Determinants of energy use in UK higher education buildings using statistical and artificial neural network methods

Multiscale convolutional recurrent neural network for residential building electricity consumption prediction

The prediction of residential building electricity consumption can help provide an early warning regarding abnormal energy use and optimize energy supply. In this study, a multiscale convolutional recurrent neural network (MCRNN) is proposed to predict residential building electricity consumption. The MCRNN model uses multiscale convolutional units to collect different information on environmental factors, such as temperature, air pressure, light, and uses a bidirectional recurrent neural network (Bi-RNN) to extract the long-term dependence information of these factors. In addition, a recurrent convolutional connection is used to filter the most useful multiscale and long-term information in the MCRNN model. The accuracy of MCRNN is evaluated through an experiment using real data. The results show that MCRNN performs better than the other models. For instance, compared with the support vector regression (SVR) and random forest (RF) models, the MCRNN model has a 47.83% and 38.72% lower root mean square error (RMSE), respectively. The MCRNN model also shows a 37.81% and 70.38% higher accuracy, respectively, compared to the SVR and RF models.

Development of an Electrical Energy Consumption Model for Malaysian Households, Based on Techno-Socioeconomic Determinant Factors

Energy-saving strategies are required to address the increasing global CO2 and electrical energy consumption problems. Therefore, the determinant factors of electrical energy consumption consist of socio-demographic changes, occupant behavior, house and appliance characteristics, or so-called techno-socioeconomic factors, which all need to be assessed. Statistics models, such as the artificial neural network (ANN), can investigate the relationship among those factors. However, the previous ANN model only used limited factors and was conducted in the developed countries of subtropical regions with different determinant factors than those in the developing countries of tropical regions. Furthermore, the previous studies did not investigate the various impacts of techno-socioeconomic factors concerning the performance of the ANN model in estimating monthly electrical energy consumption. The current study develops a model with a more-in depth architecture by examining the effect of additional factors such as socio-demographics, house characteristics, occupant behavior, and appliance characteristics that have not been investigated concerning the model performance. Thus, a questionnaire survey was conducted from November 2017 to January 2018 with 214 university students. The best combination factors in explaining the monthly electrical energy consumption were developed from occupant behavior, with 81% of the variance and a mean absolute percentage error (MAPE) of 20.6%, which can be classified as a reasonably accurate model. The current study’s findings could be used as additional information for occupants or for companies who want to install photovoltaic or wind energy systems.

The carbon footprint of a UK University during the COVID-19 lockdown

The COVID-19 pandemic has provided a unique opportunity to compare the carbon intensity of higher education delivered on- and off-campus. This is attributed to governmental lockdown orders that have forced Universities to close their campuses, ban business travel and move all teaching and learning activities online. This study represents the first known attempt to compare the carbon footprint of a mid-sized UK University produced during the COVID-19 lockdown (April-June 2020) against that generated within the respective time period in previous years. Although the overall carbon footprint of the University decreased by almost 30% during the lockdown, the carbon intensity of online teaching and learning was found to be substantial and almost equal to that of staff and student commute in the pre-lockdown period. The study contributed to an emerging academic discourse on the carbon (dis)benefits of different models of higher education provision in the UK and beyond. The study suggested that policy and management decisions on transferring education online should carefully consider the carbon implications of this transfer.

A New Generation of Thermal Energy Benchmarks for University Buildings

In 2008, the Chartered Institution of Building Services Engineers (CIBSE TM46 UC) presented an annual-fixed thermal energy benchmark of 240 kWh/m2/yr for university campus (UC) buildings as an attempt to reduce energy consumption in public buildings. However, the CIBSE TM46 UC benchmark fails to consider the difference between energy demand in warm and cold months, as the thermal performance of buildings largely depends on the ambient temperature. This paper presents a new generation of monthly thermal energy benchmarks (MTEBs) using two computational methods including mixed-use model and converter model, which consider the variations of thermal demand throughout a year. MTEBs were generated using five basic variables, including mixed activities in the typical college buildings, university campus revised benchmark (UCrb), typical operation of heating systems, activities impact, and heating degree days. The results showed that MTEBs vary from 24 kWh/m2/yr in January to one and nearly zero kWh/m2/yr in June and July, respectively. Based on the detailed assessments, a typical college building was defined in terms of the percentage of its component activities. Compared with the 100% estimation error of the TM46 UC benchmark, the maximum 21% error of the developed methodologies is a significant achievement. The R-squared value of 99% confirms the reliability of the new generation of benchmarks.

Small power and lighting load time series data for 27 departments across 8 UK hospitals

The electricity consumption of 27 departments was measured across 8 medium to large General Acute hospitals in England (largely by the authors, some data was donated and authorised for publication by the respective hospitals). The departments fall into 6 different categories which have been selected due to their prevalence in General Acute Hospitals (wards), their high energy intensities (theatres, laboratories, imaging and radiotherapy) or their distinct operating hours (day clinics). This data article provides floor areas and the time series of departmental power loads, mostly encompassing lighting and small power (but excluding central electricity use for ventilation, pumping and medical gas services). Comparative interpretations of the data are published in doi: 10.1016/j.enbuild.2016.02.052 [1].