限速情况;
所述运行参数至少包括:列车区间运行时间;
所述算法相关参数至少包括:粒子维数、种群个数和惯性因子。
可选的,所述曲线求解模块具体用于,对多目标粒子群算法的种群进行初始化,以获得满足列车操纵序列要求的列车操纵序列集合;
根据列车运行控制模型,依据能耗适应度函数、运行时间的适应度函数、停车精确度的适应度函数以及限速的适应度评价,从所述列车操控序列集合中筛选非劣解;
对筛选获得的非劣解进行个体极值更新和群体极值更新,以获得帕累托最优解;
依据所述帕累托最优解,生成所述轨道车辆的自动驾驶曲线。
可选的,所述列车运行控制模型包括:
Figure PCTCN2021132068-appb-000004
其中,F(·)为优化目标函数,c i表示所述决策变量,f e(c i)和f t(c i)分别表示优化模型的能耗和时间目标变量,s.t.表示遵从的约束条件,v 1表示初始时刻的轨道车辆的速度,v n表示停止时刻的轨道车辆的速度。
可选的,所述能耗适应度函数包括:
Figure PCTCN2021132068-appb-000005
其中,E i表示轨道车辆在第i个小区段的牵引能耗,E表示轨道车辆在整个待运行线路的牵引能耗;
所述运行时间的适应度函数包括:
Figure PCTCN2021132068-appb-000006
其中,T i表示轨道车辆在第i个小区段的运行时间,T表示轨道车辆在整个待运行线路的运行时间。
所述停车精确度的适应度函数包括:
S a=S-S i;其中,S a表示停车精度,S i表示轨道车辆实际停车位置,S表示目标停车点;
所述限速的适应度评价包括:对所述列车操纵序列集合进行筛选,剔除不满足静态限速值要求的列车操纵序列。
一种轨道车辆的能力管理及节能辅助驾驶系统,包括:存储器和处理器;
所述存储器用于存储程序代码,所述处理器用于调用所述程序代码,所述程序代码用于执行上述任一项所述的轨道车辆的能力管理及节能辅助驾驶方法。
一种存储介质,所述存储介质上存储有程序代码,所述程序代码被执行时实现上述任一项所述的轨道车辆的能力管理及节能辅助驾驶方法。
从上述技术方案可以看出,本申请实施例提供了一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置,其中,所述轨道车辆的能力管理及节能辅助驾驶方法在初始化基本参数和、将所述轨道车辆的待运行线路划分为多个区间以及初始化各个区间的决策变量后,基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线,实现了轨道车辆自动驾驶曲线的自动生成,提高了对轨道车辆运行控制的自动化和智能化水平,减轻轨道车辆驾驶员劳动强度。
同时在多目标粒子群算法中,可将最低牵引能耗和轨道车辆运行时间等多个目标作为求解目标进行求解,可实现提升车辆经济性指标的目的。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据提供的附图获得其他的附图。
图1为本申请的一个实施例提供的一种轨道车辆的能力管理及节能辅助驾驶方法的流程示意图;
图2为本申请的一个实施例提供的在目标空间中,非支配解和支配解整体的分布情况示意图;
图3为本申请的一个实施例提供的一种轨道车辆的能力管理及节能辅助驾驶系统的结构示意图;
图4为本申请的一个实施例提供的一种具体应用系统的结构示意图;
图5为图4所示系统的外观示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请实施例提供了一种轨道车辆的能力管理及节能辅助驾驶方法,如图1所示,包括:
S101:初始化基本参数;
S102:将所述轨道车辆的待运行线路划分为多个区间;
S103:初始化各个所述区间的决策变量;
S104:基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线。
多目标优化问题(Multi-objective Optimization Problem,MOP)也可以被称作多属性优化问题或者多准则优化问题。在一般情况下,一个多目标优化问题包括:n个决策变量,m个目标函数以及k个约束条件,MOP的数学描述如下:
min f(x)=(f 1(x),f 2(x),…,f m(x)),x∈Ω        (1)
s.t.h(x)=(h 1(x),h 2(x),...,h k(x))≤0       (2);
式(1)中,x表示决策向量,Ω表示n维的决策空间,f(x)表示目标向 量,h(x)代表约束条件。除此之外,多目标问题描述有以下几个重要的定义:
定义1:可行解(Feasible Solution):
h(x)=(h 1(x),h 2(x),...,h k(x))≤0      (3);
