超(超)临界锅炉钢的相稳定性与相定量及持久寿命预测方法的研究  

Study on Stability and Quantification of Phases and Creep-rupture Life Prediction Method for(ultra) Super-critical Boiler Steels

作　　者：彭志方[1] 彭芳芳 陈方玉[3] 

机构地区：[1]武汉大学动力与机械学院,武汉市430072 [2]东方锅炉(集团)股份有限公司材料研究所,四川省自贡市643001 [3]武汉钢铁(集团)公司研究院,武汉市430080

出　　处：《电力建设》2009年第12期1-5,共5页Electric Power Construction

基　　金：国家自然科学基金项目(50674072)

摘　　要：复相分离方法可以在电子探针／扫描电镜一能谱分析条件下对超（超）临界锅炉钢块状样品的析出相成分及其数量进行定量分析，以揭示其供货态钢管样品的相稳定性，以及持久断裂试样的相状态，并可反映相成分及其数量参量与持久寿命的对应关系；人工神经网络模型可以预测多元成分与热处理工艺参量及试验条件（温度及应力）改变时，9％～12％Cr铁素体耐热钢的持久寿命。介绍了以上方法的基本原理及其作用，并列举了若干应用实例，期望能为有关工作提供参考。With the use of proposed multi-phase separation method （ MPSM）, compositions and quantities of precipitated phases in （ultra） super-critical boiler steels can be determined for phase quantification analyses under SEM-EPMA/EDS conditions, which can reveal phase stability of asreceived steel tube/pipe samples and phase status of the creep-ruptured samples, as well as reflect the corresponding relations of phase chemistries and quantities to creep-rapture lives of the steel samples investigated. Through the application of artificial neural network （ANN）, creep-rupture lives of 9 - 12% Cr ferritic heat-resistant steels can be predicted with the variation of multi-component alloying systems, heat treatment processing parameters and creep testing conditions （temperature and externally applied stress level）. Principles and functions of the above methods are introduced and some applied examples are listed, which is expected to provide useful references for related work.

关 键 词：相稳定性 相定量 材质评估 持久寿命预测 人工神经网络 

分 类 号：TG115[金属学及工艺—物理冶金]