Smart Water Purification and Quality Monitoring System for Rural and Mining-Affected Areas.

1. Executive Snapshot

This technical dossier presents an end-to-end product strategy, hardware-software architecture, data model, and implementation roadmap for SIH26040: Smart Water Purification and Quality Monitoring System for Rural and Mining-Affected Areas [PS]. Designed specifically for the Department of Higher & Technical Education, Government of Jharkhand [PS], the system addresses the severe contamination of groundwater and surface water in mining belts (such as Dhanbad, West Singhbhum, Ramgarh, and Jharia) caused by suspended coal dust, acid mine drainage, dissolved total solids (TDS), iron, heavy metals, and bacterial pathogens [RESEARCHED]. The proposed solution, named JalDrishti, combines a low-power, 4-stage energy-efficient purification hardware unit (sediment filtration, activated carbon, heavy-metal selective ion-exchange resin, and ultrafiltration with UV-C LED sterilization) [INFERENCE] with an ESP32-based IoT sensing module (pH, turbidity, TDS, temperature, flow, differential pressure) [INFERENCE]. Data is evaluated locally using the Weighted Arithmetic Water Quality Index (WAWQI) against Bureau of Indian Standards (BIS IS 10500:2012) benchmarks [KNOWN] and synchronized over LoRaWAN/GSM to a web dashboard and mobile app with offline-first caching [INFERENCE].

Metadata Summary

Fundamental System Objective: To deliver a portable, solar-powered multi-stage water purification unit equipped with real-time IoT water quality sensing and edge alert capabilities, ensuring safe drinking water compliance under BIS IS 10500:2012 for off-grid rural and mining communities [INFERENCE].

2. Problem Deconstruction

The problem statement provided by the Government of Jharkhand contains explicit requirements and contextual constraints that must be preserved without omission [PS].

Atomic Requirement Extraction

• REQ-01 (Contaminant Scope): The system must target suspended particles, excessive minerals, microbial impurities, and mining-related pollutants in rural and mining-affected surface/groundwater sources [PS].
  • Source Text: "Groundwater and surface water sources are often contaminated by suspended particles, excessive minerals, microbial impurities, and mining-related pollutants, making them unsafe for consumption." [PS]
  • Interpretation: Mining effluents in Jharkhand contain coal dust, acid mine drainage (AMD), elevated iron ($>0.3\text{ mg/L}$), manganese, fluoride, and toxic heavy metals [RESEARCHED].
  • Implementation Implication: Purification must integrate multi-stage physical and chemical treatment, not merely standard sediment filters [INFERENCE].
  • Confidence Tag: [PS]
• REQ-02 (Parameter Monitoring): The system must continuously measure key water quality parameters including pH, turbidity, Total Dissolved Solids (TDS), and temperature [PS].
  • Source Text: "The system continuously monitors key water quality parameters such as pH, turbidity, TDS, and temperature..." [PS]
  • Interpretation: Real-time probe telemetry must sample raw and purified water streams at scheduled intervals [INFERENCE].
  • Implementation Implication: Sensor array must be calibrated against standard calibration buffer solutions and read by an analog-to-digital converter (ADC) on the edge microcontroller [INFERENCE].
  • Confidence Tag: [PS]
• REQ-03 (Multi-Stage Purification): The system must utilize a multi-stage purification process to remove impurities, harmful contaminants, and pathogens [PS].
  • Source Text: "...while utilizing a multi-stage purification process to remove impurities, harmful contaminants, and pathogens." [PS]
  • Interpretation: Sequential physical filtration, chemical adsorption/ion exchange, and pathogen neutralization are mandatory [INFERENCE].
  • Implementation Implication: 4-stage physical rig: (1) 5-micron PP sediment filter, (2) Bituminous Activated Carbon + Katalox-Light media, (3) Heavy-metal selective chelating ion-exchange resin, (4) 0.01-micron Ultrafiltration (UF) membrane + 275nm UV-C LED chamber [INFERENCE].
  • Confidence Tag: [PS]
• REQ-04 (IoT & Real-Time Alerting): The system must integrate IoT capabilities to provide real-time water quality insights and generate alerts for unsafe conditions [PS].
  • Source Text: "Integrated with IoT and real-time monitoring capabilities, the system provides water quality insights, generates alerts for unsafe water conditions..." [PS]
  • Interpretation: Automated rule triggers when parameters exceed BIS 10500 threshold limits, locking downstream solenoid valves and dispatching local audio-visual alerts and remote SMS/push notifications [INFERENCE].
  • Implementation Implication: MQTT/HTTP edge telemetry with local piezo buzzer and GSM/LoRa fallbacks [INFERENCE].
  • Confidence Tag: [PS]
• REQ-05 (Portability & Cost-Effectiveness): The system must be portable, cost-effective, easy to deploy, and energy-efficient for remote operation in infrastructure-limited areas [PS].
  • Source Text: "Develop a smart, cost-effective, and portable water purification system... It should be easy to deploy, energy-efficient and suitable for operation in remote areas with limited infrastructure." [PS]
  • Interpretation: System must run on low DC power ($12\text{V}$, $<25\text{W}$) capable of solar PV panel battery operation, avoiding power-hungry high-pressure RO pumps unless high TDS strictly demands it [INFERENCE].
  • Implementation Implication: Frame built with lightweight aluminum chassis, quick-connect fittings, 50W solar panel integration, and low-power Ultrafiltration (UF) pumps [INFERENCE].
  • Confidence Tag: [PS]

