🌾 TSCRIC-LoRa: Intelligent Precision Agriculture and Smart Irrigation Platform

Minor Project · B.Tech Civil Engineering · Oriental College of Technology, Bhopal · RGPV · 2024–2027

🌐 Live Dashboard • 📱 Download Android App • 📄 Project Report

---

📌 What is TSCRIC-LoRa?

TSCRIC stands for Temporal Soil-Crop Resonance Irrigation Controller.

It is an IoT-based intelligent precision agriculture and smart irrigation platform developed by civil engineering students at Oriental College of Technology, Bhopal. The system integrates real-time multi-depth soil moisture monitoring, LoRa long-range telemetry, sensor-driven irrigation decision-making, a multi-page SCADA-inspired web dashboard, and complete offline autonomous operation — all assembled from commodity hardware at approximately ₹3,000.

This project demonstrates that classical civil engineering irrigation principles — duty of water, crop delta, net irrigation requirement, ETo, water balance modelling — can be implemented in real-time embedded firmware on an affordable microcontroller and presented through a professional web dashboard.

---

🏗️ Civil Engineering Relevance

Irrigation engineering is a core sub-discipline of civil engineering. TSCRIC-LoRa directly implements the following engineering parameters in firmware:

Parameter	Equation	Implementation
Duty of Water (D)	D = A / Q [ha/cumec]	YF-S201 flow sensor measures Q; plot area from config
Crop Delta (Δ)	Δ = 8.64 × B / D [m]	FAO-56 standard — Wheat 450 mm, Rice 1200 mm
Net Irrigation Req. (NIR)	NIR = (ETc − Pe) × A	ETc from Hargreaves-Samani × Kc; Pe from SCS CN
Gross Irrigation Req. (GIR)	GIR = NIR / Ea	Ea from applied vs delivered via flow meter
ETo (Hargreaves-Samani)	ETo = 0.0023 × Ra × √ΔT × (T+17.8)	DHT22 sensor, Ra = 10 MJ/m²/day for Central India
Root-Zone Water Balance	ΔS = I + P − ETc − DP − R	All terms computed in firmware, shown in dashboard
SCS Curve Number	Q = (P−Ia)² / (P−Ia+S); CN = 75	Tipping bucket rainfall → effective rainfall
Readily Available Water	RAW = AWC × Zr × p	CSMI threshold calibrated to RAW depletion point

Based on laboratory prototype testing, the system is estimated to have the potential to achieve 20–50% water savings compared to conventional flood irrigation if deployed at field scale.

---

✨ Key Features

🌱 Core Irrigation Intelligence

• CSMI (Composite Soil Moisture Index) — tri-depth weighted average at 15 cm, 30 cm, and 45 cm
• GDD-driven crop stage progression — weights automatically shift as the crop grows (Stage 0 to Stage 3)
• IIS (Intelligent Irrigation Score) — 0–100 composite score from CSMI + SMV + TPR + ETo − Rain Penalty
• SMV / SMA / TPR — Soil Moisture Velocity, Acceleration, and Temporal Pattern Recognition for predictive irrigation

🌧️ Three-Tier Rainfall Intelligence

Priority	Source	Internet Required?	Accuracy
1 — PRIMARY	Tipping Bucket Rain Gauge (0.2 mm/tip)	No — fully offline	WMO standard direct measurement
2 — BACKUP	OpenWeatherMap API	Online only	Regional NWP forecast
3 — ESTIMATE	BMP280 Pressure Trend	No — fully offline	±20%, 1–6 hour horizon

📡 Communication Layers

• Primary: Firebase Realtime Database — 15-second cloud sync
• Fallback: LoRa SX1278 433 MHz — 30-second broadcast, 2–5 km range
• Offline: Local WiFi hotspot (SSID: TSCRIC_AI, IP: 192.168.4.1) — full dashboard on farmer's phone

