Model fusion-based indoor air prediction method

Abstract

The invention discloses a model fusion-based indoor air prediction method. According to the method, an indoor air quality prediction model and an outdoor air quality prediction model are effectively fused through a classification and regression tree algorithm according to the behaviors and activities of an occupant; the root and each leaf node of a regression tree separately train a linear regression device, so that a plurality of linear regression devices can be obtained; and therefore, the method is suitable for indoor air quality prediction under different conditions. According to the method of the present invention, efficient multi-source data fusion and model fusion are performed on a plurality of data sources such as the activities of a person, indoor air quality, outdoor air qualityand outdoor weather, so that more accurate learning and prediction can be realized.

Description

technical field [0001] The invention relates to an indoor air prediction method, in particular to an indoor air prediction method based on model fusion, which belongs to the field of environmental engineering. Background technique [0002] At present, people spend most of their time indoors, and the indoor air environment has the most direct and close impact on people's health. Analyzing and predicting the state and trend of Indoor Air Quality (IAQ) has very important research significance and application value for human health, environmental protection, architectural design and decoration. [0003] Indoor air components mainly come from: (1) Outdoor air, including the components in the outdoor atmosphere and various emissions generated by human beings in industrial production activities, through the doors and windows, door gaps, window gaps, ventilation, and filtration of buildings. system etc. into the room. (2) Indoor pollution is mainly caused by cooking, smoking, clea...

AI Technical Summary

Problems solved by technology

The above research methods have the following problems: the complexity of physical modeling is high, there are many parameters involved, the solution accuracy is related to many factors, professional and technical personnel are needed, and there are large errors in the prediction results

To predict indoor air quality, the influence of outdoor air quality and human-made events of the occupants must be considered. If there is no air quality monitoring station near the indoor building, it is necessary to comprehensively consider the monitoring station data covered in a specific space for prediction. At the same time, meteorology is also a very important factor affecting outdoor air quality. At present, research in this area is still relatively lacking.

Images