CN116227699A - High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction - Google Patents

High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction Download PDF

Info

Publication number
CN116227699A
CN116227699A CN202310183614.7A CN202310183614A CN116227699A CN 116227699 A CN116227699 A CN 116227699A CN 202310183614 A CN202310183614 A CN 202310183614A CN 116227699 A CN116227699 A CN 116227699A
Authority
CN
China
Prior art keywords
energy
train
traction load
load prediction
traction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310183614.7A
Other languages
Chinese (zh)
Inventor
李欣
朱成琨
黄文勋
李若琼
马学东
郑鑫波
魏春宇
刘顺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Lanzhou Jiaotong University
Original Assignee
Lanzhou Jiaotong University
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 Lanzhou Jiaotong University filed Critical Lanzhou Jiaotong University
Priority to CN202310183614.7A priority Critical patent/CN116227699A/en
Publication of CN116227699A publication Critical patent/CN116227699A/en
Pending legal-status Critical Current

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
    • B61L27/10—Operations, e.g. scheduling or time tables
    • B61L27/16—Trackside optimisation of vehicle or train operation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Evolutionary Computation (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Development Economics (AREA)
  • Mechanical Engineering (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Operations Research (AREA)
  • Medical Informatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Train Traffic Observation, Control, And Security (AREA)

Abstract

The invention provides a high-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction, and relates to the technical field of train operation optimization. The energy-saving operation optimizing method for the train on the long and large ramp of the high-speed rail based on traction load prediction comprises the following steps: establishing a traction load prediction model to obtain traction load prediction data; and optimizing the energy-saving operation of the train according to the traction load prediction data. By optimizing the energy-saving operation of the high-speed rail long and large ramp train based on traction load prediction, the utilization of regenerative braking energy generated by the braking working condition train on the long ramp in the system is promoted on the basis of traction load prediction, the traction energy obtained by the traction working condition train on the long ramp from an external power grid is reduced, the operation cost of the high-speed rail is reduced, and the energy-saving operation of the system is realized.

