Development of a High-Resolution Dynamic Emulator for a Chiller Plant and Its Application to Control Improvement

¹ The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8654, Japan
² Taikisha Ltd, Sumitomo Fudosan Shinjuku Grand Tower, 8-17-1 Nishi-Shinjuku, Shinjuku-ku, Tokyo 160-6129, Japan
³ DAI-DAN CO., LTD, 1-9-25 Edobori, Nishi-ku, Osaka 550-8520, Japan
⁴ Mitsubishi Heavy Industries Ltd., Marunouchi, Chiyoda-ku, Tokyo 100-0005, Japan

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

With the rise of smart buildings enabling the scalable deployment of applications for environmental and energy system, a verification platform is essential to validate their effectiveness, advantages, and safety. However, conventional energy simulations, typically using 1 -hour intervals, fail to capture transient behaviors like hunting and overshoot, leaving risks of instability. To address this, this paper developed a high-fidelity dynamic "emulator"—a detailed simulation reproducing physical phenomena and control logic—using OpenModelica. Based on 1-second interval data from an actual building, the model incorporated chiller transients, specific control logic with deadbands, and piping thermal delays. The emulator successfully reproduced the system's dynamic behaviors and revealed actual sensor errors and control deficiencies. Furthermore, through a case study on control improvement to determine appropriate parameters for system stabilization, we demonstrated that optimizing staging thresholds could eliminate hunting and stabilize temperatures, thereby improving the operational stability of the chiller plant. In the future, this emulator serves as a critical testbed for developing and verifying smart building applications, enhancing both energy efficiency and system stability.