李汪繁, 徐佳敏, 李思琦. 汽轮机转子等效应力的计算方法对比及预测分析[J]. 动力工程学报, 2022, 42(12): 1198-1205. DOI: 10.19805/j.cnki.jcspe.2022.12.005
引用本文: 李汪繁, 徐佳敏, 李思琦. 汽轮机转子等效应力的计算方法对比及预测分析[J]. 动力工程学报, 2022, 42(12): 1198-1205. DOI: 10.19805/j.cnki.jcspe.2022.12.005
LI Wang-fan, XU Jia-min, LI Si-qi. Comparative and Predictive Analysis on Calculation Methods of Steam Turbine Rotor Equivalent Stress[J]. Journal of Chinese Society of Power Engineering, 2022, 42(12): 1198-1205. DOI: 10.19805/j.cnki.jcspe.2022.12.005
Citation: LI Wang-fan, XU Jia-min, LI Si-qi. Comparative and Predictive Analysis on Calculation Methods of Steam Turbine Rotor Equivalent Stress[J]. Journal of Chinese Society of Power Engineering, 2022, 42(12): 1198-1205. DOI: 10.19805/j.cnki.jcspe.2022.12.005

汽轮机转子等效应力的计算方法对比及预测分析

Comparative and Predictive Analysis on Calculation Methods of Steam Turbine Rotor Equivalent Stress

  • 摘要: 为提高汽轮机转子应力在线监测的计算精度,系统梳理了转子应力场和温度场的简化算法。以某300 MW级汽轮发电机组高压转子为例,采用简化算法得到冷态起动工况下的应力场和温度场,并与有限元软件模拟结果进行对比分析,同时引入修正系数预测分析转子危险部位的等效应力。结果表明:对比有限差分法和解析法的简化算法结果与有限元软件模拟结果,两者的转子外表面温度和中心温度最大偏差均出现在起动初期,外表面等效应力在起动初期和后期存在较大偏差;采用基于交叉验证的支持向量回归方法预测应力时,与模拟结果吻合良好,可满足工程实践的精度需求。

     

    Abstract: To improve the calculation accuracy of online monitoring of steam turbine rotor stress, the simplified calculation methods of the stress field and the temperature field were systematically proposed. Taking the high-pressure rotor of the 300 MW turbo-generator as an example, the temperature field and stress field under cold starting conditions were analyzed by simplified calculation methods and a finite element software. The equivalent stress of the dangerous part was predicted by introducing a correction coefficient. Results show that comparing the results obtained by simplified calculation methods with the results obtained by the finite element software, the maximum deviations of the rotor outer surface and inner temperature appear in the initial stage of cold starting, and the outer surface equivalent stress has a large deviation in the initial stage and later stage of cold starting.The predicted results were in good agreement with the simulation results of finite element software by using the support vector regression(SVR)based on cross-validation, which could meet the accuracy standards of engineering practice.

     

/

返回文章
返回