基于失效物理的风电机组多故障可靠性建模方法

风电机组由于功能结构复杂且工作环境恶劣导致其退化状态呈现多样性特点,失效形式往往是多种故障耦合作用的结果,因此单独采用某一变量的性能退化过程作为评估风电机组寿命及可靠性模型是不全面的。为了更精确反映系统内部和外部性能指标的相关性,提出了一种基于失效物理的风电机组多故障可靠性建模方法。首先分析了风电机组多应力耦合下的工作环境和故障特性,通过基于失效物理的可靠性仿真分析方法获得寿命分布函数,并将其作为边缘分布函数;其次在考虑故障模式相关性时,引入两阶段估计法对Copula函数的模型进行参数估计。通过对比不同类型Copula函数的秩相关系数和平方欧氏距离,确定最优Copula函数。最后利用Skla定理和逐步分析的思想,建立多故障模式耦合下的简化风电机组可靠性模型。结果表明,考虑可靠性指标相关性时得到的可靠性结果更加合理,同时该方法避免了风电机组寿命预测和可靠性评估的保守估计,具有较好的实用性和可操作性。

A Failure Physics-based Multi-failure Reliability Modeling Approach for Wind Turbines

Due to the complex functional structure and harsh working environment of wind turbines, their degradation states are characterized by diversity, and the failure forms are often the result of the coupling of multiple faults, so it is not comprehensive to adopt the performance degradation process of one variable alone as a model for assessing the life and reliability of wind turbines. In order to more accurately reflect the correlation between internal and external performance indicators of the system, a failure physics-based multi-fault reliability modeling method for wind turbines is proposed. Firstly, the operating environment and fault characteristics of wind turbine under multi-stress coupling are analyzed, and the life distribution function is obtained by the failure physics-based reliability simulation analysis method and used as the marginal distribution function; secondly, the two-stage estimation method is introduced to estimate the parameters of the model of Copula function when considering the failure mode correlation. The optimal Copula function is determined by comparing the rank correlation coefficient and squared Euclidean distance of different types of Copula functions. Finally, a simplified wind turbine reliability model under multi-failure mode coupling is established by using Skla"s theorem and the idea of stepwise analysis. The results show that the reliability results obtained when considering the correlation of reliability indexes are more reasonable, while the method avoids the conservative estimation of wind turbine life prediction and reliability assessment, and has better practicality and operability.