🖥️ Multi-Page Dashboard

• System Dashboard with IIS gauge, CSMI sparkline, Decision Explanation Engine
• Irrigation Control — pump panel, event log, seasonal water budget
• Sensor Network — 12-node LoRa monitoring, tri-depth VWC, environmental sensors
• Analytics — 7-day IIS history, water balance trend, ETo chart, soil moisture trends
• Soil Health — DS18B20 soil temperature, EC, pH, indicative NPK trend
• Smart Farm Advisory Assistant — rule-based, works completely offline in Hinglish

⚡ Other Highlights

• Pulse irrigation — 30-second ON pulses prevent surface ponding and runoff (application rate 16.7 mm/hr, below infiltration capacity)
• Pump health monitoring — motor condition, seal, bearing, cavity, performance index
• Sensor failure fallback matrix — SAFE MODE, OWM fallback, weight rescaling on partial sensor failure
• EEPROM offline storage — 20-event ring buffer, auto-syncs to Firebase on reconnection
• millis() overflow bug fix — prevents pump re-trigger after 49.7 days

---

🔧 Hardware Components

Component	Specification	Est. Cost (₹)
ESP8266 NodeMCU V3	ESP-12E, 4MB flash, 80MHz, WiFi	280
Capacitive Soil Moisture Sensor v2.0 (×3)	3.3V, corrosion-resistant	150
CD4051 8-Channel MUX	Analog multiplexer for 3 sensors on 1 ADC	25
DHT22 Sensor	±0.5°C, ±2% RH	130
BMP280 Module	I2C, 0–1100 hPa, ±1 hPa	120
LoRa SX1278 Module	433 MHz, 17 dBm, SPI	420
YF-S201 Flow Sensor	1–30 L/min, 7.5 pulses/litre	250
5V Relay Module	Optoisolated, Active LOW	80
Mini DC Water Pump	2.5–6V submersible, 80–120 L/hr	150
DS18B20 (optional)	Waterproof soil temperature	120
EC Sensor (optional)	0–6 mS/cm	200
Soil pH Sensor (optional)	0–14 pH range	280
Tipping Bucket Rain Gauge (optional)	0.2 mm/tip, reed switch	480
Breadboard + Wires + Resistors	830-point, Dupont, 4.7kΩ pull-ups	175
Total (core)		~₹1,785
Total (full system with optionals)		~₹3,045

---

🧠 Algorithm Overview

CSMI — Composite Soil Moisture Index

CSMI = w₁×VWC_15cm + w₂×VWC_30cm + w₃×VWC_45cm

Weights shift automatically by GDD-determined crop stage:

Stage	GDD Range	w₁ (15cm)	w₂ (30cm)	w₃ (45cm)	Threshold
Germination	0–150	0.50	0.30	0.20	25%
Tillering	150–400	0.40	0.35	0.25	28%
Grain-Fill	400–800	0.30	0.40	0.30	30%
Maturation	>800	0.25	0.38	0.37	22%

IIS — Intelligent Irrigation Score

IIS = moistureScore + velocityScore + tprBonus + etoScore − rainPenalty

Component	Max Points	Rationale
Moisture Score	52.5	CSMI deficit below threshold
Velocity Score (SMV)	20.0	Rate of moisture depletion
TPR Bonus	15.0	Cosine similarity of drying pattern
ETo Score	15.0	Atmospheric evaporative demand
Rain Penalty	−40	Suppresses irrigation before rain

Irrigation triggers when: CSMI < threshold AND IIS > 65 AND rain probability < 75% AND cooldown elapsed AND not in SAFE MODE

ETo — Hargreaves-Samani (1985)

ETo = 0.0023 × Ra × √(tRange) × (T_mean + 17.8)   [mm/day]
ETc = ETo × Kc

SCS Curve Number (CN = 75)

S = 25400/CN − 254
Ia = 0.2 × S
Q = (P − Ia)² / (P − Ia + S)     [when P > Ia]
Pe = P − Q   [effective rainfall]

---

🗂️ Repository Structure

Sitamarhi/
├── index.html        # Main multi-page dashboard (HTML5 + CSS3 + JS)
├── style.css         # Dark-theme SCADA-inspired stylesheet
├── app.js            # Firebase integration, sensor logic, AI assistant, charts
├── tscric.ino        # ESP8266 Arduino firmware (CSMI, IIS, ETo, LoRa, EEPROM)
└── README.md         # This file

