基于改进小波包结合CS-BP的地面驱动螺杆泵故障诊断

针对目前地面驱动螺杆泵故障诊断存在效率不高、精度不足、损耗资源的问题，提出通过引入功率谱细化的思想改进小波包变换，再结合布谷鸟搜索(cuckoo search, CS)优化反向传播(back propagation, BP)神经网络的诊断方法。首先，通过改进的小波变换对螺杆泵有功功率分解重构得到特征向量；其次，与瞬时流量、进口回压等参数进行归一化处理，作为BP神经网络的输出层信息；再次，使用布谷鸟搜索寻优得到BP神经网络的权值和阈值，建立CS-BP故障诊断模型；最后，应用于螺杆泵不同故障类型的诊断，并通过与目前的主流诊断方法进行诊断效果的分析比较。结果表明，对于螺杆泵不同类型故障诊断的平均精度达到95.6%,对比分析证明了所提方法的可行性与优越性。

Fault diagnosis of ground-driven screw pump based on improved wavelet packet combined with CS-BP

Aiming at the problems of low efficiency, low precision and resource loss in the fault diagnosis of ground-driven screw pump, this paper proposes a diagnosis method of BP neural network optimized by introducing the idea of power spectrum refinement to improve wavelet packet transform and cuckoo search. Firstly, the active power of the screw pump is decomposed and reconstructed by the improved wavelet transform to obtain the feature vector, and then it is normalized with the parameters such as instantaneous flow and inlet back pressure as the output layer information of the BP neural network. Then the weights and thresholds of BP neural network are obtained by cuckoo search, and the CS-BP fault diagnosis model is established. Finally, it is applied to the diagnosis of different fault types of screw pump, and the average accuracy reaches 95.6 %. The feasibility and superiority of the proposed method are proved by comparing with the current mainstream diagnosis methods.