Prediction of Copper Mineralization Based on Semi-Supervised Neural Network
-
摘要: 将人工智能技术引入成矿预测研究中，可以提高预测效率，挖掘探测数据与结果之间的隐藏信息.利用半监督学习方法对样本构建要求低的优点，结合其在异常识别方面的应用效果，设计了基于分割准则的孤立森林与深度自编码网络的神经网络结构；基于西藏冈底斯地区的化探元素数据，对研究区内的铜矿进行了成矿预测工作，预测结果与已知矿区数据叠加效果较好，说明本文的神经网络结构能够完成成矿远景区的预测工作.Abstract: The introduction of artificial intelligence technology into metallogenic prediction can improve the prediction efficiency and research the hidden information between the exploration data and the results. In this paper, it designs a brand new semi-supervised neural network structure based on isolation forest with split-selection criterion(SCiForest) and autoencoder, using the advantage of semi-supervised learning method that the samples are easy to construct and the outstanding application effect in anomaly recognition. Based on the geochemical element data of Gangdise area in Tibet, the copper deposits in the study area is predicted. The prediction results are basically within the known mining area, which shows that the neural network structure in this paper can be used to predict the metallogenic prospective area.
-
Key words:
- copper deposit /
- metallogenic prediction /
- deep learning /
- semi-supervised neural network
-
图 1 基于非监督学习成矿预测流程图
Fig. 1. Flow chart of metallogenic prediction based on unsupervised learning method
图 5 基于SCiForest预测值小于0.4的数据训练的自编码器的迭代训练损失
Fig. 5. Iterative training loss of Autoencoder based on data training with SCiForest prediction value less than 0.4
图 6 自编码器生成对抗网络的重构误差分布
Fig. 6. Reconstruction error distribution of the encoder-decoder GAN
图 7 基于SCiForest预测值小于0.4的数据训练的半监督神经网络预测结果的插值可视化
a.未插值优化；b.插值优化
Fig. 7. Interpolation visualization of prediction results of semi-supervised neural network based on data training with SCiForest prediction value less than 0.4
图 8 预测结果叠加矿区示意
a.基于SCiForest预测值小于0.3的数据训练的半监督神经网络预测结果的插值可视化；b.SCiForest预测值小于0.4；c.SCiForest预测值小于0.5
Fig. 8. Schematic diagrams of prediction results superimposed mining area
图 9 基于DBSCAN的半监督神经网络预测结果的插值可视化
Fig. 9. Interpolation visualization of prediction results of semi-supervised neural network based on DBSCAN
-
[1] Allais, M., 1957.Method of Appraising Economic Prospects of Mining Exploration over Large Territories:Algerian Sahara Case Study.Management Science, 3(4):285-347. https://doi.org/10.1287/mnsc.3.4.285 [2] Cargill, S.M., Clark, A.L., 1978.Report on the Activity of IGCP Project 98.Mathematical Geology, 10(5):7. doi: 10.1007/BF02461973 [3] Carranza, E.J.M., Laborte, A.G., 2015.Data-Driven Predictive Mapping of Gold Prospectivity, Baguio District, Philippines:Application of Random Forests Algorithm.Ore Geology Reviews, 71:777-787. https://doi.org/10.1016/j.oregeorev.2014.08.010 [4] Carranza, E.J.M., Laborte, A.G., 2016.Data-Driven Predictive Modeling of Mineral Prospectivity Using Random Forests:A Case Study in Catanduanes Island (Philippines).Natural Resources Research, 25(1):35-50. https://doi.org/10.1007/s11053-015-9268-x [5] Chen, Y.L., Lu, L.J., Li, X.B., 2014.Application of Continuous Restricted Boltzmann Machine to Identify