Abstract
The present work aims to improve the human living environment by using Digital Twins (DTs) technology. The DTs technology is applied to sewage treatment infrastructure to ameliorate the problems of sewage discharge management in current environmental protection management. First, the energy consumption of the water system of central air-conditioning is studied, and the DTs model of central air-conditioning is established. Then, a model parameter identification framework is constructed based on the genetic algorithm and multistrategy initial solution space optimization (MISSO). Moreover, a model prediction interval estimation method based on K-means clustering is proposed, and an artificial network model is used to compensate for the prediction results. The experimental results demonstrate that accurate models of chilled water pump and cooling water pump can be obtained based on the MISSO and genetic algorithm. Besides, the absolute relative errors of predicted and measured power consumption per hour of most water pump models are kept within 5%. The mean value of the prediction interval of chilled water outlet temperature and cooling water outlet temperature of the chiller model is 0.45% and 0.29%, respectively. In addition, the mean value of the absolute value of the Adaptive Communication Environment (ACE) of the prediction interval of chilled water pump's hourly power consumption is 1.21%; the average absolute value of ACE of the predicted range of cooling pump power consumption per hour is 1.74%. After error compensation, the error between the predicted value and measured value of chilled water outlet temperature by the Department of Energy (DOE-2) model decreases significantly. Therefore, the uncertainty estimation method of the central air-conditioning water system proposed here has good performance. This method can reduce air-conditioning energy consumption, effectively save energy, and reduce emissions, which is conducive to environmental governance, thus improving human health.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have