3. Stakeholders & User Needs

4. Requirements Traceability

5. Constraints & Environment Realities

• Geography & Terrain: Mining regions in Jharkhand (e.g., East/West Singhbhum, Dhanbad, Giridih) feature rugged plateau terrain, acidic mine runoff, and elevated groundwater iron ($>3\text{ mg/L}$) and fluoride ($>1.5\text{ mg/L}$) [RESEARCHED].
• Connectivity Realities: Rural habitations suffer from weak or patchy 2G/4G cellular coverage. Broadband fiber is nonexistent in remote hamlets [KNOWN]. Edge devices must buffer up to 30 days of sensor readings locally in SPIFFS/Flash memory and sync via MQTT upon reconnection [INFERENCE].
• Power Supply: Rural grid power availability is intermittent (8–12 hours/day with frequent brownouts) [KNOWN]. System must be powered by a $12\text{V}$, $24\text{Ah}$ LiFePO4 battery recharged by a $50\text{W}$ solar panel [INFERENCE].
• Digital Literacy & Language: Local operators (Jal Sahiyas) may have limited literacy. The physical interface must use clear color-coded status lights (Green = Safe, Yellow = Service Required, Red = Unsafe/Cut-off) and voice alerts in Hindi/Santhali via an onboard audio module [INFERENCE].
• Water Matrix Variability: Feedwater turbidity fluctuates drastically during monsoon seasons (rising from $10\text{ NTU}$ to $>300\text{ NTU}$) due to mine runoff silt [RESEARCHED]. System requires a washable/reusable pre-filter screen to prevent rapid clogging of fine cartridges [INFERENCE].

6. Research Questions

1. What are the exact dominant heavy metal species present in groundwater across Jharkhand's coal and iron ore mining belts?
   • Findings: Research indicates high levels of dissolved Iron ($\text{Fe}^{2+}/\text{Fe}^{3+}$), Manganese ($\text{Mn}^{2+}$), Fluoride ($\text{F}^-$), Lead ($\text{Pb}^{2+}$), and total suspended solids from coal dust washings [RESEARCHED].
2. What is the power consumption of a high-pressure RO pump versus a low-power Ultrafiltration (UF) pump system?
   • Findings: Standard RO pumps draw $60\text{W}$–$100\text{W}$ AC and require $50\text{–}80\text{ PSI}$, whereas UF operates under gravity or low-power $12\text{V}$ DC booster pumps ($12\text{W}$–$18\text{W}$, $15\text{ PSI}$), reducing solar panel array sizing by 75% [KNOWN].
3. How frequently do low-cost analog pH and turbidity probes drift in field conditions?
   • Findings: Analog glass pH electrodes and optical turbidity sensors suffer calibration drift within 14–30 days under high-turbidity mine water exposure, requiring automated software calibration offsets and routine maintenance protocol [RESEARCHED].

7. Data Requirements & Sources

8. Existing Solutions & Gap Analysis

9. Candidate Solution Approaches

Approach 1: Solar-Powered High-Pressure Reverse Osmosis (RO) + Cellular IoT Transducer

• Mechanism: Feedwater is pressurized to $70\text{ PSI}$ using a heavy-duty DC diaphragm pump through a 0.0001-micron RO membrane, paired with continuous sensor telemetry.
• Pros: Removes 98% of all dissolved salts and heavy metal ions regardless of water chemistry [KNOWN].
• Cons: Consumes $80\text{W}$ power, requiring large $200\text{W}$ solar panels and heavy $50\text{Ah}$ batteries; generates $65%$ reject wastewater, which is unacceptable in water-scarce rural habitations [INFERENCE].
• Hackathon Feasibility: Low (costly, heavy, difficult to power in portable demo).

Approach 2: Low-Pressure Modular Ultrafiltration (UF) + Heavy-Metal Selective Resin + Solar IoT Edge Monitor (RECOMMENDED)

• Mechanism: Multi-stage low-pressure purification: (1) Washable 50-micron mesh pre-filter, (2) 5-micron PP sediment + Katalox-Light media (for iron/manganese/coal dust), (3) Heavy-metal selective chelating ion-exchange resin (for $\text{Pb}^{2+}$, $\text{Cd}^{2+}$, $\text{Fe}^{3+}$), (4) 0.01-micron hollow-fiber UF membrane + 275nm UV-C LED chamber. Sensors read pH, turbidity, TDS, temperature, and pressure differential.
• Pros: Operates under low pressure ($15\text{ PSI}$, $<15\text{W}$ DC power); 95% water recovery rate; highly portable; cost-effective; solar PV runnable [INFERENCE].
• Cons: TDS reduction is moderate (20-30%); highly saline water ($>2000\text{ mg/L}$) requires secondary dilution or specialized media [INFERENCE].
• Hackathon Feasibility: High (affordable, compact, low power, easily demonstrateable).

