Modeling and nonlinear predictive control of solar thermal systems in district heating

Area of Science:

• Renewable Energy Systems
• Control Engineering
• Mathematical Modeling

Background:

• District heating systems increasingly integrate solar thermal energy.
• Efficient operation of solar thermal plants is crucial for maximizing renewable energy utilization.
• Existing control strategies may not fully capture the dynamic complexities of solar thermal plants.

Purpose of the Study:

• To develop and implement a data-driven nonlinear model predictive control (NMPC) strategy for optimizing solar thermal plant operation.
• To enhance the energy efficiency and performance of solar thermal plants in district heating networks.
• To analyze the effectiveness of the proposed NMPC through simulations and real-world data.

Main Methods:

• Grey-box modeling using stochastic differential equations and real-world plant data to create a dynamic model.
• Nonlinear Model Predictive Controller (NMPC) design based on repeated trajectory linearization.
• Simulation analysis to evaluate model accuracy and controller performance under various objectives (e.g., maximizing energy or temperature).

Main Results:

• The developed dynamic model accurately represents the daytime operation of the solar thermal plant.
• The NMPC strategy significantly improves operational efficiency.
• Simulations demonstrate an increase in transported energy by up to 28%.

Conclusions:

• Data-driven modeling and NMPC are effective for optimizing solar thermal plant operations in district heating.
• The proposed control approach offers a substantial improvement in energy delivery and system efficiency.
• This methodology provides a valuable tool for enhancing the performance of renewable energy integration in heating systems.