基于动态作业载荷变化率的设备健康指标预测模型

摘要

In order to reduce the influence of the dynamic operational load on the result of the health index prediction,a health index prediction model is established based on operational load change rate.The gradient descent method is used to segment the health index curve,and establish functional dependence of deterioration acceleration of health index on change rate of operational load.And a health index prediction model is established based on this function.It is shown by the result of comparison between the predictive value of the health indices with the model presented and that with degraded hidden semi-Markov model that the method presented in this article will be even more accurate.%为了降低动态作业载荷对健康指标预测结果的影响,基于作业载荷变化率建立了健康指标预测模型.采用梯度下降法分割健康指标曲线,建立健康指标衰退加速度与作业载荷变化率的函数关系,并在此函数基础上建立健康指标预测模型.将该模型的健康指标预测值与基于退化隐式半马尔科夫模型的健康指标预测值对比,结果表明文中的方法更加准确.