摘要
为更加方便、快速、准确地预测腐蚀立管的剩余寿命,提高GM(1,1)模型在腐蚀立管剩余寿命预测方面的实用性,对传统的GM(1,1)模型进行非等间距改进并建立腐蚀立管剩余寿命预测模型。将改进的非等间距GM(1,1)模型与中心逼近法、背景值优化法、逐步优化灰导数法、非等距信息逐步优化法、非等距幂预测法和原始优化法进行对比,并进行实例应用。结果表明,改进后的非等间距GM(1,1)模型的预测精度比其他改进的GM(1,1)模型的预测精度高,而且也适用于等间距数据的预测,扩大了GM(1,1)模型的有效预测范围,避免了概率统计法需要大量样本数据的弊端,提高了计算效率。
To be able to predict remaining life of corrosion risers more easily, quickly and accurately, improve GM （ 1,1 ） model practicality in corrosion risers residual life prediction, non-equidistant improve- ment on traditional GM （ 1,1 ）. An improved corrosion riser residual life prediction model, non-equidistant GM （ 1,1 ） model was built. A comparison was made between the model and other methods, such as cen- terapproximation method, the background value optimization method, and gradually optimized gray deriva- tive method, non-isometric information gradual optimization law, non-lsometric power model prediction and original model optimization method. The results show that improved non-equidistant GM （ 1,1 ） model is more accurate in prediction than other improved GM （ 1,1 ） prediction models, and applicable to equally spaced data prediction expands the GM（ 1,1 ） model~ prediction scope, avoids the drawback that probabilistie method requires large amount of sample statistic data, and improves the computational efficiency.
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2014年第6期57-61,共5页
China Safety Science Journal
基金
国家质检公益基金资助(201210026
201310152)
甘肃省财政厅高校基本科研业务费资助(1205ZTC067)