Description

High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction
Technical Field
The invention relates to the technical field of train operation optimization, in particular to an energy-saving operation optimization method for a high-speed rail long and large ramp train with predicted traction load.
Background
Because the topography of China is high and low in east and west, the railway lines under construction in the middle and west areas represented by Sichuan Tibetan, cheng lan, yukun, xiyan and the like generally have the characteristic of dense long and large slopes, and when a train passes through the long and large slopes in the uphill direction, high enough traction power is required, and the traction energy consumption is greatly increased at the moment. When the train passes through the long ramp in the downhill direction, electric braking is continuously applied for a long time, and abundant regenerative braking energy is generated. The regenerative braking energy generated by braking the train on the long and large downhill slope can be used for supplementing the traction energy requirement of the train on the long and large uphill slope, and external power supply is saved.
Therefore, by predicting the traction load of the high-speed railway long and large ramp line, the behavior characteristics and the energy consumption requirements of the traction load in the power supply interval are known, references are provided for the establishment of a power consumption plan and the optimization of driving organizations of railway departments, the utilization of the regenerative braking energy generated by the train on the long and large ramp line in the traction power supply system is promoted, the energy-saving operation is realized, and the operation cost of the high-speed railway is reduced.
Disclosure of Invention
The invention aims to provide an energy-saving operation optimization method for a train on a high-speed rail and a long-slope road based on traction load prediction, so as to solve the problems in the prior art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the energy-saving operation optimizing method for the high-speed rail long and large ramp train based on traction load prediction is characterized by comprising the following steps of:
step one, a traction load prediction model is established, and traction load prediction data are obtained;
and step two, optimizing the energy-saving operation of the train according to the traction load prediction data.
Further, the first step is to build a traction load prediction model based on CNN-LSTM-Attention, and the input of the model is as follows: the model structure mainly comprises six layers, namely an input layer, a CNN layer, an LSTM layer, a full-connection layer, an Attention layer and an output layer.
Further, the train energy-saving operation optimization of the second step comprises single train energy-saving operation and multi-train operation matching.
Further, the single train energy-saving operation step comprises the steps of dividing an operation interval, generating an ideal operation strategy and distributing traction energy, and the multi-train operation matching step comprises the steps of determining the grade crossing train number, transmitting regenerated braking energy and overlapping braking/traction working conditions.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the invention, a CNN-LSTM-Attention traction load prediction model is established by utilizing traction load historical data and considering traction load influence factors so as to acquire high-speed rail long and large ramp traction load data; according to the traction load prediction data, a train energy-saving operation optimization scheme for coordinating and considering the saving of traction energy and peak clipping of traction load is provided, and the scheme is divided into two types of single train energy-saving operation and multi-train operation matching. By optimizing the energy-saving operation of the high-speed rail long and large ramp train based on traction load prediction, a solution is provided for solving the problem that the existing high-speed rail long and large ramp train is high in traction energy consumption and the regenerated braking energy is not fully utilized, on the basis of traction load prediction, traction energy saving and traction load peak clipping are cooperatively considered, a train energy-saving operation optimizing scheme is provided, utilization of the regenerated braking energy generated by the ramp braking working condition train in the system is promoted, traction energy consumption obtained by the ramp traction working condition train from an external power grid is reduced, and high-speed railway operation cost is reduced.
Drawings
FIG. 1 is a schematic diagram of the steps for implementing the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, an embodiment of the present invention:
the embodiment of the invention provides a high-speed rail long and large ramp train operation optimization method based on traction load prediction, which comprises the following steps of:
step one, a traction load prediction model is established, and traction load prediction data are obtained;
because the traction load data is not only related to the train running process, but also influenced by factors such as line conditions, train parameters, operation organizations and the like, the input of the prediction model is a multivariate time sequence of traction load historical data and traction load influencing factors, and the input variables have correlation. The characteristic information of traction load historical data and traction load influence factors is automatically extracted by utilizing a Convolutional Neural Network (CNN), then the time dependence among the extracted characteristic information is captured by utilizing a long-short-term memory (LSTM) network, the characteristic information is processed into a one-dimensional traction load characteristic vector through a full-connection layer, the one-dimensional traction load characteristic vector is input into an Attention (Attention) model for training, and corresponding weights are distributed for different input characteristics, so that high-precision traction load prediction is realized.
The input of the traction load prediction model based on the CNN-LSTM-attribute is as follows: traction load historical data, line condition influence factors, train parameter influence factors and operation organization influence factors are connected in series to form a vector representation. The inputs at every 4 moments constitute a characteristic vector of the unit traction load. And constructing a training sample of the traction load prediction model by utilizing the sliding window. The time step of the sliding window is set to 1, the sliding window slides along the time axis by the size of 4*n, and 1 training sample of the traction load prediction model can be generated every 1 time step.
