Intelligent fault diagnosis methods based on deep learning becomes a research hotspot in the fault diagnosis field. Automatically and accurately identifying the incipient micro-fault of rotating machinery, especially for fault orientations and severity degree, is still a major challenge in the field of intelligent fault diagnosis. The traditional fault diagnosis methods rely on the manual feature extraction of engineers with prior knowledge. To effectively identify an incipient fault in rotating machinery, this paper proposes a novel method, namely improved the convolutional neural network-support vector machine (CNN-SVM) method. This method improves the traditional convolutional neural network (CNN) model structure by introducing the global average pooling technology and SVM. Firstly, the temporal and spatial multichannel raw data from multiple sensors is directly input into the improved CNN-Softmax model for the training of the CNN model. Secondly, the improved CNN are used for extracting representative features from the raw fault data. Finally, the extracted sparse representative feature vectors are input into SVM for fault classification. The proposed method is applied to the diagnosis multichannel vibration signal monitoring data of a rolling bearing. The results confirm that the proposed method is more effective than other existing intelligence diagnosis methods including SVM, K-nearest neighbor, back-propagation neural network, deep BP neural network, and traditional CNN.

A novel convolutional neural network namely the modified CNN-GAP model is proposed for fast fault diagnosis of the DC-DC inverter. This method improves the model structure of the traditional CNN by using a global average pooling layer to replace the fully connected layer of 2~3 layers. The improved CNN-GAP method mainly contains an input layer, a feature extraction layer, a global average pooling (GAP) layer, and a Softmax output layer. Firstly, the raw 1-D time-series data directly input into the input layer of the established CNN-GAP diagnosis model. The 2-D feature maps are reconstructed in the input layer. Secondly, the representative features are automatically extracted from the 2-D feature maps by using multiple convolutional layers and pooling layers. Thirdly, the dimension transformation and size compression of the output image of the feature extraction layer is completed by the GAP layer. Finally, the fault diagnosis result of the DC-DC inverter is automatically output in the Softmax output layer. The proposed method is used for diagnosing the open-circuit fault of the IGBT in the isolated DC-DC inverter. The proposed method is more accurate and effective than other mainstream intelligent diagnosis methods including the SVM, KNN, DNN, and traditional CNN. The experiment results show that the diagnostic accuracy is up to 99.95%, and the testing time can reduce by more than 15%. The improved CNN-GAP method could greatly reduce the model parameter quantity of the traditional CNN more than 80%, which is more suitable for rapid fault diagnosis in electronic devices.

The recent appearance of Internet of Things (IoT) technologies applied in the maritime industry has introduced the Internet of Ships (IoS) paradigm. By leveraging IoS and deep learning (DL), various DL-based fault diagnosis methods have been proposed to improve shipping companies’ maintenance performance and reduce operational costs. However, the traditional centralized learning approach (CL), which centralizes the data resources of different shipping companies to a cloud server for model training, is restricted in real industrial scenarios due to privacy concerns and business competitions. In this article, we propose a novel adaptive privacy-preserving federated learning approach, named AdaPFL, for fault diagnosis in IoS, which can organize different shipping agents to collaboratively develop a model by sharing model parameters with no risk of data leakage. First, we use two common tasks as examples to demonstrate that a small part of the model parameters might reveal the shipping agents’ raw information. Based on this, the Paillier-based communication scheme is designed to preserve the raw information of the shipping agents. Furthermore, to deal with the harsh marine environment, a control algorithm is proposed to adaptively change the model aggregation interval during the training process for reducing cryptography computation and communication costs. Theoretical analysis and experiments prove the high effectiveness of the AdaPFL on a real nonindependent and identically distributed (non-i.i.d) fault data set.

In this article, a modified convolutional neural network (CNN) algorithm, namely 1D-GAPCNN-SVM, is proposed to address the early anomaly diagnosis problem. Considering the fact that traditional 2D-CNN based approaches contain too many model parameters and are not suitable for fast diagnosis applications using multisensor 1-D time-series measurements, 1D-CNN is introduced to deal with this problem. To reduce the number of parameters, a 1-D global average pooling layer is designed to substitute the fully connected layer with two or three layers. In order to further improve the diagnosis accuracy, a nonlinear multiclass support vector machine (SVM) is adopted to replace the traditional Softmax classifier as the final discriminator. Raw multisensor 1-D time-series data are directly fed into the diagnosis model, then the diagnosis result can be automatically generated. Two experiments, which are rolling bearing and GPS anomaly detection, have been conducted to demonstrate the effectiveness and the superior performance of the proposed method compared to the conventional SVM, K-nearest neighbor, deep neural network (DNN), and traditional 2D-CNN.

In order to optimize the energy management strategy and solve the problem of the power quality degradation of fuel cell hybrid electric ships, a particle swarm optimization algorithm based energy management strategy is proposed in this paper. Taking a fuel cell ship as the target ship, a system simulation model is built in Matlab/Simulink to verify the proposed energy management strategy. Through simulations and comparisons, the bus voltage curve of the optimized hybrid power system fluctuates more gently, and the voltage sag is smaller. The amplitude of the voltage fluctuation under maneuvering conditions is reduced by 55% compared with that of the original ship. The charging and discharging process of the composite energy storage system is optimized under maneuvering conditions, the power quality of the marine power grid is improved, and the use of the energy management strategy can extend the service life of the battery.

The class imbalance problem is prevalent in the condition monitoring (CM) data of marine diesel engines. That significantly deteriorates the diagnostic performance of a data-driven fault diagnosis. In this paper, a novel fault diagnosis method named Graph Convolutional Network based on distance and probability topological graphs (DPGCN) model is proposed for solving the problem of imbalanced classification. Firstly, the collected CM dataset is transformed into two topological graphs, the distance topological graph and the probability topological graph, by exploring the similarity between samples from the linear and non-linear relationships. Graph learning is then introduced to learn and extract the correlation between adjacent samples in addition to samples’ own features, which provide more topological information for this imbalanced classification task. After multi-layer graph learning, two feature embeddings based on samples’ features and topological graphs are learned, and a self-attention mechanism substantially fuses these two embeddings into one combination embedding to derive deeper correlation information for the fault diagnosis. The collected dataset of a marine diesel engine shows that the proposed DPGCN model can improve the classification accuracy and stability under the imbalanced datasets.