摘要
HVAC领域的节能设计和最优化控制一直是目前研究领域的热门话题 ,尤其由于直燃式溴化锂机组运行中的能耗问题 ,使得对它的模型研究更显得重要。由于这个系统高度的非线性 ,耦合性 ,多变量性 ,传统的辨识方法不能很好的解决这个问题 ,所以本文提出采用神经网络方法来辨识这样一种模型系统 ,并且结合实际的实验数据和厂家提供的性能数据 。
Energy saving and optimal control are always hot topics in HVAC studies.One drawback of direct-fired LiBr absorption chiller lies in its excessive fuel consumption, thus the research in modeling this kind of chiller is becoming an imperative issue.The traditional identification methods based on linear system are not appropriate to this kind of chiller system.This system is known to be complicated, non-linear, with multi-variables and complicated coupling between system components. In this paper,the neural network identification method is presented to solve this problem. Combined with the practical data from experiment sites, the forward system identification approach is proved to predict the system performance well.
出处
《制冷学报》
CAS
CSCD
北大核心
2001年第1期35-42,共8页
Journal of Refrigeration
关键词
神经网络
直燃式溴化锂系统
系统辨识
Neural network,Direct-fired absorption chiller,System identification