Faradai is an industrial energy intelligence platform designed to optimize electricity consumption, reduce tariff penalties, orchestrate distributed energy assets, and transform industrial energy infrastructure into an intelligent, real-time optimization network.
The platform combines:
- Industrial IoT
- Smart metering
- Real-time streaming systems
- Optimization engines
- Energy analytics
- Demand response orchestration
- Tariff intelligence
- Distributed energy management
The core business thesis:
Indian factories are losing massive amounts of money due to:
- demand charge spikes
- poor power factor
- tariff inefficiencies
- solar mismanagement
- energy visibility gaps
- battery underutilization
- billing errors
- lack of real-time optimization
Faradai becomes the operating system for industrial electricity.
India’s industrial sector consumes approximately 42% of national electricity.
Most industrial energy management today is:
- manual
- spreadsheet-driven
- reactive
- non-real-time
- operationally fragmented
Factories receive electricity bills worth crores annually but lack:
- real-time visibility
- optimization capability
- predictive intelligence
- load coordination
- tariff understanding
The result:
- avoidable penalties
- inefficient operations
- wasted renewable energy
- poor asset utilization
- hidden energy leakage
Factories pay penalties based on peak demand.
Simultaneous startup of:
- compressors
- furnaces
- chillers
- motors
causes sudden spikes.
One 15-minute spike can increase the monthly bill significantly.
Poor PF leads to recurring DISCOM penalties.
Most factories do not monitor PF continuously.
Factories consume expensive electricity during peak hours despite optimization opportunities.
Factories are rapidly adopting:
- rooftop solar
- BESS systems
- EV charging
But lack orchestration intelligence.
Industrial electricity billing is highly complex.
Errors include:
- incorrect demand calculations
- tariff misclassification
- reactive energy mistakes
- metering anomalies
Reduce industrial electricity costs using:
- real-time monitoring
- demand spike prediction
- load optimization
- tariff intelligence
Create an industrial energy optimization network integrating:
- solar
- batteries
- smart loads
- energy markets
- carbon systems
Become India’s:
Coordinating:
- industrial demand response
- distributed energy assets
- utility load balancing
- virtual power plants
- carbon intelligence
Faradai consists of:
| Layer | Function |
|---|---|
| Metering Layer | Collect electrical data |
| Edge Layer | Gateway + local processing |
| Cloud Layer | Data ingestion + storage |
| Analytics Layer | Monitoring + pattern analysis |
| Optimization Layer | Cost minimization engine |
| Intelligence Layer | Forecasting + ML |
| Experience Layer | Dashboard + alerts |
| Utility Layer | Demand response + grid integration |
Collect high-frequency electrical data from industrial systems.
| Parameter | Description |
|---|---|
| Voltage | Line voltage |
| Current | Load current |
| Active Power | kW |
| Reactive Power | kVAR |
| Apparent Power | kVA |
| Frequency | Grid frequency |
| Power Factor | PF |
| Harmonics | THD |
| Energy | kWh |
| Demand | Peak demand |
Initial deployments use third-party industrial meters.
Examples:
- Schneider
- Janitza
- L&T
- Secure
- Genus
Bridge industrial hardware with cloud infrastructure.
- Poll smart meters
- Read Modbus registers
- Timestamp data
- Buffer locally
- Handle connectivity failures
- Push data to cloud
- Edge preprocessing
- Raspberry Pi
- Industrial mini PC
- ARM gateways
- Custom industrial edge gateway
| Protocol | Purpose |
|---|---|
| Modbus RTU | Meter communication |
| RS485 | Physical industrial communication |
| MQTT | Cloud streaming |
| TCP/IP | Internet communication |
Move industrial telemetry reliably from factories to cloud.
Smart Meter → Gateway → MQTT Broker → Backend Services
- MQTT
- FastAPI
- PostgreSQL/TimescaleDB
- Kafka
- Redis Streams
- Kubernetes
- Event-driven microservices
Store and process high-frequency industrial telemetry.
- TimescaleDB
- InfluxDB
- ClickHouse
Example:
| Timestamp | Factory | Meter | Power |
|---|---|---|---|
| 10:00:01 | F1 | M1 | 120 |
| 10:00:02 | F1 | M1 | 124 |
| 10:00:03 | F1 | M1 | 130 |
Millions of rows/day.
The optimization engine predicts and prevents expensive demand spikes.
- Real-time load data
- Historical patterns
- Tariff structure
- Machine priority
- Production schedules
- Load staggering suggestions
- Demand risk alerts
- Automated scheduling recommendations
Current demand: 1.3 MW
Contract limit: 1.5 MW
Prediction: 1.68 MW in 4 minutes.
Suggested action: Delay Compressor-2 startup by 7 minutes.
Understand and optimize Indian industrial tariff structures.
