Abstract :

The Rapid Adoption Of Electric Vehicles (EVs) Has Created A Growing Need For Efficient Battery Management Systems Capable Of Ensuring Reliability, Safety, And Longevity. One Of The Most Critical Aspects Of EV Performance Is Battery Health, Commonly Represented By The State Of Health (SOH) And Remaining Capacity. Accurate Prediction Of Battery Degradation Enables Predictive Maintenance, Reduces Operational Risks, And Enhances Lifecycle Management. This Research Presents A Deep Learning-based EV Battery Health Prognostics System Utilizing A Hybrid Convolution Neural Network (CNN) And Long Short-Term Memory (LSTM) Architecture. The Proposed System Integrates Data-driven Modeling With A User-friendly Graphical Interface Developed Using Tkinter, Allowing Users To Generate Datasets, Train Models, And Predict Battery Health In Real Time. The Dataset Consists Of Key Battery Parameters Such As Voltage, Current, Temperature, And State Of Charge (SOC), Which Are Essential Indicators Of Battery Performance. Synthetic Data Generation Is Also Incorporated To Simulate Battery Degradation Patterns For Testing And Demonstration Purposes. The CNN Component Of The Model Is Responsible For Extracting Local Temporal Features And Identifying Complex Patterns In The Input Data. Subsequently, The LSTM Layer Captures Long-term Dependencies And Temporal Correlations Within The Time-series Data, Which Are Crucial For Modeling Battery Degradation Trends. The Model Is Trained Using Normalized Data Processed Through MinMax Scaling, And Performance Is Optimized Using The Adam Optimizer With Mean Squared Error (MSE) As The Loss Function. The System Allows Users To Load Real-world Datasets Or Generate Synthetic Data, Preprocess The Input, Train The Model, And Perform Predictions Through An Interactive Interface. The Output Is Presented As Predicted Battery Capacity, Which Directly Correlates With SOH Estimation. The Hybrid CNN-LSTM Approach Significantly Improves Prediction Accuracy Compared To Traditional Machine Learning Methods, As It Effectively Captures Both Spatial And Temporal Characteristics Of The Data. This Work Contributes To The Field Of Intelligent Battery Management Systems By Providing A Scalable, Efficient, And Easy-to-use Solution For Battery Health Monitoring. The Proposed System Can Be Extended To Real-time IoTbased EV Monitoring And Integrated With Onboard Vehicle Systems For Continuous Diagnostics. Overall, This Research Demonstrates The Potential Of Deep Learning Techniques In Enhancing The Reliability And Efficiency Of EV Battery Systems.

Published:

07-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

1457-1467

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite