We build intelligent software solutions that address key challenges in battery data analysis, modelling, and validation. Our technologies enable faster decision-making, improved performance insights, and more efficient battery development workflows.
Technical Lead in Battery Modelling
Dr. Benyamin Ebrahimpour is the Technical Lead in Battery Modelling at DeepPSI, specialising in AI-driven modelling for lithium-ion batteries. He holds a PhD from the University of Portsmouth and focuses on combining physics-based modelling with AI to develop scalable engineering solutions.
AI-Powered Battery Data Diagnostics
NoahCells is an AI-powered battery data diagnostics platform developed by DeepPSI to transform how lithium-ion battery datasets are analysed in industrial and research environments. The platform enables rapid, secure, and standardised interpretation of battery test data, allowing engineers to assess data quality, extract key performance indicators, and determine dataset readiness before advanced modelling or simulation. Designed for deployment in confidential and controlled environments, NoahCells addresses a critical bottleneck in battery validation workflows by replacing manual, script-based data cleaning and analysis with automated, physics-constrained diagnostics. To visit the product, please click here: https://noahcells.com/
Collaboration
The project is led by DeepPSI Ltd, with industrial validation in collaboration with Breathe Battery Technologies Ltd. It is part of the Women in Innovation Awards 2025/26, supporting the development and commercialisation of advanced digital technologies in the UK.
Impact
- Reduces manual data cleaning and analysis time by up to 60–70%
- Enables faster and more reliable battery validation workflows
- Improves data standardisation, auditability, and engineering confidence
- Supports battery safety, performance assessment, and second-life evaluation
- Strengthens the UK battery ecosystem through secure, scalable digital tools
AI-Enhanced Surrogate Model
DeepPSI is developing an AI-enhanced, physics-constrained surrogate modelling framework to accelerate engineering simulation and digital design across advanced manufacturing supply chains. The platform combines first-principles physics models with machine learning to replicate high-fidelity electro-thermal simulations in seconds rather than hours, enabling real-time design exploration and optimisation. Initially focused on lithium-ion battery systems, the framework is designed to be transferable to automotive, aerospace, and energy applications.
Collaboration
The project is led by DeepPSI Ltd, with:
- Industrial validation partner
- Technical consultancy and independent validation from academia
This project is part of the Battery Innovation Feasibility Studies, Round 2
programme, supporting UK innovation in battery technologies for electrification.
Impact
- Reduces simulation time from hours to seconds
- Enables rapid virtual prototyping and design optimisation
- Lowers computational cost and energy consumption
- Reduces material waste through early-stage digital validation
- Strengthens UK supply chain resilience and digital sovereignty
- Supports innovation across battery, automotive, and aerospace industries