GridDB Logo
GridDB Developers
Download
Back to Gallery

Esmart

By * Rohan Kumar Shah. Nishant Chhetri, Manish Dey Health
Esmart
Health

Esmart: AI-Driven Prosumer Energy Platform

Inspiration

India’s grid is currently drowning in green energy it cannot manage. [cite_start]This isn’t a shortage of power, but a crisis of predictive intelligence where the lack of real-time AI coordination forces massive energy wastage and financial instability [cite: 378-379].

[cite_start]We noticed that clean energy is frequently “curtailed”—essentially thrown away—because grid operators cannot predict surges in generation or shifts in demand in real-time[cite: 426]. [cite_start]Esmart bridges the gap between the “prosumer” and a smarter, more resilient grid[cite: 428].

What it does

Esmart is an AI-driven platform designed to solve the predictive intelligence crisis in modern energy grids. [cite_start]Unlike current software designed for damage control, Esmart is built for occurrence prevention[cite: 381, 404].

The 6-Step Core Engine:

  1. [cite_start]Solar/Wind Forecasting: Uses weather APIs and historical datasets to predict local generation[cite: 409].
  2. [cite_start]Load Prediction: Analyzes consumer demand patterns to understand exactly when energy is needed[cite: 410].
  3. [cite_start]Congestion Check: Monitors transmission line strength to identify potential grid bottlenecks[cite: 411].
  4. [cite_start]Storage Health (SoH/SoC): Real-time monitoring of battery capacity and health to determine usable backup[cite: 412, 416].
  5. [cite_start]Curtailment Prediction: Synthesizes the first four steps to identify when clean energy will be wasted[cite: 418].
  6. [cite_start]Action Recommendation: Suggests the best path—storing in batteries, shifting industrial loads, or trading with neighbors[cite: 419].

Key Features

  • [cite_start]Peer-to-Peer (P2P) Trading: Decentralized marketplace for neighbors to buy/sell excess solar energy, bypassing utility markups[cite: 421].
  • [cite_start]Grid Stability Monitoring: Tracks voltage (e.g., 228V) and frequency (e.g., 50.2 Hz) to warn of impending outages[cite: 422].
  • [cite_start]Waste Reduction: Targets “curtailment” to monetize surplus green energy[cite: 423].
  • [cite_start]Environmental Impact Analytics: Translates data into metrics like CO₂ saved and trees planted[cite: 424].

How we built it

We developed the platform using a modern web stack designed for real-time data handling:

  • [cite_start]Frontend (React): Component-based architecture for handling live data streams without page reloads[cite: 430].
  • [cite_start]Styling: CSS3 with Custom Properties (Variables) for an “electric” aesthetic[cite: 431].
  • [cite_start]State Management: Tracks global states like battery capacity (SoC) and trading offers[cite: 433].
  • [cite_start]Predictive Logic: A custom engine integrating weather forecasting, demand patterns, and congestion data[cite: 435].
  • [cite_start]Simulation: Vanilla JavaScript used to simulate real-time energy flow and metric updates[cite: 436].

Challenges we ran into

  • [cite_start]Data Synchronization: Coordinating unpredictable weather data with fluctuating consumer demand patterns[cite: 438].
  • [cite_start]UI Complexity: Designing a dashboard that makes complex technical data (Voltage, Frequency, SoH) intuitive for homeowners[cite: 439].
  • [cite_start]Real-time Alerts: Engineering the system to trigger “Power Cut” predictions with high confidence based on instability markers[cite: 440].

How we use GridDB

[cite_start]The Esmart platform uses GridDB, a high-performance time-series database optimized for IoT, to store real-time energy readings [cite: 441-442].

  • [cite_start]Data Storage: Stores continuous streams of voltage, current, solar output, and battery levels collected every few seconds[cite: 443].
  • [cite_start]Data Management: Manages user profiles, AI-generated power cut predictions, and P2P trading offers[cite: 447].
  • [cite_start]Integration: The React frontend queries GridDB through REST APIs to fetch live grid status and energy flow charts[cite: 448].
  • [cite_start]Analytics: Leverages GridDB’s time-series capabilities for fast retrieval of historical data for trend analysis and AI model training[cite: 449].

Built With

css3, griddb, javascript, react, rest-api

Team Fakerz

  • Rohan Kumar Shah
  • Nishant Chhetri
  • Manish Dey [cite_start][cite: 453-455]
Team Members: * Rohan Kumar Shah. Nishant Chhetri, Manish Dey