Multivariate Geochemical Anomaly.Journal of Geochemical Exploration, 140:56-63. https://doi.org/10.1016/j.gexplo.2014.02.013 [6] Cheng, G.X., Niu, R.Q., Zhang, K.X., et al., 2018.Opencast Mining Area Recognition in High-Resolution Remote Sensing Images Using Convolutional Neural Networks.Earth Science, 43(Suppl.2):256-262(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTotal-DQKX2018S2021.htm [7] Hariharan, S., Tirodkar, S., Porwal, A., et al., 2017.Random Forest-Based Prospectivity Modelling of Greenfield Terrains Using Sparse Deposit Data:An Example from the Tanami Region, Western Australia.Natural Resources Research, 26(4):489-507. https://doi.org/10.1007/s11053-017-9335-6 [8] Huang, Y., Fu, J.G., Li, G.M., et al., 2019.Determination of Lalong Dome in South Tibet and New Discovery of Rare Metal Mineralization.Earth Science, 44(7):2197-2206(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTotal-DQKX201907001.htm [9] Jiao, Y.J., Huang, X.R., Li, G.M., et al., 2019.Deep Structure and Mineralization of Zhaxikang Ore-Concentration Area, South Tibet:Evidence from Geophysics.Earth Science, 44(6):2117-2128(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTotal-DQKX201906026.htm [10] LeCun, Y., Bengio, Y., Hinton, G., 2015.Deep Learning.Nature, 521:436-444. https://doi.org/10.1038/nature14539 [11] Li, G.M., Wang, G.M., Gao, D.F., et al., 2002.Perspective of Copper Ore Deposit and Its Exploration Direction in Gangdise Metallogenic Belt, Tibet.Mineral Deposits, 21(Suppl.1):144-147(in Chinese with English abstract). [12] Liu, F.T., Ting, K.M., Zhou, Z.H., 2010.On Detecting Clustered Anomalies Using SCiForest.Machine Learning and Knowledge Discovery in Databases.Springer, Berlin Heidelberg, 274-290. https://doi.org/10.1007/978-3-642-15883-4_18 [13] Liu, W.L., Li, S.T., Sun, D.J., et al., 2000.Prediction of Pore-Fluid Pressure in Deep Strata of Songliao Basin.Earth Science, 25(2):137-142(in Chinese with English abstract). http://www.researchgate.net/publication/304822307_Prediction_of_pore_fluid_pressure_in_deep_strata_of_Songliao_Basin [14] Liu, X.Z., 2007.Copper Resources in China:Today and Tomorrow.Northwestern Geology, 40(1):83-88(in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTotal-XBDI200701008.htm [15] Ma, Z.Z., 2009.Thoughts on the Guarantee of Mineral Resources in China.Land and Resources Information, (3):2-7(in Chinese with English abstract). [16] Moser, G., Serpico, S.B., Benediktsson, J.A., 2013.Land-Cover Mapping by Markov Modeling of Spatial-Contextual Information in Very-High-Resolution Remote Sensing Images.Proceedings of the IEEE, 101(3):631-651. https://doi.org/10.1109/JPROC.2012.2211551 [17] Pan, H.P., Liu, G.Q., 1997.Applying Back-Propagation Artificial Neural Networks to Predict Coal Quality Parameters and Coalbed Gas Content.Earth Science, 22(2):210-214(in Chinese with English abstract). http://www.zhangqiaokeyan.com/academic-journal-foreign_other_thesis/020418035049.html [18] Quan, X.D., Wang, Y.J., Wang, B., et al., 2013.Application of ANN and GIS in Uranium Metallization