City Research Online

Abstract

Building cooling load prediction is one of the key elements in the energy conservation achievements. Most of the mathematical models using in the industry nowadays include forward and inverse modeling approaches. However, these models consume much computer resources and require a longer computational time.

Multi-layer perceptron (MLP) model of artificial neural network (ANN) is adopted in this thesis. The model is widely used in engineering approaches that render good performance in adaptability, nonlinearity and mapping. It also has good ability in predicting the cooling energy consumption of buildings. It is reported that the occupants’ activities inside the buildings can have significant impact on the accuracy of the model. The existing input parameters used for the ANN models could not represent the complexity of the activities inside the buildings well. Most of the traditional ANN models adopted fixed profile or historic load data to represent building occupancy in simulating building cooling energy consumption. However, building occupancy is never still. The dynamic changes occurred in the occupancy of the buildings therefore make the forecasting of building cooling load difficult and less accurate. This thesis aims at (i) introducing a novel model to represent occupants’ presence and activities; and (ii) investigating the effect of using the novel model on improving the predictive accuracy of building cooling energy consumption.

The simulation results demonstrate that building occupancy data play a significant role in building cooling energy consumption prediction and the use of the novel approach significantly improves the predictive accuracy of building cooling energy consumption model.

Publication Type:	Thesis (Doctoral)
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Departments:	School of Science & Technology > Engineering Doctoral Theses School of Science & Technology > School of Science & Technology Doctoral Theses

Export

Downloads