基于GM(1,1)-Markov模型的桥梁耐久性预测

Research on GM(1,1)-Markov Model for Prediction of Bridge Durable Condition

为了使桥梁管养部门能够以最低的成本进行养护维修,提出了将GM(1,1)-Markov模型应用于桥梁耐久性预测中,GM(1,1)-Markov模型将灰色理论与马尔可夫链有机结合起来,利用二者的优点充分反映桥梁耐久性状况的整体退化趋势与呈现波动的特性,最大限度地利用历史信息,解决了传统的退化模型和概率模型无法考虑不确定因素对桥梁耐久性状况的影响和局限性问题。最后通过对江苏省高速公路上某桥的耐久性状况进行预测分析,介绍了其在项目实践中的应用。

With increasing age, the bridge durable condition prediction is becoming essential for bridge management authority to enable bridge to be maintained at minimum overall cost. The GM（ 1 , 1）-Markov Model is a model combining Markov Model with Grey Model（1,1）. When predicting, GM（ 1,1 ） represents the total deterioration tendency of bridge. Markov Model will take into account other uncertain factors leading to bridge condition improvement such as maintenance. This model, to some extent, settled the limitation of conventional regression model only for total tendency and probabilistic model merely used in bridge stock. Finally condition predicting and gets good results.