Long Short-Term Memory Networks and Bayesian Optimization for Predicting the Time-Weighted Average Pressure of Shield Supporting Cycles

Abstract

Characteristic parameters of shield supporting in fully mechanized mining, especially time-weighted average pressure (TWAP), are crucial for the analysis and prediction of roof weightings in longwall panels. Despite the leap-forward development of underground data collection and transmission, mining and regional correlation analysis of massive shield data remains challenging. In this study, a hybrid machine learning model integrating the long short-term memory (LSTM) networks and the Bayesian optimization (BO) algorithm was developed to predict TWAP based on the setting pressure (SP), revised setting pressure (RSP), final pressure (FP), number of yielding (NY), TWAP in the last supporting cycle (TWAP (last)), and loading rate in each period. Statistical measures including the mean square error and mean absolute error were used to validate and compare the prediction performances of the BP model, the LSTM model, and the BO-LSTM model. Furthermore, sensitivity studies were carried out to evaluate the importance of input parameters. The results show that the BO-LSTM model is robust in predicting TWAP. FP and TWAP (last) are the most important input parameters in TWAP prediction, followed by RSP and NY. Moreover, the total importance scores of loading rates reach 0.229, indicating the necessity of including these parameters into the dataset. The proposed BO-LSTM model is capable of predicting TWAP which serves for shield-roof status intelligent perception.

A Project Risk Prediction Model Based on Bayesian Networks

“…The project risk management model based on Bayesian networks proposed in this article is shown in figure 1. During the inference process, the bucket elimination method is used for inference [8,9]. The main idea is to use the chain product rule and conditional independence at the symbolic level to perform elimination operations on formulas.…”

Research on the Integration Teaching Mode of Traditional Music Elements and College Piano under the Background of Information Technology

“…2) Short-time average over-zero rate When the discrete-time signal adjacent to the two sample points of the positive and negative sign of the sign, called "over zero", that is, at this time, the signal time waveform across the horizontal axis of the zero level. The average over-zero rate can be obtained by counting the number of times the sample point values change sign per unit time [18]. It is defined as:…”

A Daily Air Pollutant Concentration Prediction Framework Combining Successive Variational Mode Decomposition and Bidirectional Long Short-Term Memory Network

“…It is widely recognized that the structure and model parameters of artificial neural networks significantly impact the model's performance. Therefore, the BO algorithm was employed to optimize rarely selected parameters such as the number of neurons in the BiLSTM layer, the number of hidden layers, and the learning rate of optimization [41]. The mean square error (MSE) was adopted as the loss function, and the hyperparameters were optimized within the following ranges: the number of neurons in the BiLSTM layer ranged from 50 to 200, the number of hidden layers ranged from 1 to 4, and the learning rate of the optimizer ranged from 0.001 to 0.1.…”