Pavement grade identification method based on reverse analysis of vehicle vibration response

Here, a pavement grade identification method based on Hilbert-Huang transformation and probabilistic neural network being used for reverse analysis of vehicle vibration response was proposed to provide effective information for active suspension control. Firstly, vehicle model and random input pavement model were established. Hilbert-Huang transformation was used to decompose and transform simulation data of vehicle vibration responses under different grades of pavements, and obtain instantaneous energy of vehicle vibration responses. Then, sensitive characteristic parameters of instantaneous energy were extracted, the probabilistic neural network was used to train pavement classifier, the mapping relation among pavement grade and ranges of characteristic parameters was determined to complete design of pavement grade classifier. Finally, acceleration sensors were used to collect vehicle vibration response data under typical road surfaces, extract characteristic parameters of test data, and input them into the trained road surface classifier for realizing grade identification of the tested road surface. The identification results showed that the reverse analysis method based on vehicle vibration response, combined with Hilbert-Huang transformation and probabilistic neural network, can identify the current vehicle driving pavement grade. © 2022 Chinese Vibration Engineering Society. All rights reserved.