The development of a solar irradiance nowcasting method for buildings

Chen, Lei 2024. The development of a solar irradiance nowcasting method for buildings. PhD Thesis, Cardiff Univerisity. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (10MB) | Preview
	PDF (Cardiff University Electronic Publication Form) - Supplemental Material Restricted to Repository staff only Download (357kB)

Abstract

To control the environmental problems caused by the ever-growing global greenhouse gas emissions, reducing carbon emissions from buildings is necessary. Although solar energy has been widely used in buildings to reduce carbon emissions, its instability poses significant challenges to its utilisation and thus a large number of solar irradiance forecasting methods have been developed to anticipate its fluctuations. However, these methods are usually not customised for buildings and are not developed based on the needs of future building intelligence. Consequently, the aim of this research is to develop a solar irradiance nowcasting method with high spatial-temporal resolution based on low-cost equipment and user-friendly programming to achieve reliable nowcasting of Global Horizontal Irradiance (GHI), Direct Normal Irradiance (DNI), and Diffuse Horizontal Irradiance (DHI) for buildings and their future development. In general, this research first applied a clear systematic review methodology to identify appropriate forecasting methods for buildings. Secondly, through using low-cost equipment and user-friendly programming, this research develops a solar irradiance nowcasting method with very short-term forecasting horizons and high spatial-temporal resolution based on a promising Residual Neural Network (ResNet-152) model to achieve 10-sec, 1-min, 5-min, and 10-min nowcasting of GHI, DNI and GHI. Thirdly, a series of comparative tests are conducted to explore the effect of different factors on the nowcasting performance and the results of these comparative tests are verified by selected evaluation metrics. Finally, this research discusses the current and potential applications of the solar irradiance nowcasting method on buildings to demonstrate the practicality this research. Based on the results verification and analysis of comparative tests, the developed nowcasting method demonstrates evident reliability under the influence of different factors, including various time intervals, forecasting horizons, sky conditions, forecasting models and datasets. Ph.D. Thesis Lei Chen Cardiff University ii Welsh School of Architecture In conclusion, the significance of this research is to innovatively explore a solar irradiance nowcasting method from an architectural perspective and achieve reliable nowcasting of GHI, DNI and DHI for optimising the operational efficiency and safety, occupant comfort and design of buildings. Also, this research establishes an interdisciplinary methodology integrating the knowledge of meteorology, imagery, computer science and architecture, which can increase interdisciplinary communication and cooperation, as well as provide more research directions. Ultimately, the specific approaches, tools and outcomes of the research, as well as the discussion of the possible applications of solar irradiance nowcasting on buildings, provide initial inspiration for future research and the future development of buildings.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Architecture
Date of First Compliant Deposit:	8 January 2025
Last Modified:	08 Jan 2025 14:56
URI:	https://orca.cardiff.ac.uk/id/eprint/175112

Actions (repository staff only)

Edit Item