基于粗糙集-人工神经网络的建筑施工安全评价及应用

本文依据建筑施工过程安全状态特征,从人、机、环境、管理四个方面进行分析,构建了建筑施工安全评价指标体系,并将粗糙集理论与人工神经网络模型两种方法相结合,建立了基于粗糙集-人工神经网络的建筑施工安全评价模型,对建筑施工过程安全状况进行评价,即:首先通过专家评议同粗糙集理论相结合,对评价指标体系中的各指标进行约简,并确定影响建筑施工安全的核心因素;然后将经粗糙集约简后的评价指标采用人工神经网络方法进行处理和计算。通过实际应用验证表明,经过训练后的神经网络所得的预测值同理论值相近,证明此种评价模型能为建筑施工安全评价提供合理的依据。

Safety Evaluation of Construction Technology Based on Rough Sets-neural Network

Safety criteria system for construction technology is established on the basis of the features of construction technology procedure,from the aspects of people,machine,environment and management.The safety evaluation is processed by the combination of rough sets and artificial neural network.At first,the expert evaluation method is combined with rough sets,the elements of safety criteria are reduced,and the core elements that influence the construction safety are determined.Then,the evaluation indexes reduced by rough sets are used for the treatment and calculation by artificial neural network.The results show that the prediction values by artificial neural network after training are similar to theoretical values,indicating that this evaluation model has rationality in the safety evaluation of construction technology.