Weakly Cemented Sandstone Abrasivity Test Data for Neural Network Modeling

Validation group training data samples for predicting rock abrasivity indices. The dataset was created by Yafeng Yao and published on figshare in April 2026. It is a small dataset at 5.5 KB, stored in an XLS file.

Use Cases

• Train an improved fuzzy stochastic RBF neural network to predict the Cerchar Abrasivity Index (CAI) using inputs like hardness, wave velocity, porosity, and equivalent quartz content.
• Validate model performance for predicting the Cerchar Abrasivity Ratio (CAR) of weakly cemented sandstones.
• Compare prediction accuracy of the improved RBF model against traditional methods like MLR and SVR using metrics such as MSE, MAE, and R-squared.
• Analyze how rock abrasivity changes across different physical states (dry, water-saturated, mud-saturated) using the CAI measurements.
• Study the relationship between microstructure characteristics and the CAI value in different saturation states.

Strengths

• Model validation shows a high coefficient of determination (R²) of 0.967 for predictions.
• The improved RBF model reduced MSE by up to 65.7% and MAE by up to 47.4% compared to baseline models.

Limitations

• The dataset is very small at 5.5 KB, indicating a limited sample size for training and validation.
• Geographic scope is limited to weakly cemented sandstones from western China.
• Specific row count and column details are unknown.

Provenance

Source

Yafeng Yao via figshare.

Collection Method

Data generated from Cerchar abrasivity characteristic tests on rock samples.

Time Range

null

Freshness

Last updated in April 2026.

Geography

Weakly cemented sandstones from western China.