Efficient numerical solution of boundary identification problems: MFS with adaptive stochastic optimization
作者：
Highlights：
• Demonstration of a novel (heuristic) computational technique for the efficient numerical solution of inverse boundary identification problems in two dimensions.
• Derivation of a posteriori error estimates for inverse boundary identification.
• An adaptive stochastic algorithm is developed for the first time in the MFS literature to determine its parameters, although it is addressed in the context of boundary identification.
• Substantial improvement on previously existing non-linear optimization algorithms.
摘要
•Demonstration of a novel (heuristic) computational technique for the efficient numerical solution of inverse boundary identification problems in two dimensions.•Derivation of a posteriori error estimates for inverse boundary identification.•An adaptive stochastic algorithm is developed for the first time in the MFS literature to determine its parameters, although it is addressed in the context of boundary identification.•Substantial improvement on previously existing non-linear optimization algorithms.
论文关键词：A posteriori error estimator,Fundamental solutions,Inverse boundary identification,Adaptive stochastic strategy
论文评审过程：Received 9 December 2020, Revised 24 March 2021, Accepted 17 May 2021, Available online 8 June 2021, Version of Record 8 June 2021.
论文官网地址：https://doi.org/10.1016/j.amc.2021.126402
