Uranium Potential Evaluation of Zhuguangshan Granitic Pluton in South China Based on Machine Learning
-
摘要: 地学大数据和机器学习的结合，为矿床勘查提供了新的发展方向.华南广泛发育花岗岩体，是花岗岩型铀矿的重要产区，因此如何判断特定花岗岩体是否具有产铀矿的潜力，对于指导华南花岗岩型铀矿勘查具有重要意义.系统收集了前人已发表的华南花岗岩地球化学元素含量数据（不包括待评价的诸广山地区的九峰岩体、红山岩体和茶山岩体），共获得1 711条数据.然后按照7∶3的比例划分为训练集和测试集，进而分别建立了随机森林（random forest，RF）算法和K近邻（K-nearest neighbor，KNN）算法分类模型，并对两种分类模型的精确度、召回率、ROC（receiver operating characteristic curve）曲线进行评价，选出泛化能力较好的模型，最后利用泛化能力较好的模型对诸广山地区九峰岩体、红山岩体和茶山岩体进行成矿潜力评价.结果表明，随机森林分类模型对测试集的分类精确度、预测结果可靠度均高于K近邻分类模型，随机森林分类模型对测试集上的数据分类精确度达到了93%，利用上述创建的随机森林分类模型对九峰、红山和茶山岩体进行预测.预测结果表明，红山岩体和茶山岩体含矿的概率较高，而九峰岩体含矿概率较低.该研究为进一步缩小地质找矿勘查范围提供了可靠的依据，并且该模型可以作为地质找矿工作者的辅助工具.Abstract: The combination of geological data and machine learning provides a new direction for mineral exploration. The granitic pluton is widely developed in South China, which is an important producing area for granite-type uranium deposits. Therefore, whether the granitic pluton has the potential to produce uranium deposits is of great significance for guiding the exploration of granite-type uranium deposits in South China. In this paper, the geochemical data of granites in South China are systematically collected (excluding the Jiufeng, Hongshan and Chashan granite plutons to be evaluated in Zhuguangshan area) from previous published papers, and a total of 1 711 data pieces are obtained. They are further divided into training set and test set according to the ratio of 7∶3. Then, the random forest (RF) algorithm and K-nearest neighbor (KNN) algorithm classification models were established respectively, and the accuracy, recall rate and ROC (receiver operating characteristic curve) curve of the two classification models were evaluated, and the models with good generalization ability were selected. Finally, the metallogenic potential of the Jiufeng pluton, Hongshan pluton and Chashan pluton in the Zhuguangshan area were evaluated using the models with good generalization ability. The results show that the classification accuracy and reliability of prediction results of random forest classification model are higher than those of K-nearest neighbor classification model, and the classification accuracy of the random forest classification model on the test set reached 93%. The random forest classification model created above was used to evaluate the metallogenetic potential of the Jiufeng, Hongshan and Chashan plutons. The prediction results show that the probability of metallogenetic potentiality in the Hongshan and Chashan plutons is high, whereas the probability in the Jiufeng pluton is low. This study provides a reliable basis for further geological prospecting, and the model can be used as an auxiliary tool for geological prospecting.
-
图 1 诸广山岩体分布及采样点位置图
1.中粒-细粒二云母花岗岩；2.中-中粗粒（斑状）黑云母花岗岩；3.粗粒斑状黑云母花岗岩；4.印支期花岗岩；5.海西期黑云母闪长岩；6.加里东期花岗闪长岩；7.条带状混合花岗岩；8.条带状混合岩；9.取样点及编号；10.铀矿床及编号；11.断裂；图改自朱捌（2010）
Fig. 1. Zhuguangshan pluton distribution and sampling point location map
表 1 九峰、红山和茶山岩体随机森林模型预测结果
Table 1. Jiufeng, Hongshan and Chashan plutons random forest model prediction results
编号 不含矿概率(%) 含矿概率(%) 预测结果 九峰岩体 06168 64 36 0 06170 86 14 0 06171 89 11 0 06172 93 7 0 06173 84 16 0 0629 17 83 1 红山岩体 0631 11 89 1 0632 10 90 1 0633 10 90 1 0635 9 91 1 06184 4 96 1 茶山岩体 06185 3 97 1 06186 1 99 1 
下载: 导出CSV
-
