High-precision localization of pipeline damage using acoustic emission technology in the marine environment remains challenging. A hybrid path optimization-based localization method for structural damage source localization is proposed. First, the short-time Fourier transform (STFT) maps acoustic signals into the time-frequency domain, with K-means cluster grouping signals to separate different damage sources. Subsequently, the ROTH dual-weighted cross-correlation algorithm is employed to accurately estimate the time difference of acoustic wave arrivals at sensors under strong noise conditions, extracting temporal characteristics of the acoustic waves. Next, an initial propagation path is generated via the Rapidly-exploring Random Tree (RRT) algorithm, and further optimized through an improved D* dynamic path optimization algorithm to determine the optimal propagation path. Finally, acoustic emission source locations are calculated by using hyperbolic equations and weighted least squares based on geometric localization methods. Experimental results demonstrate that this method can accurately locate the structural damage source in complex environments, significantly outperforming traditional approaches.