首页|两阶段注采管损伤仿真识别方法研究

两阶段注采管损伤仿真识别方法研究

扫码查看
注采管的微裂缝或微损伤会逐渐发展为疲劳断裂破坏,给储气库的安全运营带来巨大挑战.于是,以注采管为对象,将改进灰关联度与柔度曲率差相结合形成新的损伤识别指标,提取出流固耦合作用下注采管有限元模型的固有频率和振型,以计算改进灰关联柔度曲率差指标(IGMFC)值,对结构损伤进行定位,并利用改进后的粒子群算法(IPSO)优化极限学习机(ELM)对结构进行损伤定量.结果表明指标IGMFC具有很好的准确度、灵敏度和抗噪性,且优化后的极限学习机量化损伤程度的误差在2%以内,损伤识别结果误差仅为 2.92%,有一定的应用价值.
Research on Simulation Identification Method of Two-Stage Injection-Production Pipe Damage
The micro-cracks or micro-damages of the injection-production pipes will gradually develop into fa-tigue fracture damage,which brings great challenges to the safe operation of the gas storage.Based on this,taking the injection-production pipe as the object,a new damage identification index was formed by combining the improved grey correlation degree with the flexibility curvature difference.The natural frequency and mode shape of the injection-production pipe finite element model under the action of fluid-structure interaction were extracted to calculate the improved grey-related compliance curvature difference index(IGMFC)value for structural damage localization.Addi-tionally,the improved particle swarm optimization(IPSO)was used to optimize the extreme learning machine(ELM)to quantify the structural damage.The results show that the index IGMFC has good accuracy,sensitivity and noise re-sistance,and the error of the optimized extreme learning machine to quantify the damage degree is within 2%,and the error of the damage identification result is only 2.92%,which has certain application value.

Compliance curvature differenceParticle swarm algorithm(PSO)Extreme learning machine(ELM)Injection-production stringDamage identification

骆正山、张轩博、王小完

展开 >

西安建筑科技大学管理学院,陕西 西安 710055

柔度曲率差 粒子群算法 极限学习机 注采管柱 损伤识别

国家自然科学基金陕西省社会科学基金

418775272018S34

2024

计算机仿真
中国航天科工集团公司第十七研究所

计算机仿真

CSTPCD
影响因子:0.518
ISSN:1006-9348
年,卷(期):2024.41(5)
  • 12