Soft sensor approach for optimal operation of cooling tower for energy conservation

This paper presents a data-driven adaptive soft sensor approach to investigate the performance and optimal operation of cooling tower for energy conservation. To achieve this aim, first, the cooling tower process is characterized by an adaptive soft sensor with nonnegative garrote (NNG) variable selection procedure. Then, based on the statistics result of NNG variable selection, the effect of environment temperature and humidity on the cooling capacity was investigated by principal component analysis (PCA). At last, the optimal strategy of fan operation mode was presented and the implementation of the optimal strategy was investigated for a cooling tower, which showed that there was large room for energy conservation.