Approach 3: Chemical Coagulation-Flocculation Dosing Unit + Microcontroller Sensor Array

• Mechanism: Automated micro-dosing of alum/polyaluminium chloride (PAC) to settle mining silt, followed by gravity sand filter and chlorination.
• Pros: Effective for extremely turbid surface mine pit water ($>500\text{ NTU}$) [KNOWN].
• Cons: Requires handling liquid chemical reagents in remote villages; risk of chemical over-dosing; high mechanical maintenance [INFERENCE].
• Hackathon Feasibility: Medium (chemical handling risks during live demonstration).

10. Recommended Approach & Why

Selected Approach: Approach 2 (Hybrid Modular Ultrafiltration + Heavy-Metal Ion-Exchange Resin + Low-Power Solar ESP32 IoT Edge System) [INFERENCE].

Justification

1. Energy Efficiency & Portability: Mining-affected rural areas in Jharkhand frequently suffer from total grid failure. Approach 2 operates on a small $12\text{V}$ DC booster pump ($12\text{W}$) and UV-C LED ($3\text{W}$), totaling $<15\text{W}$. The entire system can run continuously off a compact $50\text{W}$ solar panel and $12\text{V}$ LiFePO4 battery pack [INFERENCE].
2. Specific Contaminant Targeting: Mining runoff in Jharkhand contains high suspended solids, iron, manganese, and heavy metals rather than extreme sea-water salinity [RESEARCHED]. Katalox-Light media and selective chelating resin directly eliminate iron ($>95%$) and heavy metals without the massive water wastage of RO systems [INFERENCE].
3. Water Conservation: Generates $<5%$ flush waste, compared to RO systems that waste $65\text{–}70%$ of raw water [KNOWN].
4. Rejection of Alternatives: Approach 1 (RO) was rejected due to high power consumption ($80\text{W}+$), high cost, and unacceptable water wastage in drought-prone rural mining blocks. Approach 3 was rejected due to the severe logistical hazard of supplying and handling chemical liquid reagents in unassisted rural hamlets [INFERENCE].

11. Technical Architecture

Layered System Architecture Diagram

+-----------------------------------------------------------------------------------+
|                               1. HARDWARE LAYER                                  |
| +---------------------+  +----------------------+  +----------------------------+ |
| |   Purification Rig  |  |    IoT Sensor Array  |  |   Control & Actuation      | |
| | 5u PP -> Katalox    |  |  pH Probe (Analog)   |  | Solenoid Valve (12V DC)    | |
| | Heavy Metal Resin   |  | Turbidity Probe      |  | Piezo Alarm / Status LEDs  | |
| | UF Membrane         |  | TDS Probe            |  | 12V DC Booster Pump        | |
| | UV-C LED (275nm)    |  | DS18B20 Temp Probe   |  | Relays & MOSFET Drivers    | |
| +---------------------+  +----------------------+  +----------------------------+ |
+-----------------------------------------------------------------------------------+
                                         | ADC / GPIO / I2C
                                         v
+-----------------------------------------------------------------------------------+
|                             2. EDGE COMPUTING LAYER                               |
| +-------------------------------------------------------------------------------+ |
| | Microcontroller: ESP32-WROOM-32U                                              | |
| | - Local Processing: Raw Voltage -> Calibrated Units Conversion                 | |
| | - Local Computation: Weighted Arithmetic WQI Calculation (BIS IS 10500)       | |
| | - Rule Engine: Fail-Safe Auto Valve Shutoff (if pH < 6.5 or Turbidity > 5NTU) | |
| | - Offline Buffer: SPIFFS Flash Memory (30-day telemetry logging)               | |
| +-------------------------------------------------------------------------------+ |
+-----------------------------------------------------------------------------------+
                                         | MQTT over Cellular / LoRaWAN
                                         v
+-----------------------------------------------------------------------------------+
|                             3. CLOUD BROKER & BACKEND                             |
| +-------------------+  +------------------------+  +----------------------------+ |
| |  Mosquitto MQTT   |  | Node.js / Express API  |  | Time-Series & Relational   | |
| |  Message Broker   |  | Telemetry & Alert Engine|  | PostgreSQL + TimescaleDB   | |
| +-------------------+  +------------------------+  +----------------------------+ |
+-----------------------------------------------------------------------------------+
                                         | REST APIs / WebSockets
                                         v
+-----------------------------------------------------------------------------------+
|                            4. APPLICATION & UX LAYER                              |
| +--------------------------------------+  +-------------------------------------+ |
| | Admin Web Dashboard (React + Leaflet)|  | Mobile App (Flutter - Android/iOS)  | |
| | - Live Regional GIS Heatmaps         |  | - Jal Sahiya Local Monitoring       | |
| | - District PHED Water Quality Index  |  | - Bluetooth Calibration Interface   | |
| | - Automated Service Ticket Generation|  | - Push Notifications & SMS Alerts   | |
| +--------------------------------------+  +-------------------------------------+ |
+-----------------------------------------------------------------------------------+

