论文部分内容阅读
结合隧道施工的特点,提出了基于径向量基函数神经网络修正的模糊期望值决策法。首先采用三角模糊数的形式给出评价指标的取值以及评价者的主观经验值,基于模糊期望值决策法得到隧道施工安全的评价期望值;然后构造适用于高维输入的径向量基函数神经网络算法,建立网络自组织调整隐节点优化规则,采用RBF神经网络修正模糊决策得到期望值,从而建立了RBF神经网络修正模糊期望值的安全评价方法。从安全管理、环境条件等8个方面建立了隧道工程安全评价指标体系。结合工程案例,运用该方法对隧道工程的施工安全进行评价。结果表明,总体上该方法与模糊评价法结果一致,但更具合理性和准确性。
Combining the characteristics of tunnel construction, a fuzzy expectation value decision method based on RBF neural network is proposed. Firstly, the value of evaluation index and the subjective experience value of the evaluator are given by using the triangular fuzzy number, and the evaluation expectation of tunnel construction safety is obtained based on the fuzzy expectation value decision method. Then, a radial basis function neural network algorithm suitable for high-dimensional input is constructed , The establishment of network self-organization to adjust the hidden node optimization rules, the use of RBF neural network to correct the fuzzy decision to obtain the expected value, thus establishing a fuzzy RBF neural network to amend the expected value of the safety evaluation method. From the aspects of safety management and environmental conditions, the tunnel project safety evaluation index system has been established. Combined with the project case, this method is used to evaluate the construction safety of the tunnel project. The results show that, in general, the method is consistent with the fuzzy evaluation method, but more reasonable and accurate.