基于RCGA的PPC模型在化探异常识别与提取中的应用

发布时间：2018-03-01 21:43

本文关键词： 投影寻踪分类实数编码遗传算法化探数据处理异常识别提取云南个旧　出处：《吉林大学学报(地球科学版)》2017年04期 　论文类型：期刊论文

【摘要】：为了进行地球化学异常的识别与提取,针对化探数据的特点,本文提出了一种将高维降维技术——投影寻踪分类(PPC)模型与实数编码遗传算法(RCGA)相结合的计算方法,分析了运用RCGA-PPC模型进行化探异常识别与提取的关键技术问题,并在MATLAB环境下开发了该方法的软件应用模块。以云南个旧地区水系沉积物地球化学数据为例,选取区域内Sn、Cu、Pb、Zn、As、Cd等主要成矿元素及与成矿关系密切的9种元素作为计算变量,利用RCGA-PPC模型对其进行处理和异常识别。研究表明:RCGAPPC模型中最佳投影值较高的地区与该区域实际矿床(点)吻合情况较好。该模型对化探异常的识别能力较强,是一种有效的化探多元素综合异常识别与提取方法。
[Abstract]:In order to identify and extract geochemical anomalies and in view of the characteristics of geochemical data, this paper presents a calculation method which combines the high-dimensional reduced-dimension technique, projection pursuit classification (PPC) model, with real coded genetic algorithm (RCGA). This paper analyzes the key technical problems of geochemical anomaly identification and extraction using RCGA-PPC model, and develops the software application module of this method under the MATLAB environment. Taking the geochemical data of water system in Gejiu area of Yunnan Province as an example, In this paper, the main metallogenic elements, such as PbPbPbPbPbPbPbPbPbPbPbPbPb, and 9 elements closely related to mineralization, are selected as calculation variables in the region. The RCGA-PPC model is used to deal with and identify anomalies. The results show that the best projection value of the WRCGAPPC model is in good agreement with the actual ore deposits (points) in the region, and the model has a strong ability to identify geochemical anomalies. It is an effective method to identify and extract multi-element anomaly in geochemical exploration.
【作者单位】：中山大学地球科学与工程学院;中山大学广东省地质过程与矿产资源探查重点实验室;中国地质大学(武汉)资源学院;中国地质大学地质过程与矿产资源国家重点实验室;
【基金】：国家自然科学基金项目(41502310) 广东省自然科学基金博士启动项目(2015A030310246)~~
【分类号】：P632

【参考文献】