Step-by-Step Data Flow Narrative

1. Water Ingress & Sensing: Raw water enters the intake port. Before entering the primary purification stack, a pre-sensing chamber measures raw pH, turbidity, TDS, and temperature using analog probes connected to the ESP32 ADC pins [INFERENCE].
2. Edge Calibration & WAWQI Calculation: The ESP32 converts raw analog voltages into physical units using stored polynomial calibration curves. The embedded C++ firmware calculates the Weighted Arithmetic Water Quality Index (WAWQI) score in real time [INFERENCE].
3. Safety Rule Execution & Actuation:
   • If WAWQI is within safe drinking limits ($WQI \le 50$, pH $6.5\text{--}8.5$, Turbidity $<1\text{ NTU}$, TDS $<500\text{ mg/L}$), the ESP32 energizes the inline $12\text{V}$ solenoid valve, allowing water to flow through the 4-stage filter stack and UV-C LED chamber to the outlet tap [INFERENCE].
   • If any parameter breaches critical BIS 10500 limits (e.g., Turbidity $>5\text{ NTU}$ or pH $<6.5$), the valve automatically closes, preventing unsafe consumption. A red LED illuminates and an onboard piezo alarm sounds [INFERENCE].
4. Telemetry Sync & Offline Buffering: Telemetry packets formatted in JSON are transmitted via MQTT to the cloud broker. If cellular signal is lost, telemetry records are appended to an offline SPIFFS flash buffer on the ESP32. Upon cellular link restoration, the buffer flushes sequentially [INFERENCE].
5. Cloud Ingestion & Visualization: The Node.js backend ingests MQTT payloads, writes time-series logs to PostgreSQL/TimescaleDB, and pushes live updates via WebSockets to the React GIS Web Dashboard and Flutter Mobile App [INFERENCE].

12. Technology Stack

13. AI/ML Justification

AI/ML Evaluative Assessment

Is AI strictly required for basic water potability classification? NO. Water potability classification is governed deterministically by Bureau of Indian Standards (BIS IS 10500:2012) mathematical rules using the Weighted Arithmetic Water Quality Index (WAWQI) [KNOWN]. Employing an LLM or complex neural network for simple potability classification introduces hallucination risks, latency, and unnecessary computational cost [INFERENCE].

Justified AI Component 1: Anomaly Detection & Sensor Calibration Drift Prediction

• Purpose: Detect physical sensor fouling, cable degradation, or artificial probe tampering by flagging abnormal parameter combinations (e.g., sudden TDS drops to zero while conductivity remains high) [INFERENCE].
• Model Type: Lightweight Isolation Forest or Microcontroller TinyML (e.g., TensorFlow Lite for Microcontrollers) [INFERENCE].
• Input: Multi-variable time-series window (pH, TDS, Turbidity, Temp, Delta Time, Flow Rate) [INFERENCE].
• Output: Anomaly Score ($0\text{--}100$) and Sensor Health State (OK / DRIFT_DETECTED / FAULTY) [INFERENCE].
• Data Required: 5,000 synthetic and historical sensor drift sequences [ASSUMPTION].
• Why AI is Preferable: Fixed static threshold rules fail to detect subtle multivariate sensor calibration drift over time [INFERENCE].
• Evaluation Metric: F1-score ($>0.90$) on synthetic test datasets [INFERENCE].
• Failure Mode Handling: If anomaly score $>80%$, flag reading for manual re-calibration; do NOT trigger false water valve shut-off unless primary raw thresholds (BIS IS 10500) are explicitly breached [INFERENCE].
• Human Oversight: Local Jal Sahiya performs physical buffer-solution calibration when flagged by the anomaly alert [INFERENCE].

Justified AI Component 2: Predictive Filter Clogging & Remaining Useful Life (RUL)

• Purpose: Predict when the sediment filter and heavy-metal resin cartridges will clog, enabling preemptive maintenance dispatch [INFERENCE].
• Model Type: Linear Regression / Gradient Boosting Regressor (LightGBM) running on the cloud backend [INFERENCE].
• Input: Accumulated flow volume (Liters), average inlet turbidity, total differential pressure ($\Delta P$), and operating hours [INFERENCE].
• Output: Estimated Remaining Useful Life (RUL) in days and Liters [INFERENCE].
• Data Required: Bench test pressure drop curves across varying turbidity levels [INFERENCE].
• Why AI is Preferable: Clogging rate is non-linear and depends heavily on monsoon turbidity spikes and usage throughput [INFERENCE].
• Evaluation Metric: Mean Absolute Error (MAE $<3\text{ days}$) [INFERENCE].
• Failure Mode Handling: Fall back to a fixed volumetric replacement limit ($5,000\text{ Liters}$) if predictive model confidence drops [INFERENCE].
• Human Oversight: PHED district engineer validates replacement schedule before dispatching spare cartridges [INFERENCE].