称满足式(3)中约束条件的决策变量x∈Ω为可行解。
定义2:可行解集(Feasible Solution Set):在决策空间Ω中,称由所有的可行解x构成的集合为可行解集,用Xρ表示,Xρ∈X。
定义3:帕累托支配(帕累托Dominance):
Figure PCTCN2021132068-appb-000007
对于可行解集Xρ中任一两个可行解x、y,当满足式(4)条件时,成为x支配y,记作
Figure PCTCN2021132068-appb-000008
定义4:帕累托最优解集(帕累托Dominance Set):
Figure PCTCN2021132068-appb-000009
如果满足式(5),即决策向量x不被任何决策空间中的向量支配,那么x是一个帕累托最优解,所有的帕累托最优解的集合共同构成了帕累托最优解集。
定义5:帕累托前沿(或称帕累托前端)(帕累托Front)
ρf *={f(x)|x∈ρ *}      (6);
式(6)表示帕累托最优解集在目标空间中的映射行车了帕累托前沿。
考虑优化两个目标的优化问题,求解目标函数的最小化,参考图2,图2体现了在目标空间中,非支配解和支配解整体的分布情况。
图2中,实线表示的是帕累托前沿,实心圆点A、B在帕累托前沿上,因此都是最优解,是非支配解;空心点C,D,E虽然在搜索空间的范围内,但 是并不位于帕累托前沿上,因此它们不是最优解,处于被支配的关系,是劣于帕累托前沿上的解。
相较于单目标优化问题,多目标优化问题要复杂的多,因为后者需要同时优化多个目标。当改良其中的某一个目标时,很有可能会造成其他的目标变差,因此要考虑各个目标之间的关系并加以权衡。一般情况下,一个多目标的优化问题并不会像单目标优化问题那样存在着一个绝对的最优解,多目标优化问题的求解结果一般是一组帕累托最优解的集合。解决实际问题时,应当结合实际问题的情况以及决策者的选择偏好,在最优解集中选择一些合适的解作为求解的问题最优解。因此对于一个多目标寻优问题来说,在求解过程中最重要的是解出尽可能多的,分布比较均匀的帕累托最优解。
与基本粒子群算法相比,多目标粒子群算法不仅能够很好的继承基本粒子群算法简单、快速收敛的优点,而且还解决了多目标进化算法收敛速度慢,易于陷入局部最优解的问题,因此在本实施例中,基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线。
下面对本申请实施例提供的轨道车辆的能力管理及节能辅助驾驶方法的各个步骤的具体可行执行方式进行描述。
可选的,在本申请的一个实施例中,所述初始化基本参数包括:
初始化列车参数、线路参数、运行参数和算法相关参数;
其中,所述列车参数至少包括:列车型号、编组情况、列车质量、列车牵引和制动系数;
所述线路参数至少包括:线路长度、坡度、曲度和限速情况;
所述运行参数至少包括:列车区间运行时间;
所述算法相关参数至少包括:粒子维数、种群个数和惯性因子。
列车参数描述:
列车参数包括很多方面,其中与本文研究对象相关的参数主要包括:列车编组方式、车体长度、车辆自重、定员和超员情况下的载客量以及列车的牵引和制动特性等参数。下面将给出部分参数。
1)列车车型包括:带司机室的拖车TC、带受电弓拖车TP、两端动车M、中间动车M1。这四类车型自重分别为:1M
TC车:约重33吨;
TP车:约重33吨;
M车:约重35吨;
M1车:约重35吨。
2)乘客人均重量按照60Kg/人计算。
3)列车的最大加速度为;1m/s 2
4)列车的最大减速度为;1m/s 2
其他列车参数在具体的案例分析中将结合具体的情况给出。
可选的,所述将所述轨道车辆的待运行线路划分为多个区间包括:
将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的坡度,以获得坡度离散化合集;
根据所述轨道车辆的待运行线路的静态限速,将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的静态限速值,以获得限速 离散化合集。
在进行运行线路设计和施工中,为了满足规划等需要,线路情况往往比较复杂,就某一区间而言,可能存在多个坡度不同的坡道,曲率不同的弯道等情况。同时,受到环境、土建条件等限制,线路存在一定的静态限速。求解对象受到线路参数的影响,为方便求解,先将部分参数进行离散化处理。
1)线路坡度离散化:将某一区间作为研究对象,区间长度是固定的,根据线路数据可知整个区间包括多个坡度不同的坡道,坡道长度和坡度已知。在进行离散化处理时,将整个轨道区段划分为n个等长的的小区段,则坡度离散化可表示为合集G={g 1,g 2,...,g n};
