Analysis method and experimental study of ice accumulation detection signal based on Lamb waves

A quantitative identification method for in-flight icing has the capability to significantly enhance the safety of aircraft operations.Ultrasonic guided waves have the unique advantage of detecting icing in a relatively large area,but quantitative identification of ice layers is a challenge.In this paper,a quantitative identification method of ice accumulation based on ultrasonic guided waves is proposed.Firstly,a simulation model for the wave dynamics of piezoelectric coupling in three dimensions is established to analyze the propagation characteristics of Lamb waves in a structure consisting of an aluminum plate and an ice layer.The wavelet transform method is utilized to extract the Time of Flight(ToF)or Time of Delay(ToD)of S_(0)/B_(1) mode waves,which serves as a characteristic parameter to precisely determine and assess the level of ice accumulation.Then,an experimental system is developed to evaluate the feasibility of Lamb waves-based icing real-time detection in the presence of spray conditions.Finally,a combination of the Hampel median filter and the moving average filter is developed to analyze ToF/ToD signals.Numerical simulation results reveal a positive correlation between geometric dimensions(length,width,thickness)of the ice layer and ToF/ToD of B1 mode waves,indicating their potential as indicators for quantifying ice accumulation.Experimental results of real-time icing detection indicate that ToF/ToD will reach greater peak values with the growth of the arbitrary-shaped ice layer until saturation to effectively predict the simulation results.This study lays a foundation for the practical application of quantitative icing detection via ultrasonic guided waves.