模糊权马尔可夫模型在桥梁状况预测中的应用

On application of fuzzy weight Markov model in bridge condition state predict

以马尔可夫链方法为基础,针对桥梁技术状况值为相依随机变量的特点,采取以规范化的各阶自相关系数为权重,用加权马尔可夫链来预测未来技术状况,在此基础上,根据模糊集理论中的级别特征值,计算出具体的技术状况值,最后,以一实桥前20年的检测资料为实例,对该方法进行了具体的应用,并与灰色模型及灰色马尔可夫链模型等的计算结果进行对比分析,证实了该模型的预测精度。

Based on the Markov chain theory. Then this paper presents a method which is called fuzzy Markov chain with weights to predicted the future bridge condition state by regarding the standardized self-coefficients as weights based on the special characteristics of bridge condition state being a dependent stochastic variable. Base on this, a bridge condition state index is obtained by using the level characteristics value of fuzzy sets. This method is applied to a concrete bridge with 20 years detection data. And then, the GM（ 1 ,1 ）and Markov-GM（ 1,1 ）models are applied to the same bridge and get its future condition index. Though the comparison of the results of the three methods, the prediction results of the fuzzy Markov chain model are good.