An application status review of computational intelligence algorithm in GSHP field

Zhou, Shiyu; Liu, Dawei; Cao, Shiwei; Li, Xiaoping; Zhou, Yucheng

doi:10.1016/j.enbuild.2019.109424

Cited by 16 publications

(1 citation statement)

References 89 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If any of the required input factors are missing, the predicted COP will be inaccurate or unpredictable. Predictive models using machine learning utilize various theoretical methods to find and predict the relationship between input and output data, and compared to mathematical methods, they can predict with relatively few input factors and require big data [57]. In building energy systems, various data are collected using a building automation system (BAS) for system operation, and monitoring and can be utilized to develop predictive models through machine learning.…”

Section: Prediction Of Cop For Water Flow Controlmentioning

confidence: 99%

Variable Water Flow Control of Hybrid Geothermal Heat Pump System

Shin,

Kim,

Lee

et al. 2023

Energies

View full text Add to dashboard Cite

Ground heat accumulation caused by imbalanced heating and cooling loads in a building can cause the heat-source temperature to increase as the operating age of a geothermal heat pump (GHP) system increases. An alternative system to improve upon this situation is the hybrid GHP system. This study reviews existing research on GHP systems and hybrid GHP systems, variable water flow (VWF) control, and coefficient of performance (COP) prediction. Generally, constant flow control is applied to the circulating pump to provide a flow rate according to the maximum load. The need for VWF control was identified because the hybrid GHP system is used mainly as a heating and cooling heat source system for partial loads rather than the entire building load. Previous studies on predicting the COPs of GHP systems developed prediction models by selecting input values based on mathematical models, collecting data through multiple measurement points, and utilizing data from production environments. The model can be limited by the field environment, and it is necessary to predict the COP using machine learning based on existing field monitoring data.

show abstract

Section: Prediction Of Cop For Water Flow Controlmentioning

confidence: 99%