Techno economic integrated planning of solar integrated electric vehicle charging infrastructure in India using an AI enabled multi objective planning framework.

India's transition to electric mobility demands charging infrastructure that is cost-efficient, grid-compatible, and capable of integrating solar generation. Existing studies typically examine demand forecasting, PV utilisation, charging-topology behaviour, and economic viability in isolation, limiting their relevance for large-scale deployment. This work proposes a unified co-design framework that jointly optimises charging-station siting, charger sizing, PV allocation, and operational economics under India's tariff structure. Hourly EV demand is predicted using a hybrid forecasting model that combines Temporal Fusion Transformers with Graph Neural Networks to capture spatial and temporal variations. Solar-generation modelling, topology-based charger efficiencies, and distribution-grid constraints are incorporated into a techno-economic formulation. A multi-objective optimisation approach (NSGA-II) identifies configurations that minimise cost, reduce peak grid loading, and maximise solar utilisation. The framework is demonstrated using a representative mixed urban-highway region. Results show a 28-35% reduction in peak grid load, a 40-70% improvement in utilisation, and a 12-18% decrease in the levelised cost of charging compared with non-optimised deployments. The findings highlight the importance of integrated planning that aligns solar availability, demand behaviour, and tariff incentives. The proposed methodology offers a scalable decision-support tool for policymakers, utilities, and private developers planning future EV charging networks in India.