Open-source tool-box for identifying envelope dynamics of detached residential buildings.

Summary

The rapid electrification of residential buildings has created rising concerns about the ability of the power grid to deal with seasonal electric space conditioning peaks. Better control of home heating and cooling systems can help solve this problem. However, unlike other control areas where extensive data are often available, each home presents a unique plant for which data are limited. This complicates the development of advanced control systems. This paper presents an open-source toolbox that researchers and industry can use as a plug-and-play tool for identifying the thermal dynamics of detached single-family homes. The tool supports identification of 3R2C and 2R1C thermal circuit models and can be trained using either thermal or electrical data alongside temperature measurements. The 2R1C method uses a computationally inexpensive constrained least squares formulation, while the 3R2C used a meta-heuristic genetic algorithm. The thermal mass temperature is treated as a hidden state with a Kalman Filter in place. The system is tested on a real-world building for prediction accuracy of indoor temperature and heat supply. In these tests, the root-mean-square hour-ahead temperature prediction errors are 0.8 ℃ and 0.57 ℃ for the 2R1C and 3R2C models, respectively. Under highly transient indoor conditions, such as fast indoor temperature set-point changes, the 3R2C performs better, while under more steady conditions the two model structures perform similarly.

Available documents

Format PDF

Pages: 10 p.

Available

Free

Details

  • Original title: Open-source tool-box for identifying envelope dynamics of detached residential buildings.
  • Record ID : 30032969
  • Languages: English
  • Source: 2024 Purdue Conferences. 8th International High Performance Buildings Conference at Purdue.
  • Publication date: 2024/07/15

Links


See other articles from the proceedings (63)
See the conference proceedings