Aerodynamic Performance Optimization Of A High-Speed Train Based On Surrogate Model

With their convenience and comfort,high-speed trains are increasingly becoming a focus of attention for many governments and have become a popular means of transportation.China is a vast land,within which climate varies significantly between regions.Different climatic environments have different effects on train performance.The northeastern region and the Qinghai-Tibet Plateau are relatively cold,the northwestern region is relatively windy with sandstorms and the climate to the south of the Yangtze River is temperate.For high-speed trains operating in different climatic environments,the focus on their operational performance is different.In order to improve economy,comfort and safety,vehicle optimization work for specific climate types is ongoing.In engineering optimization,methods based on a surrogate model are commonly used.This thesis conducts research on the related problems that may affect the train in varied climatic environments and proposes reasonable optimization measures.Aiming at the intensified nonlinearity of the vehicle system in the crosswind environment,a multi-layer neural network surrogate model is proposed to deal with the windinduced safety optimization problem of high-speed trains.The main research contents and conclusions of the thesis are as follows:First,the effect of the shape of the streamlined area of the train on the aerodynamic performance of the head and tail cars is studied in an ideal climatic environment.The traditional aerodynamic optimization work on high-speed trains is usually based on the assumption that the shapes of the head and tail cars are the same.In this case,the differences in effect on the aerodynamic performance of the head car as opposed to the tail car,caused by each geometrical feature of the streamlined region,is ignored.The streamlined shape affects the aerodynamic characteristics of the head and tail cars differently.In this thesis,the effects of five shape parameters in the streamlined region,on the drag of the head car,the drag of the tail car and the lift of the tail car are studied,and the aerodynamic performance of the two optimization schemes is compared.The research results show that the aims of reducing the drag of the head car and reducing the aerodynamic force of the tail car are in opposition.However,the two objectives of reducing drag and lift force of the tail car are more consistent.Therefore,the optimization mode that assumes that the head and tail cars are the same shape,limits the improvement of the aerodynamic performance of high-speed trains.In the future,trains can further optimize their running performance by adjusting the shape of the head and tail cars independently.After optimization,the drag force of the head car,the drag force of the tail car and the lift force of the tail car decrease obviously,and the optimization effect of the new scheme is more than twice that of the traditional scheme under the same situation.Second,the characteristics of ice accretion in the bogie area of the head car of a highspeed train in a cold environment are studied.Combining the computational fluid dynamics method with the Euler multiphase flow model,the flow field characteristics,the distribution characteristics of droplets and ice crystals,and the ice accretion process in the bogie region of the head car are simulated.The results of the study show that locations with higher positive pressure in the bogie area are more prone to ice accretion.The amount of ice accretion in the first bogie area of the head car is much larger than that in the second bogie area.When the train runs at minus ten degrees Celsius for 15 minutes,the ice accumulated heavily in the two bogie areas of the head car.In view of this situation,a deflector is designed to improve it.When the head car equipped with optimized deflector runs under the same conditions,the ice accumulation of the two bogies decreases obviously.With the passage of time,the deflector can still play the effect of anti-ice accumulation,and effectively reduce the running resistance of the head car.Third,the influence of the working angle and installation position of the braking plate on the aerodynamic characteristics of the train under crosswind is studied.Taking the threecarriage train as the research object,the influence of a change in angle of the braking plate on the aerodynamic force of the car body was studied under the condition of crosswind.Then,based on the surrogate model method,the influence of the position of the braking plate on the aerodynamic force of the head car is studied,and the functional relationship between the position and the aerodynamic force is determined.The research results show that opening the braking plate in the crosswind situation not only increases the aerodynamic resistance of the train,but also significantly changes the lift and lateral force of the train.With the increase of the braking plate angle,the drag and lift of the head car,the middle car and the tail car show a trend of first increasing and then decreasing.The lateral forces on the head car and the middle car show a trend of decreasing first and then increasing.In the tail car,a side force in the opposite direction appeared,and showed a trend of increasing first and then decreasing.When the first braking plate is located near the end of the streamlined area of the head car,the aerodynamic drag gain of the head car is larger,but if both braking plates are close to the streamlined area,the lift force of the head car may increase sharply.When the two braking plates,especially the second braking plate,are located at the rear end of the passenger compartment,the side force of the head car may be large.Therefore,it is recommended that the first braking plate be arranged at the end of the streamlined area,and that the second braking plate be arranged in the middle of the passenger compartment.Fourth,the multi-objective optimization of wind-induced safety of high-speed trains is carried out based on the multi-layer neural network surrogate model.In view of the fact that the increasing nolinearity of vehicle systems in crosswind conditions leads to the decline of the accuracy of the surrogate model,this thesis proposes the multi-layer neural network surrogate model to deal with such problems,and conducts optimization research on the vehicle wind-induced safety problem in two cases.The research results show that compared with the traditional Kriging model,the neural network model has more adjustable parameters and the multi-layer neural network surrogate model has advantages in mapping accuracy over the traditional Kriging model.It is more suitable for optimization of vehicle system parameters in crosswind.In case 1,the crosswind speed is 20 m/s and the train running speed is 350 km/h.In this case,the operation safety of the head car deteriorates very seriously,in which the derailment coefficient,wheel load reduction rate and wheelset lateral force exceed the safety limits.After the optimization,the operation safety of the vehicle in the same environment is improved,and the derailment coefficient,the wheel load reduction rate and the wheelset lateral force are greatly reduced.In case 2,the crosswind speed is 15 m/s,the train running speed is350 km/h,and the lift wings are installed on the top of the train.In this case,the derailment coefficient,wheel load reduction rate and the wheelset lateral force all exceed the safety limits.After optimization,train’s operation safety is also significantly improved.Combined with the characteristics of various climates,this thesis optimizes the operation performance of high-speed trains in different climates by using numerical simulation methods.In view of the fact that the increasing nolinearity of vehicle systems in crosswind conditions leads to the decline of the accuracy of the surrogate model,this thesis proposes the multi-layer neural network surrogate model to deal with such problems.The reliability and validity of the surrogate model are verified.The relevant research conclusions can provide a reference for the future optimal design of high-speed trains.