2)静态限速离散化:对于一个固定的列车运行区间,由于土建条件等因素的限制,线路存在着一定的静态限速,列车在运行过程中一定不能超过该静态限速的值,线路不同位置对应的静态限速值已知,同样将线路划分为n个等长的小区段,则静态限速离散化可表示为合集V={v 1,v 2,...,v m}。
可选的,所述初始化各个所述区间的决策变量包括:
根据所述坡度离散化合集和所述限速离散化合集,建立距离约束条件、限制速度约束条件和加速度约束条件;
所述距离约束条件包括:0<s<S,其中,s表示所述轨道车辆在小区段中运行过程中任意时刻的运行距离,S表示小区段的总长度;
所述限制速度约束条件包括:0≤v i<v lim,其中,v i表示轨道车辆在小区段中任意一个位置的速度,v lim表示小区段的静态限速值;
所述加速度约束条件包括:a min≤a i≤a max,其中,a i表示轨道车辆在加速度,a min和a max分别表示列车的最小制动加速度和最大牵引加速度。
可选的,所述基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线包括:
对多目标粒子群算法的种群进行初始化,以获得满足列车操纵序列要求的列车操纵序列集合;
根据列车运行控制模型,依据能耗适应度函数、运行时间的适应度函数、停车精确度的适应度函数以及限速的适应度评价,从所述列车操控序列集合中筛选非劣解;
对筛选获得的非劣解进行个体极值更新和群体极值更新,以获得帕累托最优解;
依据所述帕累托最优解,生成所述轨道车辆的自动驾驶曲线。
其中,所述列车运行控制模型包括:
Figure PCTCN2021132068-appb-000010
其中,F(·)为优化目标函数,c i表示所述决策变量,f e(c i)和f t(c i)分别表示优化模型的能耗和时间目标变量,s.t.表示遵从的约束条件,v 1表示初始时刻的轨道车辆的速度,v n表示停止时刻的轨道车辆的速度。
所述能耗适应度函数包括:
Figure PCTCN2021132068-appb-000011
其中,E i表示轨道车辆在第i个小区段的牵引能耗,E表示轨道车辆在整个待运行线路的牵引能耗;
在能耗适应度评价时,将轨道车辆在整个待运行线路的牵引能耗作为代表 列车牵引能耗的评价指标,即f e=E。
所述运行时间的适应度函数包括:
Figure PCTCN2021132068-appb-000012
其中,T i表示轨道车辆在第i个小区段的运行时间,T表示轨道车辆在整个待运行线路的运行时间。
在运行时间的适应度评价时,基于函数f t=T 0-T进行评价,其中,T 0表示轨道车辆时刻表运行时间,f t表示轨道车辆在区间运行的准点率评价指标。
所述停车精确度的适应度函数包括:
S a=S-S i;其中,S a表示停车精度,S i表示轨道车辆实际停车位置,S表示目标停车点;
所述限速的适应度评价包括:对所述列车操纵序列集合进行筛选,剔除不满足静态限速值要求的列车操纵序列。
下面对本申请实施例提供的轨道车辆的能力管理及节能辅助驾驶系统进行描述,下文描述的轨道车辆的能力管理及节能辅助驾驶系统可与上文描述的轨道车辆的能力管理及节能辅助驾驶方法相互对应参照。
相应的,本申请实施例还提供了一种轨道车辆的能力管理及节能辅助驾驶系统,如图3所示,所述轨道车辆的能力管理及节能辅助驾驶系统包括:
参数初始化模块100,用于初始化基本参数;
区间划分模块200,用于将所述轨道车辆的待运行线路划分为多个区间;
变量初始化模块300,用于初始化各个所述区间的决策变量;
曲线求解模块400,用于基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线。
参考图4,图4示出了轨道车辆的能力管理及节能辅助驾驶系统应用于轨道车辆时的场景示意图。
在图4中,HMI(Human Machine Interface)表示人机交互单元,WTD(Wireless Transmission Device)表示车载信息无线传输设备,CCU(Center Control Unit)表示中央控制单元,EDAS(Energy Drive Advice System)主机即为集成有本申请实施例提供的轨道车辆的能力管理及节能辅助驾驶系统的设备,HMI、WTD、CCU和EDAS通过以太网交换机建立通信连接,通过MVB(Multifunction Vehicle Bus,多功能车辆总线)与轨道车辆的其他设备建立通信连接。
参考图5,图5示出了图4所示的系统的外观示意图。
可选的,所述区间划分模块200具体用于将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的坡度,以获得坡度离散化合集;
