A greenhouse.
A connected system.

A greenhouse technology provider required a centralized platform to monitor environmental conditions, automate supplemental lighting, and integrate heterogeneous sensors and industrial control equipment.

The solution needed to connect physical greenhouse infrastructure with a modern cloud platform, enabling real-time monitoring, configurable lighting strategies, remote equipment control, and data-driven operational decisions.

The project involved designing and engineering an end-to-end IoT ecosystem, spanning embedded edge devices, industrial communication protocols, cloud infrastructure, backend services, and a real-time web application.

Primary objective: Transform independently operating greenhouse devices into a unified, cloud-connected automation platform capable of monitoring and controlling supplemental lighting based on actual environmental conditions and configurable cultivation targets.

Greenhouse lighting management presents several technical and operational challenges.

Fragmented hardware ecosystem

Sensors, lighting controllers, and other industrial devices frequently rely on different communication interfaces and protocols.

A unified platform must integrate these devices without requiring greenhouse operators to interact with each manufacturer's software independently.

Dynamic natural lighting

Natural sunlight varies throughout the day.

Supplemental lighting systems must adapt to changing daylight conditions rather than operate exclusively on fixed schedules.

Photoperiod and light exposure targets

Different cultivation strategies require specific lighting periods and target levels of accumulated light.

Maintaining these targets requires continuous measurement, aggregation, and control decisions.

Operational visibility

Operators need centralized access to environmental measurements, equipment status, automation settings, and historical performance.

Reliable edge-to-cloud communication

The platform must coordinate cloud services with devices operating inside greenhouses, where network conditions and hardware capabilities can vary.

The solution was designed as a distributed IoT architecture with four principal layers.

Layer 1 — Greenhouse Sensors & Equipment

Integration with hardware devices used for environmental measurement and lighting control.

The lighting system incorporates measurements such as:

The platform supports multiple device integration mechanisms to accommodate different hardware capabilities.

Layer 2 — IoT Edge Gateway

Python-based gateway services run on Raspberry Pi devices deployed within the greenhouse environment.

Responsibilities:

Containerized deployment using Docker enables consistent installation and maintenance across edge devices.

Ansible supports automated gateway deployment and configuration.

Layer 3 — Azure Cloud Platform

Microsoft Azure provides centralized application hosting, device communication, data persistence, and automation infrastructure.

Core components:

Infrastructure provisioning is automated through Terraform, while CI/CD pipelines support application delivery.

Layer 4 — Monitoring & Control Application

A browser-based interface provides operational visibility and configuration capabilities.

Technology stack:

React 19, TypeScript, Vite, Tailwind CSS, SWR, Recharts, and Server-Sent Events.

The frontend enables greenhouse operators to:

A central capability of the platform is the automation of supplemental greenhouse lighting.

Rather than relying solely on predefined on/off schedules, the platform evaluates measured light conditions against cultivation targets.

Photoperiod management

Operators configure lighting periods and operational windows.

The automation system uses these settings to determine when supplemental lighting is permitted or required.

Daily Light Integral tracking

The platform calculates accumulated light exposure using PPFD measurements over time.

DLI represents the total photosynthetically active light delivered per unit area during a defined period, expressed in mol/m²/day.

To estimate accumulated light, the system integrates PPFD measurements over time.

This enables lighting decisions to consider the light already delivered by natural sunlight and artificial sources.

Daylight and supplemental lighting

The automation distinguishes between contributions from daylight and supplemental lighting to support more informed decisions about artificial light requirements.

Dynamic light adjustment

The control logic evaluates measured conditions relative to configured lighting targets.

The objective is to supply additional light when required while avoiding unnecessary lighting operation when sufficient natural light is available.

Control stability

Lighting automation must also avoid excessive switching and rapid oscillations caused by short-term sensor fluctuations.

The solution incorporates control logic intended to maintain stable operation while responding to relevant changes in greenhouse conditions.

The backend was developed using .NET 10 and ASP.NET Core Web API, following Clean Architecture principles.

The implementation uses:

The backend coordinates configuration, sensor information, lighting control workflows, and communication with the IoT infrastructure.

This architecture separates business rules and automation decisions from device-specific implementation details.

The frontend was designed to support both operational monitoring and system administration.

Monitoring dashboards

Operators can inspect sensor measurements, lighting-related indicators, and equipment information from a centralized interface.

Historical visualization

Time-series charts provide visibility into changes in light exposure and system behavior.

Live updates

Server-Sent Events enable backend-driven updates to connected clients, reducing the need for continuous frontend polling.

Device management

The platform supports the configuration and monitoring of connected greenhouse equipment through the cloud application.

Together, these capabilities provide a unified operational interface across previously fragmented hardware systems.

The solution includes cloud and edge deployment workflows.

Cloud infrastructure

Terraform defines and provisions the Azure infrastructure required by the application.

Application delivery

CI/CD pipelines automate building, testing, and deploying application components.

Edge deployment

Python gateway applications are containerized using Docker and deployed to Raspberry Pi devices.

Ansible automates gateway setup and deployment operations.

This approach supports repeatable releases across cloud services and distributed edge installations.

The engagement involved extensive end-to-end technical ownership.

Responsibilities included:

The work combined solution architecture, hands-on software engineering, industrial protocol integration, and cloud infrastructure management.

The platform was designed to deliver several key operational advantages.

Centralized visibility

A single application for monitoring greenhouse lighting and connected equipment.

Automated lighting decisions

Reduced reliance on manual control through configurable lighting targets and sensor-driven automation.

Improved resource utilization

The ability to adapt supplemental lighting to measured daylight conditions creates opportunities to reduce unnecessary electricity consumption.

Hardware interoperability

A gateway-based approach makes it possible to integrate equipment with different communication protocols into a common cloud architecture.

Scalable architecture

The separation of edge gateways, cloud services, and web applications provides a foundation for deployment across additional greenhouse installations.

Foundation for advanced optimization

Historical data, configurable cultivation parameters, and automated device control create opportunities for future predictive analytics and optimization.

These benefits describe the intended and enabled capabilities of the platform. Quantified energy savings, productivity gains, and financial returns have not been established in this case study.

This project demonstrates the engineering of a production-oriented industrial IoT platform that connects physical greenhouse infrastructure with a scalable cloud application.

Its distinguishing characteristics include:

The core engineering achievement was turning heterogeneous greenhouse equipment and sensor measurements into a unified, configurable automation system.

The architecture also establishes a technical foundation for more advanced agricultural decision-support capabilities, including predictive lighting optimization and data-driven resource management.