Automated design exploration for EVs and xEVs

Overview

‍

In the shift toward electrified mobility, the battery is no longer a subsystem, it is the linchpin of performance, integration, and cost strategy. However, its design remains a deeply complex, high-stakes process where mechanical, thermal, electrical, and packaging constraints intersect.

‍

This case study demonstrates how Dessia’s  AI generative design platform and library enables OEMs and Tier-1 suppliers to systematically explore and validate optimal battery layouts across both EV and xEV platforms. By automating architecture generation and integration analysis, Design teams can replace iterative guesswork with intelligent exploration — ensuring their battery systems meet performance targets, fit within constrained spaces, and are ready for industrialization.

‍

Context & challenge

‍

Electrification is fundamentally reshaping vehicle architectures. As the demands on EV and xEV battery systems intensify—higher energy density, faster charging, and precise thermal control—design teams face an exponentially growing solution space.

‍

Battery architecture is defined early in the development cycle but impacts nearly every downstream decision: structural integration, thermal strategy, weight distribution, and even cost of ownership. Yet traditional development methods rely on linear, experience-driven iterations that are inherently limited in scope. Design convergence is often reached under severe time pressure—limiting innovation and introducing late-stage risks.

‍

To address this, a paradigm shift is needed: from manual iteration to AI-augmented design space exploration.

‍

The approach: Generative engineering methodology

‍

Dessia’s AI solution leverages a formalized systems design approach combined with generative algorithms. At its core is a domain-specific model that captures both the functional behavior and 3D constraints of the battery system. This model serves as a foundation for exhaustive, constraint-aware architecture exploration.

‍

Key capabilities:

• Multilevel abstraction: Batteries are modeled across hierarchies, from high-level power interfaces to low-level module/cell arrangements.
• System knowledge embedding: Engineering rules (electrical, thermal, mechanical) are codified to guide feasible architecture generation.
• Automated exploration: AI algorithms systematically generate and validate a wide range of battery layouts, electrical topologies, and integration options within available vehicle packaging.
• Concurrent wiring & cooling design: Once valid architectures are generated, the system simultaneously proposes wiring harnesses and thermal routing aligned with spatial, safety, and performance criteria.

‍

Design execution: From requirements to realistic architectures

‍

A customized AI-App can be built from Dessia’s library to transform system requirements into realistic, validated battery architectures—ready for integration.

‍

1. Translating system requirements into design inputs

The process begins by capturing battery-level requirements stemming from the broader EV or xEV powertrain context—such as nominal voltage, capacity, power profile, and environmental constraints. Spatial integration limits are also defined, based on the available volume within the vehicle chassis. These parameters establish the foundation for downstream architectural exploration.

‍

2. Automated generation of functional configurations

Dessia’s platform leverages an object-oriented system modeling approach to define battery architectures at both the module and cell levels. Guided by embedded engineering design rules, the solution generates a wide range of valid electrical configurations—spanning conventional series-parallel arrangements to more intricate topologies. Each variant complies with electrical feasibility, thermal considerations, and structural logic.

‍

3. Intelligent layout construction within packaging constraints

For each generated electrical architecture, the platform evaluates its ability to be physically integrated into the available design volume. Layouts are built based on precise dimensional rules and geometric compatibility, taking into account accessibility, serviceability, and mass distribution. Only physically viable arrangements progress through the design flow.

‍

4. Embedded generation of supporting subsystems

In parallel with core architecture development, the platform automatically proposes auxiliary system designs to complete the integration:

• High-voltage wiring paths: Routed for minimal length, reduced electromagnetic interference, and compliance with safety margins.
• Cooling pipe layouts: Positioned to enable efficient thermal transfer while respecting structural and manufacturability constraints.

The final outcome is a library of fully defined battery layouts—each matched with its corresponding wiring and cooling infrastructure, ready for evaluation or handover.

‍

Strategic outcomes and benefits

‍

Battery layout decisions have a ripple effect—impacting integration, safety, performance, and supplier coordination. Yet these decisions are often made with limited time, incomplete information, and far too many unknowns. Dessia’s approach changes that, bringing structure, speed, and visibility to a process that’s typically opaque and slow.

‍

Turning complexity into clarity

Designing battery systems for EVs and xEVs means juggling tight packaging, evolving power requirements, and growing thermal demands—all under aggressive timelines. Traditionally, engineers rely on experience, limited tooling, and manual iterations to reach a feasible layout.

Dessia brings a new level of clarity and speed to this process. By automatically generating valid battery configurations and checking them against real-world constraints from the start, it empowers teams to explore more options, avoid late-stage surprises, and move forward with confidence.

‍

Faster design convergence

Instead of spending weeks building, testing, and adjusting battery layouts by hand, engineers using Dessia’s platform can generate and compare high-quality designs in just a few hours. This dramatically speeds up the development process without sacrificing quality or accuracy.

‍

Smarter exploration, beyond manual limits

Design engineers often have to make decisions based on just a few design options because exploring more would take too much time. Dessia changes that. The platform explores hundreds of valid configurations automatically, uncovering solutions that may not have been considered, and helping teams make better, more informed decisions.

‍

Integration confidence from the start

One of the biggest risks in battery design is realizing, too late, that a chosen layout doesn’t actually fit in the vehicle, or causes problems with wiring or cooling. With Dessia, every design is checked early on for physical feasibility. That means no unpleasant surprises later, and fewer delays during integration.

‍

Better communication with suppliers

Dessia doesn’t just help internal teams. It also makes it easier to collaborate with battery and component suppliers. By providing clear, detailed layouts, including exact space limits and connection points, OEMs can ensure suppliers fully understand what’s needed, leading to smoother development and fewer misunderstandings.

‍

‍

Conclusion: Design certainty in an uncertain landscape

‍

In an era where the boundaries of electrification are defined by battery innovation, the ability to explore is a competitive differentiator. Dessia’s generative design methodology introduces rigor, repeatability, and computational depth into what was once an intuitive and fragmented process.

‍

By transforming design engineering knowledge into structured algorithms and coupling it with high-fidelity system models, Dessia empowers automotive players to approach battery layout design with precision, scale, and confidence.

‍

This is not just about automation. It’s about augmenting engineering capability, so that design engineers can focus on strategic trade-offs, not manual permutations.

‍