边坡稳定性预测的混沌神经网络方法

CHAOTIC NEURAL NETWORK METHOD FOR SLOPE STABILITY PREDICTION

边坡系统本身的各种参量是不确定的和随机的,在其演化过程中,表现出复杂的非线性行为,会发生一系列的混沌现象。运用现代混沌理论与神经网络方法的基本原理,把混沌理论与神经网络结合起来,建立了边坡稳定性预测的混沌神经网络模型,从新的角度研究边坡稳定性预测问题,对该问题进行了探索性研究。对该理论的建立及预测方法进行了系统的讨论,为该领域的研究提供了完整的技术方法。对64个典型滑坡实例的研究结果表明,混沌神经网络模型具有较高的精度。

The parameters in the slope system are uncertain and random due to the complexity of slope,and a series of chaos will appear in the course of the slope evolution. In order to provide a nonlinear method for slope stability analysis,the chaotic neural network method is established based on the combination of the chaotic theory and the neural network technology. The establishment of the prediction model and the process of application are discussed in detail. The application of the method to 64 cases of slopes shows its high precision.