Publications

My research papers published in peer-reviewed journals

A Robust Adapted Flexible Parallel Neural Network Architecture for Early Prediction of Lithium Battery Lifespan
EnergyIF=9.32024

A Robust Adapted Flexible Parallel Neural Network Architecture for Early Prediction of Lithium Battery Lifespan

Lidang Jiang, Zhuoxiang Li, Changyan Hu, Junxiong Chen, Qingsong Huang, Ge He

This paper proposes a Flexible Parallel Neural Network (FPNN) architecture for early prediction of lithium battery lifespan, demonstrating superior prediction performance on the MIT dataset. The architecture integrates InceptionBlock, 3D CNN, 2D CNN, and dual-stream network modules, extracting electrochemical features from video-format data through 3D CNN and achieving multi-scale feature abstraction via InceptionBlock. Using the first 10-100 cycles of data from the MIT dataset, the MAPE values are 1.26%, 0.41%, 0.37%, 0.33%, 0.32%, 0.32%, 0.31%, 0.31%, 0.22%, and 0.34% respectively.

View Paper PDF
Generating Comprehensive Lithium Battery Charging Data with Generative AI
Applied EnergyIF=10.92025

Generating Comprehensive Lithium Battery Charging Data with Generative AI

Lidang Jiang, Changyan Hu, Sibei Ji, Hang Zhao, Junxiong Chen, Ge He

This paper introduces a generative AI model for synthesizing complete lithium battery charging data, addressing data incompleteness and the challenge of generating high-quality datasets. Using EOL and ECL as supervisory conditions, an embedding layer is integrated into the CVAE model to develop the Refined Conditional Variational Autoencoder (RCVAE). By preprocessing data into a video-like format, RCVAE can generate charging data in real-time (voltage, current, temperature, charging capacity).

View Paper PDF