报告时间：2024年12月27日（星期五）9:00-10:00

报告地点：理学楼609报告厅

报告题目：Reconstruction of sound sources and obstacles based on neural networks

报告摘要: In this talk, we explore the mechanism of a new audio life detector that enables the localization of trapped people and the identification of the shape of buried debris. Based on the practical situation, the above process can be modeled as a double inverse model. The far-field data of different sound sources and obstacles are collected in both finite aperture and full aperture, and a parallel neural network is constructed to solve the problem of decreased computational accuracy caused by the lack of information in the double inverse model. Numerical experiments show the effectiveness of the method and the robustness of the network to noise.

报告人简介：长春理工大学数学与统计学院教授，硕士研究生导师。主要研究兴趣是数学物理反问题、机器学习算法的设计与理论分析等。在JCAM，OE，CICP，IPI等期刊发表论文20余篇，主持并参与国家自然科学基金、军委科技委领域基金、吉林省科技厅基金、吉林省教育厅基金8项。目前担任中国仿真学会不确定系统分析与仿真专委会委员、IPIA会员、吉林省运筹学会理事。