根据所述轨道车辆的待运行线路的静态限速,将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的静态限速值,以获得限速离散化合集。
可选的,所述变量初始化模块300具体用于,根据所述坡度离散化合集和所述限速离散化合集,建立距离约束条件、限制速度约束条件和加速度约束条件;
所述距离约束条件包括:0<s<S,其中,s表示所述轨道车辆在小区段中运行过程中任意时刻的运行距离,S表示小区段的总长度;
所述限制速度约束条件包括:0≤v i<v lim,其中,v i表示轨道车辆在小区段中任意一个位置的速度,v lim表示小区段的静态限速值;
所述加速度约束条件包括:a min≤a i≤a max,其中,a i表示轨道车辆在加速度,a min和a max分别表示列车的最小制动加速度和最大牵引加速度。
可选的,所述参数初始化模块100具体用于,初始化列车参数、线路参数、运行参数和算法相关参数;
其中,所述列车参数至少包括:列车型号、编组情况、列车质量、列车牵引和制动系数;
所述线路参数至少包括:线路长度、坡度、曲度和限速情况;
所述运行参数至少包括:列车区间运行时间;
所述算法相关参数至少包括:粒子维数、种群个数和惯性因子。
可选的,所述曲线求解模块400具体用于,对多目标粒子群算法的种群进行初始化,以获得满足列车操纵序列要求的列车操纵序列集合;
根据列车运行控制模型,依据能耗适应度函数、运行时间的适应度函数、停车精确度的适应度函数以及限速的适应度评价,从所述列车操控序列集合中筛选非劣解;
对筛选获得的非劣解进行个体极值更新和群体极值更新,以获得帕累托最优解;
依据所述帕累托最优解,生成所述轨道车辆的自动驾驶曲线。
可选的,所述列车运行控制模型包括:
Figure PCTCN2021132068-appb-000013
其中,F(·)为优化目标函数,c i表示所述决策变量,f e(c i)和f t(c i)分别表示优化模型的能耗和时间目标变量,s.t.表示遵从的约束条件,v 1表示初始时刻的轨道车辆的速度,v n表示停止时刻的轨道车辆的速度。
可选的,所述能耗适应度函数包括:
Figure PCTCN2021132068-appb-000014
其中,E i表示轨道车辆在第i个小区段的牵引能耗,E表示轨道车辆在整个待运行线路的牵引能耗;
所述运行时间的适应度函数包括:
Figure PCTCN2021132068-appb-000015
其中,T i表示轨道车辆在第i个小区段的运行时间,T表示轨道车辆在整个待运行线路的运行时间。
所述停车精确度的适应度函数包括:
S a=S-S i;其中,S a表示停车精度,S i表示轨道车辆实际停车位置,S表示目标停车点;
所述限速的适应度评价包括:对所述列车操纵序列集合进行筛选,剔除不满足静态限速值要求的列车操纵序列。
相应的,本申请实施例还提供了一种轨道车辆的能力管理及节能辅助驾驶系统,包括:存储器和处理器;
所述存储器用于存储程序代码,所述处理器用于调用所述程序代码,所述程序代码用于执行上述任一实施例所述的轨道车辆的能力管理及节能辅助驾驶方法。
相应的,本申请实施例还提供了一种存储介质,所述存储介质上存储有程序代码,所述程序代码被执行时实现上述任一实施例所述的轨道车辆的能力管理及节能辅助驾驶方法。
综上所述,本申请实施例提供了一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置,其中,所述轨道车辆的能力管理及节能辅助驾驶方法在初始化基本参数和、将所述轨道车辆的待运行线路划分为多个区间以及初始化各个区间的决策变量后,基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线,实现了轨道车辆自动驾驶曲线的自动生成,提高了对轨道车辆运行控制的自动化和智能化水平,减轻轨道车辆驾驶员劳动强度。
同时在多目标粒子群算法中,可将最低牵引能耗和轨道车辆运行时间等多个目标作为求解目标进行求解,可实现提升车辆经济性指标的目的。
本说明书中各实施例中记载的特征可以相互替换或者组合,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似部分互相参 见即可。
对所公开的实施例的上述说明,使本领域专业技术人员能够实现或使用本申请。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的,本文中所定义的一般原理可以在不脱离本申请的精神或范围的情况下,在其它实施例中实现。因此,本申请将不会被限制于本文所示的这些实施例,而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (16)

