论文部分内容阅读
在桥梁损伤评估系统模型的基础上,提出了基于神经网络的桥梁损伤评估方法。该法通过神经网络对已有的桥梁损伤评估实例的学习,可使神经网络较好地表达评估结果与评价因素之间的关系,从而代替评估群体进行评估。这不仅可以减少评估的工作量,节约人力物力,还可以较好地利用和积累专家经验知识,降低评估过程中人为因素的影响,保证评估结果的客观性。同时,建立的评估系统具有自学习功能,可以在没有专家或缺少专家的条件下,模拟有经验的评估机理,完成对旧桥损伤的评估。实例计算表明,各网络模型花费很少的时间完成对样本的训练后,便可利用训练好的隐含层权值与阈值对实际桥梁进行评估。
Based on the bridge damage assessment system model, a bridge damage assessment method based on neural network is proposed. This method learns the existing example of bridge damage assessment through neural network, which can make the neural network better express the relationship between evaluation results and evaluation factors, so as to replace evaluation groups for evaluation. This not only can reduce the workload of assessment, save manpower and material resources, but also make good use of and accumulation of expert experience, reduce the impact of human factors in the assessment process and ensure the objectivity of assessment results. At the same time, the established evaluation system has the function of self-learning, and can simulate an experienced assessment mechanism and evaluate the damage of the old bridge without experts or lack of experts. The case study shows that the network model can train the actual bridge with the trained hidden layer weights and thresholds after spending a little time training the samples.