- State tariff database
- ToD tariff engine
- PF penalty calculation
- Demand charge simulation
- Open Access analysis
- Billing error detection
Initial:
- Maharashtra
- Gujarat
- Tamil Nadu
- Karnataka
Deliver actionable operational alerts.
- SMS
- Mobile app
- Dashboard notifications
⚠ Demand Spike Warning
Predicted Peak: 1.62 MW Possible Penalty: ₹2.4 lakh
Suggested Action: Delay Compressor-2 startup by 7 minutes.
- Real-time demand
- Power quality
- Load distribution
- Solar generation
- Demand savings
- PF savings
- ToD optimization
- Carbon reductions
- Load curves
- Peak analysis
- Equipment utilization
Optimize distributed energy resources.
- Solar utilization tracking
- Export optimization
- Curtailment detection
- Forecasting
- Charge/discharge scheduling
- Peak shaving
- Arbitrage optimization
- Backup reserve management
Coordinate industrial load flexibility during grid stress.
DISCOM requests: Reduce 20 MW from 6–8 PM.
Platform coordinates:
- HVAC load reduction
- battery discharge
- compressor staggering
- non-critical load shifting
Across multiple factories.
Factories get paid. Grid stabilizes. Platform takes orchestration fee.
DISCOM Grid
↓
Factory HT Panel
↓
Industrial Loads
↓
Smart Meters
↓
RS485 / Modbus
↓
IoT Gateway
↓
MQTT / Internet
↓
Cloud Backend
↓
Streaming + Analytics
↓
Optimization Engine
↓
Alerts + Dashboard
Machines consume electricity.
Meters continuously measure:
- voltage
- current
- kW
- PF
- harmonics
Values stored internally in Modbus registers.
Example:
| Register | Value |
|---|---|
| 1001 | Voltage |
| 1002 | Current |
| 1003 | Power |
Gateway requests values periodically.
Example:
Read register 1001.
Meter returns: 440V.
Gateway creates JSON payload.
Example:
{
"factory_id": "F101",
"meter_id": "M12",
"timestamp": "2026-05-27T10:00:01",
"voltage": 440,
"current": 220,
"power_kw": 125,
"pf": 0.92
}Gateway publishes packet.
Cloud subscribes.
Backend validates:
- schema
- device identity
- timestamps
- integrity
Data stored in time-series DB.
Platform identifies:
- trends
- anomalies
- demand spikes
- power quality events
Optimization engine determines:
- which loads to delay
- when to use battery
- when to consume solar
System sends operational recommendations.
Monthly reports quantify:
- money saved
- penalties avoided
- optimization gains
- React
- Next.js
- TypeScript
- TailwindCSS
- Recharts
- Python
- FastAPI
- Node.js (optional services)
- MQTT
- Kafka
- TimescaleDB
- PostgreSQL
- Redis
- AWS
- GCP
- Azure
- Raspberry Pi
- Industrial Linux gateway
- Modbus RTU
- RS485
- MQTT
- TCP/IP
- OPC-UA (future)
- Python
- Pandas
- NumPy
- CVXPY
- OR-Tools
- PyTorch
Examples:
- demand threshold rules
- PF alerts
- ToD recommendations
- demand forecasting
- load prediction
- anomaly detection
- battery orchestration
- solar forecasting
- MPC (Model Predictive Control)
- reinforcement learning
- time-series forecasting
- anomaly detection
- demand prediction
- battery dispatch optimization
- load orchestration
- tariff forecasting
- virtual power plant coordination
Analyze historical electricity bills.
Install smart meters and gateway.
Collect 30–60 days baseline patterns.
System starts:
- spike prediction
- alerting
- recommendations
Generate documented M&V reports.
Add:
- solar
- battery
- multiple plants
- demand response
- Meter deployment
- Gateway installation
- Electrical installation
Real-time monitoring.
Demand optimization + savings.
Solar + battery + demand response.
15–25% of proven savings.
- carbon credit commission
- demand response aggregation fees
- utility partnerships
- Open Access advisory
- ₹3–25Cr annual electricity bill
- HT connection
- rooftop solar installed
- energy-intensive operations
- textile dyeing
- cold storage
- paper mills
- steel rerolling
- packaging
- chemicals
- Maharashtra
- Gujarat
- Tamil Nadu
- Karnataka
Sell directly using:
- electricity bill analysis
- savings audit
- free 60-day pilot
Channel partners:
- energy auditors
- solar EPC companies
- tariff consultants
Strategic partnerships:
- industrial estates
- DISCOM DSM cells
- carbon markets
Customers do not buy:
- dashboards
- AI buzzwords
- graphs
Customers buy:
Core value proposition:
"You saved ₹4.2 lakh this month."
Industrial load behavior dataset.
Demand reduction algorithms.
State-specific energy economics.
Gateway + metering integration.
DISCOM partnerships.
Aggregated industrial flexibility.
