论文部分内容阅读
研究了一种超球体平方根无迹Kalman滤波算法用来有效跟踪涡扇发动机气路部件发生渐变性和突变性故障的健康参数.该算法通过超球体单形采样来降低算法的计算量,采用测量残差协方差阵的平方根代替方差阵进行递推运算,提高了算法的计算效率和数值稳定性.分别采用扩展Kalman滤波算法、无迹Kalman滤波算法和超球体平方根无迹Kalman滤波算法对某型涡扇发动机进行仿真,结果表明:超球体平方根无迹Kalman滤波算法的滤波时间减少50%左右,能够实现渐变性和突变性故障中健康参数的准确估计,是一种有效的涡扇发动机气路部件参数估计和故障诊断方法.
A hyper-sphere square root-free kerman filter algorithm is used to effectively track the health parameters of the turbofan engine air-circuit components with gradual and sudden failure.The algorithm reduces the computational complexity of the algorithm by using single-sphere hyperspherical sampling, Residual error covariance matrix instead of the square root of variance recurrence calculation, improve the computational efficiency and numerical stability of the algorithm.Extended Kalman filter algorithm, unscented Kalman filter algorithm and the hypersphere square root no trace Kalman filter algorithm for a certain type The results show that the filtering time of the square root-free Kalman filter of hypersphere is reduced by about 50%, which can accurately estimate the health parameters in the fading and mutation faults. It is an effective way for turbofan engine Part Parameter Estimation and Fault Diagnosis Methods.