Altmann, A., Toloşi, L., Sander, O., et al., 2010. Permutation Importance: A Corrected Feature Importance Measure. Bioinformatics, 26(10): 1340-1347. https://doi.org/10.1093/bioinformatics/btq134 Breiman, L., 2001. Random Forest. Machine Learning, 45: 5-32. doi: 10.1023/A:1010933404324 Breiman, L., 2004. RFtools-for Predicting and Understanding Data. Technical Report, Berkeley University, Berkeley, USA. Brown, W.M., Groves, D.I., Gedeon, T.D., 2003. An Artificial Neural Network Method for Mineral Prospectivity Mapping: A Comparison with Fuzzy Logic and Bayesian Probability Methods. In: Sandham, W.A., Leggett, M., eds., Geophysical Applications of Artificial Neural Networks and Fuzzy Logic. Modern Approaches in Geophysics, 21. Springer, Dordrecht. https://doi.org/10.1007/978-94-017-0271-3_12 Chehreh Chelgani, S., Matin, S.S., Hower, J.C., 2016. Explaining Relationships between Coke Quality Index and Coal Properties by Random Forest Method. Fuel, 182: 754-760. https://doi.org/10.1016/j.fuel.2016.06.034 Chen, Y.L., Zhou, B., Li, X.B., 2012. Mineral Target Prediction Based on Boltzmann Machines. Progress in Geophysics, 27(1): 179-185(in Chinese with English abstract). Dong, S.S., Huang, Z.X., 2013. A Brief Theoretical Overview of Random Forests. Journal of Integration Technology, 2(1): 1-7(in Chinese with English abstract). Hao, H.Z., Gu, Q., Hu, X.M., 2021. Research Advances and Prospective in Mineral Intelligent Identification Based on Machine Learning. Earth Science, 46(9): 3091-3106(in Chinese with English abstract). Harris, D., Zurcher, L., Stanley, M., et al., 2003. A Comparative Analysis of Favorability Mappings by Weights of Evidence, Probabilistic Neural Networks, Discriminant Analysis, and Logistic Regression. Natural Resources Research, 12(4): 241-255. https://doi.org/10.1023/b:narr.0000007804.27450.e8 Hong, J., Gan, C.S., Liu, J., 2018. Preliminary Study on the Relationship between Trace and Major Elements in Rocks Based on Machine Learning: A Case Study of Zr in OIB. Chinese Journal of Geology, 53(4): 1285-1299(in Chinese with English abstract). Hong, S., Zuo, R.G., Huang, X.W., et al., 2021. Distinguishing IOCG and IOA Deposits via Random Forest Algorithm Based on Magnetite Composition. Journal of Geochemical Exploration, 230: 106859. https://doi.org/10.1016/j.gexplo.2021.106859 Izadi, H., Sadri, J., Mehran, N.A., 2013. Intelligent Mineral Identification Using Clustering and Artificial Neural Networks Techniques. 2013 First Iranian Conference on Pattern Recognition and Image Analysis (PRIA), IEEE. Birjand, Iran. https://doi.org/10.1109/pria.2013.6528426 Lan, H.F., Wang, H.Z., Ling, H.F., et al., 2020. Petrogenesis of the Chashan Granite in the Northern Guangdong Province and Its Implication for the Metallogenic Potential of Tungsten and Uranium Mineralization. Acta Geologica Sinica, 94(4): 1143-1165(in Chinese with English abstract). doi: 10.3969/j.issn.0001-5717.2020.04.009 Matin, S.S., Chelgani, S.C., 2016. Estimation of Coal Gross Calorific Value Based on Various Analyses by Random Forest Method. Fuel, 177: 274-278. https://doi.org/10.1016/j.fuel.2016.03.031 Nicodemus, K.K., Malley, J.D., 2009. Predictor Correlation Impacts Machine Learning Algorithms: Implications for Genomic Studies. Bioinformatics, 25(15): 1884-1890. https://doi.org/10.1093/bioinformatics/btp331 Rodriguez-Galiano, V., Sanchez-Castillo, M., Chica-Olmo, M., et al., 2015. Machine Learning Predictive Models for Mineral Prospectivity: An Evaluation of Neural Networks, Random Forest, Regression Trees and Support Vector Machines. Ore Geology Reviews, 71: 804-818. https://doi.org/10.1016/j.oregeorev.2015.01.001 Shao, F., Xu, J.J., Shao, S., et al., 2014. Geological Characteristics and Mineralization of the Granite-Type Uranium Deposits in South China. Resources Survey and Environment, 35(3): 211-217(in Chinese with English abstract). Song, Y., Huang, J., Zhou, D., et al., 2007. IKNN: Informative K-Nearest Neighbor Pattern Classification. In: Kok, J.N., Koronacki, J., Lopez de Mantaras, R., eds., Knowledge Discovery in Databases: PKDD 2007, PKDD 2007. Lecture Notes in Computer Science, 4702. