基于非线性有源自回归模型的船用凝汽器故障早期预警

针对故障数据稀缺的现实情况，为实现船用凝汽器性能的准确评估以及灵敏故障早期预警,提出了基于非线性有源自回归模型的故障早期预警方法。考虑到凝汽器参数间存在的时序特性以及非线性特性，本文采用非线性有源自回归模型对参数间关系进行刻画并建立了面向故障早期预警的常模式模型。利用凝汽器物理模型的故障仿真数据进行试验，结果表明提出方法对正常数据和故障早期数据的检测精度分别达到98.13%与100%。对比实验证明了考虑时序特性在船用凝汽器故障早期预警中的必要性。

Fault Early Warning Method for Marine Condenser Based on Nonlinear Autoregressive Model with Exogenous Inputs

Facing the scarcity of actual fault data in practice, a fault early warning method based on a nonlinear autoregressive model with exogenous inputs is proposed to realize accurate performance evaluation and sensitive fault early warning for marine condenser. Considering the time-series information and nonlinear properties among relevant parameters of the marine condenser, the nonlinear autoregressive model with exogenous inputs is adopted to characterize the inherent relationship, and the normal pattern model for fault early warning is established. The feasibility of the proposed method is verified by experiments with data from an established condenser physical model, and the results show that the proposed method achieves 98.13% and 100% detection accuracy for normal data and fault data respectively. The comparison experiment further proves the necessity of the consideration of the time-series information in the marine condenser fault early warning.