Copper Content Inversion of Copper Ore Based on Reflectance Spectra and the VTELM Algorithm

Area of Science:

• Geochemistry
• Analytical Chemistry
• Machine Learning

Background:

• Copper is a vital national resource with widespread industrial applications.
• Traditional copper ore analysis methods are often slow and expensive.
• Existing methods present significant drawbacks for efficient resource management.

Purpose of the Study:

• To develop a novel, efficient method for detecting copper content in copper ore.
• To address the limitations of traditional, time-consuming copper detection techniques.
• To leverage spectral information for accurate copper ore analysis.

Main Methods:

• Chemical analysis to determine precise copper content in ore samples.
• Spectrometric testing to acquire detailed spectral data from copper ores.
• Development of a two-hidden-layer extreme learning machine (ELM) algorithm with variable nodes and regularization.

Main Results:

• A feasible method for copper content detection using spectral information was established.
• The proposed ELM algorithm demonstrated enhanced speed and accuracy in copper detection.
• Experimental validation confirmed the effectiveness of the spectral-based approach.

Conclusions:

• Spectral information analysis provides a viable alternative for copper ore content determination.
• The novel ELM algorithm offers a significant improvement over conventional methods.
• This research facilitates faster, more accurate, and cost-effective copper resource assessment.