数据驱动的冷却水塔建模与节能优化操作

Data-driven modeling and optimal operation simulation of a cooling tower

针对冷却水塔的节能操作给出了一种数据驱动的建模与优化方法。首先,基于冷却水塔实际运行数据,应用非负绞杀变量选择方法给出一个自适应模型用于描述冷却水塔过程,该模型对于冷却水塔出口水温具有良好的预测精度。根据变量选择结果,分析了外界空气温度与湿度对冷却能力的影响。然后,提出了基于模型的冷却水塔风机的优化操作策略,并进行实验将之应用于冷却水塔的操作。研究结果显示,基于模型的优化操作具有较大的节能空间。

This paper proposed a data-driven modeling and optimalization approach to investigate the operation of a cooling tower for energy conservation. Firstly, based on data from the actual operation of a cooling tower, the cooling tower process was characterized by an adaptive model with nonnegative garrote （NNG） variable selection procedure. On the basis of the statistical results of variable selection, the effects of ambient air temperature and humidity on the cooling capacity of the tower were investigated. Subsequently, the model-based optimal strategy of fan operation was proposed, and its implementation was virtually studied. The research results showed that there was considerable room for energy conservation.