Explicit Exclusion of AI

AI/ML must NOT be used to alter, bypass, or predict physical water purification processes or override safety shut-off valves. Safety valve shut-off MUST remain governed by deterministic C++ firmware logic running locally on the ESP32 [INFERENCE].

14. Data Architecture & Offline/Low-Network Design

Database Schema (PostgreSQL / TimescaleDB)

-- Devices Table
CREATE TABLE devices (
    device_id VARCHAR(64) PRIMARY KEY,
    location_name VARCHAR(128) NOT NULL,
    latitude DOUBLE PRECISION NOT NULL,
    longitude DOUBLE PRECISION NOT NULL,
    district VARCHAR(64) NOT NULL,
    installation_date DATE NOT NULL,
    status VARCHAR(20) DEFAULT 'ACTIVE'
);

-- Time-Series Water Telemetry Table (Hypertable)
CREATE TABLE water_telemetry (
    timestamp TIMESTAMPTZ NOT NULL,
    device_id VARCHAR(64) REFERENCES devices(device_id),
    ph NUMERIC(4,2) NOT NULL,
    turbidity NUMERIC(6,2) NOT NULL,
    tds NUMERIC(6,2) NOT NULL,
    temperature NUMERIC(4,2) NOT NULL,
    wqi_score NUMERIC(5,2) NOT NULL,
    valve_status BOOLEAN NOT NULL,
    battery_level NUMERIC(4,1) NOT NULL
);
SELECT create_hypertable('water_telemetry', 'timestamp');

-- Alerts Table
CREATE TABLE alerts (
    alert_id SERIAL PRIMARY KEY,
    device_id VARCHAR(64) REFERENCES devices(device_id),
    timestamp TIMESTAMPTZ NOT NULL,
    parameter_breached VARCHAR(32) NOT NULL,
    value_measured NUMERIC(6,2) NOT NULL,
    severity VARCHAR(20) CHECK (severity IN ('WARNING', 'CRITICAL')),
    resolved BOOLEAN DEFAULT FALSE
);

Offline & Low-Network Synchronization Design

1. Local Storage Engine: The ESP32 utilizes a $1.5\text{ MB}$ SPI Flash File System (SPIFFS) partition structured as a ring-buffer file storing binary telemetry records (Struct size: 32 bytes) [INFERENCE].
2. Capacity & Storage Calculation:
   • Sampling Rate: 1 record per 15 minutes = 96 records/day = $3.07\text{ KB/day}$ [INFERENCE].
   • $1.5\text{ MB}$ Flash memory stores up to $480\text{ days}$ of offline telemetry [INFERENCE].
3. Synchronization & Retry Protocol:
   • The edge device monitors signal availability using AT commands (AT+CSQ) [INFERENCE].
   • When signal quality is acceptable ($CSQ > 10$), the device sends buffered logs in compressed 50-record MQTT chunks [INFERENCE].
   • Exponential Backoff: Retry intervals double upon failure ($10\text{s}, 20\text{s}, 40\text{s}, \dots, \text{max } 1\text{ hour}$) to preserve battery charge [INFERENCE].
4. Bandwidth Reduction: Telemetry is encoded in compact Protocol Buffers (Protobuf) or optimized JSON keys ({"p":7.2,"t":1.1,"d":240,"w":18}) rather than verbose XML/REST payloads, reducing payload size by $80%$ [INFERENCE].

15. Security, Privacy & Governance

• Data Sensitivity Classification: Environmental water quality parameters (pH, turbidity, TDS) are non-personal, open public health data [KNOWN]. However, exact GPS coordinates of critical village drinking sources and user access logs must be protected against malicious tampering [INFERENCE].
• Authentication & Device Identity:
  • Every IoT device is provisioned with a unique X.509 client certificate or HMAC-SHA256 device key flashed into ESP32 secure storage [INFERENCE].
  • MQTT connections require TLS 1.2 encryption over port 8883 [KNOWN].
• Authorization & Access Control: Role-Based Access Control (RBAC):
  • Public User: View anonymized village WQI scores and public safety maps.
  • Jal Sahiya / Field Operator: Read/Write local BLE diagnostic logs and request re-calibration.
  • PHED Administrator: Full access to configure alert thresholds, update firmware over-the-air (OTA), and manage hardware assets.
• Governance & Misuse Prevention: Firmware updates must be cryptographically signed (RSA-2048) by the backend build pipeline. Unsigned binaries will be rejected by the ESP32 bootloader, preventing malicious firmware flashing [INFERENCE].
• Liability Disclaimer: The system acts as a real-time field screening and purification aid. Official legal potability certifications remain subject to NABL-accredited laboratory validation under Jal Jeevan Mission guidelines [RESEARCHED].

