船用燃机性能降级新型预测模型建立及应用分析

船用燃气轮机气路维护是舰船动力系统健康管理当中必要的环节，采用视情维护的方法是当前十分先进的维护手段，其中对气路降级趋势的预测是维护方法的核心。气路降级数据具有明确的增长趋势与不规律的波动性，符合灰色系统的特征。本文采用无偏灰色预测模型改进灰色马尔科夫模型，在此基础上扩大原始数据的维数，采用灰色关联度模型，将类似于当前序列的参考序列引入，最终构成了一种新的混合预测模型。最后，选择燃气轮机压气机结垢的过程作为实例来验证这一新的模型。通过研究模型参数和预测准确性之间的关系，为实例提出了建议的最佳参数。从预测准确性与对波动性的预测角度与不同的预测模型的比较结果表明，新的模型优于一些其他序列预测模型。

A Novel Prognostic Model for Marine Gas Turbine Degradation and Application Analysis

Marine gas turbine maintenance is a necessary part in the health management of marine power system. Condition-based Maintenance is the most advanced maintenance method for it, in which the prediction of gas path degradation trend is the core. As the degradation data show specific increasing trend with irregular fluctuation, it meets the characteristics of the grey system. In this paper, a new hybrid prediction model is built with the unbiased grey model and the grey correlation model added into the frame of grey Markov model to expend the dimension of the original data. Finally, the fouling process of the gas turbine compressor is chosen as an example to verify the new model. By studying the relationship between the model parameters and the prediction accuracy, the optimal parameters for the model are proposed. The comparison results show that the new model is better than some other sequence prediction models in the prediction accuracy and fluctuation.