The model structure is mainly divided into six layers, and is specifically as follows:
1. an input layer for inputting traction load data x at time t s Line condition influencing factor l s Train parameter influencing factor p s Operation organization influence factor r s The data are serially connected into a vector Y t . The step length of the sliding window is set to be 1, and the new traction load and time sequence data of the influencing variable can be obtained after the sliding window slides once, and the new traction load and the time sequence data of the influencing variable are standardized and then used as the input of the CNN-LSTM-Attention model.
And 2, a CNN layer, wherein the CNN layer performs feature extraction on the input train loading data and the time sequence data of the influencing variables. The relative position of the train at 4 moments and the time sequence of the influencing variables form a traction load characteristic matrix, and the size of the traction load characteristic matrix is 4*4. This layer consists of 2 convolutional layers and 2 pooling layers. The number of convolution kernels used by each convolution layer is doubled layer by layer to increase the network depth, and the pooling layer selects the maximum pooling mode. The reLU function is selected as the activation function for the convolutional layer. The CNN layer automatically extracts the input traction load and the characteristics of time sequence data affecting the variables through convolution kernel operation.
3. The LSTM layer has strong capability of learning long-time sequences and has the capability of automatically encoding and selecting important information. And selecting a double-layer LSTM network, taking the feature vector extracted by the CNN layer as input, learning the internal change rule of the feature vector, capturing the time dependency between data, and carrying out time sequence modeling. The dual-layer LSTM network also helps reduce model training time.
4. And the full connection layer integrates the traction load feature matrix to obtain a one-dimensional traction load feature vector, and realizes the transformation of data dimension, so that the features extracted by the LSTM layer are completely transferred into the Attention layer.
5. And the Attention layer is used for carrying out Attention training on the traction load characteristic vector to obtain the factor weight for influencing the traction load prediction, and reducing the factor weight for influencing the traction load with weak relevance with the real load so as to realize the traction load prediction with higher precision.
6. And (3) outputting a layer, obtaining a traction load predicted value of t+1, and obtaining a final traction load predicted result.
Step two, optimizing energy-saving operation of the train according to traction load prediction data;
according to traction load prediction data based on CNN-LSTM-Attention, the regenerative braking energy is utilized to supplement traction energy consumption, energy-saving operation is realized, meanwhile, the traction load peak clipping effect is improved, and a train energy-saving operation optimization scheme for coordinating and considering traction energy saving and traction load peak clipping is provided and is divided into two types of single train energy-saving operation and multi-train operation matching:
1. single train energy saving operation
The traction load result based on the CNN-LSTM-attribute is used as a data base, and the aim of saving the traction energy consumption of the train is achieved by distributing the minimum energy consumption to the typical working condition stage and meeting the specified operation time division. Because the train does not consume traction energy in the acceleration-cruising-coasting-braking working condition conversion, the optimal allocation of the minimum energy consumption in the acceleration working condition and the cruising working condition is mainly considered. The method comprises the following steps:
1) Dividing a train running interval into a plurality of sections, and taking the same line gradient as the same speed limit as a dividing basis;
2) Generating an ideal operating strategy consisting of acceleration-cruising-coasting-braking in each section, as far as possible in accordance with the prescribed operating times of the section operation;
3) And distributing traction energy to acceleration and cruising working conditions according to traction load prediction data, and solving the optimal distribution amount by utilizing a heuristic genetic algorithm to obtain a single train energy-saving speed curve.
2. Multi-train operation matching
The running matching of the multiple trains can be realized, namely, the running matching of the traction working condition trains and the braking working condition trains in the same power supply interval is realized, the regenerative braking energy generated by the braking working condition trains in the long downhill slope is transferred to the trains in the same power supply interval, which are in the traction working condition of the uphill section at the moment, and specifically, the traction energy consumption net value of the trains is reduced and the traction load peak clipping effect is realized by overlapping or prolonging the traction/braking working condition time of the train workshops in the same power supply interval. The method comprises the following steps:
1) Analyzing the up-down train of the possible intersection of the ramp sections according to the train running diagram, and determining the train number with overlapped traction/braking working conditions corresponding to the load curve abrupt sections in the traction load prediction data based on the CNN-LSTM-Attention;
2) Calculating transmissible regenerative braking energy, wherein only the regenerative braking energy utilization in the same power supply interval is considered, the redundant regenerative braking energy can be dissipated through a resistor, and the traction network voltage fluctuation cannot be caused in the energy transmission process;
3) And the overlapping of the braking/traction working conditions among multiple trains is coordinated, the overlapping time of the braking/traction working conditions is prolonged as much as possible by distributing the interval redundancy time on the basis of meeting the operation rule time division, and the transmission and consumption of regenerative braking energy in the same power supply interval are promoted to the greatest extent.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a separate embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to the embodiments described in detail below, and that the embodiments described in the examples may be combined as appropriate to form other embodiments that will be apparent to those skilled in the art.