16. MVP Scope (Hackathon Build)

In Scope (Must Build for Live Demonstration)

1. Functional Hardware Prototype Rig: Fully integrated multi-stage filtration stack (Sediment filter + Activated Carbon cartridge + UF Membrane + 12V DC Pump + 275nm UV-C LED module) mounted on a portable frame [INFERENCE].
2. Sensor Telemetry Module: Live physical sensor array measuring pH, Turbidity, TDS, and Temperature using an ESP32 microcontroller [INFERENCE].
3. Edge Fail-Safe Actuation: Automated $12\text{V}$ solenoid shut-off valve that triggers physically when dipped into contaminated water (e.g., highly turbid or acidic test solution) [INFERENCE].
4. Edge WAWQI Calculation: Real-time computation of Water Quality Index directly on ESP32 firmware according to BIS IS 10500 rules [INFERENCE].
5. Cloud MQTT Pipeline & Web GIS Dashboard: Live React web dashboard displaying device location on a map, real-time sensor gauges, historical charts, and live alert banners [INFERENCE].
6. Mobile Flutter App (Android): Field operator interface displaying live BLE/Wi-Fi telemetry, green/red safety indicators, and maintenance ticket logs [INFERENCE].

Described Only / Out of Scope (Production System Extensions)

1. Custom ASIC Silicon / Industrial Enclosure: Custom IP67 injection-molded plastic chassis (standard off-the-shelf acrylic/aluminum frame used for hackathon demo) [INFERENCE].
2. Full NABL Laboratory Chemical Validation: Multi-month laboratory ICP-MS testing across all 80+ BIS IS 10500 parameters [NEEDS VALIDATION].
3. Production LoRaWAN City Gateways: Large-scale physical tower installations across all 24 districts of Jharkhand [NEEDS VALIDATION].

17. Live Demo Simulation Plan

To deliver a compelling and completely honest presentation without claiming non-existent integrations, the team will establish a live physical test bench during judging [INFERENCE].

+---------------------------------------------------------------------------------+
|                            LIVE DEMO BENCH SETUP                               |
|                                                                                 |
|  [ Beaker A: Pure Water ]             [ Beaker B: Mining Contaminated Simulation ] 
|  - pH: 7.2                            - pH: 4.5 (Dilute Acid addition)          |
|  - Turbidity: 0.8 NTU                 - Turbidity: 80 NTU (Bentonite Clay/Dirt) |
|  - TDS: 180 mg/L                      - TDS: 950 mg/L (NaCl / Mineral Salt)     |
|             |                                       |                           |
|             +-------------------+-------------------+                           |
|                                 |                                               |
|                                 v                                               |
|                  +------------------------------+                               |
|                  |   JalDrishti Sensing &       |                               |
|                  |   Purification Hardware      |                               |
|                  +------------------------------+                               |
|                                 |                                               |
|         +-----------------------+-----------------------+                       |
|         | Real-time Telemetry                           | Physical Valve        |
|         v                                               v                       |
|  +------------------------------+             +-------------------+             |
|  | Web GIS Dashboard & App      |             | Solenoid Valve    |             |
|  | - Live Gauge Updates         |             | - OPEN (Beaker A) |             |
|  | - Red Alert Banner Trigger   |             | - SHUT (Beaker B) |             |
|  +------------------------------+             +-------------------+             |
+---------------------------------------------------------------------------------+

Demonstration Execution Flow

1. Normal Condition Test (Beaker A): Probes are submerged in clean water. Dashboard displays Status: SAFE, WQI: 18 (Excellent), solenoid valve remains OPEN, and clean water flows smoothly from the outlet tap [INFERENCE].
2. Contamination Injection Test (Beaker B): Probes are transferred into simulated mine water (acidic pH, high turbidity clay mixture).
   • Within 3 seconds, raw sensor values breach limits on the ESP32 [INFERENCE].
   • The solenoid valve instantly CLICKS SHUT, stopping water flow [INFERENCE].
   • The onboard Red LED flashes and piezo buzzer sounds [INFERENCE].
   • The Web Dashboard instantly flashes a CRITICAL ALERT: Acidic pH & High Turbidity Breach banner, and the Flutter app receives a push notification [INFERENCE].
3. Simulated Disconnection Handling: The Wi-Fi router power is toggled OFF. The ESP32 logs readings to local SPIFFS flash. Upon turning Wi-Fi ON, buffered logs upload seamlessly to the dashboard without data loss [INFERENCE].

