Modelling of Chiller Performance Using Artificial Neural Networks
This study investigates the applicability of artificial neural networks (ANNs) to predict the performance of chiller systems. The chiller performance is affected by numerous parameters, and the relationships between them are highly nonlinear. ANNs are appropriate for tackle these complex nonlinear issues due to their learning ability and versatile mapping capabilities. An ANN model for one water chiller is presented in this paper with the objective of requiring only those input parameters that are readily available. For this aim, field investigations on the year-round chiller system have been conducted. With the experimental data, an ANN model for the system was developed based on the standard back propagation algorithm. The model was used for predicting the chiller performance parameters of compressor power input and coefficient of performance (COP). With the satisfactory predicting results, this study shows that ANN can be a promising tool for predicting the chillers performance for fault detection and other diagnosis purposes.
chiller coefficient of performance artificial neural network
Jia Yang K.T. Chan Xiangsheng Wu Xiaofeng Yang
Department of Building Services Engineering The Hong Kong Polytechnic University Hong Kong, China Dept of Estate Management and Environmental Engineering Logistical Engineering University Chongqing, Dept. of Estate Management and Environmental Engineering Logistical Engineering University Chongqing
国际会议
成都
英文
1694-1697
2010-12-17（万方平台首次上网日期，不代表论文的发表时间）