基于CPSO-ICA的航空瞬变电磁信号去噪方法研究  

Research on Method of Airborne Transient Electromagnetic Signal Denoising Based on CPSO-ICA

作　　者：余建国[1] 田宝 周鹏[1] 李玲玲[1] YU Jianguo;TIAN Bao;ZHOU Peng;LI Lingling(School of Intelligent Engineering,Zhengzhou University of Aeronautics,Zhengzhou 450046,China)

机构地区：[1]郑州航空工业管理学院智能工程学院,河南郑州450046

出　　处：《信阳师范学院学报（自然科学版）》2023年第4期611-617,共7页Journal of Xinyang Normal University(Natural Science Edition)

基　　金：国家自然科学基金项目(62202434);国家重点研发计划项目(2016YFC0802102);河南省杰出外籍科学家工作室项目(GZS2022011);河南省自然科学基金项目(232300420330);河南省高等学校重点科研项目(23B440001);河南省科技攻关项目(222102210093)。

摘　　要：为解决航空瞬变电磁检测信号易受到外部噪声干扰进而严重影响测量数据质量与可靠性的问题,提出使用独立成分分析(ICA)法,对检测信号进行处理,通过分离原始信号中的有效信号和噪声信号,实现信号去噪。为解决步长设定问题,将混沌粒子群优化算法(CPSO)引入到ICA中(CPSO-ICA),使用粒子群算法实时动态调整ICA的步长函数,进一步使用混沌算法优化标准粒子群算法以实现步长的全局寻优,加快收敛速度的同时减小稳态误差,增强ICA算法的去噪效果。模拟信号与实测数据的实验结果表明,CPSO-ICA算法能够在保证去噪效果的同时,完整地保留原始信号的特征,能够为后期反演提供可靠和有效的航空瞬变电磁数据.To address the issue external noise interference on airborne transient electromagnetic detection signals,which significantly impacts the quality and reliability of measurement data,the utilization of independent component analysis(ICA)is proposed for signal processing.The ICA method separates the effective signal from the noisy signal in order to achieve denoising.Additionally,to overcome step size setting challenges,a chaotic particle swarm optimization algorithm(CPSO)is introduced into ICA(CPSO-ICA).This approach dynamically adjusts the ICA step size function in real time using particle swarm optimization algorithm and further optimizes global optimization of step size using standard particle swarm optimization algorithm.As a result,it accelerates convergence speed,reduces steady-state error,and enhances denoising effect of ICA algorithm.Experimental results with simulated signals and measured data demonstrate that CPSO-ICA algorithm not only ensures deno effectiveness but also fully retains original signal features while providing reliable and effective airborne transient electromagnetic data for subsequent inversion.

关 键 词：航空瞬变电磁 去噪 独立成分分析 混沌粒子群 

分 类 号：TN911.4[电子电信—通信与信息系统]