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74976-9_25 Strobl, C., Boulesteix, A.L., Kneib, T., et al., 2008. Conditional Variable Importance for Random Forests. BMC Bioinformatics, 9: 307. https://doi.org/10.1186/1471-2105-9-307 Sun, Z.Y., Liu, H.Y., Ju, H.Y., et al., 2021. Assessment of Importance-Based Machine Learning Feature Selection Methods for Aggregate Size Distribution Measurement in a 3D Binocular Vision System. Construction and Building Materials, 306: 124894. https://doi.org/10.1016/j.conbuildmat.2021.124894 Tian, Z.J., 2014. Comparative Study on Chronology, Geochemistry and Mineralogical Characteristics of Uranium-Producing and Uranium-Non-Producing Granites in Zhiguang Mountain (Dissertation). China University of Geosciences, Beijing(in Chinese with English abstract). Vincenzi, S., Zucchetta, M., Franzoi, P., et al., 2011. Application of a Random Forest Algorithm to Predict Spatial Distribution of the Potential Yield of Ruditapes Philippinarum in the Venice Lagoon, Italy. Ecological Modelling, 222(8): 1471-1478. https://doi.org/10.1016/j.ecolmodel.2011.02.007 Wang, H.Z., Yang, F., Luo, Z.Y., 2016. An Experimental Study of the Intrinsic Stability of Random Forest Variable Importance Measures. BMC Bioinformatics, 17: 60. https://doi.org/10.1186/s12859-016-0900-5 Wang, K.X., Sun, T., Yu, J.H., et al., 2020. Provenances of the Ediacaran Sedimentary Rocks in the Zhuguangshan Area and Their Implications for Granitoid-Related Uranium Mineralization in South China. Ore Geology Reviews, 124: 103588. https://doi.org/10.1016/j.oregeorev.2020.103588 Wu, H., Xia, Y., Zhou, K.K., et al., 2020. Highly Fractionated Granite Magmas maybe the Main Uranium Source of Granite-Type Uranium Deposits in South China: Evidence from the Uranium Content of Zircon in Southern Zhuguangshan Granitic Composite. Acta Petrologica Sinica, 36(2): 589-600 (in Chinese with English abstract). doi: 10.18654/1000-0569/2020.02.16 Xiao, Z.H., Xiong, S.B., Li, C.H., et al., 2020. Types of Uranium Deposits in Central Zhuguang Mountains in Hunan Province, South China and Their Metallogenic Regularity and Prospecting Directions. China Geology, 3(3): 411-424. https://doi.org/10.31035/cg2020040 Yin, X.Z., 2009. An Ameliorated SVM Classifying Algorithm Combined with KNN. Journal of Image and Graphics, 14(11): 2299-2303(in Chinese with English abstract). doi: 10.11834/jig.20091117 Youn, E., Jeong, M.K., 2009. Class Dependent Feature Scaling Method Using Naive Bayes Classifier for Text Datamining. Pattern Recognition Letters, 30(5): 477-485. https://doi.org/10.1016/j.patrec.2008.11.013 Zhang, B.Y., Sun, J.K., Luo, X., et al., 2019. Data Analysis of Major and Trace Element of Gabbro Clinopyroxene from Different Tectonic Setting. Earth Science Frontiers, 26(4): 33-44(in Chinese with English abstract). Zhang, L., 2017. Mineralogical Characteristics of Representative Yanshanian Granites in Southern Zhuguang and Their Implications for Petrogenesis and Metallogenic Potential (Dissertation). Nanjing University, Nanjing(in Chinese with English abstract). Zhang, L., Chen, Z.Y., Li, S.R., et al., 2017. Isotope Geochronology, Geochemistry, and Mineral Chemistry of the U-Bearing and Barren Granites from the Zhuguangshan Complex, South China: Implications for Petrogenesis and Uranium Mineralization. Ore Geology Reviews, 91: 