Local parameter identification with neural ordinary differential equations

The data-driven methods extract the feature information from data to build system models, which enable estimation and identification of the systems and can be utilized for prognosis and health management(PHM). However, most data-driven models are still black-box models that cannot be interpreted. In this study, we use the neural ordinary differential equations(ODEs), especially the inherent computational relationships of a system added to the loss function calculation, to approximate the governing equations. In addition, a new strategy for identifying the local parameters of the system is investigated, which can be utilized for system parameter identification and damage detection. The numerical and experimental examples presented in the paper demonstrate that the strategy has high accuracy and good local parameter identification. Moreover, the proposed method has the advantage of being interpretable. It can directly approximate the underlying governing dynamics and be a worthwhile strategy for system identification and PHM.

Design and Experiment of PZT Network-based Structural Health Monitoring Scanning System

The active Lamb wave and piezoelectric transducer （PZT）-based structural health monitoring （SHM） technology is a kind of efficient approach to estimate the health state of aircraft structure. In practical applications, PZT networks are needed to monitor large scale structures. Scanning many of the different PZT actuator-sensor channels within these PZT networks to achieve on-line SHM task is important. Based on a peripheral component interconnect extensions for instrumentation （PXI） platform, an active Lamb wave and PZT network-based integrated multi-channel scanning system （PXI-ISS） is developed for the purpose of practical applications of SHM, which is compact and portable, and can scan large numbers of actuator-sensor channels and perform damage assessing automatically. A PXI-based 4 channels gain-programmable charge amplifier, an external scanning module with 276 actuator-sensor channels and integrated SHM software are proposed and discussed in detail. The experimental research on a carbon fiber composite wing box of an unmanned aerial vehicle （UAV） for verifying the functions of the PXI-ISS is mainly discussed, including the design of PZTs layer, the method of excitation frequency selection, functional test of damage imaging, stability test of the PXI-ISS, and the loading effect on signals. The experimental results have verified the stability and damage functions of this system.