Share

Energy-saving control method of NH3-CO2 cascade refrigeration system driven by a hybrid mechanism-based and data-based approach.

Number: 2185

Author(s) : FENG Y., WANG C., QU S., LI Y., REN D., XING Z.

Summary

System-level intelligent and intensive control is crucial for achieving energy efficiency upgrades in existing refrigeration systems. This study proposed an energy-saving control method for refrigeration systems based on a hybrid mechanism-based and data-based model. The aim of this study is to address the drawbacks of high calculation errors in single mechanism-driven control method and high dependence on historical data in single data-driven control method. Firstly, a big database of system operating parameters was established, which consisted of data collected from sensors and calculation results derived from the mechanism-based system simulation model. This extensive database can cover the entire operating range of the system. Based on this database, a data-driven system simulation model was developed to accurately predict the operating status and energy consumption of the system. Then, in order to achieve real-time energy-saving optimization of system control parameters, a nonlinear multivariate function optimization algorithm was adopted. The application of this energy-saving control method in NH3-CO2 cascade refrigeration system significantly reduces energy consumption by 10%-26%. Furthermore, when compared to data-driven models that entirely establish on historical senor data and mechanism-based models that heavily depend on assumptions, the hybrid-based model demonstrated improved system simulation accuracy by more than 2% and 10%, respectively. In conclusion, this study provides a promising direction for improving system performance in refrigeration systems.

Available documents

Format PDF

Pages: 10 p.

Available

Free

Download

Details

  • Original title: Energy-saving control method of NH3-CO2 cascade refrigeration system driven by a hybrid mechanism-based and data-based approach.
  • Record ID : 30033069
  • Languages: English
  • Subject: Technology
  • Source: 2024 Purdue Conferences. 20th International Refrigeration and Air-Conditioning Conference at Purdue.
  • Publication date: 2024/07/17

Links


See other articles from the proceedings (187)
See the conference proceedings

Indexing