To address the problem, it is difficult to extract fault features for neutral-point-clamped (NPC) three-level inverters under nonstationary conditions. This article proposes a multi-information feature fusion diagnosis method based on attention collaborative stacked long short-term memory (ASLSTM) neural networks. First, parallel structural stacked LSTM networks are constructed, which are used to automatically extract features from multisource time-series data. Then, the attention mechanism is applied to weight these features adaptively. Finally, the fault is identified by features that integrate multiple sources of information. In addition, the quantum particle swarm optimization (QPSO) algorithm is used to intelligently tune the hyperparameters of ASLSTM to improve the reasonableness of hyperparameter selection for the diagnostic model. The simulation results of the fault diagnosis of the NPC three-level inverter circuit show that the proposed method is able to extract high-discriminative features from raw data and has better diagnostic results under various conditions compared with other schemes.
In practical fault diagnosis, the monitoring fault data is accumulated incrementally, it is necessary to detect the newly added fault data. To this end, this paper proposed a broad residual network (BRES) fault diagnosis method with incremental learning capability. Firstly, the deep feature representation of the raw data is obtained by the residual network, and the obtained features and corresponding labels are then updated to the BLS. For the newly collected data, the incremental learning of new fault modes is achieved by automatic feature extraction of the ResNet and the node expansion of the BLS. The effectiveness of the proposed method is verified by motor-driven converters fault diagnosis. Experimental results indicate that the method can effectively update the diagnosis model to incrementally learn new fault categories and new fault modes.