  1. 一种轨道车辆的能力管理及节能辅助驾驶方法,其特征在于,包括:
    初始化基本参数;
    将所述轨道车辆的待运行线路划分为多个区间;
    初始化各个所述区间的决策变量;
    基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线。
  2. 根据权利要求1所述的方法,其特征在于,所述将所述轨道车辆的待运行线路划分为多个区间包括:
    将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的坡度,以获得坡度离散化合集;
    根据所述轨道车辆的待运行线路的静态限速,将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的静态限速值,以获得限速离散化合集。
  3. 根据权利要求2所述的方法,其特征在于,所述初始化各个所述区间的决策变量包括:
    根据所述坡度离散化合集和所述限速离散化合集,建立距离约束条件、限制速度约束条件和加速度约束条件;
    所述距离约束条件包括:0<s<S,其中,s表示所述轨道车辆在小区段中运行过程中任意时刻的运行距离,S表示小区段的总长度;
    所述限制速度约束条件包括:0≤v i<v lim,其中,v i表示轨道车辆在小区段中任意一个位置的速度,v lim表示小区段的静态限速值;
    所述加速度约束条件包括:a min≤a i≤a max,其中,a i表示轨道车辆在加速度,a min和a max分别表示列车的最小制动加速度和最大牵引加速度。
  4. 根据权利要求3所述的方法,其特征在于,所述初始化基本参数包括:
    初始化列车参数、线路参数、运行参数和算法相关参数;
    其中,所述列车参数至少包括:列车型号、编组情况、列车质量、列车牵引和制动系数;
    所述线路参数至少包括:线路长度、坡度、曲度和限速情况;
    所述运行参数至少包括:列车区间运行时间;
    所述算法相关参数至少包括:粒子维数、种群个数和惯性因子。
  5. 根据权利要求4所述的方法,其特征在于,所述基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线包括:
    对多目标粒子群算法的种群进行初始化,以获得满足列车操纵序列要求的列车操纵序列集合;
    根据列车运行控制模型,依据能耗适应度函数、运行时间的适应度函数、停车精确度的适应度函数以及限速的适应度评价,从所述列车操控序列集合中筛选非劣解;
    对筛选获得的非劣解进行个体极值更新和群体极值更新,以获得帕累托最优解;
    依据所述帕累托最优解,生成所述轨道车辆的自动驾驶曲线。
  6. 根据权利要求5所述的方法,其特征在于,所述列车运行控制模型包括:
    Figure PCTCN2021132068-appb-100001
    其中,F(·)为优化目标函数,c i表示所述决策变量,f e(c i)和f t(c i)分别表示优化模型的能耗和时间目标变量,s.t.表示遵从的约束条件,v 1表示初始时刻的轨道车辆的速度,v n表示停止时刻的轨道车辆的速度。
  7. 根据权利要求5所述的方法,其特征在于,所述能耗适应度函数包括:
    Figure PCTCN2021132068-appb-100002
    其中,E i表示轨道车辆在第i个小区段的牵引能耗,E表示轨道车辆在整个待运行线路的牵引能耗;
    所述运行时间的适应度函数包括:
    Figure PCTCN2021132068-appb-100003
    其中,T i表示轨道车辆在第i个小区段的运行时间,T表示轨道车辆在整个待运行线路的运行时间;
    所述停车精确度的适应度函数包括:
    S a=S-S i;其中,S a表示停车精度,S i表示轨道车辆实际停车位置,S表示目标停车点;
    所述限速的适应度评价包括:对所述列车操纵序列集合进行筛选,剔除不满足静态限速值要求的列车操纵序列。
  8. 一种轨道车辆的能力管理及节能辅助驾驶系统,其特征在于,包括:
    参数初始化模块,用于初始化基本参数;
    区间划分模块,用于将所述轨道车辆的待运行线路划分为多个区间;
    变量初始化模块,用于初始化各个所述区间的决策变量;
    曲线求解模块,用于基于多目标粒子群算法,利用所述基本参数和各个所述区间的决策变量求解所述轨道车辆的自动驾驶曲线。
