Traditional machine learning algorithms have made great achievements in data-driven fault diagnosis. However, they assume that all the data must be in the same working condition and have the same distribution and feature space. They are not applicable for real-world working conditions, which often change with time, so the data are hard to obtain. In order to utilize data in different working conditions to improve the performance, this paper presents a transfer learning approach for fault diagnosis with neural networks. First, it learns characteristics from massive source data and adjusts the parameters of neural networks accordingly. Second, the structure of neural networks alters for the change of data distribution. In the same time, some parameters are transferred from source task to target task. Finally, the new model is trained by a small amount of target data in another working condition. The Case Western Reserve University bearing data set is used to validate the performance of the proposed transfer learning approach. Experimental results show that the proposed transfer learning approach can improve the classification accuracy and reduce the training time comparing with the conventional neural network method when there are only a small amount of target data.

Intelligent condition monitoring and fault diagnosis by analyzing the sensor data can assure the safety of machinery. Conventional fault diagnosis and classification methods usually implement pretreatments to decrease noise and extract some time domain or frequency domain features from raw time series sensor data. Then, some classifiers are utilized to make diagnosis. However, these conventional fault diagnosis approaches suffer from the expertise of feature selection and they do not consider the temporal coherence of time series data. This paper proposes a fault diagnosis model based on Deep Neural Networks (DNN). The model can directly recognize raw time series sensor data without feature selection and signal processing. It also takes advantage of the temporal coherence of the data. Firstly, raw time series training data collected by sensors are used to train the DNN until the cost function of DNN gets the minimal value; Secondly, test data are used to test the classification accuracy of the DNN on local time series data. Finally, fault diagnosis considering temporal coherence with former time series data is implemented. Experimental results show that the classification accuracy of bearing faults can get 100%. The proposed fault diagnosis approach is effective in recognizing the type of bearing faults.

The bearing is the key component of rotating machinery, and its performance directly determines the reliability and safety of the system. Data-based bearing fault diagnosis has become a research hotspot. Naive Bayes (NB), which is based on independent presumption, is widely used in fault diagnosis. However, the bearing data are not completely independent, which reduces the performance of NB algorithms. In order to solve this problem, we propose a NB bearing fault diagnosis method based on enhanced independence of data. The method deals with data vector from two aspects: the attribute feature and the sample dimension. After processing, the classification limitation of NB is reduced by the independence hypothesis. First, we extract the statistical characteristics of the original signal of the bearings effectively. Then, the Decision Tree algorithm is used to select the important features of the time domain signal, and the low correlation features is selected. Next, the Selective Support Vector Machine (SSVM) is used to prune the dimension data and remove redundant vectors. Finally, we use NB to diagnose the fault with the low correlation data. The experimental results show that the independent enhancement of data is effective for bearing fault diagnosis.

Lithium-ion batteries offer excellent advantages of high efficiency, small size, and low cost, but their instability and inconformity remain challenging. Sudden failure of batteries may cause serious accidents, endangering the safety of people’s lives and properties. Advanced remaining useful life prediction methods for batteries can effectively avoid those accidents. In this article, we proposed a novel hybrid method with the purpose of enhancing battery remaining useful life prediction precision and robustness. Based on improved complete ensemble empirical mode decomposition with adaptive noise algorithm, utilizing a special-designed interpolation reconstruction mechanism, the battery capacity degradation series was decomposed into a trend subseries and several fluctuation subseries. Weighted least square support vector machine and long short-term memory network are then established to perform prediction for the trend subseries and the fluctuation ones, respectively. A complementary series of experiments is designed to verify the effectiveness of the proposed method. The simulation results represent that the proposed method achieves higher prediction accuracy and robustness over other comparison models. The proposed approach provides a promising and effective alternate for lithium-ion batteries remaining useful life prediction without relying on the cell dynamic process, which is meaningful for cases with limited measurement parameters or limited computational power.

Lithium-ion batteries are widely applied in many fields. It is important for predicting battery life (RUL). It is randomly discharged that the lithium-ion battery under random conditions. The experiment of constant current discharge cannot simulate the discharge state under working conditions. Based on the data collection of the NASA dataset, the DGWO-ELM algorithm is proposed to predict lithium-ion battery. The DGWO-ELM is composed of Extreme Learning Machine (ELM), Grey Wolf Optimization (GWO), and Differential Evolution (DE) for the purpose of improving the accuracy of prediction. The algorithm uses GWO algorithm to optimize the weight and threshold of ELM and improves the three deficiencies in the GWO algorithm. The DGWO-ELM algorithm is proved preferably than ELM predictor improved by particle swarm optimization (PSO-ELM) and SVM predictor improved by Grey Wolf Optimization (GWO-SVM). The algorithm is verified by NASA’s lithium-ion battery constant current discharge data, and then used to predict the RUL of the lithium-ion battery in a random discharge environment. The results show that the DGWO-ELM performs well on improving the accuracy of prediction.