本文關(guān)鍵詞： 獨立分量分析 盲信號分離 位場分離 弱異常提取　出處：《石油地球物理勘探》2017年06期 　論文類型：期刊論文

【摘要】：獨立分量分析(ICA)能夠通過發(fā)掘和去除數(shù)據(jù)間的高階相關(guān)成分,使輸出分量之間相互獨立,對獨立源信號進行分離和提取,是盲源分離(BSS)的核心技術(shù)。首先介紹了ICA的基本理論,運用基于負(fù)熵最大的FastICA(固定點算法)對算法原理進行說明和處理實驗;然后根據(jù)重磁信號的特點對獨立分量分析的應(yīng)用進行初步探索,并用冪次方迭代的ICA算法進行了重力異常仿真實驗,驗證了方法在重力異常的分離和弱異常提取中的有效性;最后應(yīng)用于實際磁測資料處理,有效識別了不同異常體。
[Abstract]:Independent component analysis (ICA) can separate and extract independent source signals by exploiting and removing high-order correlation components between the data and making the output components independent of each other. BSS is the core technology of blind source separation. Firstly, the basic theory of ICA is introduced, and the principle of the algorithm is explained and processed by FastICA (fixed Point algorithm) based on maximum negative entropy. Then according to the characteristics of gravity and magnetic signals, the application of independent component analysis (ICA) is preliminarily explored, and the gravity anomaly simulation experiment is carried out with the ICA algorithm of power power iteration. The validity of the method in gravity anomaly separation and weak anomaly extraction is verified. Finally, it is applied to the actual magnetic data processing, and the different abnormal bodies are effectively identified.
【作者單位】： 青海省第三地質(zhì)礦產(chǎn)勘查院;中國科學(xué)院青藏高原研究所大陸碰撞與高原隆升重點實驗室;青海省第一地質(zhì)礦產(chǎn)勘查院;
【基金】：國土資源部公益性行業(yè)科研專項經(jīng)費項目(201411025) 青海省科技廳重大科技專項(2016-SF-A3)聯(lián)合資助
【分類號】：P631
【正文快照】： *青海省西寧市城西區(qū)西川南路61號青海省第三地質(zhì)礦產(chǎn)勘查院物探所,810029。Email:493316908@qq.com本項研究受國土資源部公益性行業(yè)科研專項經(jīng)費項目(201411025)和青海省科技廳重大科技專項(2016-SF-A3)聯(lián)合資助。馬龍,孟軍海,付強,山中雪,李寶蘭.獨立分量分析在重磁數(shù)據(jù)處理，

本文編號：1480032