SmartWatt

SmartWatt is a full-stack residential energy estimation platform for Kerala households. The project combines a machine learning inference layer with physics-based constraints to produce appliance-level consumption estimates, bill projections, and usage insights.

About

SmartWatt is designed to provide practical, software-only household energy intelligence without requiring additional IoT hardware. The platform estimates appliance-wise consumption from user-provided inputs, reconciles predictions with physics constraints, and produces transparent reports aligned with Kerala tariff structures. It is intended for student research, engineering validation, and real-world decision support for electricity cost optimization.

Product Preview

Application Home	User Workflow

Dashboard	Results Report

If images are not visible yet, add files to docs/screenshots/ as described in docs/screenshots/README.md.

Repository Structure

• Backend/: FastAPI-based inference and billing service.
• Frontend/: Next.js-based web application and dashboard.
• start_backend.bat: convenience launcher for backend development.
• Run.txt: minimal local run reference.

Key Features

• Hybrid AI and deterministic physics pipeline.
• Appliance-level and batch prediction endpoints.
• KSEB slab-aware bill calculation.
• Scenario simulation for potential savings.
• Dashboard reporting and persisted assessment history.

System Highlights

Hybrid Inference Pipeline

• Routes complex appliances through trained neural models.
• Uses deterministic physics constraints for stable outputs.
• Provides confidence and anomaly metadata for transparency.

End-to-End User Flow

• Guided multi-step input on the frontend.
• Batch prediction and reconciliation in backend services.
• KSEB tariff-aware billing and optimization insights.

Quick Start

1. Start backend:

cd Backend
python -m uvicorn main:app --reload --port 8000

2. Start frontend in a second terminal:

cd Frontend
npm install
npm run dev

3. Open the application:

• Frontend: http://localhost:3000
• Backend API: http://localhost:8000
• Backend docs: http://localhost:8000/docs

Configuration

• Keep local secrets only in .env or .env.local files.
• Use .env.example files as templates for required keys.

Development Standards

• TypeScript and Python code should remain lint- and test-friendly.
• API schema and frontend field mappings should be updated together.
• Documentation changes should accompany architectural or interface changes.

License

This repository is licensed under the MIT License. See LICENSE.