  9. 根据权利要求8所述的系统,其特征在于,所述区间划分模块具体用于将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的坡度,以获得坡度离散化合集;
    根据所述轨道车辆的待运行线路的静态限速,将所述轨道车辆的待运行线路离散化为多个等长的小区段,并记录每个小区段的静态限速值,以获得限速离散化合集。
  10. 根据权利要求9所述的系统,其特征在于,所述变量初始化模块具体用于,根据所述坡度离散化合集和所述限速离散化合集,建立距离约束条件、限制速度约束条件和加速度约束条件;
    所述距离约束条件包括:0<s<S,其中,s表示所述轨道车辆在小区段中运行过程中任意时刻的运行距离,S表示小区段的总长度;
    所述限制速度约束条件包括:0≤v i<v lim,其中,v i表示轨道车辆在小区段中任意一个位置的速度,v lim表示小区段的静态限速值;
    所述加速度约束条件包括:a min≤a i≤a max,其中,a i表示轨道车辆在加速度,a min和a max分别表示列车的最小制动加速度和最大牵引加速度。
  11. 根据权利要求10所述的系统,其特征在于,所述参数初始化模块具体用于,初始化列车参数、线路参数、运行参数和算法相关参数;
    其中,所述列车参数至少包括:列车型号、编组情况、列车质量、列车牵引和制动系数;
    所述线路参数至少包括:线路长度、坡度、曲度和限速情况;
    所述运行参数至少包括:列车区间运行时间;
    所述算法相关参数至少包括:粒子维数、种群个数和惯性因子。
  12. 根据权利要求11所述的系统,其特征在于,所述曲线求解模块具体用于,对多目标粒子群算法的种群进行初始化,以获得满足列车操纵序列要求的列车操纵序列集合;
    根据列车运行控制模型,依据能耗适应度函数、运行时间的适应度函数、停车精确度的适应度函数以及限速的适应度评价,从所述列车操控序列集合中筛选非劣解;
    对筛选获得的非劣解进行个体极值更新和群体极值更新,以获得帕累托最优解;
    依据所述帕累托最优解,生成所述轨道车辆的自动驾驶曲线。
  13. 根据权利要求12所述的系统,其特征在于,所述列车运行控制模型包括:
    Figure PCTCN2021132068-appb-100004
    其中,F(·)为优化目标函数,c i表示所述决策变量,f e(c i)和f t(c i)分别表示优化模型的能耗和时间目标变量,s.t.表示遵从的约束条件,v 1表示初始时刻的轨道车辆的速度,v n表示停止时刻的轨道车辆的速度。
  14. 根据权利要求12所述的系统,其特征在于,所述能耗适应度函数包括:
    Figure PCTCN2021132068-appb-100005
    其中,E i表示轨道车辆在第i个小区段的牵引能耗,E表示轨道车辆在整个待运行线路的牵引能耗;
    所述运行时间的适应度函数包括:
    Figure PCTCN2021132068-appb-100006
    其中,T i表示轨道车辆在第i个小区段的运行时间,T表示轨道车辆在整个待运行线路的运行时间;
    所述停车精确度的适应度函数包括:
    S a=S-S i;其中,S a表示停车精度,S i表示轨道车辆实际停车位置,S表示目标停车点;
    所述限速的适应度评价包括:对所述列车操纵序列集合进行筛选,剔除不满足静态限速值要求的列车操纵序列。
  15. 一种轨道车辆的能力管理及节能辅助驾驶系统,其特征在于,包括:存储器和处理器;
    所述存储器用于存储程序代码,所述处理器用于调用所述程序代码,所述程序代码用于执行权利要求1-7任一项所述的轨道车辆的能力管理及节能辅助驾驶方法。
  16. 一种存储介质,其特征在于,所述存储介质上存储有程序代码,所述程序代码被执行时实现权利要求1-7任一项所述的轨道车辆的能力管理及节能辅助驾驶方法。
PCT/CN2021/132068 2021-05-13 2021-11-22 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置 WO2022237115A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110523564.3 2021-05-13
CN202110523564.3A CN113147841A (zh) 2021-05-13 2021-05-13 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置

