Prediction of ash content in coal flotation tailings using multispectral image features: a double-layer random forest method

Key Points

• Ash content prediction achieves a correlation coefficient of 0.92, enhancing flotation efficiency.
• Utilizing multispectral image features, the model improves accuracy in predicting coal tailings composition.
• Double-layer random forest analysis employs advanced machine learning techniques for robust predictions.
• Findings indicate potential for optimizing coal processing methods based on accurate ash measurements.