灰色预测理论在船舶机械故障诊断中的应用

摘要

In order to improve the accuracy of the grey prediction model in the mechanical fault diagnosis of ships,the tradition model of GM (1,1) is established,and its deficiencies are listed.To overcome the deficiencies,the improved Euler algorithm is applied to the solution of the grey prediction model.Through verification,the absolute correlation degree of the improved model is 0.995,the variance ratio is 12.61％,and the small error probability is 100％,which meet the level of precision grade 1.Combining the oil spectroscopic analysis and the engineering threshold setting,the improved model is applied to the reliability test of the comprehensive transmission device of a ship.According to the variation of Fe mass concentration in oil,the abnormal symptom information of its fault is successfully captured,which is effective to prevent the fault occurrence.%为提高利用灰色预测模型在船舶机械故障诊断中的精度,先建立传统的GM(1,1)模型,并找出其不足.针对此不足,提出将改进欧拉算法应用于灰色预测模型的求解.经计算验证,改进的灰色预测模型的绝对关联度为0.995,方差比为12.61％,小误差概率为100％,均符合一级精度等级.结合油液光谱分析和工程阈值制定,将改进的灰色预测模型应用到某船综合传动装置的可靠性检验中,根据预测油液中Fe质量浓度的变化,成功地监测到综合传动装置的故障异常征兆信息,有效地防止了故障的发生.