Publications (1)

Publication Number Publication Date
WO2022237115A1 true WO2022237115A1 (zh) 2022-11-17

Family

ID=76875274

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/132068 WO2022237115A1 (zh) 2021-05-13 2021-11-22 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置

Country Status (2)

Country Link
CN (1) CN113147841A (zh)
WO (1) WO2022237115A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113147841A (zh) * 2021-05-13 2021-07-23 中车长春轨道客车股份有限公司 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置
CN113997915B (zh) * 2021-11-26 2022-06-24 北京大象科技有限公司 基于大数据的列车自动驾驶ato精确停车控制方法
CN114379617A (zh) * 2022-02-23 2022-04-22 苏州知时节交通科技有限公司 一种列车节能控制方法

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104766484A (zh) * 2015-03-23 2015-07-08 南京邮电大学 基于进化多目标优化和蚁群算法的交通控制与诱导系统和方法
CN105930937A (zh) * 2016-05-07 2016-09-07 浙江大学 一种考虑地铁快慢线的列车运行调度一体化优化方法
US20180196436A1 (en) * 2017-01-10 2018-07-12 GM Global Technology Operations LLC Methods and apparatus for optimizing a trajectory for an autonomous vehicle
CN109625028A (zh) * 2018-10-26 2019-04-16 北京全路通信信号研究设计院集团有限公司 一种列车进站或出站时空路径规划方法及装置
US20190193765A1 (en) * 2016-06-13 2019-06-27 Siemens Industry, Inc. Method and system for train route optimization
CN110187639A (zh) * 2019-06-27 2019-08-30 吉林大学 一种基于参数决策框架的轨迹规划控制方法
CN110428161A (zh) * 2019-07-25 2019-11-08 北京航空航天大学 一种基于端边云架构的无人驾驶矿车云智能调度方法
CN110568760A (zh) * 2019-10-08 2019-12-13 吉林大学 适用于换道及车道保持的参数化学习决策控制系统及方法
CN110749333A (zh) * 2019-11-07 2020-02-04 中南大学 基于多目标优化的无人驾驶车辆运动规划方法
CN113147841A (zh) * 2021-05-13 2021-07-23 中车长春轨道客车股份有限公司 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106777752B (zh) * 2016-12-30 2019-04-02 华东交通大学 一种高速列车追踪运行曲线优化设定方法
CN110539780B (zh) * 2018-11-27 2022-02-11 中车长春轨道客车股份有限公司 列车行驶提示方法及装置
CN109591861B (zh) * 2018-11-29 2020-05-29 中车长春轨道客车股份有限公司 一种实现智能定速和节能驾驶的方法及装置
CN110990950B (zh) * 2019-12-04 2021-12-21 北京交通大学 基于混合运行模式的多目标列车节能优化方法
CN111976693A (zh) * 2020-07-28 2020-11-24 中车长春轨道客车股份有限公司 一种高速轨道列车制动力优化分配方法