---

🚀 Getting Started

Dashboard (Web)

Visit the live dashboard directly:

https://amanvyahut.github.io/Sitamarhi/

For offline use, connect to the ESP8266 hotspot (TSCRIC_AI, password: 12345678) and open:

http://192.168.4.1

Firmware (ESP8266)

1. Install Arduino IDE with ESP8266 board support
2. Install required libraries: Firebase ESP8266, DHT sensor library, Adafruit BMP280, LoRa, OneWire, DallasTemperature
3. Open tscric.ino in Arduino IDE
4. Configure your WiFi credentials and Firebase project URL in the firmware
5. Flash to ESP8266 NodeMCU via USB

Firebase Setup

1. Create a Firebase project at console.firebase.google.com
2. Enable Realtime Database (set rules to allow read/write for testing)
3. Copy your database URL into tscric.ino and app.js

---

📊 System Performance (7-Day Lab Testing)

Metric	Result
Irrigation Efficiency	78% (Good — above 70% benchmark)
Water Savings (estimated vs flood baseline)	~28–30% in test window
Projected seasonal savings (estimated)	20–50% at field scale
Water Use Efficiency	87%
LoRa Network Uptime	99.2%
LoRa Packet Delivery Rate	98.7%
Firebase Push Latency	avg 380 ms (4G)
Sensor Reading Cycle	10 seconds (±50 ms)

⚠️ Note: Results above are from laboratory prototype testing with a 6 m² plot. Field-scale validation has not yet been conducted. Water saving percentages are estimates based on a traditional flood irrigation baseline.

---

🔮 Future Scope

• Solar power integration — 5–10W panel + LiFePO4 + MPPT (PM-KUSUM compatible)
• PCB + IP67 enclosure — field-grade production deployment
• DS3231 RTC — accurate offline event timestamping
• ESP32 upgrade — dual-core, dual UART, Bluetooth
• CNN crop disease detection — ESP32-CAM + TensorFlow Lite (Phase 3)
• Reinforcement Learning — optimal irrigation policy from multi-season data (Phase 5)
• LoRaWAN gateway — command-area-wide monitoring at gram panchayat level
• PM-KUSUM / PMKSY integration — government scheme alignment

---

👥 Project Team

Name	Role
Aman Kumar (0126CE243D04)	Team Lead · Project Developer
Akash Kumar (0126CE243D03)	Team Member
Aditya Kumar (0126CE243D01)	Team Member
Akash Khargande (0126CE243D02)	Team Member
Dr. Yogesh Iyer Murthy	Project Guide · Dept. of Civil Engineering, OCT Bhopal

Institution: Oriental College of Technology, Bhopal · Department of Civil Engineering
University: Rajiv Gandhi Proudyogiki Vishwavidyalaya (RGPV), Bhopal
Academic Year: 2024–2027

---

📚 Key References

1. Allen et al. — FAO Irrigation and Drainage Paper 56 (1998) — Crop Evapotranspiration Guidelines
2. Hargreaves & Samani — Reference Crop ETo from Temperature (1985)
3. A.M. Michael — Irrigation Theory and Practice (Vikas Publishing, 2010)
4. S.K. Garg — Irrigation Engineering and Hydraulic Structures (Khanna Publishers, 2012)
5. WMO-No. 8 — Guide to Meteorological Instruments and Methods of Observation (2018)
6. Hsiao (1990) — Root-zone weighted averaging at multiple depths
7. Semtech — SX1276/77/78/79 LoRa Transceiver Datasheet

---

📄 License

This project is submitted as a B.Tech Minor Project at Oriental College of Technology, Bhopal under RGPV. The code and documentation are made available for academic reference. Commercial use is not permitted without prior permission from the project team and institution.

---

TSCRIC-LoRa · Oriental College of Technology, Bhopal · Civil Engineering 2024–2027
Guide: Dr. Yogesh Iyer Murthy · Team: Aman Kumar · Akash Kumar · Aditya Kumar · Akash Khargande

🌐 Live Dashboard | 📱 Android App