BMS Thermal Runaway ML Model
Budget: ₹600 – ₹1,500 INR
I am looking to push the state-of-the-art in battery safety by building a machine-learning model that can spot the earliest signs of thermal runaway inside a Battery Management System. Your job begins with research and data preparation, moves through model design and training in Python, and ends with a clear, publication-ready report that can be shared with engineers and stakeholders.
Here’s what I need from you:
• A working ML model for real-time thermal runaway detection, coded in Python and fully commented.
• Rigorous validation showing accuracy, latency, and robustness benchmarks, accompanied by plots or tables that make results easy to assess.
• A professionally formatted technical document (Word or LaTeX is fine) that covers the problem background, data preprocessing, model architecture, training procedure, evaluation, and recommendations for deployment.
Acceptance criteria
• Model achieves clearly stated performance metrics on withheld test data.
• All code executes end-to-end on my hardware using only open-source libraries.
• The report follows standard R&D structure, includes references, and is ready for peer review.
If you have prior experience with battery systems, anomaly detection, or physics-informed ML, please mention it when you respond.
Here’s what I need from you:
• A working ML model for real-time thermal runaway detection, coded in Python and fully commented.
• Rigorous validation showing accuracy, latency, and robustness benchmarks, accompanied by plots or tables that make results easy to assess.
• A professionally formatted technical document (Word or LaTeX is fine) that covers the problem background, data preprocessing, model architecture, training procedure, evaluation, and recommendations for deployment.
Acceptance criteria
• Model achieves clearly stated performance metrics on withheld test data.
• All code executes end-to-end on my hardware using only open-source libraries.
• The report follows standard R&D structure, includes references, and is ready for peer review.
If you have prior experience with battery systems, anomaly detection, or physics-informed ML, please mention it when you respond.