The accurate estimation of battery health conditions is a crucial challenge for development of battery management systems due to the degradation of cathode and anode materials. In this paper, a fusion of deep learning model and feature analysis methods is employed to approach accurate estimation for state of health (SOH) and remaining useful life (RUL). The differential thermal voltammetry (DTV) signal analysis is executed to pre-process the datasets from Oxford University. A deep learning model is constructed with LSTM network as the core, combined with Bayesian optimization and dropout technique. This work shows that the deep learning model could approach the SOH and RUL early estimation with the mean absolute error of RUL maintained around 0.5%. It is potential that this deep learning model, combined with DTV signal analysis methods, could approach early prediction and estimation of battery SOH and RUL, contributing to the development of the next-generation high-energy-density and highly safety commercial batteries.

Battery parameters identification is crucial for accurate prediction of battery life in electric vehicles. In order to optimize battery parameter identification, an improved PSO (particle swarm optimization) algorithm was proposed based on the use of Thevenin battery model to abstract the problem into an optimization problem. The experimental results show that the computational accuracy of the improved PSO algorithm is higher than that of the genetic algorithm and the original PSO algorithm, and the battery parameters calculated by the improved PSO algorithm are also more accurate.

Deep learning-based models have been diffusely utilized in industrial virtual sensing tasks. However, these models built on neural networks are faced with poor interpretability, i.e., they cannot explain why these features are used, why they can improve model performance, and whether they are reliable in building models, etc. In this paper, an interpretable neural network (INN) based on Generalized Additive Model (GAM) structure is proposed to build a succinct virtual sensor with selected efficient input factors. The proposed INN is a disentangled model with multiple subnetworks of GAM, and each subnetwork is meant to discover either one main factor or one interactive factor for the output prediction. In this model, both the standalone and the interactive input variables are taken as the original input factor. Meanwhile, the time delays are also considered in the variables, since the time-delay factors commonly exist between the industrial process variables. Through the sparsity constraint learning, the input factor selection can be explained by the importance, and the model performance is also improved with a tight structure. A numerical example and a real industrial case are presented to verify the effectiveness and superiority of the proposed virtual sensor.

To improve the reliability and interpretability of industrial process monitoring, this article proposes a causal graph spatial-temporal autoencoder (CGSTAE). The network architecture of CGSTAE combines two components: a correlation graph structure learning module based on spatial self-attention mechanism (SSAM) and a spatial-temporal encoder-decoder module utilizing graph convolutional long short-term memory (GCLSTM). The SSAM learns correlation graphs by capturing dynamic relationships between variables, while a novel three-step causal graph structure learning algorithm is introduced to derive a causal graph from these correlation graphs. The algorithm leverages a reverse perspective of causal invariance principle to uncover the invariant causal graph from varying correlations. The spatial-temporal encoder-decoder, built with GCLSTM units, reconstructs time series process data within a sequence-to-sequence framework. The proposed CGSTAE enables effective process monitoring and fault detection through two statistics in the feature space and residual space. Finally, we validate the effectiveness of CGSTAE in process monitoring through the Tennessee Eastman process (TEP) and a real-world air separation process (ASP).