Rockburst state prediction method based on comprehensive CNN-LSTM

Abstract

The invention provides a rockburst state prediction method based on comprehensive CNN-LSTM, and the method comprises the steps: carrying out the phase space reconstruction of a rockburst state variable according to the time sequence data of the rockburst state variable, and obtaining a phase space; inputting the phase space into a convolutional neural network CNN to obtain a time sequence with high-dimensional feature information; inputting the time sequence into a deep learning LSTM model, and carrying out feature time sequence prediction; dividing the rockburst state variable time series data into training set data and test set data, performing learning training on a CNN-LSTM model by using the training set data, and extracting time features evolved by the phase space data to obtain a trained CNN-LSTM model. According to the method provided by the invention, the advantages of the data feature high expression capability shown by the CNN and the deep learning LSTM model in continuity time series data prediction are combined to carry out rock burst state prediction at the t + 1 moment, so that the prediction error is reduced, and the prediction precision is improved.

Description

technical field [0001] The invention relates to the fields of underground excavation engineering and mine excavation engineering, in particular to a rockburst state prediction method based on a comprehensive CNN-LSTM. Background technique [0002] Mine rockburst has always been one of the major safety threats in the mining industry, and its occurrence is highly certain and unpredictable. Accurate prediction of mine rockburst is the basis of prevention and control of ground pressure disasters. Experts and scholars have paid attention to this issue and carried out a lot of related research work. At the same time, many monitoring resources have been invested in rockburst disasters at home and abroad, and a large amount of multi-source heterogeneous monitoring data related to rockbursts has been obtained. [0003] In the field of rockburst and rockburst prediction and early warning, a large number of scholars have carried out a series of prediction studies based on the theory o...

AI Technical Summary

Problems solved by technology

[0006] The present invention provides a rockburst state prediction method based on comprehensive CNN-LSTM, the purpose of which is to solve the problem that the rockburst state prediction error is large and the accuracy is not high

Images