|
| Quantitative prediction model of polymetallic mineral resources under loess overlying area based on big data |
|
Received:November 15, 2022
Revised:December 06, 2022
Accepted:December 07, 2022
Published Online:November 01, 2023
|
| View Full Text View/Add Comment Download reader |
| DOI： |
| KeyWord:Big data; Loess covered area; Metal mineral resources; Quantitative prediction; Favorable metallogenic information; Nonnegative matrix factorization |
| Author | Institution |
| Zhang Yueqiu |
JIANGSU GEOLOGICAL BUREAU |
| Niu Lin |
JIANGSU GEOLOGICAL BUREAU |
| Ju Weiwei |
JIANGSU GEOLOGICAL BUREAU |
| Liu Xiaohu |
JIANGSU GEOLOGICAL BUREAU |
| XIAO E |
JIANGSU GEOLOGICAL BUREAU |
|
| Hits: 1500 |
| Download times: 632 |
| Abstract: |
| In order to reduce the difficulty of mining polymetallic ore under loess overlying area, a quantitative prediction model of polymetallic mineral resources under loess overlying area based on big data was proposed to reasonably predict the distribution of metal mineral resources in hidden mining area. Build the model of quantitative prediction of mineral resources, to the model input 2 d GIS layer with 3 d geological model, entity mineralization favorable information extracted from input of huge amounts of data, using principal component analysis algorithm to multi-type mineralization favorable information for the parallel processing of unified format, and according to the type of data stored in the corresponding distributed database in turn, A non-negative matrix factorization algorithm is used to add non-negative constraints to the data in the database to realize the decomposition of mineral resources data. The fuzzy C-means classifier based on NMF algorithm is used to classify the data, obtain the best sample data of mineral resources, and complete the quantitative prediction of polymetallic mineral resources under loess overlying area. The experimental results show that the model can accurately predict the metal content and metal density of each measuring point in the polymetallic mining area under the loess overlying area, and can also divide the scale of the target area according to the total metal content, and predict the metal mineral reserves and metallogenic favorable degree under each measuring point. |
| Close |
|
|
|