风力发电机齿轮箱优化逐层故障诊断方法

风力发电机齿轮箱的故障诊断在风力发电机组正常运行中起着重要作用,除了识别故障类型外,故障的严重程度对风机的维护也具有指导意义,因此,一种优化堆叠诊断结构(OSDS)被提出以识别故障类型和严重性。首先对原始振动信号进行压缩采样,然后将压缩样本分别输入第1层和第2层深度信任网络(DBN),对故障类型和严重性进行识别,同时采用混沌量子粒子群优化算法(CQPSO)对每个DBN进行优化。通过两组实验得到的结果表明,故障类型诊断准确率分别达到99.24%和97.21%,故障严重程度诊断准确率达到99.06%,同时诊断时间仅为1.493和2.176 s。     

A flat microwave photonic filter based on M-Z modulator and fiber Bragg grating

A new multiple-taps and flat microwave photonic filter, which is composed of fiber Bragg grating, M-Z modulator and erbium-doped fiber, is put forward. The flat band-pass or flat band-stop response can be realized by adjusting the coupler＇s factor and the reflectivity of the fiber Bragg grating or the gain of the erbium-doped fiber. The free spectral range of the filter can be tuned by controlling the length of the erbium-doped fiber. The potential and feasibility of the proposed filtering structures have been demonstrated by simulation.