Smart Home Energy Monitoring System

Smart Home Energy Monitoring System using Python-based virtual sensors and a real-time web dashboard. Designed to demonstrate IoT concepts including ESP32, MQTT, energy analytics, alerts, and cloud dashboards without requiring physical hardware.

---

Problem Statement

Homes and small businesses have no visibility into where electricity goes. This project delivers utility-grade per-circuit energy insights using low-cost, non-invasive sensors — no electrician required for the electronics side.

---

---

Project Overview

This project simulates a smart home energy monitoring system that tracks electrical consumption, estimates cost, detects overload conditions, and visualizes data on a live dashboard.

Although no physical hardware is used during execution, the project includes complete ESP32 firmware, MQTT architecture, and circuit diagrams showing how the system would work in a real deployment.

---

Architecture

SCT-013 Clamp
     │
     ▼
Bias + RC Filter
     │
     ▼
ESP32 ADC (RMS sampling @ 4 kHz)
     │
     ▼
JSON payload via MQTT (PubSubClient)
     │
     ├──► Home Assistant (alerts, automations)
     │
     └──► Telegraf → InfluxDB → Grafana (historical charts)

Python Simulation (no hardware needed):
     simulator.py → MQTT / CSV → report_generator.py → PDF

---

Screenshots

Virtual Circuit Diagram (Wokwi)

Energy Analytics Dashboard

Power Consumption Graphs

Running Simulation

Components

Hardware	Purpose
ESP32-DevKitC	MCU — Wi-Fi, ADC, MQTT
SCT-013-050 current clamp	Non-invasive AC current sensing
2× 10kΩ, 100kΩ, 100Ω	Bias divider + protection
10µF + 0.1µF capacitors	Bias decoupling + anti-alias filter
3.5mm audio jack	Clamp connector
ADS1115 (optional)	16-bit ADC for multi-channel
Buzzer / LED (optional)	Local alert indicator

Software	Purpose
Arduino ESP32 core	Firmware
PubSubClient	MQTT publishing
Eclipse Mosquitto	MQTT broker
InfluxDB v2	Time-series storage
Telegraf	MQTT → InfluxDB bridge
Grafana	Dashboard & charts
Home Assistant	Automations & alerts
Python 3.10+	Simulation & reports
paho-mqtt	Python MQTT client
reportlab	PDF report generation

---

📁 Folder Structure

Smart-Home-Energy-Monitoring-System/
│
├── arduino_code/
│   └── main.cpp
│
├── dashboard/
│   ├── energy.html
│   ├── docker-compose.yml
│   ├── home_assistant_sensors.yaml
│   ├── telegraf.conf
│   └── mosquitto/
│
├── data/
│   ├── energy_data.json
│   └── energy_log.csv
│
├── python_simulation/
│   ├── simulator.py
│   ├── live_simulator.py
│   └── report_generator.py
│
├── outputs/
│   └── energy_report.pdf
│
├── circuit_diagram/
│   └── wiring_guide.md
│
├── docs/
│   └── interview_qa.md
│
├── main.py
├── requirements.txt
└── README.md

---

🚀 Setup & Execution

Option A — Python Simulation (no hardware)

# 1. Clone
git clone https://github.com/Sonia068/Smart-Home-Energy-Monitoring-System
cd Smart-Home-Energy-Monitoring-System

# 2. Install dependencies
pip install -r requirements.txt

# 3. Run normal simulation (30 readings, ~30 seconds)
python main.py

# 4. Run overload scenario + generate PDF report
python main.py --mode overload --readings 20 --report

# 5. All modes
python main.py --mode normal    # everyday household
python main.py --mode high      # heavy appliances
python main.py --mode overload  # triggers alerts
python main.py --mode multi     # 6 mixed appliances

# 6. Report only (from saved CSV)
python main.py --report-only

Option B — Full Stack with Docker

# Start Mosquitto + InfluxDB + Grafana + Telegraf
cd dashboard
docker compose up -d

# Check all services running
docker compose ps

# View Grafana at http://localhost:3000  (admin/admin)
# View InfluxDB at http://localhost:8086

# Run simulator pointing to local broker
cd ..
python main.py --mode multi --readings 120

Option C — Real Hardware (ESP32)

See detailed steps in Hardware Execution section below.

