An energy node-based modelling method for indoor energy flow state of industrial buildings

The energy consumption of industrial buildings continues to increase. Improving the energy efficiency of buildings and effectively reducing energy consumption are urgent problems to be solved. In order to effectively evaluate the energy consumption of industrial buildings, this paper proposes a method for constructing the indoor energy flow state model of industrial buildings. The state of energy flow in the building is described by the state of temperature, humidity and wind speed. The control space adopts PCA, and the global space adopts KPCA to extract energy flow state characteristic parameters, then K-means cluster it. The clustering areas are defined as energy nodes, then extracted fusion points and stable points from energy nodes to describe energy flow state. Comparative experiments were carried out with the original temperature and humidity monitoring model of industrial buildings. Based on CFD, the model ensure the accuracy, the effective coverage is increased by 20%, and the amount of information feedback is reduced by 55%.