X80管线钢疲劳裂纹扩展速率声发射预测方法
Acoustic emission prediction method for fatigue crack propagation rate of X80 pipeline steel
摘要
[目的]随着我国油气管网规模的不断扩大和管线材料的老化,疲劳裂纹引发的失效事故日益频发,对管道安全运行构成严峻挑战.现有疲劳裂纹监测技术在实时评估裂纹演化行为方面存在不足.[方法]为此,该文提出一种融合声发射传感技术与贝叶斯统计建模的 X80管线钢疲劳裂纹扩展速率预测方法.通过开展三点弯曲疲劳裂纹扩展试验,研究 X80管线钢在不同应力比和最大载荷条件下的裂纹扩展行为,并分析裂纹在稳态与失稳扩展阶段内的声发射应力波特征.在此基础上,通过贝叶斯统计建模方法构建一种基于声发射信号幅度变化率特征的疲劳裂纹扩展速率预测模型.[结果]试验结果表明,该模型能够在无需依赖应力强度因子幅值的先验信息的条件下,准确判定裂纹从稳态到失稳阶段的转折点,并精确预测裂纹在稳态扩展阶段下的裂纹扩展速率和扩展长度,预测值与实际测量值的平均误差落在 95%置信区间内.[结论]该方法可为声发射技术在油气管道疲劳裂纹动态监测中的应用提供有力的技术支持,进一步提升其在结构健康监测中的应用潜力.
[Objective]With the expansion of China's oil and gas pipeline networks and the aging of pipeline materials,fatigue crack-induced failure incidents have become increasingly frequent,posing severe challenges to pipeline safety.Current technologies exhibit limitations in monitoring crack propagation behavior in real-time.[Methods]This paper proposes a method for predicting the fatigue crack propagation rate(FCPR)of X80 pipeline steel by integrating acoustic emission(AE)sensing with Bayesian statistical modeling.Through three-point bending fatigue crack propagation tests,the crack propagation behavior of X80 pipeline steel was investigated under varying loading conditions.The characteristics of the AE signal were analyzed across both stable and unstable crack propagation stages.A Bayesian statistical model was developed to predict the FCPR,leveraging AE signal features related to the rate of change in amplitude.[Results]Experimental results indicate that this model can accurately identify the transition point from stable to unstable crack propagation without prior information on the amplitude of the stress intensity factor.It precisely forecasts the crack propagation rate,with the predicted values falling within the 95%confidence interval of actual measured values.[Conclusion]This method provides robust technical support for the application of AE technology in the monitoring of fatigue cracks in oil and gas pipelines,further enhancing its potential in structural health monitoring.
年,卷(期):
2026, 52(9)
分类号:
TB9
TP2
机标分类号:
TG156
U661.41
TH162
在线出版日期:
2026年10月14日
基金项目:
62103082:国家自然科学基金2024JC-YBQN-0545:陕西省自然科学基础研究计划资助项目:中国石油集团工程股份有限公司关键技术2023ZZ30:装备研发项目,