基于BP神经网络的高含水岩石爆破震动参数预报

Forecast of Blasting Vibration Parameters in High Water Cut Rock based on BP Neural Network

运用现场试验震动监测得出水介质对高含水岩石爆破地震波传播规律的影响,再结合BP神经网络理论,将水介质纳入网络模型,建立爆破震动参数预报的BP神经网络模型。采用高含水岩石爆破现场监测数据对网络模型进行训练。把训练达到最优后的预报结果与实测结果作对比,发现BP神经网络模型预报参数与实测值较为接近。

The influence of water medium on the propagation law of blasting vibration in high water cut rock is studied by vibration monitoring in field test. Combining with the BP neural network theory, the water medium is incor- porated into the network model to establish the blasting vibration parameter forecasting model. The network model is trained by using site monitoring data of high water cut rock blasting. The optimum results of BP neural network model after training are compared with the measured data. It is found that the BP neural network model is close to the measured value, which is instructive for such engineering blasting.