According to statistic data, machinery faults contribute to largest proportion of High-voltage circuit breaker failures, and traditional maintenance methods exist some disadvantages for that issue. Therefore, based on the wavelet packet decomposition approach and support vector machines, a new diagnosis model is proposed for such fault diagnoses in this study. The vibration eigenvalue extraction is analyzed through wavelet packet decomposition, and a four-layer support vector machine is constituted as a fault classifier. The Gaussian radial basis function is employed as the kernel function for the classifier. The penalty parameter c and kernel parameter δ of the support vector machine are vital for the diagnostic accuracy, and these parameters must be carefully predetermined. Thus, a particle swarm optimization-support vector machine model is developed in which the optimal parameters c and δ for the support vector machine in each layer are determined by the particle swarm algorithm. The validity of this fault diagnosis model is determined with a real dataset from the operation experiment. Moreover, comparative investigations of fault diagnosis experiments with a normal support vector machine and a particle swarm optimization back-propagation neural network are also implemented. The results indicate that the proposed fault diagnosis model yields better accuracy and efficiency than these other models.

In view of the wide application of wind power, it is critically essential to develop solutions to the high fault rate and the long mean time to repair (MTTR). Multi-sensor fusion methods have been widely applied in condition monitoring and anomaly detection of industrial installation. Aimed at the wind turbine driveline, a new method utilizing the deep residual LSTM network with Attention model (ResLSTM-AM) is proposed in this paper. Firstly, data from multi-sensor fusion systems like supervisory control and data acquisition (SCADA) is preprocessed by quartile data cleaning and corresponding selection using the calculation of maximal information coefficient (MIC) to enhance its reliability. Secondly, we establish the neural network by the deep residual network (ResNet), long short-term memory network (LSTM), and attention mechanism (AM) for time series forecasting of wind turbines. The newly added residual connecting architecture in LSTM layers ensures a higher learning rate in feature extraction and provides a more flexible data flow both in forward and backward signals. Additionally, the attention mechanism after a LSTM layer amplifies the influence of vital data. The model gets validated based on historical SCADA data from actual wind turbines containing healthy conditions and anomaly conditions, showing ResNet contributes more. Based on the root-mean-square error (RMSE), the alarming threshold is calculated by an exponential weighted moving average (EWMA). For two chosen variables, the proposed method is confirmed to be efficient and reliable with the accuracy of 0.9945 and 0.9880 in the results of trials and comparisons with other existing models.

Backlash is a key internal excitation on the dynamic response of planetary gear transmission. After the gear transmission running for a long time under load torque, due to tooth wear accumulation, the backlash between the tooth surface of two mating gears increases, which results in a larger and irregular backlash. However, the increasing backlash generated by tooth accumulated wear is generally neglected in lots of dynamics analysis for epicyclic gear trains. In order to investigate the impact of backlash generated by tooth accumulated wear on dynamic behavior of compound planetary gear set, in this work, first a static tooth surface wear prediction model is incorporated with a dynamic iteration methodology to get the increasing backlash generated by tooth accumulated wear for one pair of mating teeth under the condition that contact ratio equals to one. Then in order to introduce the tooth accumulated wear into dynamic model of compound planetary gear set, the backlash excitation generated by tooth accumulated wear for each meshing pair in compound planetary gear set is given under the condition that contact ratio equals to one and does not equal to one. Last, in order to investigate the impact of the increasing backlash generated by tooth accumulated wear on dynamic response of compound planetary gear set, a nonlinear lumped-parameter dynamic model of compound planetary gear set is employed to describe the dynamic relationships of gear transmission under the internal excitations generated by worn profile, meshing stiffness, transmission error, and backlash. The results indicate that the introduction of the increasing backlash generated by tooth accumulated wear makes a significant influence on the bifurcation and chaotic characteristics, dynamic response in time domain, and load sharing behavior of compound planetary gear set.