Prognosis Of Lithium-Ion Battery Health with Hybrid EKF-CNN+LSTM Model Using Differential Capacity

Battery degradation is a major challenge in electric vehicles (EV) and energy storage systems (ESS). However, most degradation investigations focus mainly on estimating the state of charge (SOC), which fails to accurately interpret the cells' internal degradation mechanisms. Differential capacity analysis (DCA) focuses on the rate of change of cell voltage about the change in cell capacity, under various charge/discharge rates. This paper developed a battery cell degradation testing model that used two types of lithium-ions (Li-ion) battery cells, namely lithium nickel cobalt aluminium oxides (LiNiCoAlO2) and lithium iron phosphate (LiFePO4), to evaluate internal degradation during loading conditions. The proposed battery degradation model contains distinct charge rates (DCR) of 0.2C, 0.5C, 1C, and 1.5C, as well as discharge rates (DDR) of 0.5C, 0.9C, 1.3C, and 1.6C to analyze the internal health and performance of battery cells during slow, moderate, and fast loading conditions. Besides, this research proposed a model that incorporates the Extended Kalman Filter (EKF), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) networks to validate experimental data. The proposed model yields excellent modelling results based on mean squared error (MSE), and root mean squared error (RMSE), with errors of less than 0.001% at DCR and DDR. The peak identification technique (PIM) has been utilized to investigate battery health based on the number of peaks, peak position, peak height, peak area, and peak width. At last, the PIM method has discovered that the cell aged gradually under normal loading rates but deteriorated rapidly under fast loading conditions. Overall, LiFePO4 batteries perform more robustly and consistently than (LiNiCoAlO2) cells under varying loading conditions.

A Multi-Layer CNN-GRUSKIP model based on transformer for spatial TEMPORAL traffic flow prediction

Traffic flow prediction remains a cornerstone for intelligent transportation systems ITS, influencing both route optimization and environmental efforts. While Recurrent Neural Networks RNN and traditional Convolutional Neural Networks CNN offer some insights into the spatial temporal dynamics of traffic data, they are often limited when navigating sparse and extended spatial temporal patterns. In response, the CNN-GRUSKIP model emerges as a pioneering approach. Notably, it integrates the GRU-SKIP mechanism, a hybrid model that leverages the Gate Recurrent Unit of GRU capabilities to process sequences with the SKIP feature of ability to bypass and connect longer temporal dependencies, making it especially potent for traffic flow predictions with erratic and extended patterns. Another distinctive aspect is its non-standard 6-layer CNN, meticulously designed for in-depth spatiotemporal correlation extraction. The model comprises (1) the specialized CNN feature extraction, (2) the GRU-SKIP enhanced long-temporal module adept at capturing extended patterns, (3) a transformer module employing encoder-decoder and multi-attention mechanisms to hone prediction accuracy and trim model complexity, and (4) a bespoke prediction module. When tested against real-world datasets from California of Caltrans Performance Measurement System PeMS, specifically PeMS districts 4 and 8, the CNN-GRUSKIP consistently outperformed established models such as ARIMA, Graph Wave Net, HA, LSTM, STGCN, and APTN. With its potent predictive prowess and adaptive architecture, the CNN-GRUSKIP model stands to redefine ITS applications, especially where nuanced traffic dynamics are in play.