⚡ Campus Energy Monitor
Cloud Dashboard • Cloudflare Pages • CPUT Final Year Project
Phase L1
Phase L2
Phase L3
🌏 Commands sent via MQTT → HiveMQ Cloud → ESP32 in lab. Response within ~1 second.
🌀 Fan Control
💡 Enclosure Light
🔦 IR LED (GPIO 32)
🔔 Buzzer (GPIO 27)
⬛ Emergency
📷 IR Remote Reference (NEC remote, GPIO 14 — hardware-side only)
1 → Fan OFF | 2 → 30%
3 → 60% | 4 → 100%
5 → AUTO (temp-based)
VOL+ → Light ON (full)
VOL- → Light OFF
100+ → LED ON
200+ → LED OFF
CH+ → Beep 1×
CH → Beep 2×
CH- → Buzzer stop
0 → ALL OFF
PREV → Force temp read
EQ → Force PZEM read
⚡ Phase L1 (0x01) N/A
⚡ Phase L2 (0x02) N/A
⚡ Phase L3 (0x03) N/A
📈 Cloud Data Platforms
📊 ThingSpeak — Historical Trends
Data from ThingSpeak Ch-A (Voltage/Current/Power) & Ch-B (PF/Freq/kWh). 6 charts — 15 s updates from ESP32.
📱 MQTT Connection (HiveMQ Cloud)
📈 Grafana
🧠 AI Brain — Keys & Fallback
🎤 AI Voice Settings
🌟 ElevenLabs → ☁ Azure → ⭐ StreamElements → 🔊 VoiceRSS → 🌐 Browser
⚠ Danger Zone
Reset all dashboard configuration to factory defaults. Clears API keys, MQTT credentials, TTS settings, Personalize/schedule settings, and chat history. The page will reload.
💡 PIR Motion Light
🌀 Fan Temperature Thresholds
⏰ LED Light Scheduler
📷 Camera — TTGO OV2640
📧 Email Reports
🔔 System Alerts
ESP32: /fan off • /fan low • /fan med • /fan high • /fan auto • /light on • /light off • /led on • /led off • /beep • /all off | Camera: /cam on • /cam off • /cam snap • /pir on • /pir off
ℹ About This System
Campus Energy Monitor is an ESP32-based 3-phase energy monitoring system built as a CPUT Final Year Project. It delivers real-time energy data via MQTT, full remote hardware control, and an AI assistant with live sensor context — all accessible from anywhere in the world through this cloud dashboard.
Key Features
- 3-Phase Energy Metering — PZEM-004T on L1, L2, L3 (voltage, current, power, PF, frequency, kWh)
- Temperature Monitoring — DS18B20 with automatic PWM fan control (4 threshold levels)
- Remote Fan / Light / LED / Buzzer control via MQTT command topics
- PIR Motion-activated enclosure light — HC-SR501 with brightness fade
- Real-time MQTT sync — HiveMQ Cloud broker, sensor state published every 5 seconds
- AI Energy Assistant — Gemini 2.5 Flash with live MQTT sensor context in system prompt
- AI Voice Call — full-duplex conversation, continuous speech recognition + TTS
- AI Video Call — camera feed with automatic frame analysis every 30 seconds
- Multi-provider TTS — Browser, Azure Neural, ElevenLabs, VoiceRSS (configurable in Settings)
- Slash commands routed via MQTT: /fan off /fan auto /light on /led off /beep /all off
- Grafana Cloud tab — embed historical charts and alerts dashboard
- Cloudflare Pages — global HTTPS, zero cold-start, custom domain ready
Computer Engineer specializing in software development, scientific computing, networking, electronics, and embedded systems with extensive experience in technical tutoring. I believe technology should solve real problems. My journey across tech — from electronics and industrial to computer engineering — taught me that the best solutions come from curiosity, persistence, and the courage to take unconventional paths.
Today I bridge hardware and software, turning ideas into working systems — from IoT devices monitoring energy consumption to web applications solving client challenges. I also teach, because sharing knowledge deepens understanding. When a student grasps a complex concept, that is engineering success too.
📷 Lab Camera — TTGO OV2640
💡 Lab LED — GPIO 32
⏱ Auto Snapshot
🎦 Last Snapshot
No snapshot yet🖼 Snapshot Gallery
Snapshots are saved locally in this browser only (not synced across devices) — up to 20 most recent. Click a thumbnail to view full size, ⬇ to download, × to delete.