Energy optimization of bitcoin mining integrated greenhouse with model predictive control

This study presents a novel control framework that integrates bitcoin mining waste heat with greenhouse climate regulation, leveraging model predictive control to enhance energy efficiency and economic viability. While the feasibility of using bitcoin mining waste heat for greenhouse heating through thermal modeling and parametric studies has been explored, this work focuses on real-time predictive control, dynamically regulating temperature, humidity, and CO₂ concentration based on external weather conditions and crop requirements. The MPC framework manages the dynamic and interconnected climate conditions of the greenhouse by leveraging real-time weather data and predictive models. By dynamically controlling actuators such as fans, blinds, heating systems, and CO₂ injection, the framework ensures optimal conditions for crop growth while minimizing energy costs. Simulations are conducted for greenhouses of varying sizes (individual, semi-commercial, and commercial) across eight cities representing diverse climate zones. Results show that MPC outperforms On-Off control by reducing energy consumption by up to 15 %, while maintaining climate conditions within target ranges over 95 % of the time. Furthermore, economic analysis demonstrates that integrating bitcoin mining waste heat results in significant profitability, with net annual profits reaching up to $1.5 million for commercial-scale greenhouses under high bitcoin prices. The findings establish a scalable and intelligent control strategy for integrating bitcoin mining with agriculture, advancing sustainable food production while optimizing energy use and reducing greenhouse gas emissions in agricultural operations.