Research On Energy-Saving Operation Optimization Of Urban Rail Train Based On Artificial Bee Colony Algorithm

Urban rail transit has more convenience than road transportation.In recent years,urban rail transit has developed vigorously.More and more cities dedicated to the construction and development of rail transit.The total mileage and passenger volume of operations have also hit record highs.Behind the rapid development has also brought huge energy consumption and high costs that cannot be ignored.In the context of implementing sustainable development policies,it is not only necessary to continue to improve the ability to transport passengers,but it is also urgent to do a good job in energy saving and emission reduction of urban rail transit and to provide a satisfactory answer for the sustainable development of urban rail transit.Traction energy consumption occupies the vast majority of train energy consumption.By taking measures to reduce traction energy consumption,the goal of optimizing train energy saving can be effectively achieved.The thesis mainly studies the optimization of the recommended speed curve of the train,and also takes into account the multi-objective optimization with energy conservation as the main goal and the requirement of quickly planning the next inter-station running curve during the stop time.Combined with the current research situation,the two optimization algorithms of artificial bee colony algorithm are applied to the optimization of urban rail trains for comparative analysis.main tasks as follows:Firstly,analyze the horizontal loading of the following vehicles in different operating conditions,establish a train dynamics model,and compare the single-point and multi-point physical models of the train.Four target models were established: energy consumption model,comfort model,parking distance model and parking time model.Multiple objective functions are converted into single objective functions through the weight coefficient method.Considering the premise of safety,the value of each weighting coefficient is determined on the basis of multiple experimental simulations.At the same time,in order to improve the calculation speed,the principle of "turning zero into the whole" is adopted,and the total mileage between stations is divided into several sections to solve with the speed limit and slope of the line,and different speed code strategies are used for different sections to achieve The purpose of energy-saving optimization of urban rail train operation.Secondly,the principles,steps,advantages and disadvantages of the basic artificial bee colony algorithm are analyzed separately.Considering the problems of premature maturity and local optimization that are prone to appear in the actual engineering,adaptability evaluation and search methods are used to improve.Compared with the original algorithm,the improved algorithm's convergence speed and optimization effect have certain performance improvements.Using the data of Yizhuang Line for simulation,it is concluded that when the algorithm is applied to the optimization of the train running curve,all indicators have been improved.Finally,a combination of the artificial bee colony and simulated annealing algorithm is proposed to propose a nested simulated annealing-artificial bee colony algorithm(SA-ABC).The internal is simulated annealing algorithm and the external is the artificial bee colony algorithm.The algorithm combines the advantages of the two algorithms,and the simulation results of the four test functions of Sphere,Rastrigin,Griewank and Schaffer show that the nested algorithm is not only feasible but also has better performance.Using the data simulation between the Jinghai Road station and Ciqu Nan station of the Yizhuang Line for comparison,the four indicators of energy consumption,comfort,parking distance and parking time have different levels of optimization effects,which proves that SA-ABC algorithm is used in the problem of train curve optimization has more advantages.