超临界锅炉高温管内氧化皮堆积量快速探测方法  

Rapid Detection Method of Oxide Skin Scale Accumulation in High Temperature Tube of Supercritical Boiler

作　　者：林世治[1] LIN Shi-zhi(Fujian College of Water Conservancy and Electric Power,Sanming Fujian 366000,China)

机构地区：[1]福建水利电力职业技术学院,福建三明366000

出　　处：《粘接》2020年第1期155-158,共4页Adhesion

摘　　要：超临界及超超临界机组管道的金属氧化已成为影响热电厂安全运行的主要因素之一,而对氧化皮堆积量的快速、精准的测量是十分有必要的。论文通过对伴生的氧化皮外加强稳磁场,找出磁感应强度变化与管内相应氧化皮堆积状况的耦合关系,提出一种全新的氧化皮快速无损探测方法。为提高检测方法的快速性及鲁棒性,论文将Fletcher和Reeves(FR)共轭梯度法通过有限的标量因子和动态步长进行改进,并分别采用平均值误差(MAE),均方根误差(RMSE),平均偏差(MBE)和威尔莫特指数(D)为指标对所建立的测量模型进行了验证。结果表明,改进的非线性共轭梯度法在检测精度及鲁棒性方面均具有较好的性能,而且具有快速收敛的特性。Metal oxidation of the pipelines of supercritical and ultra-supercritical units has become one of the main factors affecting the safe operation of thermal power plants,which is necessary to measure and predict the oxide skin scale accumulation at the right time.In this paper,by stabilizing the magnetic field outside the associated scale,the coupling relationship between the change in magnetic induction intensity and the corresponding scale accumulation in the tube is found,and a new rapid and non-destructive method for scale detection is proposed.The main work of this paper is to improve the conjugate gradient method of Fletcher and Reeves(FR) through limited scalar factor and dynamic step size to improve the fastness and robustness of the detection method.The model is established by using the mean absolute error(MAE),root mean square error(RMSE),mean bias error(MBE) and Wilmot index(D) as the indexes respectively.The results show that the improved nonlinear conjugate gradient method has good performance in prediction accuracy and robustness with fast convergence speed.

关 键 词：超临界锅炉 氧化皮 共轭梯度法 测量模型 

分 类 号：TK229.2[动力工程及工程热物理—动力机械及工程]