基于径向基神经网络的机车牵引能耗计算模型  

Study on Locomotive Traction Energy Consumption Calculation Based on RBF Neural Network

作　　者：李志勇[1,2] 文睿[1] 危韧勇[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410083 [2]株洲南车时代电气股份有限公司技术中心,湖南株洲412309

出　　处：《铁道学报》2011年第9期27-30,共4页Journal of the China Railway Society

基　　金：铁道部科技研究开发计划(2008Z003-D)

摘　　要：机车牵引能耗与机车属性、列车编组情况、线路条件、操作方式等诸多因素密切相关,是运输组织和调配机车的重要依据。本文构建了机车牵引能耗计算模型。首先考虑机车的实际牵引操控情况,建立列车运动方程模型,求解机车在不同运行工况下的运动变化和牵引做功;然后建立径向基神经网络,拟合机车总效率与不同牵引工况之间的关系模型,选择机车型式试验的实际数据样本训练径向基神经网络;最后由牵引做功和机车总效率求解机车牵引能耗。将所建模型应用于SS9型机车进行能耗计算,表明该模型能准确反映机车运用过程中的牵引能耗数据。Locomotive traction energy consumption, which is closely connected with locomotive properties, train formation, railway line conditions and field operation, is an important basis for optimization of transport organization and locomotive deployment. In consideration of the actual locomotive operating conditions, this paper establishes the train motion equation model to reveal the motion changes and traction work under different traction conditions. Furthermore, this paper builds the RBF neural network model to fit the relationship between the total locomotive efficiency and different traction conditions and to train the RBF neural network with locomotive type test data. On the basis of the above two models, locomotive traction energy consumption values can be obtained from the traction work and total locomotive efficiency. Taking locomotive SS9 as an example, the proposed models yield results in complete agreement with actual traction energy consumption data generated in locomotive operation.

关 键 词：机车 牵引能耗计算 机车总效率 径向基神经网络 

分 类 号：U260.153[机械工程—车辆工程]