Abstract：The gold mining resource holds significant economic and financial value, providing precious metal resources for the country, driving economic growth, and enhancing currency stability and hedging capabilities in the international financial market. However, while precise, the current chemical analysis methods for measuring gold ore grades in mines face issues such as long processing times, high costs, and reagent pollution, hindering the automation of ore grade and beneficiation method adjustments based on real-time grade information. In contrast, due to its efficiency, eco-friendliness, and in-situ measurement advantages, visible-near infrared spectroscopy is gradually becoming an effective alternative for estimating metal grades in mining areas. First, the raw spectral data were processed using Savitzky-Golay (SG) smoothing to reduce noise, and the spectral characteristics of gold ores were analyzed. It was found that reflectance correlates with gold grade, and a gold absorption feature is present at 455 nm. Based on this finding, dimensionality reduction was performed on the raw spectral data using principal component analysis (PCA), isometric feature mapping (ISOMAP), and locally linear embedding (LLE), with the resulting dimensions reduced to 6, 5, and 5, respectively. Finally, prediction models for gold grade were established using random forest (RF), extremely randomized trees (ET), decision trees (DT), gradient boosting decision tree (GBDT), adaptive boosting (Adaboost), extreme gradient boosting (XGBoost), and stacking ensemble learning algorithms on the dimensionally reduced data.Results indicated that the Stacking ensemble learning method outperformed single models in all aspects. Among them, the LLE-Stacking combined model achieved the highest accuracy, withR2 of 0.972, RPD of 5.935, and an average relative error of 0.231 between predicted and actual values. The method proposed in this study allows for rapid and accurate predictions of gold content in ore, significantly improving the inversion accuracy compared to traditional models, providing new technological means for the rapid and in-situ measurement of gold grades in mines, and holding great significance for efficient gold extraction.
毛亚纯，夏安妮，曹 旺，刘 晶，文 杰，贺黎明，陈煊赫. 基于高光谱数据和Stacking集成学习算法的金矿品位快速反演[J]. 光谱学与光谱分析, 2025, 45(07): 2061-2067.
MAO Ya-chun, XIA An-ni, CAO Wang, LIU Jing, WEN Jie, HE Li-ming, CHEN Xuan-he. Rapid Inversion of Gold Ore Grades Based on Hyperspectral Data and Stacking Ensemble Learning Algorithm. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(07): 2061-2067.
[1] Shang J, Hamori S. Energy Economics, 2024, 132: 107473.
[2] Mao Y C, Huang J Q, Cao W, et al. Infrared Physics & Technology, 2021, 115: 103717.
[3] Xie H F, Mao Z Z, Xiao D, et al. Measurement, 2022, 203: 112003.
[4] Wang Y Y, Niu R Q, Lin G, et al. Environmental Geochemistry and Health, 2023, 45(12): 9103.
[5] Li X, Wen J, Fu Y, et al. Infrared Physics & Technology, 2022, 123: 104170.
[6] Wang J, Gao J. Chemometrics and Intelligent Laboratory Systems, 2025, 257: 105308.
[7] Xu M Y, Mao Y C, Zhang M Q, et al. Remote Sensing, 2023, 15(16): 4100.
[8] Cui S C, Zhou K F, Ding R F, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2022, 275: 121190.
[9] Wu M H, Dou S, Lin N, et al. Remote Sensing, 2023, 15(19): 4713.
[10] Tan K, Ma W B, Chen L H, et al. Journal of Hazardous Materials, 2021, 401: 123288.
[11] Lin N, Jiang R Z, Li G Z, et al. Ecological Indicators, 2022, 143: 109330.
[12] Ouyang H K, Tang L L, Ma J L, et al. Plants, 2024, 13(8): 1163.
[13] Rahman L-U，Shah A, Qureshi R, et al. Advances in Materials Science and Engineering, 2015, 2015: 638629.
[14] Anowar F, Sadaoui S, Selim B. Computer Science Review, 2021, 40: 100378.
[15] Wang J L, Gao J X, Huang J Q, et al. Canadian Journal of Remote Sensing, 2022, 48(5): 592.
[16] Fu B L, He X, Yao H, et al. International Journal of Applied Earth Observation and Geoinformation, 2022, 112: 102890.
[17] Xie B H, Chen B J, Ma J F, et al. Food Analytical Methods, 2024, 17(3): 416.
