Research On Ventilation Epidemic Prevention And Energy Saving Regulation Based On Multi-resolution Detection Of Occupancy

With the spread of the COVID-19 epidemic,a large number of infections have been reported at domestic and international.The viruses are often transmitted indoors,in densely populated and poorly ventilated building environments.Ventilation plays an important role in reducing the risk of infection as a central means of controlling the spread of indoor air pollution.In public buildings,the dynamic and non-uniform spatial distribution of occupants poses a significant challenge to building ventilation during an outbreak.Current ventilation systems in public buildings are still largely empirical and predetermined,and are based on low resolution of occupancy,which does not respond efficiently and accurately to the dynamic distribution and changes in occupants.Indoor occupant information includes a variety of resolution levels such as number,location,density,and metabolic rate.Density distribution and metabolic rate are important factors from the perspective of ventilation control and epidemic prevention.Therefore,this thesis is based on the detection of multi-resolution occupancy in the following three aspects.(1)In response to the problem that the traditional ventilation control strategy cannot respond to the dynamic change of occupants efficiently and accurately,a ventilation control strategy based on multi-resolution occupancy is proposed.At the same time,computer vision technology is introduced into the field of ventilation control,and the widely distributed surveillance cameras in public places are used to achieve multiresolution occupancy detection.(2)To address the current problem of using high ventilation rate to reduce the risk of infection in high-density places,a ventilation control strategy based on zonal occupant density is proposed.The strategy takes the actual fresh air demand of zoned occupants and low risk of infection as the factors for determining the fresh air volume in the room.The YOLO detection algorithm is used to obtain information on zoned occupant density and set the zoned occupant density threshold,while the Wells-Riley infection risk prediction model is combined to determine the fresh air demand of indoor occupants,enabling the ventilation control system to automatically adjust between on demand controlled ventilation mode and epidemic prevention mode,so as to meet the demand for indoor occupant ventilation and infection risk reduction with minimal energy consumption.(3)Aiming at the problem that the traditional ventilation control strategy is difficult to match reasonable ventilation volume according to the actual demand of occupants in places with high metabolic rate,a ventilation control strategy based on occupant metabolic rate is proposed.This strategy uses the metabolic rate,which is high resolution of occupancy,as a factor to determine the demand for fresh air in the room,so as to match the reasonable ventilation quantity for the occupant in the room.Using pixel statistics theory and neural network algorithms,the metabolic rate characteristics are extracted and the metabolic rate prediction model is developed.Finally,the validity of the occupant metabolic rate prediction model was verified experimentally,and the ventilation strategy was evaluated in terms of both energy consumption and infection risk.The experiments show that the occupant metabolism-based ventilation control strategy can achieve better results in terms of infection risk control compared to the traditional ventilation control strategy,while saving 13% of energy.