---

🔧 Hardware Execution (Real ESP32)

Step 1 — Install Arduino IDE + ESP32 board

Arduino IDE → File → Preferences → Board Manager URLs:
https://raw.githubusercontent.com/espressif/arduino-esp32/gh-pages/package_esp32_index.json

Tools → Board → Boards Manager → search "esp32" → Install

Step 2 — Install Libraries

Sketch → Include Library → Manage Libraries:
- PubSubClient  (knolleary)
- Adafruit ADS1X15  (optional, for external ADC)

Step 3 — Configure Firmware

Edit arduino_code/main.cpp:

const char* WIFI_SSID = "YOUR_HOME_WIFI";
const char* WIFI_PASS = "YOUR_PASSWORD";
const char* MQTT_HOST = "192.168.1.10";   // your broker IP

Step 4 — Upload

Tools → Board → ESP32 Dev Module
Tools → Port  → COM3 (or /dev/ttyUSB0 on Linux)
Sketch → Upload

Step 5 — Wire the Circuit

Follow circuit_diagram/wiring_guide.md.
Ask a licensed electrician to place the clamp on the mains wire.

Step 6 — Calibrate

1. Plug in a known load (e.g., a 60 W bulb).
2. Open Serial Monitor (115200 baud).
3. Expected current = 60 / 230 = 0.261 A.
4. If reading is off, type: CAL:0.XXXXX in Serial Monitor to set new factor.

Step 7 — Start Stack

cd dashboard
docker compose up -d

Step 8 — Add HA Sensors

Paste dashboard/home_assistant_sensors.yaml into Home Assistant's configuration.yaml and restart HA.

---

📊 Sample Console Output

=== Smart Home Energy Monitor ===
Active Appliances:
  • LED Lights             40W  PF=0.95
  • Ceiling Fan            75W  PF=0.85
  • Laptop                 65W  PF=0.90
  TOTAL                   180W

[  1/ 30] 2025-08-14T10:00:01  MODE=NORMAL   V= 229.3V  I=  0.791A  P=   181.4W  ██  ΣkWh=0.0001  ₹0.0004
[  2/ 30] 2025-08-14T10:00:02  MODE=NORMAL   V= 231.1V  I=  0.778A  P=   179.9W  █   ΣkWh=0.0001  ₹0.0008
...
[  8/ 30] ... ⚠ ALERT
         ↳ SPIKE: Transient load spike detected

---

📈 Formulas

Parameter	Formula
RMS Current	Irms = √( Σ(i²) / N )
Apparent Power	S (VA) = Irms × Vnominal
Energy (interval)	ΔWh = S(W) × Δt(s) / 3600
Cumulative kWh	kWh = ΣΔWh / 1000
Cost	₹ = kWh × tariff (₹8/kWh)

---

🌍 Industry Relevance

Sector	Application
Smart Homes	Per-appliance usage, bill splitting
Hostels / PGs	Tenant sub-metering
Commercial Buildings	HVAC & lighting baselines, peak shaving
Factories	Motor health monitoring, overload detection
Solar Companies	Net metering, consumption vs. generation
Building Automation	SCADA-like dashboards at low cost

---

🔮 Future Improvements

• Appliance-level NILM (non-intrusive load monitoring)
• Mobile app (Flutter / React Native)
• AI/ML energy forecasting
• Solar generation vs. consumption overlay
• Voice assistant integration (Alexa / Google)
• Automated load shedding relay control
• OTA firmware updates (ArduinoOTA)

🎥 Demo Video

👉 Click the image above to watch the full demo of the Smart Home Energy Monitoring System

---

Execution Steps

Install Dependencies

pip install -r requirements.txt

Generate Live Data

cd python_simulation
python live_simulator.py

The simulator continuously updates:

• energy_data.json
• energy_log.csv

Energy Calculations

Power:

Power (W) = Voltage × Current

Energy:

Energy (Wh) = Power × Time / 3600

Cost:

Cost = kWh × ₹8 per unit

---

Features

Live dashboard update Power analytic Voltage monitorin Energy calculatio Cost estimatio Overload detectio Alert loggin CSV expor PDF report generatio ESP32 firmware include Circuit diagram included

---

Real Hardware Implementation (Future Deployment)

The project includes complete implementation files for:

SCT-013 Current Sensor
        │
        ▼
ESP32 ADC
        │
        ▼
MQTT Broker
        │
        ▼
InfluxDB
        │
        ▼
Grafana Dashboard
        │
        ▼
Home Assistant Alerts

---

👨‍💻 Author

Sonia Thakur

• GitHub: @Sonia068
• LinkedIn: https://www.linkedin.com/in/sonia-thakur-6ab93b349/

---

⭐ If you found this project helpful, please give it a star!