论文部分内容阅读
针对支持向量机算法存在的不足,研究了一种基于稀疏贝叶斯框架的机器学习方法——相关向量机在航空发动机故障检测中的应用。排气温度是进行发动机监控与故障诊断的重要依据,应用相关向量机对其进行预测。通过仿真实验,证明了相关向量机方法在样本数据较少的情况下只产生了很少的相关向量,并且能够及时准确地预测出发动机排气温度;同时可以使用真实值与预测值的相对误差作为系统是否发生故障的判断依据。
Aiming at the deficiency of support vector machine algorithm, a machine learning method based on sparse Bayesian framework is studied. The application of correlation vector machine in aeroengine fault detection. Exhaust gas temperature is an important basis for engine monitoring and fault diagnosis, and it is predicted by relevant vector machine. Through simulation experiments, it is proved that the correlation vector machine (SVM) method produces only a few relevant vectors with less sample data, and can predict the engine exhaust temperature timely and accurately. At the same time, the relative error between the real value and the predicted value As a basis for judging whether the system is faulty or not.