基于参数优化GM-Markov模型的海底管道腐蚀预测

为了避免由海底管道腐蚀导致的穿孔泄漏等事故,及时对海底管道进行维修和防护,利用传统灰色系统处理少数据与贫数据的特点、马尔科夫理论预测未来状态的特点,提出基于参数优化GM-Markov模型的海底管道剩余寿命预测方法。首先分析了灰色GM(1,1)模型构建的可行性,随后建立参数优化的GM(1,1)模型,改变模型初始条件,对海底管道腐蚀深度进行预测。根据预测的腐蚀深度,利用Markov模型对海底管道未来腐蚀状态作定量分析,预测其剩余寿命。以某海底管道试验段为例,预测了该管道的剩余寿命。结果表明:改变初始条件后的模型可在样本数据少的情况下达到更高的预测精度,推广应用性较强。(图3,表6,参31)

Prediction of submarine pipeline corrosion based on parameter optimized GM-Markov model

To avoid accidents such as perforation leakage caused by submarine pipeline corrosion,and to timely maintain and protect submarine pipelines,the prediction method for submarine pipeline residual life based on parameter-optimized GM-Markov model was proposed in combination with the characteristics of traditional grey system to process little data,poor data and Markov theory to predict the future state.The feasibility of building the grey GM(1,1) model was firstly analyzed,then the GM(1,1) model with optimized parameters was built,and the initial conditions of the model were changed to predict the corrosion depth of submarine pipelines.According to the predicted corrosion depth,Markov model was used to quantitatively analyze the future corrosion state of submarine pipelines and predict their residual life.Some submarine pipeline test section was selected as an example to predict the residual life of the pipeline.The results show that the model with the initial conditions changed enables higher prediction accuracy with only a small amount of sample data,demonstrating its worth of popularization and application.