论文部分内容阅读
针对航空发动机非线性、非高斯的特点,提出一种用于航空发动机气路故障诊断的自适应神经网络权值调整粒子滤波(SANNWA-PF)算法。该算法根据粒子分布情况确定分裂和调整的粒子数目,进而根据粒子权重采用正态分布的方式进行分裂,采用反向传插(BP)神经网络进行权值调整,缓解了粒子的退化和贫化,具有更强的自适应性能和跟踪能力。通过一维非线性跟踪模型和航空发动机气路故障诊断仿真研究表明:SANNWA-PF算法具有良好的非高斯性能,相对粒子滤波一维非线性追踪模型估计精度提高约21%,航空发动机气路故障诊断在高斯噪声和非高斯噪声下分别提高约30%和26%,诊断速度分别提高约7倍和10倍。
Aiming at the non-linear and non-Gaussian characteristics of aero-engine, an adaptive neural network weight-adjusted particle filter (SANNWA-PF) algorithm for the fault diagnosis of aeroengine gas line is proposed. The algorithm determines the number of split and adjusted particles according to the distribution of particles, and then divides the particles according to the normal distribution by using the weight of particles. The BP neural network is used to adjust the weights to alleviate the degradation and depletion of particles , With more adaptive performance and tracking capabilities. The results show that the SANNWA-PF algorithm has good non-Gaussian performance and the estimation accuracy of one-dimensional nonlinear tracking model with relative particle filter is improved by about 21%. The aero-engine gas circuit fault The diagnosis increased by about 30% and 26% respectively under Gaussian noise and non-Gaussian noise, and the diagnostic speed increased about 7 times and 10 times respectively.