Skip to main content

System Architecture

The high level software architecture of the EMS4DC is depicted on the following figure:

Core Components​

PostgreSQL Database Cluster

The database cluster runs continuously in the background, serving as the central data repository for real-time measurement data from connected devices, optimizer engine input parameters, and optimizer engine output results.

Python Stack

A Python stack consists of four continous scripts running in parallel:

  • Data Acquisition Process measure.py: Executes every 10 seconds to collect real-time measurements from energy assets
  • Optimizer Trigger optimizer.py: Runs every 15 minutes to perform energy optimization calculations
  • Metrics Calculation Process metrics.py: Runs every hour to calculate metrics and KPIs for the passed hour
  • Forecasting Module forecast.py: Runs a forecast generation, handles model retraining.

Backend Server (Express.js)

Provides RESTful API endpoints that process and serve data to the frontend application, handling requests for all application features and tabs.

Frontend Server (Vite)

Delivers the SCADA (Supervisory Control and Data Acquisition) interface, providing users with real-time visualization and control capabilities through a modern web application.