论文部分内容阅读
定量分析滑油滤磨屑是提高航空发动机内部磨损故障诊断成功率的有效技术途径。为满足某型航空发动机使用维护中滑油滤磨屑快速定量分析和磨损状态诊断的实际需求,设计制造了油滤磨屑自动清洗收集装置,采用电荷耦合器件(CCD)获取油滤磨屑数字图像,运用二维最大熵遗传算法对磨屑图像进行阈值分割后提取磨屑目标的特征量,将磨屑特征量与反映铁磁性磨屑当量质量的磨损烈度值进行关联,用以确定发动机内部磨损状态。实际分析过程中,首先针对大量已知磨损烈度值的油滤磨屑图像案例样本,通过多元回归分析法确定磨屑图像特征量和磨损烈度值之间的线性关系;然后针对待测磨屑图像样本,运用已得到的回归公式计算出磨损烈度值,与设定的磨损烈度阈值进行比较,实现对发动机磨损状态的分析诊断;最后,通过某型发动机维护过程中滑油滤磨屑定量分析诊断的实际案例验证了所提方法的有效性。
Quantitative analysis of oil filter debris is to improve the aeroengine internal wear failure diagnosis of the success rate of effective technical ways. In order to meet the actual demand of rapid quantitative analysis and wear status diagnosis of lubricating oil filter in use and maintenance of aeroengine, oil filter debris automatic cleaning and collecting device is designed and manufactured, and the charge-coupled device (CCD) Image, using the two-dimensional maximum entropy genetic algorithm for debris image segmentation Threshold extraction of the characteristics of the debris object, the wear debris characteristics and reflect the equivalent quality of wear-resistant ferrite wear intensity values associated with the engine to determine the internal Wear status. In the actual analysis process, a large number of image samples of oil filter debris with known wear intensity values are firstly used to determine the linear relationship between the amount of wear debris image and the feature value of wear debris image by multiple regression analysis. Then, Samples, using the regression formula has been obtained to calculate the value of wear intensity, and set the threshold of wear intensity compared to achieve the diagnosis of the engine wear state diagnosis; Finally, through the maintenance of a certain type of engine oil lubrication quantitative analysis of the diagnosis The actual cases verify the effectiveness of the proposed method.