基于SAO-VMD-FFT的激光诱导荧光光谱信号信噪比提升方法

The Methodology for Enhancing the Signal-to-Noise Ratio of Laser-Induced Fluorescence Spectroscopy Based on SAO-VMD-FFT

针对基于激光诱导荧光光谱法(LIFs)的便携式场地地下水检测技术装备在原位检测地下水中典型重金属时,光传感信号含噪声严重,影响后续定量分析准确度的问题,提出了一种雪消融优化器(SAO)寻优算法优化的变分模态分解(VMD)结合快速傅里叶变换(FFT)的LIFs光传感信号信噪比提升算法。利用SAO算法对VMD分解关键参数分解模态个数K和惩罚因子α进行优化,实现含噪信号的自适应最佳VMD分解,得到一系列本征模态分量。进一步结合FFT算法寻找频率突变点从而筛选出相关模态。累计相关模态得到信噪比提升后的信号。通过仿真和实测信号分析,结果表明,与EMD、WTD和S-G等常用算法相比,建立的算法能更好地去除LIFs信号中的噪声分量,保留有用信息。对方法的可行性和有效性进行了验证,可为后续利用LIFs传感信号原位定量分析地下水中重金属元素含量提供支持。

The portable field groundwater detection technique based on laser-induced fluorescence spectroscopy(LIFs)faces significant challenge due to high levels of noise in the optical sensing signal.This noise adversely affects the accuracy of subsequent quantitative analysis of typical heavy metals in situ.To mitigate this issue,a novel algorithm that combined the snowmelt optimization(SAO)algorithm with variational mode decomposition(VMD)optimized using fast fourier transform(FFT)was proposed.The algorithm aims was to improve the signal-to-noise ratio of the optical sensing signals generated by LIFs.The SAO algorithm optimized the essential parameters of VMD,including the number of decomposition modes(K)and the penalty factor(α),therefore an adaptive and optimal decomposition of the noisy signal was achieved,resulting in a set of intrinsic mode components.Furthermore,the FFT algorithm was utilized to detect frequency change points and select relevant modes.The accumulation of relevant modes contributed to enhancing the signal-to-noise ratio.Through simulation and experimental signal analysis,it was demonstrated that the proposed algorithm surpassed commonly used algorithms in efficiently removing noise components from LIFs signals while preserving valuable information,such as empirical mode decomposition(EMD),wavelet transform(WTD),and the Savitzky-Golay(S-G)algorithm.This study validates the feasibility and effectiveness of the proposed method,providing substantial support for the accurate quantitative analysis of heavy metal content in groundwater through the utilization of LIFs sensing signals.