基于LIBS技术与WOA-Staking集成模型的粉煤灰Fe元素定量分析  

Quantitative Analysis of Fe in Fly Ash Based on LIBS Technology and WOA-Stacking Integrated Model

作　　者：魏鑫巍 Wei Xinwei(Xi'an Shiyou University,Shaanxi,710065)

机构地区：[1]西安石油大学,陕西710065

出　　处：《当代化工研究》2025年第6期61-63,共3页Modern Chemical Research

摘　　要：粉煤灰作为燃煤过程的固体残余物,具有广泛的应用价值,尤其是其中的Fe元素在催化、土壤改良和环境修复中发挥重要作用。传统的元素分析方法如电感耦合等离子体发射光谱法(Inductively Coupled Plasma Optical Emission Spectrometer,ICP-OES)需要样品预处理,成本高且耗时长。激光诱导击穿光谱(Laser-induced Breakdown Spectroscopy,LIBS)技术因其无需样品预处理、检测速度快和成本低等优势,逐渐成为粉煤灰元素分析的重要工具。通过结合LIBS技术与鲸鱼优化算法(Whale Optimization Algorithm,WOA)优化的堆叠集成模型,实现了粉煤灰中Fe元素的高效定量分析。实验结果表明,优化后的GBDT模型预测精度最高(R^(2)=0.9567,RMSEp=0.0246),其他模型如RF、AdaBoost和KNN也表现出色。一次导数(D1st)预处理方法进一步提升了堆叠集成模型的预测性能。该方法为粉煤灰资源化利用提供了技术支持,在环境监测和土壤改良领域具有广泛应用潜力。Fly ash,a solid residue from coal combustion,has extensive application value,particularly due to the significant role of its Fe element in catalysis,soil improvement,and environmental remediation.Traditional elemental analysis methods such as inductively coupled plasma optical emission spectroscopy(ICP-OES)require sample pretreatment,which is costly and time-consuming.Laser-induced breakdown spectroscopy(LIBS)technology has emerged as a vital tool for fly ash elemental analysis due to its advantages of no sample pretreatment,fast detection speed,and low cost.By integrating LIBS technology with a stacking integrated model optimized by the whale optimization algorithm(WOA),efficient quantitative analysis of Fe in fly ash was achieved.Experimental results demonstrated that the optimized GBDT model exhibited the highest prediction accuracy(R^(2)=0.9567,RMSEp=0.0246),while other models such as RF,AdaBoost,and KNN also performed well.The first derivative(D1st)preprocessing method further enhanced the predictive performance of the stacking integrated model.This method provides technical support for the resource utilization of fly ash,with broad application potential in environmental monitoring and soil improvement.

关 键 词：激光诱导击穿光谱 鲸鱼优化算法 Stacking集成学习 粉煤灰 化学计量学 

分 类 号：TQ536[化学工程—煤化学工程]