基于T-S模型的模糊神经网络在边坡稳定评价中的应用

受地质、工程等众多因素的影响，岩土质边坡稳定性具有未确知性、随机性、模糊性、可变性等特点，很难用简单的力学、数学模型描述。提出了用基于Takagi-Sugeno模型的模糊神经网络来对边坡稳定性进行评价，该模型同时兼具神经网络和模糊逻辑二者的优点，既可以比较容易地处理模糊性的实际问题，又具有较好的学习能力。将此模型与BP神经网络模型同时应用于80个实际边坡样本进行训练和预测，结果表明该模型具有预测精度更高、收敛速度更快、预测结果与实际结果吻合度更高的特点

Application of fuzzy-neural network based on T-S model to estimation of slope stability

With the influence of various factors such as geology and engineering etc., the stability of slope engineering is characterized by uncertainness, randomness, fuzziness, and alterableness. Therefore, it is hard to estimate by using simple mechanical and/or mathematic model. It is proposed to solve this problem by using fuzzy-neural network based on Takagi-Sugeno model, which has advantages of neural network and fuzzy logic model. It not only can easily solve the actual engineering problem with fuzziness but also has excellent learning ability. Application results reveal that this model has high accuracy and higher efficiency in convergence compared with the BP neural network model.