• Electronics and Telecommunications Trends
• /
• v.38 no.4
• /
• pp.95-103
• /
• 2023

The importance of the decentralized training with decentralized execution (DTDE) framework is well-known in the study of multiagent reinforcement learning. In many real-world environments, agents cannot share information. Hence, they must be trained in a decentralized manner. However, the DTDE framework has been less studied than the centralized training with decentralized execution framework. One of the main reasons is that many problems arise when training agents in a decentralized manner. For example, DTDE algorithms are often computationally demanding or can encounter problems with non-stationarity. Another reason is the lack of simulation environments that can properly handle the DTDE framework. We discuss current research trends in the DTDE framework.

• KIEAE Journal
• /
• v.16 no.6
• /
• pp.13-19
• /
• 2016

Purpose: In this study, we examined outdoor air fraction using historical data of actual Air Handling Unit (AHU) in the existing building during intermediate season and analyzed optimal outdoor air fraction by control types for economizer. Method: Control types for economizer which was used in analysis are No Economizer(NE), Differential Dry-bulb Temperature(DT), Diffrential Enthalpy(DE), Differential Dry-bulb Temperature+Differential Enthalpy(DTDE), and Differential Enthalpy+Differential Dry-bulb Temperature (DEDT). In addition, the system heating and cooling load were analyzed by calculating the outdoor air fraction through existing AHU operating method and control types for economizer. Result: Optimized outdoor air fraction through control types was the lowest in March and distribution over 50% was shown in May. In case of DE control type, outdoor air fraction was the highest of other control types and the value was average 63% in May. System heating load was shown the lowest value in NE, however, system cooling load was shown 1.7 times higher than DT control type and 5 times higher than DE control type. For system heating load, DT and DTDE is similar during intermediate season. However, system cooling load was shown 3 times higher than DE and DEDT. Accordingly, it was found as the method to save cooling energy most efficiently with DE control considering enthalpy of outdoor air and return air in intermediate season.