Skip to content
View yasirusama61's full-sized avatar
πŸ’­
I may be slow to respond.
πŸ’­
I may be slow to respond.

Block or report yasirusama61

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
yasirusama61/README.md

Header Image

πŸ‘‹ Hi there, I'm Usama Yasir Khan!

I'm an AI Engineer at XING Mobility with a passion for advancing technology in battery management systems, machine learning, and AI-driven solutions for real-world challenges. I specialize in temperature prediction, state estimation, and digital twin development for energy storage systems and electric vehicles.


πŸ’Ό About Me

  • πŸ”­ Current Role: AI Engineer at XING Mobility, working on AI Battery Management Systems
  • πŸš— Industry: Electric vehicles, Battery Management Systems (BMS), Renewable Energy
  • 🌱 Interests: Battery state estimation, predictive maintenance, digital twins, sustainable tech
  • πŸ’¬ Ask me about: Machine Learning, Battery Systems, Data Analysis, Predictive Modeling
  • πŸ› οΈ Agile Development: Experienced in Agile methodologies, using Tuleap for Sprint planning, tracking tasks, and facilitating daily stand-up meetings for smooth and efficient development.

πŸš€ Agile/Sprint Planning Expertise

I thrive on Agile methodologies to ensure smooth, efficient, and collaborative workflows. Leveraging tools like Tuleap, I manage Sprints with precision and flexibility. Here's how I bring Agile into action:


πŸ› οΈ Key Practices:

  • πŸ“ Backlog Refinement: Prioritize tasks to align with upcoming Sprint goals.
  • πŸ—“οΈ Sprint Planning: Define SMART goals and allocate tasks effectively across the team.
  • ⏱️ Daily Stand-ups: Foster transparency with brief updates and tackle blockers head-on.
  • πŸ”„ Retrospectives: Reflect, learn, and enhance workflows for continuous improvement.

πŸ“Š Sprint Workflow Overview

Below is a snapshot of a typical Sprint workflow, showcasing how tasks progress from Backlog to Completion:

Working on AI BMS
**Dynamic Development**
Kanban Workflow Animation
**Agile Sprint Workflow**
Team Collaboration
**Team Collaboration**

🌟 Why Agile Matters in Data Science?

Agile methodologies allow me to:

  1. Deliver iterative and incremental value with rapid prototyping and testing.
  2. Adapt quickly to shifting project requirements or data insights.
  3. Maintain collaboration and transparency across technical and non-technical stakeholders.

πŸ’‘ Quote:
"Data Science thrives in Agileβ€”an iterative approach keeps innovation and insights flowing."


🌟 Let's build smarter workflows, one Sprint at a time!


πŸ”½ Highlights / Proficiencies / Interests / Beliefs

Highlights

  • Extensive experience in AI and machine learning for battery management systems.
  • Proven expertise in developing predictive models for temperature and state estimation.

Proficiencies

  • Programming Languages: Python, MATLAB
  • Frameworks: TensorFlow, PyTorch, Scikit-Learn
  • Deployment: AWS Cloud, NXP N97 MCU, i.MX RT MCU

Interests

  • Battery state estimation, digital twins, renewable energy, and sustainable tech.

Beliefs

  • Passionate about advancing technology for a sustainable future.
  • Believes in leveraging AI to create practical, real-world solutions.

πŸš€ Skills & Technologies

Programming Languages

Python R MATLAB

Machine Learning & AI

TensorFlow PyTorch Scikit-Learn PyBAMM

Deployment & Cloud

AWS Docker GitHub Codespaces NXP N97 MCU i.MX RT MCU

Data Processing

Pandas NumPy

Data Visualization

Plotly Matplotlib Dash

Optimization Techniques

PSO (Particle Swarm Optimization) PBnB (Pruned Branch and Bound) Grid Search Random Search

Agile Tools

Tuleap


🌟 Featured Projects

1. Advanced SOC Estimation using Transfer Learning

  • Implemented a State of Charge (SOC) estimator using transfer learning, pre-training on LG 18650HG2 Li-ion battery data and fine-tuning it for a specific electric vehicle application or a new battery chemistry.
  • Technologies: Python, TensorFlow
  • GitHub Repository

2. Battery SOC and Temperature Estimation

  • Developed models to estimate battery SOC and temperature in immersion-cooled battery packs, incorporating time-series analysis for accurate predictions.
  • Technologies: Python, TensorFlow
  • GitHub Repository

3. Battery Temperature Prediction

  • Developed a model to predict battery temperature using unique features from the LG 18650HG2 dataset, leveraging transfer learning to enhance predictive accuracy.
  • Technologies: Python, TensorFlow, Scikit-Learn
  • GitHub Repository

