基于PRGM-Markov链的管道腐蚀动态预测技术

Dynamic Prediction Technology of Pipeline Corrosion Based on GM-Markov Chain

为研究管道腐蚀发展规律,对管道剩余寿命进行有效预测,结合灰色理论(GM)和马尔科夫(Markov)链模型,从无偏优化、等维信息更新等方面对GM模型进行完善,采用粒子群(PSO)算法对背景权值进行优化,再根据预测时产生的相对误差范围划分状态区间,并引入Markov链模型对预测值进行修正。结果表明,背景权值的选取对于模型预测精度影响最大,优化后的PRGM-Markov链模型预测结果的平均相对误差为1.02%,与RGM (1,1)模型相比,精度提高了47.42%,最大相对误差仅为2.89%,说明该模型可有效改善原灰色模型的预测结果,并适用于中长期存在较大波动的管道腐蚀深度预测。

In order to study the development law of pipeline corrosion and predict the remaining life effectively,combining the Grey Theory Model(GM) and Markov Chain model,the GM model is improved from the aspects of unbiased optimization and equidimensional information update,and the Particle Swarm Optimization(PSO) algorithm is used to optimize the background weight value.Then,the state interval is divided according to the relative error range generated during the prediction,and the Markov Chain model is introduced to correct the predicted value.The results show that the selection of background weight value has the greatest influence on the prediction accuracy of the model.The average relative error of the optimized PRGM-Markov Chain model is 1.02%.Compared with RGM(1,1) model,the accuracy of the optimized PRGM-Markov Chain model is improved by 47.42%,and the maximum relative error is only 2.89%.It shows that this model can effectively improve the prediction results of the Grey model,and is suitable for predicting the corrosion depth of pipelines with large fluctuations in the medium and long term.