Experimental Dataset and Source Code for Multi-Model Analysis of Coal Ash Composition and Fusion Characteristics

This dataset contains experimental data on the chemical composition (SiO₂, Al₂O₃, Fe₂O₃, CaO, MgO, SO₃, TiO₂, K₂O, Na₂O, P₂O₅) and flow temperature (FT) of multiple coal ash samples, along with complete Python source code for multi-model analysis.The dataset has been standardized and can be directly used for research on coal ash fusion characteristics prediction, coal-based solid waste resource utilization, CO₂ mineralization, copper-molybdenum separation, and related fields. The source code implements the training, prediction, and performance evaluation of various machine learning models, including Random Forest, XGBoost, ExtraTrees, and GA-BP, with a complete workflow of SHAP feature importance analysis and visualization chart generation, ensuring the reproducibility of research results.Data file description:- coal_ash_data.xlsx: Excel format for manual viewing, filtering, and analysis- coal_ash_data.csv: Original CSV format for code reproduction and batch processing- main.py: Complete Python source code for multi-model analysis, covering the entire process of data reading, preprocessing, model training, result evaluation, and visualization- All files are packaged into a project archive, which can be directly run.This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Please cite properly when using.