Four measurable targets, a three-layer software platform, and a set of applications designed to reshape how HEP ODS plans, connects and maintains the grid.

This is a research project: it aims to discover hidden correlations that could drive future business improvements at HEP ODS. Concrete outcomes can only be approximated in advance — but with high probability of being achieved.

The architecture is built for adaptability across diverse EU regulatory and technical contexts, so other DSOs can integrate it into their existing environments – cloud-based or on-premise.

LAYER 01

Central Data Warehouse (DWH)

A unified data lake integrating SCADA, GIS, smart-metering and external sources — ensuring interoperability and scalability as new data types and volumes are added over time.

LAYER 02

Predictive Middleware Engine

AI and statistical models — including deep learning, time-series forecasting, regression and transformer-based architectures — performing correlation analysis, anomaly detection and scenario simulation across heterogeneous datasets.

LAYER 03

User Interface & Visualization Layer

Interactive dashboards, financial modules and configurable tools that support planning, maintenance and day-to-day decision-making for grid operators.

Network Planning

RES Integration

Outage Management

Maintenance Optimization

While the pilot focuses on Croatia’s electricity distribution system, the architecture is fully compatible with European DSO frameworks such as the EU DSO Entity. Adaptation to local regulatory standards and data protocols is supported through modular APIs and configurable data models.

This means the platform can be deployed by other DSOs across the EU, as well as in emerging markets with growing renewable penetration – turning a national pilot into a template for the sector.