- demand optimization
- tariff engine
- alerts
- savings reports
- solar optimization
- battery orchestration
- advanced forecasting
- carbon modules
- DISCOM integrations
- demand response aggregation
- virtual power plant layer
Faradai evolves from:
Energy monitoring → Industrial optimization → Distributed energy orchestration → Grid-edge infrastructure → Virtual power plant network
Ultimate vision:
- industrial protocol complexity
- data reliability
- optimization accuracy
- field deployments
- hardware failures
- installation quality
- DISCOM bureaucracy
- long industrial sales cycles
- savings attribution disputes
- factories connected
- MW monitored
- demand spikes prevented
- uptime
- customer savings
- ARR
- savings-share revenue
- churn
- DISCOM partnerships
- demand response MW
- managed electricity spend
- carbon credits processed
Faradai is NOT:
- dashboard software
- energy monitoring SaaS
- generic IoT company
Faradai IS:
The company sits at the intersection of:
- industrial systems
- distributed energy
- optimization engines
- grid infrastructure
- utility intelligence
- climate technology
Over time, every connected factory improves:
- optimization models
- forecasting accuracy
- tariff intelligence
- demand response capability
creating a compounding network effect.
The long-term outcome is a national-scale grid-edge intelligence network coordinating industrial electricity usage in real time.
faradai/
│
├── apps/
│ ├── api/
│ ├── dashboard/
│ ├── gateway/
│ ├── optimization-engine/
│ ├── analytics-engine/
│ ├── alert-engine/
│ ├── ml-engine/
│ ├── report-engine/
│ ├── tariff-engine/
│ └── admin-panel/
│
├── services/
│ ├── auth-service/
│ ├── user-service/
│ ├── factory-service/
│ ├── meter-service/
│ ├── telemetry-service/
│ ├── websocket-service/
│ ├── billing-service/
│ ├── notification-service/
│ └── audit-service/
│
├── packages/
│ ├── shared-types/
│ ├── shared-utils/
│ ├── energy-calculations/
│ ├── tariff-rules/
│ ├── mqtt-client/
│ ├── modbus-client/
│ ├── logging/
│ ├── config/
│ └── ui-components/
│
├── infrastructure/
│ ├── docker/
│ ├── kubernetes/
│ ├── terraform/
│ ├── nginx/
│ ├── monitoring/
│ ├── mqtt/
│ ├── kafka/
│ └── scripts/
│
├── databases/
│ ├── migrations/
│ ├── seeds/
│ ├── schemas/
│ └── backups/
│
├── docs/
│ ├── architecture/
│ ├── api/
│ ├── protocols/
│ ├── deployment/
│ ├── optimization/
│ └── security/
│
├── tests/
│ ├── integration/
│ ├── load/
│ ├── e2e/
│ └── performance/
│
├── tools/
│ ├── simulators/
│ ├── billing-parsers/
│ ├── tariff-importers/
│ └── data-generators/
│
├── .github/
├── .env
├── docker-compose.yml
├── pnpm-workspace.yaml
├── README.md
└── Makefile
apps/api/
│
├── src/
│ ├── routes/
│ ├── controllers/
│ ├── services/
│ ├── middleware/
│ ├── schemas/
│ ├── validators/
│ ├── websocket/
│ ├── events/
│ ├── repositories/
│ ├── models/
│ ├── config/
│ └── main.py
│
├── tests/
├── Dockerfile
└── requirements.txt
apps/dashboard/
│
├── src/
│ ├── app/
│ ├── components/
│ ├── modules/
│ ├── hooks/
│ ├── services/
│ ├── charts/
│ ├── store/
│ ├── layouts/
│ ├── styles/
│ └── utils/
│
├── public/
├── package.json
└── next.config.js
apps/gateway/
│
├── src/
│ ├── modbus/
│ ├── mqtt/
│ ├── polling/
│ ├── buffering/
│ ├── device-drivers/
│ ├── local-storage/
│ ├── retry-engine/
│ ├── health-monitor/
│ ├── configs/
│ └── main.py
│
├── tests/
└── requirements.txt
apps/optimization-engine/
│
├── src/
│ ├── algorithms/
│ ├── constraints/
│ ├── solvers/
│ ├── schedulers/
│ ├── models/
│ ├── rules/
│ ├── forecasting/
│ ├── simulations/
│ ├── dispatch/
│ └── engine.py
apps/optimization-engine/
│
├── src/
│ ├── algorithms/
│ ├── constraints/
│ ├── solvers/
│ ├── schedulers/
│ ├── models/
│ ├── rules/
│ ├── forecasting/
│ ├── simulations/
│ ├── dispatch/
│ └── engine.py
apps/tariff-engine/
│
├── src/
│ ├── states/
│ │ ├── maharashtra/
│ │ ├── gujarat/
│ │ ├── tamilnadu/
│ │ └── karnataka/
│ │
│ ├── billing/
│ ├── penalties/
│ ├── tod/
│ ├── open-access/
│ └── engine.py