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104766484A (zh) * 2015-03-23 2015-07-08 南京邮电大学 基于进化多目标优化和蚁群算法的交通控制与诱导系统和方法
CN105930937A (zh) * 2016-05-07 2016-09-07 浙江大学 一种考虑地铁快慢线的列车运行调度一体化优化方法
US20190193765A1 (en) * 2016-06-13 2019-06-27 Siemens Industry, Inc. Method and system for train route optimization
US20180196436A1 (en) * 2017-01-10 2018-07-12 GM Global Technology Operations LLC Methods and apparatus for optimizing a trajectory for an autonomous vehicle
CN109625028A (zh) * 2018-10-26 2019-04-16 北京全路通信信号研究设计院集团有限公司 一种列车进站或出站时空路径规划方法及装置
CN110187639A (zh) * 2019-06-27 2019-08-30 吉林大学 一种基于参数决策框架的轨迹规划控制方法
CN110428161A (zh) * 2019-07-25 2019-11-08 北京航空航天大学 一种基于端边云架构的无人驾驶矿车云智能调度方法
CN110568760A (zh) * 2019-10-08 2019-12-13 吉林大学 适用于换道及车道保持的参数化学习决策控制系统及方法
CN110749333A (zh) * 2019-11-07 2020-02-04 中南大学 基于多目标优化的无人驾驶车辆运动规划方法
CN113147841A (zh) * 2021-05-13 2021-07-23 中车长春轨道客车股份有限公司 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置

Also Published As

Publication number Publication date
CN113147841A (zh) 2021-07-23

Similar Documents

Publication Publication Date Title
WO2022237115A1 (zh) 一种轨道车辆的能力管理及节能辅助驾驶方法及相关装置
WO2021136130A1 (zh) 一种轨迹规划方法及装置
CN110450825B (zh) 城市轨道交通列车运行控制方法
Li et al. Coordinated cruise control for high-speed train movements based on a multi-agent model
WO2021114742A1 (zh) 一种混合动力电动汽车综合预测能量管理方法
Nie et al. Real-time dynamic predictive cruise control for enhancing eco-driving of electric vehicles, considering traffic constraints and signal phase and timing (SPaT) information, using artificial-neural-network-based energy consumption model
Wang et al. A survey on optimal trajectory planning for train operations
CN114241778B (zh) 高速公路网联车协同匝道汇入多目标优化控制方法和系统
Yin et al. Smart train operation algorithms based on expert knowledge and ensemble CART for the electric locomotive
CN109978350A (zh) 一种基于工况分解动态规划算法的地铁列车节能优化方法
CN111619624B (zh) 一种基于深度强化学习的有轨电车运行控制方法和系统
CN112307564A (zh) 列车ato目标运行速度曲线优化方法及装置
CN113788021B (zh) 一种结合前车速度预测的自适应跟车巡航控制方法
CN106056238B (zh) 列车区间运行轨迹的规划方法
CN111959492B (zh) 一种网联环境下考虑换道行为的hev能量管理分层控制方法
CN112686453B (zh) 机车能耗智能预测方法及系统
Bai et al. Cooperative control of metro trains to minimize net energy consumption
CN112660130A (zh) 基于智能网联信息的新能源汽车滑行控制系统、方法及新能源汽车
CN108883784A (zh) 向列车驾驶员提供驾驶建议的方法和列车驾驶员咨询系统
CN110304066A (zh) 自动驾驶模式下的路线选择方法、系统、设备及存储介质
CN110532846A (zh) 自动换道方法、装置及存储介质
CN113487857B (zh) 一种区域多路口可变车道协同控制决策方法
CN113867354A (zh) 一种自动驾驶多车智能协同的区域交通流导引方法
Meng et al. Combining the matter-element model with the associated function of performance indices for automatic train operation algorithm
CN114074680B (zh) 基于深度强化学习的车辆换道行为决策方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21941679

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21941679

Country of ref document: EP

Kind code of ref document: A1