模糊理论在岩质边坡稳定性评价中存在的知识获取和自适应能力较低等方面不足,而神经网络在模糊推理方面欠缺,因此,提出基于自适应神经元模糊推理系统的边坡稳定性评价方法。通过模型结构的建立、模型训练和测试,得到可用于边坡稳定性评价的基于自适应神经元模糊推理系统模型。测试结果表明,该模型计算结果与边坡实际稳定系数十分接近,对边坡稳定性的预测结果也与实际相符。与基于神经网络方法的计算结果比较,该方法在建模简便程度及计算精度等方面明显具有优势。 Fuzzy theory is not enough to acquire knowledge and low adaptability in the evaluation of rock slope stability. However, neural network is lacking in fuzzy reasoning. Therefore, the slope stability based on adaptive neuron fuzzy inference system is proposed Evaluation method. Through the establishment of model structure, model training and testing, an adaptive neuron fuzzy inference system model that can be used for slope stability evaluation is obtained. The test results show that the calculated result of the model is very close to the actual stability coefficient of the slope, and the prediction result of slope stability is also in line with the actual one. Compared with the calculation results based on neural network method, this method has obvious advantages in modeling convenience and calculation accuracy.