18. Implementation Roadmap

+-----------------------------------------------------------------------------------+
|                        SIH 2026 IMPLEMENTATION TIMELINE                           |
|                                                                                   |
|  PHASE 1: Hardware Rig & Sensing (May 2026 - Jun 2026)                            |
|  [||||||||||||||||||||] 100%                                                      |
|  - Multi-stage filter assembly & probe calibration                                |
|                                                                                   |
|  PHASE 2: Firmware & Edge Logic (Jul 2026)                                        |
|  [||||||||||||||||||||] 100%                                                      |
|  - ESP32 C++ WAWQI engine, valve shutoff & SPIFFS buffering                       |
|                                                                                   |
|  PHASE 3: Cloud Backend & Dashboard (Aug 2026)                                    |
|  [||||||||||||||||||||] 100%                                                      |
|  - MQTT broker, PostgreSQL database, React GIS map & Flutter app                  |
|                                                                                   |
|  PHASE 4: Bench Validation & Pitch Prep (Sep 1 - Sep 30, 2026)                    |
|  [||||||||||||||||||||] 100%                                                      |
|  - Live beaker test simulation, stress testing & final dossier                    |
+-----------------------------------------------------------------------------------+

Milestone Schedule (Targeting 30 September 2026 Deadline) [PS]

• Milestone 1 (30 June 2026): Mechanical assembly of 4-stage filter rig and electrical wiring of ESP32, pH, TDS, turbidity, and temperature probes [INFERENCE].
• Milestone 2 (31 July 2026): Firmware completion: Local WAWQI calculation, automated solenoid shut-off, SPIFFS ring-buffer logging, and MQTT transmission [INFERENCE].
• Milestone 3 (31 August 2026): Cloud platform & mobile app deployment: Node.js backend, PostgreSQL/TimescaleDB hypertable setup, React GIS map UI, and Flutter Android app BLE diagnostic mode [INFERENCE].
• Milestone 4 (20 September 2026): End-to-end physical integration testing with simulated mine water samples; fault-injection testing; demo script optimization [INFERENCE].
• Final Project Delivery (30 September 2026): Complete solution submission for SIH judging [PS].

19. Testing & Validation Plan

20. Risk Register

21. Cost & Scalability

Hackathon Prototype Build Cost

Production Scale Cost (1,000+ Units Deployment)

• Hardware Bill of Materials (BOM): Reduced to ₹6,500 – ₹7,500 per unit through custom PCB manufacturing, bulk media procurement, and direct OEM sensor sourcing [INFERENCE].
• Cloud Infrastructure: Approximately ₹45 per unit/month (AWS/GCP PostgreSQL, IoT Core, and SMS gateway costs) [INFERENCE].
• Scalability Path: Modular hardware cartridges allow deployment across diverse water matrices (e.g., swapping heavy-metal resin for fluoride-selective activated alumina in fluoride-heavy districts like Palamu) [RESEARCHED].

22. Differentiation

23. Future Scope

1. Multi-Spectrophotometric Heavy Metal Sensing: Integrate low-cost optical colorimetric sensors ($520\text{nm}/660\text{nm}$) for quantitative detection of dissolved Iron ($\text{Fe}^{3+}$) and Fluoride ($\text{F}^-$) concentrations [INFERENCE].
2. Community Water ATM Integration: Incorporate RFID card readers and pay-per-liter dispensing logic connected to local Gram Panchayat accounts [INFERENCE].
3. Automated Electro-Chemical Backwashing: Implement self-cleaning backwash solenoid cycles driven by pressure-differential triggers to extend UF membrane life from 6 months to 2 years without manual intervention [INFERENCE].

24. Assumptions & Open Items Log

Consolidated Assumptions & Open Items

• [ASSUMPTION] Feedwater total dissolved solids (TDS) in target rural mining villages in Jharkhand remain predominantly below $1,500\text{ mg/L}$, allowing effective treatment via Ultrafiltration and selective resin without requiring high-pressure RO [INFERENCE].
• [ASSUMPTION] Local Jal Sahiyas or village caretakers will be available to clean the 50-micron mesh pre-filter screen on a bi-weekly basis during peak monsoon months [INFERENCE].
• [NEEDS VALIDATION] Exact long-term degradation rates of analog optical turbidity sensors when continuously submerged in highly abrasive coal-dust water channels [NEEDS VALIDATION].
• [NEEDS VALIDATION] Official PHED approval procedures for installing inline IoT shut-off valves on public Jal Jeevan Mission piped water supply (PWS) schemes in Jharkhand [NEEDS VALIDATION].

Final Recommendation

Adopt Approach 2 (Hybrid Modular Ultrafiltration + Heavy-Metal Ion-Exchange Resin + Solar ESP32 IoT Edge System). This approach directly fulfills every explicit requirement of problem statement SIH26040 [PS] while avoiding the excessive power consumption and high water wastage of reverse osmosis systems [INFERENCE].

Recommended Stack

• Hardware: ESP32-WROOM-32U, DFRobot Analog Probes (pH, TDS, Turbidity), DS18B20 Temp Probe, SIM800L GSM, $12\text{V}$ DC Solenoid Valve, $12\text{V}$ DC Booster Pump, 275nm UV-C LED [INFERENCE].
• Purification Stack: 50-micron Mesh Pre-filter -> 5-micron PP Sediment -> Katalox-Light Media -> Heavy-Metal Selective Ion-Exchange Resin -> 0.01-micron Hollow-Fiber UF Membrane -> UV-C LED [INFERENCE].
• Backend & Cloud: Node.js, Express, Mosquitto MQTT, PostgreSQL + TimescaleDB [KNOWN].
• Frontend & Mobile: React.js + Leaflet Maps (Web), Flutter (Mobile App) [KNOWN].

