基于灰色神经网络组合模型的航空发动机磨损趋势预测

航空发动机的磨损机理十分复杂，且受到诸多不确定因素影响，单一预测模型难以对其变化趋势进行有效预测。针对该问题提出了一种BP网络与改进灰色模型相融合的组合预测模型，并引入混沌理论的C—C方法确定BP网络的嵌入参数和时间延时。仿真结果显示，该组合模型相比单一的神经网络模型和灰色模型精度更高，更客观地反映出发动机滑油中金属含量的变化趋势，可为科学制定发动机维修决策提供重要依据。

Grey neural network forecasting method of aero-engine wear trend

The wear mechanism of aero-engine is complex and affected by many complicated factors, there- fore the single model is difficult to forecast the trend effectively. To solve this problem, a combined fore- casting model of improved grey model and neural network is proposed in this paper, in addition, phase space reconstruction is carded out by C-C method to determine the input samples and output samples of the BP network. Results show that the combined model bears higher prediction accuracy than the single model and reflects more objectively the metal content in engine oil. The method is of significance for engine main- tenance decision.