首页|基于门控循环单元的动车组制动盘温度模型构建与预测研究

基于门控循环单元的动车组制动盘温度模型构建与预测研究

扫码查看
文中提出了一种基于人工神经网络的动车组制动盘温度预测模型.该模型利用门控循环单元获取参数的时序特征,预测高速动车组紧急制动过程中制动盘温度变化.以某动车组制动运行试验的实测数据进行模型训练,预测不同初速度制动工况下制动盘温度变化情况.同时将预测结果与其他 3 种人工神经网络模型进行比较.结果表明,基于门控循环单元的制动盘温度预测模型效果显著,能有效提取数据特征,为制动盘温度预测提供了一种新的思路.
Research on Model Construction and Prediction of Brake Disc Temperature of EMU Based on Gated Recurrent Unit
This paper proposes a model for predicting the temperature of EMU brake disc based on neural network.In this model,the time series characteristics of parameters obtained by gating cycle unit are used to predict the temperature of brake disc during emergency braking of high-speed EMU.The model training is carried out by the measured data of an EMU braking test and predict the temperature changes of the brake disc under different initial speed braking conditions.The results were compared with those of other three artificial intelligence prediction methods,demonstrating the superiority of the proposed method,and provides a new idea for brake disc temperature prediction.

EMUdisc brakeneural networktemperature predictionGated Recurrent Unit(GRU)

刘洋、郭奇宗、贾志东、杨亦铮

展开 >

中国铁道科学研究院集团有限公司 机车车辆研究所,北京 100081

动车组和机车牵引与控制国家重点实验室,北京 100081

北京纵横机电科技有限公司,北京 100094

动车组 制动盘 神经网络 温度预测 门控循环单元

2024

铁道机车车辆
中国铁道科学研究院 中国铁道学会牵引动力委员会

铁道机车车辆

北大核心
影响因子:0.254
ISSN:1008-7842
年,卷(期):2024.44(5)