Loading...
AI-driven zero trust and blockchain framework for secure electric vehicle infrastructure
Daah, Clement Fallot, Y. Konur, Savas
Daah, Clement
Fallot, Y.
Konur, Savas
Publication Date
2026-05-25
End of Embargo
Supervisor
Rights
© 2026 The Author(s). This is an Open Access article distributed under the Creative Commons CC-BY license (https://creativecommons.org/licenses/by/4.0/)
Peer-Reviewed
Yes
Open Access status
openAccess
Accepted for publication
2026-02-05
Institution
Department
Awarded
Embargo end date
Files
Loading...
qureshi_et_al_2026.pdf
Adobe PDF, 10.78 MB
Additional title
Abstract
Electric vehicle (EV) charging infrastructures are increasingly exposed to sophisticated cyber threats, including replay, spoofing, privilege escalation, and geolocation-based attacks. While standards such as ISO 15118 and OCPP 2.0.1 provide interoperability and cryptographic guarantees, they rely on static policies or isolated detection mechanisms, leaving gaps against adaptive adversaries. This paper presents an AI-driven Zero Trust Blockchain (AI-ZTB) framework whose novelty lies in the system-level integration of identity and access management, AI-based risk assessment, and blockchain-backed decentralized auditability with IPFS-based evidence storage, while operational governance remains centrally managed by the service provider. Unlike prior AI-only or blockchain-only frameworks, AI-ZTB introduces a fully integrated and enforceable Zero Trust control loop in which AI-generated risk scores are operationally bound to access enforcement decisions through smart contracts, enabling adaptive, auditable, and context-aware security governance in real time. The framework was implemented in Python with Solidity smart contracts and evaluated through a large-scale network simulation involving batches of 10,000 EV-charging sessions, trained on a dataset of 50,000 legitimate and adversarial behaviours using Random Forest, Autoencoder, and Isolation Forest models. Results demonstrate that AI-ZTB achieves access-decision accuracy above 95%, reducing false acceptance and rejection rates to approximately 3%. A comparative analysis evaluates AI-ZTB against industry standards (ISO 15,118 and OCPP 2.0.1) as secure communication baselines, and against prior integrated frameworks from the literature, highlighting differences in architectural scope, policy enforceability, and auditability rather than protocol-level performance. Despite modest inference and logging overheads, performance remained within real-time operational tolerances. The framework establishes a robust foundation for securing EV infrastructures, with extensibility to smart grids and other cyber-physical environments.
Version
Published version
Citation
Daah C, Fallot Y, Qureshi A et al (2026) AI-driven zero trust and blockchain framework for secure electric vehicle infrastructure. Expert Systems with Applications. 312: 131577.
Link to publisher’s version
Link to published version
Link to Version of Record
Type
Article