Analytical Accuracy Prediction Model for Fourier Transform-Based Coating Thickness Estimation in Eddy Current Testing of Coated Steel Sheets Using a Multi-Frequency Approach

Accurate thickness determination of thin electrically conductive coatings on ferromagnetic steel substrates using eddy current testing is crucial in industrial applications. In the literature, exact analytical models have been described for selected eddy current problems. This makes it particularly suitable to use a model-based estimation approach to determine unknown parameters, such as coating thickness. However, real-world measurement data are always affected by noise, which negatively impacts the achievable parameter estimation accuracy. Accordingly, signal processing of the measured data, e.g., based on Fourier transform, is crucial to determine how particular disturbances affect the parameter estimation. In this letter, the concept of the Cramér–Rao bound is applied to develop a prediction model for the achievable accuracy of the coating thickness estimate. This model is based on an analytical eddy current model for a differential coil system and a simplified noise model, incorporating sinusoidal excitation currents in a multifrequency approach.