一种两阶段优化的声速剖面自适应简化算法OA

To address the excessive computational load caused by high-resolution sound speed profile(SSP)in underwater acoustic propagation modeling,an adaptive SSP simplification algorithm is pro‑posed,which balances global optimality with computational efficiency.In the algorithm,profile simpli‑fication is formulated as an optimization model to find the minimum set of characteristic points subject under a user-defined root mean square error(RMSE)constraint.To achieve this efficiently,a two-stage strategy is designed in this paper.In the first stage the Visvalingam-Whyatt(VW)algorithm is em‑ployed to rapidly screen high-quality candidate points,significantly pruning the search space.In the second stage the dynamic programming(DP)is utilized to perform a global search within an adaptive binary search framework,precisely identifying the minimum number of points and their distribution satisfying the error constraint.Experimental results show that at the same compression ratio,the pro‑posed algorithm outperforms traditional methods in terms of both RMSE and the relative error of acous‑tic ray travel time,which provides an efficient and reliable sound speed profile simplification solution for high-precision underwater acoustic applications.