1040-1065. https://doi.org/10.1016/j.oregeorev.2017.07.017 Zhang, L., Chen, Z.Y., Li, X.F., et al., 2018. Zircon U-Pb Geochronology and Geochemistry of Granites in the Zhuguangshan Complex, South China: Implications for Uranium Mineralization. Lithos, 308-309: 19-33. https://doi.org/10.1016/j.lithos.2018.02.029 Zhang, Q., Zhou, Y.Z., 2017. Big Data will Lead to a Profound Revolution in the Field of Geological Science. Chinese Journal of Geology, 52(3): 637-648(in Chinese with English abstract). Zheng, Z.Y., 2019. Comparison of Several Machine Learning Methods for Identification of Multiple Geochemical Anomalies in Helong Area, Jilin Province (Dissertation). Jilin University, Changchun (in Chinese with English abstract). Zhou, Y.Z., Chen, S., Zhang, Q., et al., 2018a. Advances and Prospects of Big Data and Mathematical Geoscience. Acta Petrologica Sinica, 34(2): 255-263(in Chinese with English abstract). Zhou, Y.Z., Wang, J., Zuo, R.G., et al., 2018b. Machine Learning, Deep Learning and Python Language in Field of Geology. Acta Petrologica Sinica, 34(11): 3173-3178(in Chinese with English abstract). Zhou, Y.Z., Li, P.X., Wang, S.G., et al., 2017. Research Progress on Big Data and Intelligent Modelling of Mineral Deposits. Bulletin of Mineralogy, Petrology and Geochemistry, 36(2): 327-331, 344 (in Chinese with English abstract). doi: 10.3969/j.issn.1007-2802.2017.02.016 Zhou, Y.Z., Zuo, R.G., Liu, G., et al., 2021. The Great-Leap-Forward Development of Mathematical Geoscience during 2010—2019: Big Data and Artificial Intelligence Algorithm are Changing Mathematical Geoscience. Bulletin of Mineralogy, Petrology and Geochemistry, 40(3): 556-573, 777(in Chinese with English abstract). Zhu, B., 2010. Research on Mantle Fluid and Uranium Mineralization (Dissertation). Chengdu University of Technology, Chengdu(in Chinese with English abstract). Zuo, R.G., Peng, Y., Li, T., et al., 2021. Challenges of Geological Prospecting Big Data Mining and Integration Using Deep Learning Algorithms. Earth Science, 46(1): 350-358(in Chinese with English abstract). 陈永良, 周斌, 李学斌, 2012. 基于Boltzmann机的矿产靶区预测. 地球物理学进展, 27(1): 179-185. 董师师, 黄哲学, 2013. 随机森林理论浅析. 集成技术, 2(1): 1-7. 郝慧珍, 顾庆, 胡修棉, 2021. 基于机器学习的矿物智能识别方法研究进展与展望. 地球科学, 46(9): 3091-3106. doi: 10.3799/dqkx.2020.360 洪瑾, 甘成势, 刘洁, 2018. 基于机器学习的岩石微量元素与主量元素关系初探: 以洋岛玄武岩中锆元素为例. 地质科学, 53(4): 1285-1299. 兰鸿锋, 王洪作, 凌洪飞, 等, 2020. 粤北茶山岩体岩石成因与铀、钨成矿潜力探讨. 地质学报, 94(4): 1143-1165. doi: 10.3969/j.issn.0001-5717.2020.04.009 邵飞, 许健俊, 邵上, 等, 2014. 华南花岗岩型铀矿地质特征及成矿作用. 资源调查与环境, 35(3): 211-217. doi: 10.3969/j.issn.1671-4814.2014.03.009 田泽瑾, 2014. 诸广山产铀与不产铀花岗岩的年代学, 地球化学及矿物学特征对比研究(硕士学位论文). 北京: 中国地质大学. 伍皓, 夏彧, 周恳恳, 等, 2020. 高分异花岗岩浆可能是华南花岗岩型铀矿床主要铀源: 来自诸广山南体花岗岩锆石铀含量的证据. 岩石学报, 36(2): 589-600. 殷小舟, 2009. 一种改进的结合K近邻法的SVM分类算法. 中国图象图形学报, 14(11): 2299-2303. doi: 10.11834/jig.20091117 郑泽宇, 2019. 吉林省和龙地区多元地球化学异常识别的几种机器学习方法比较(硕士学位论文). 长春: 吉林大学. 章宝月, 孙建鹍, 罗熊, 等, 2019. 三类构造背景辉长岩单斜辉石主量元素和微量元素的数据分析研究. 地学前缘, 26(4): 33-44. 张丽, 2017. 诸广南部燕山期代表性花岗岩的矿物学特征及对岩石成因和成矿潜力的指示意义(硕士学位论文). 南京: 南京大学. 张旗, 周永章, 2017. 大数据正在引发地球科学领域一场深刻的革命: 《地质科学》2017年大数据专题代序. 地质科学, 52(3): 637-648. 周永章, 陈烁, 张旗, 等, 2018a. 大数据与数学地球科学研究进展: 大数据与数学地球科学专题代序. 岩石学报, 34(2): 255-263. 周永章, 王俊, 左仁广, 等, 2018b. 地质领域机器学习、深度学习及实现语言. 岩石学报, 34(11): 3173-3178. 周永章, 黎培兴, 王树功, 等, 2017. 矿床大数据及智能矿床模型研究背景与进展. 矿物岩石地球化学通报, 36(2): 327-331, 344. 周永章, 左仁广, 刘刚, 等, 2021. 数学地球科学跨越发展的十年: 大数据、人工智能算法正在改变地质学. 矿物岩石地球化学通报, 40(3): 556-573, 777. 朱捌, 2010. 地幔流体与铀成矿作用研究(博士学位论文). 成都: 成都理工大学. 左仁广, 彭勇, 李童, 等, 2021. 基于深度学习的地质找矿大数据挖掘与集成的挑战. 地球科学, 46(1): 350-358. doi: 10.3799/dqkx.2020.111 -
点击查看大图
图(11) / 表(1)
计量
- 文章访问数: 449
- HTML全文浏览量: 465
- PDF下载量: 71
- 被引次数: 0