4. Battery Aging Classification

  • Created a classification model to assess and predict battery aging based on various operational parameters.
  • Technologies: Python, Scikit-Learn
  • GitHub Repository

5. Quality Prediction App

  • Developed an application to predict and monitor quality metrics in manufacturing, using machine learning for proactive quality assurance.
  • Technologies: Python, Dash, XGBoost
  • GitHub Repository

6. Early Risk Detection System

  • Created a predictive maintenance system to detect risks early in the battery manufacturing process, allowing for timely interventions.
  • Technologies: Python, Scikit-Learn
  • GitHub Repository

Feel free to explore my other repositories for more insights into my work and contributions.

πŸ“ˆ GitHub Stats and Most Used Languages

GitHub Stats Most Used Languages

GitHub Activity Graph

Usama's github activity graph

πŸ”₯ GitHub Streak Stats

GitHub Streak

πŸ“« Connect with Me

LinkedIn Email Me


πŸ” Current Focus

  • Dashboard Creation: Developing a visualization tool for real-time temperature monitoring in batteries.
  • Explainable AI (XAI): Adding SHAP-based interpretability to machine learning models for enhanced transparency.
  • Edge Deployment: Exploring deployment options for BMS applications on resource-constrained devices.
  • Agile Collaboration: Using tools like Tuleap for Sprint planning and task tracking to ensure efficient project progression.

🌱 Future Goals

I'm excited to continue expanding my expertise in:

  • Advanced state estimation techniques for electric vehicle batteries
  • Integration of AI with IoT for smarter energy management solutions
  • Building scalable predictive models for renewable energy applications

πŸ† References and Acknowledgments

  1. Dataset: LG 18650HG2 Li-ion Battery Data and Example Deep Neural Network xEV SOC Estimator Script, by Philip Kollmeyer, Carlos Vidal, Mina Naguib, and Michael Skells at McMaster University. DOI: 10.17632/cp3473x7xv.3

  2. PyBAMM: The open-source Python Battery Mathematical Modeling (PyBAMM) library, developed by the PyBAMM community, provides a flexible framework for battery simulations. For more information, visit the official PyBAMM repository: PyBAMM GitHub or the project website: PyBAMM.org.

  3. Battery Dataset: Additional datasets from the Stanford DAWN Benchmarking Suite and NASA Prognostics Data Repository have also informed aspects of my battery research. These repositories provide public datasets for battery degradation and health estimation. Links to the repositories:

  4. Methodologies and Algorithms: Several optimization and machine learning techniques used in my work were inspired by widely recognized research papers and methods in the field of AI-driven battery management. Special thanks to the machine learning and battery modeling research communities for their contributions.

  5. Acknowledgments: Special thanks to the contributors and researchers who have inspired my work in battery technology, including the creators of PyBAMM and the authors of publicly available datasets. Their work has greatly enriched my ability to develop predictive models and digital twin systems for battery management.


Thank you for visiting my profile! Feel free to reach out if you'd like to discuss AI, battery management, Agile practices, or potential collaborations. Let’s build the future of sustainable energy together!

Pinned Loading

  1. Battery_SOC_Temp_Estimation Battery_SOC_Temp_Estimation Public

    A repository for the development of machine learning models focused on predicting battery State of Charge (SOC) and temperature in immersion-cooled battery packs, incorporating time series analysis…

    Python 3

  2. Advanced-SOC-Estimation-using-Transfer-Learning Advanced-SOC-Estimation-using-Transfer-Learning Public

    Implement a transfer learning-based SOC estimator, pre-train a model on LG 18650HG2 Li-ion Battery Data and fine-tune it for a specific electric vehicle application or new battery chemistry.

    Python 2

  3. AI_BMS_Optimization AI_BMS_Optimization Public

    An AI-driven solution to optimize battery performance, efficiency, and longevity. Features include real-time mode switching (Performance, Eco, Balanced, Custom), SOC and temperature predictions, dy…

    Python 2

  4. battery-aging-classification battery-aging-classification Public

    This repository simulates battery aging data and applies machine learning models to classify battery health into aging stages such as Healthy, Moderate Aging, and Aged. The project uses real-world …

    Python 1

  5. early-risk-detection early-risk-detection Public

    A machine learning classification approach for risk detection in battery cell manufacturing

    Python 1

  6. Quality_APP Quality_APP Public

    A repo designed to digitalize quality control in battery manufacturing with real-time monitoring and predictive analytics. Quality_APP leverages data from IPQC, FQC, and OQC stages to assess produc…

    Python 1