Abstract: Carbon fiber reinforced polymer (CFRP) bogies offer advantages such as low density and high specific strength, making them highly significant for reducing the weight of electric multiple units. However, the anisotropic nature of CFRP makes it prone to internal defects like voids and delamination, which severely compromise the structural performance. Thermal wave radar is a non-destructive testing technique based on active thermography capable of performing both qualitative and quantitative analyses of defects in composites. This technology detects internal defects by employing broadband frequency-modulated thermal excitation combined with advanced signal processing algorithms. However, existing research has primarily focused on qualitative defect detection such as defect contrast enhancement, with limited quantitative analysis of defect depth. This study proposes a novel method for estimating internal defect depth based on thermal wave radar technology. Firstly, phase contrast peak value is proposed as a new feature for estimating defect depth. Both theoretical analysis and numerical simulation indicate that the peak phase contrast exhibits a linear relationship with defect depth. Secondly, it reveals the influence of defect size on the linear relationship between phase contrast peak values and defect depth. The defect diameter is induced as a correction factor to construct a composite variable. This addresses significant estimation errors for defects of different sizes at the same depth. Thirdly, a quantitative workflow is established for estimating the depth of defects using infrared thermal wave radar technology. This enables the precise estimation of the depth of defects of different sizes and depths under a single thermal loading. This enhances both detection efficiency and engineering applicability. Six defects with varying sizes and depths in carbon fiber reinforced polymer laminate yielded an average measurement error of only 8%. The influence of material thermal properties and thermal wave radar parameters are also discussed. This study will enhance quantitative non-destructive testing methods for CFRP bogie defects, providing a safe and reliable foundation for lightweighting high-speed train bogies.