Claims (4)

1. The energy-saving operation optimizing method for the high-speed rail long and large ramp train based on traction load prediction is characterized by comprising the following steps of:
step one, a traction load prediction model is established, and traction load prediction data are obtained;
and step two, optimizing the energy-saving operation of the train according to the traction load prediction data.
2. The method for optimizing the energy-saving operation of the high-speed rail long and large ramp train based on traction load prediction according to claim 1, wherein the first step is to build a traction load prediction model based on CNN-LSTM-Attention, and the input of the model is as follows: the model structure mainly comprises six layers, namely an input layer, a CNN layer, an LSTM layer, a full-connection layer, an Attention layer and an output layer.
3. The method for optimizing the energy-saving operation of the train on the long and large ramp of the high-speed rail based on traction load prediction according to claim 1, wherein the energy-saving operation optimization of the train in the second step comprises single-train energy-saving operation and multi-train operation matching.
4. The method for optimizing energy-saving operation of a long and large high-speed rail road train based on traction load prediction according to claim 3, wherein the step of single train energy-saving operation comprises dividing an operation interval, generating an ideal operation strategy, distributing traction energy, and the step of multi-train operation matching comprises determining road crossing times, transmitting regenerative braking energy and overlapping braking/traction working conditions.
CN202310183614.7A 2023-03-01 2023-03-01 High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction Pending CN116227699A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310183614.7A CN116227699A (en) 2023-03-01 2023-03-01 High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310183614.7A CN116227699A (en) 2023-03-01 2023-03-01 High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction

Publications (1)

Publication Number Publication Date
CN116227699A true CN116227699A (en) 2023-06-06

Family

ID=86572738

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310183614.7A Pending CN116227699A (en) 2023-03-01 2023-03-01 High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction

Country Status (1)

Country Link
CN (1) CN116227699A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117829378A (en) * 2024-03-04 2024-04-05 华东交通大学 Track traffic energy consumption prediction method based on space-time data

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117829378A (en) * 2024-03-04 2024-04-05 华东交通大学 Track traffic energy consumption prediction method based on space-time data
CN117829378B (en) * 2024-03-04 2024-05-14 华东交通大学 Track traffic energy consumption prediction method based on space-time data

Similar Documents

Publication Publication Date Title
CN102360401B (en) Method for designing urban rail transit energy-saving run chart based on genetic algorithm
CN110717218B (en) Electric drive vehicle distributed power drive system reconstruction control method and vehicle
CN110550034A (en) two-gear AMT comprehensive gear shifting method for pure electric vehicle
CN109583125B (en) Optimization method for improving utilization rate of regenerative braking energy of subway train
CN108791367A (en) The energy saving method of operating of train
CN107180282A (en) A kind of peak period many train energy conservation optimizing methods
CN112116207A (en) Multi-constraint-condition train operation adjustment calculation method and device
CN106004518B (en) A kind of electric automobile energy management prognostic control method based on car networking
CN109992843B (en) Urban rail transit train operation diagram construction method based on operation grade
CN107368920A (en) A kind of off-peak period multi-train movement energy conservation optimizing method
CN116227699A (en) High-speed rail long and large ramp train energy-saving operation optimization method based on traction load prediction
CN110990950B (en) Multi-target train energy-saving optimization method based on hybrid operation mode
CN114298510A (en) Time schedule and speed curve optimization method based on NSPSO algorithm
Montazeri-Gh et al. Intelligent approach for parallel HEV control strategy based on driving cycles
CN111680413A (en) Tramcar timing energy-saving operation optimization method and system based on double-layer algorithm
CN109376437A (en) A kind of vehicle-mounted energy storage optimal control method of new energy based on multi-objective genetic algorithm
Chuang et al. Design of optimal coasting speed for saving social cost in mass rapid transit systems
CN108985662A (en) A kind of train operation optimization method based on parallel immunity particle cluster algorithm
Cheng et al. Intelligent driving methods based on sparse LSSVM and ensemble CART algorithms for high-speed trains
CN113276829B (en) Vehicle running energy-saving optimization weight-changing method based on working condition prediction
CN113479187A (en) Layered different-step-length energy management method for plug-in hybrid electric vehicle
Li et al. Neural network-based subway regenerative energy optimization with variable headway constraints
CN117458477A (en) Electric vehicle scheduling method considering participation of load aggregators in grouping optimization mode
Xie et al. Control strategy of hybrid power system for Fuel Cell Electric Vehicle based on neural network optimization
CN115571125A (en) Model prediction control algorithm for PCC function of commercial vehicle

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