Prediction of Electric Load Neural Network Prediction Model for Big Data

Abstract

Rock bursts are one of the most common and hazardous disasters encountered in coal mining operations, particularly in China. The occurrence of rock bursts poses significant risks to the safety of miners and the structural integrity of mining operations. Consequently, coal mines are required to conduct extensive and frequent rock burst risk assessments, which contribute to the increasing workload and operational pressures on mine personnel. Addressing this challenge is critical for enhancing the safety and efficiency of mining activities. In response to this issue, this paper proposes a novel approach to predicting rock bursts by preprocessing electromagnetic radiation (EMR) and acoustic emission (AE) signals to remove noise and interfering signals. The preprocessing step is crucial as it helps in isolating the relevant features that are indicative of potential rock bursts. By extracting both the interference signal features and the signal precursor features, a comprehensive training set for the predictive model is established. To further enhance the accuracy of predictions, the study employs advanced machine learning techniques, specifically random forests and decision trees, to train and analyze the signal features. These models are chosen for their robustness and ability to handle complex data patterns, making them well-suited for the prediction task. The experimental results are promising, demonstrating that the model achieves a high prediction accuracy of 99.39%, with an Area Under the Curve (AUC) value of 0.99, indicating excellent performance. This high level of accuracy suggests that the model is highly effective in predicting rock bursts, thereby offering a valuable tool for improving mine safety and reducing the workload associated with manual risk assessments.