Long-Term Prediction Model of Rock Mass Displacement Based on Chaotic-Dynamic Time Delay BP Neutral Network

The chaos-dynamic time delay BP neural network model is built to realize long-term prediction of rock mass displacement of large underground grave, and fast analyze long-time stability of surrounding rock through optimized structure of BP neural network coupling with chaotic-dynamic parameters of displacement. Embedding dimension m is set as the number of input layer, and predicting feedback mode of neural network is added, and prediction training samples are generated dynamically, so long-term forecasting are enabled. The relative bigger delay time τ is selected, then time delay between adjacent phase points is hτ when forecasting step is h, so the long-time prediction is realized by limited number of forecasting steps. The instances show that computational stability of built prediction model is preferable with faster calculating speed, and prediction precision is all within 10% when predicting step h is not more than 5, and the number of forecasting displacements is not more than 10. Therefore the forecasting results are real time and effective, the long-term prediction of rock mass displacement is realized.