A5069555179- Advanced Battery Technologies Research
- Advancements in Battery Materials
- Electric Vehicles and Infrastructure
- Reliability and Maintenance Optimization
- Fault Detection and Control Systems
- Modeling and Simulation Systems
- Real-time simulation and control systems
- Simulation Techniques and Applications
- Electric and Hybrid Vehicle Technologies
- Control Systems and Identification
- Sensor Technology and Measurement Systems
- Traffic Prediction and Management Techniques
- Time Series Analysis and Forecasting
- Data Visualization and Analytics
Southwest University of Science and Technology
2021-2024
Sichuan University
2022
To accurately evaluate the state of charge (SOC) and health (SOH) Li-ion battery, second-order RC equivalent-circuit model is used to characterize battery performance, a novel dual adaptive Kalman filtering algorithm presented by adding double cycles noise steps realize joint estimation SOC internal resistance. The variables can be corrected with each other as go through cycle under three operating conditions. accuracy method proposed in this paper significantly improved compared extended...
For the battery management system, accurate estimation of state charge and health is great significance. Herein, ternary Li-ion taken as research object; second-order resistor-capacitor (RC) equivalent circuit advantage to characterize performance. A method for calculating batteries based on capacity fading was established. novel forgetting factor dual particle filter algorithm proposed co-estimation by combining algorithm. The under Beijing Bus Dynamic Stress Test conditions are evaluated....
The remaining useful life (RUL) is a core parameter of the battery management system. To realize accurately predict RUL, paper takes National Aeronautics and Space Administration test data set as research object, capacity degradation model based on an exponential growth built to characterize aging process. A novel cuckoo search optimization particle filtering algorithm proposed for RUL prediction by transferring particles in prior distribution region maximum likelihood region. initial cycle...
Abstract For the lithium battery management system and real‐time safety monitoring, two issues are of great significance, namely, ability to accurately update model parameters in real time estimate state charge health. In this context, thesis adopts second‐order RC equivalent circuit forgetting factor recursive least squares ‐ double extended Kalman filtering (FFRLS‐DEKF) algorithm with multi‐time scales low‐pass filter. Forgetting is applied conduct online parameter identification,...
Recent advances in foundation models have emphasized the need to align pre-trained with specialized domains using small, curated datasets. Studies on these underscore importance of low-data training and fine-tuning. This topic, well-known natural language processing (NLP), has also gained increasing attention emerging field scientific machine learning (SciML). To address limitations fine-tuning, we draw inspiration from Heavy-Tailed Self-Regularization (HT-SR) theory, analyzing shape...
Data-centric methods have shown great potential in understanding and predicting spatiotemporal dynamics, enabling better design control of the object system. However, pure deep learning models often lack interpretability, fail to obey intrinsic physics, struggle cope with various domains. While geometry-based methods, e.g., graph neural networks (GNNs), been proposed further tackle these challenges, they still need find implicit physical laws from large datasets rely excessively on rich...
The remaining useful life is the core parameters of battery management system. To realize accurately predict life, paper takes National Aeronautics and Space Administration test data set as research object, a capacity degradation model based on an exponential growth built to characterize aging process. A novel cuckoo search optimization particle filtering algorithm proposed for prediction by transferring particles in prior distribution region maximum likelihood region. initial cycle numbers...
State of health evaluation lithium-ion batteries has become a significant research direction in related fields attributed to the crucial impact on reliability and safety electric vehicles. In this research, dynamic adaptive cuckoo search optimization long short-term memory neural network algorithm was proposed. The aging mechanism battery described effectively by extracting selecting high correlation indicators including voltage, current, charging time, etc. A strategy introduced stabilize...