Design of a Smart Chili Mini Greenhouse with ESP32 and Model Predictive Control

Abstract

Chili cultivation requires stable environmental conditions to achieve optimal plant growth. However, conventional control methods such as On-Off control are reactive and less efficient in handling dynamic environmental changes. This study develops a mini greenhouse system based on the Internet of Things (IoT) integrated with Model Predictive Control (MPC) to maintain the stability of the microclimate for chili plants. The system uses an ESP32 microcontroller as the main controller, DHT22 and soil moisture sensors to monitor temperature, air humidity, and soil moisture, as well as actuators in the form of a fan and water pump. The MPC algorithm is applied to predict system conditions and determine optimal control actions. The system is also integrated with an IoT platform and Telegram notifications for real-time monitoring. Simulation results show that the MPC method provides more stable temperature control compared to the On-Off control method, with overshoot reduced from approximately ±2°C to ±0.5°C and a faster settling time. Therefore, the developed system can improve microclimate control stability and operational efficiency in a mini greenhouse.