Biggest Dependency

Availability of calibrated, reliable analog sensor probes (pH, TDS, turbidity) with stable ADC conversion logic on the ESP32 microcontroller platform [INFERENCE].

Biggest Risk

Probe surface fouling caused by fine coal dust accumulation, resulting in false sensor drift readings over extended field operations [INFERENCE].

Biggest Technical Challenge

Implementing low-noise ADC sampling and reliable polynomial software calibration on the ESP32 to prevent false-positive solenoid valve shut-offs [INFERENCE].

First 5 Actions

1. Procure ESP32 development board, analog sensor kit (pH, TDS, turbidity, temp), $12\text{V}$ solenoid valve, and Relay module [INFERENCE].
2. Wire sensors to ESP32 ADC pins and calibrate readings against certified buffer solutions ($\text{pH } 4.01, 7.00, 10.01$) [INFERENCE].
3. Assemble 4-stage physical filter rig (Sediment + Carbon/Katalox + Heavy Metal Resin + UF + UV-C LED) on a portable frame [INFERENCE].
4. Write ESP32 C++ firmware for local WAWQI calculation, SPIFFS logging, and automated solenoid valve shut-off logic [INFERENCE].
5. Deploy Node.js MQTT server and React web GIS dashboard to demonstrate real-time telemetry streaming and live alert triggers [INFERENCE].

Validation Before Coding

• Verify ADC voltage stability on ESP32 GPIO pins using an oscilloscope/multimeter under active sensor load [INFERENCE].
• Perform benchtop flow testing to confirm that the $12\text{V}$ DC booster pump achieves $>1.5\text{ Liters/minute}$ flow through the complete 4-stage filter stack [INFERENCE].

Buildability Score /100

88 / 100 (Highly feasible for a 4–6 student engineering team within the available timeframe using commercially off-the-shelf sensors, modular filtration components, standard ESP32 microcontrollers, and modern web frameworks) [INFERENCE].

Confidence

High (0.92) – Fully supported by problem statement requirements [PS], published BIS IS 10500 standards [KNOWN], and groundwater research data from Central Ground Water Board reports in Jharkhand mining belts [RESEARCHED].

Research Sources

1. Title: Bureau of Indian Standards - Drinking Water Specification (IS 10500:2012) [KNOWN]
   • URL: https://infralens.in/code/IS-10500-2012
   • Source Type: Official National Standard Organization.
   • Supported Claim: Acceptable and permissible limits for pH (6.5–8.5), Turbidity (1–5 NTU), TDS (500–2000 mg/L), Iron (0.3 mg/L), Fluoride (1.0–1.5 mg/L), and Nitrate (45 mg/L).
   • Reliability: High (Official BIS standard).
2. Title: Central Ground Water Board (CGWB) Ground Water Quality Reports - Jharkhand [RESEARCHED]
   • URL: https://cgwb.gov.in/en/ground-water-quality
   • Source Type: Official Government Portal / Academic Report.
   • Supported Claim: Presence of heavy metals (Iron, Manganese, Lead), fluoride, and nitrate contamination across groundwater sources in Jharkhand mining districts.
   • Reliability: High (Official Government Hydrogeological Agency).
3. Title: Jal Jeevan Mission - Water Quality Monitoring and Information System (WQMIS) [RESEARCHED]
   • URL: https://ejalshakti.gov.in/
   • Source Type: Official Government Portal.
   • Supported Claim: National framework for rural water testing laboratories and community monitoring under Jal Jeevan Mission.
   • Reliability: High (Ministry of Jal Shakti, Govt of India).
4. Title: A Study on Groundwater Quality Based on Major Ion Chemistry of Jharkhand State in India [RESEARCHED]
   • URL: https://www.orientjchem.org/vol37no4/a-study-on-groundwater-quality-based-on-major-ion-chemistry-of-jharkhand-state-in-india-a-review/
   • Source Type: Peer-Reviewed Academic Research Paper.
   • Supported Claim: Characterization of water pollution, coal mine runoff, and elevated iron/fluoride in West Bokaro and Dhanbad regions.
   • Reliability: High (Peer-reviewed publication).
5. Title: Application of the Weighted Arithmetic Water Quality Index (WAWQI) in Assessing Water Quality [RESEARCHED]
   • URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC10006569/
   • Source Type: Peer-Reviewed Scientific Literature (PMC).
   • Supported Claim: Mathematical formulation and rating scales for Weighted Arithmetic Water Quality Index (WAWQI) evaluation.
   • Reliability: High (PubMed Central / Scientific Literature).