doi: 10.3850/S201042862015001155

Mapping Prediction Method and Application on Random
Vibration Environment of Vehicle Structure

Dong Longlei^a, Zhang Jingjing^b and Yan Guirong^c State Key Laboratory for strength and Vibration of Mechanical Structures,
Xi'an Jiaotong University, No. 28, Xianning West Road, Xi'an, Shanxi,
Xi'an, 710049, China ^adongll@xjtu.edu.cn ^bjingjing2012@stu.xjtu.edu.cn ^cyangr@xjtu.edu.cn

ABSTRACT

In view of the problem of the vibration response of the new vehicle, the mapping prediction method of random vibration environment is proposed. Firstly, the dynamic models of the same structure under different system conditions are established. Then, the vibration response of the structure is calculated based on the load equivalent principle, and the response learning samples are constructed in the frequency domain. Next, the mapping relation model of this structure is established by using the machine learning method. Finally, the response of the structure under certain kind of condition is used as the inputting working sample of the mapping relation model, and then the response of the structure under the other conditions is predicted. Taking a vehicle engine test as an example, using the prediction method of random vibration environment, the vibration response of the engine structure under different test conditions was carried out. Compared with the test results, the prediction results were satisfied with the requirement of engineering precision. Furthermore, the vibration response prediction of the vehicle in flight environment was realized, which provided the basis for the development of the environment test conditions for the key parts of the vehicle.

Keywords: Vehicle, Random vibration, Mapping prediction method, Machine learning.

---

Back to TOC