| The automatic driving of freight train can be realized by tracking the target speed curve through the control strategy.Therefore,the performance of the target speed curve will ultimately affect the automatic driving function of the train.The purpose of this paper is to solve the multi-objective optimization model of freight train and obtain a target speed curve with superior performance and meeting various indexes.The multi-objective optimization problem of freight train belongs to the nonlinear programming problem with multiple constraints.At present,there are many methods used to solve the multi-objective optimization model of freight train,including analytical method,numerical method and intelligent optimization algorithm.Among them,analytical method and numerical method rely on accurate mathematical model and are difficult to calculate,while the intelligent algorithm has low dependence on the model,is easy to program,and can find the approximate global optimal solution within the effective number of iterations.Therefore,sparrow optimization algorithm is introduced into the multi-objective optimization problem of freight train in this paper.The work of this paper can be roughly divided into three points:(1)Build the multi-objective optimization model of freight train automatic drivingAccording to the principle of automatic driving,the traction force,resistance and braking force of freight train are analyzed.Based on the single mass point and multi mass point models used in the traction calculation of freight train,considering their respective advantages and disadvantages,combined with the research content of this paper,a hybrid traction calculation model is proposed.Taking the reduction of train operation time error and traction energy consumption as the optimization objective,and the speed limit and working condition distance as the constraints,the multi-objective optimization model of freight train is established.The time objective is added to the objective function as the penalty function,and the adaptive weighted particle swarm optimization algorithm is used to solve it,and the target speed curve meeting the requirements of energy-saving punctuality is obtained.(2)Determine the working condition sequence according to the line speed limit and rampThis paper studies the four-stage energy-saving strategy widely used in rail transit,analyzes the reasons why it is not suitable for freight trains,and studies the operation strategy suitable for complex lines of freight trains.Firstly,the working condition sequence under the time-saving strategy is obtained according to the line speed limit,and then the working condition corresponding to the original position is replaced by coasting and traction working condition according to the position of large up ramp and down ramp in the line,On other ramps,the original working conditions will remain unchanged,and finally a working condition sequence with better comprehensive performance than the four-stage strategy is obtained.(3)Improved sparrow algorithm for multi-objective optimization of freight trainsAiming at the defects of unbalanced global search ability and local development ability and easy to fall into local solutions of sparrow algorithm,an improved sparrow algorithm based on adaptive weight and Levy flight disturbance is proposed.Aiming at the problems that random generation of initial population may produce a large number of infeasible solutions and slow calculation speed,a method of searching working condition conversion points in a fixed range is proposed;Using the linear weighted summation of time and energy consumption,the multi-objective optimization problem is transformed into a single objective optimization problem.The simulation results of single objective optimization with different weights are obtained.A group of weights that can meet the punctuality are selected for subsequent optimization,and the improved sparrow algorithm is used to solve the train multi-objective optimization problem. |
PDF Full Text Request