Prediction:A Case Study of Northern Tarim Basin.Uranium Geology, 29(6):374-379. http://en.cnki.com.cn/Article_en/CJFDTotal-YKDZ201306009.htm [19] Shao, Y.J., He, H., Zhang, Y.Z., et al., 2007.Metallogenic Prediction of Xiangxi Gold Deposit Based on BP Neural Networks.Journal of Central South University (Science and Technology), 38(6):1192-1198. http://en.cnki.com.cn/Article_en/CJFDTOTAL-ZNGD200706032.htm [20] Tessema, A., 2017.Mineral Systems Analysis and Artificial Neural Network Modeling of Chromite Prospectivity in the Western Limb of the Bushveld Complex, South Africa.Natural Resources Research, 26(4):465-488. https://doi.org/10.1007/s11053-017-9344-5 [21] Wang, J., Song, J.W., Chen, M.Q., et al., 2015.Road Network Extraction:A Neural-Dynamic Framework Based on Deep Learning and a Finite State Machine.International Journal of Remote Sensing, 36(12):3144-3169. https://doi.org/10.1080/01431161.2015.1054049 [22] Xiong, Y.H., Zuo, R.G., 2016.Recognition of Geochemical Anomalies Using a Deep Autoencoder Network.Computers & Geosciences, 86:75-82. https://doi.org/10.1016/j.cageo.2015.10.006 [23] Xiong, Y.H., Zuo, R.G., 2018.GIS-Based Rare Events Logistic Regression for Mineral Prospectivity Mapping.Computers & Geosciences, 111:18-25. https://doi.org/10.1016/j.cageo.2017.10.005 [24] Xu, Y.Y., Chen, Z.L., Xie, Z., et al., 2017.Quality Assessment of Building Footprint Data Using a Deep Autoencoder Network.International Journal of Geographical Information Science, 31(10):1929-1951. https://doi.org/10.1080/13658816.2017.1341632 [25] Xu, Y.Y., Wu, L., Xie, Z., et al., 2018.Building Extraction in Very High Resolution Remote Sensing Imagery Using Deep Learning and Guided Filters.Remote Sensing, 10(1):144. https://doi.org/10.3390/rs10010144 [26] Zhao, P.D., 2007.Quantitative Mineral Prediction and Deep Mineral Exploration.Earth Science Frontiers, 14(5):1-10. https://doi.org/10.1007/s11442-007-0020-2 [27] Zuo, R.G., Xiong, Y.H., Wang, J., et al., 2019.Deep Learning and Its Application in Geochemical Mapping.Earth Science Reviews, 192:1-14. https://doi.org/10.1016/j.earscirev.2019.02.023 [28] 程国轩, 牛瑞卿, 张凯翔, 等, 2018.基于卷积神经网络的高分遥感影像露天采矿场识别.地球科学, 43(增刊2):256-262. doi: 10.3799/dqkx.2018.987 [29] 黄勇, 付建刚, 李光明, 等, 2019.藏南拉隆穹窿的厘定及其稀有多金属成矿作用新发现.地球科学, 44(7):2197-2206. doi: 10.3799/dqkx.2019.114 [30] 焦彦杰, 黄旭日, 李光明, 等, 2019.藏南扎西康矿集区深部结构与成矿:来自地球物理的证据.地球科学, 44(6):2117-2128. doi: 10.3799/dqkx.2018.352 [31] 李光明, 王高明, 高大发, 等, 2002.西藏冈底斯铜矿资源前景与找矿方向.矿床地质, 21(增刊1):144-147. https://www.cnki.com.cn/Article/CJFDTOTAL-KCDZ2002S1044.htm [32] 刘文龙, 李思田, 孙德君, 等, 2000.松辽盆地深层孔隙流体压力预测.地球科学, 25(2):137-142. http://www.earth-science.net/article/id/914 [33] 刘小舟, 2007.我国重要有色金属资源——铜矿的现状及展望.西北地质, 40(1):83-88. https://www.cnki.com.cn/Article/CJFDTOTAL-XBDI200701008.htm [34] 麻志周, 2009.我国矿产资源保障问题的思考.国土资源情报, (3):2-7. https://www.cnki.com.cn/Article/CJFDTOTAL-GTZQ200903002.htm [35] 潘和平, 刘国强, 1997.应用BP神经网络预测煤质参数及含气量.地球科学, 22(2):210-214. http://www.earth-science.net/article/id/488 -
下载:
点击查看大图
图(10)
计量
- 文章访问数: 842
- HTML全文浏览量: 107
- PDF下载量: 44
- 被引次数: 0
