Research on a Drilling Parameter Prediction Model Based on a Hybrid Model of Multi-Layer Perceptron (MLP) and Convolutional Neural Network (CNN)

This paper focuses on the drilling operation conditions at the Hongliulin Coal Mine, Yulin City, Shaanxi Province. A hybrid model combining Multi-layer Perceptron (MLP) and Convolutional Neural Network (CNN), along with a weighted mechanism, is proposed to predict drilling parameters. The Pearson correlation coefficients between the standardized features (drilling depth, torque, and current) and the target variable (pressure) are calculated using correlation analysis code. The results show that the correlation coefficients between drilling depth, torque, current, and drilling pressure are 0.3427, 0.4903, and 0.08279, respectively. Building on this, feature importance is calculated based on the Random Forest algorithm. The results show that the feature importance values of drilling depth, torque, and current with respect to drilling pressure are <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$1,0.6822$</tex>, and 0.4318, respectively. These features are assigned weights of 1, 0.6822, and 0.4318 for predicting drilling pressure, thereby improving the prediction accuracy. The model training and validation are based on drilling data, and the prediction results are quantitatively evaluated. The results show that the constructed hybrid model of Multi-layer Perceptron (MLP) and Convolutional Neural Network (CNN) demonstrates good predictive performance. It can provide data support for coal mine construction and holds significant engineering application value.