Non-invasive human body thermal comfort AI sensing method

Abstract

The invention discloses a non-invasive human body thermal comfort AI sensing method. The method includes the following steps of: acquiring and preprocessing data; defining and extracting skin sensitivity indexes; constructing a deep learning network for non-invasive skin sensing thermal comfort; and performing training to generate a network model. According to the method, the image data of human body skin are acquired through computer vision, and after the data are processed, an address mapping table between region-of-interest images and skin temperatures is constructed; and skin sensitivity differences between different human bodies for external heat and cold stimuli are adopted as weight coefficients, feature extraction is performed on the address mapping table and the SSIs (Skin Sensitivity Index), and feature fusion is performed, training is further performed, obtained network models are saved, and optimization is performed, so that a final network model is obtained for skin temperature prediction. With the AI sensing method of the invention adopted, three major challenges, namely, the micro skin changes, inter-individual differences and intra-individual time variability in human body thermal comfort detection can be overcome; and energy optimization, energy conservation and environmental protection can be realized. The method has high operability.

Description

technical field [0001] The invention belongs to the fields of computer vision, machine learning and building physics, and in particular relates to a non-invasive human body thermal comfort detection method for intelligent buildings or intelligent automatic driving. Background technique [0002] According to the annual statistical report, 21% of the world's annual energy consumption comes from commercial and residential buildings. In some countries and regions with relatively rapid urbanization, energy consumption is increasing at a rate of 32% per year. In building energy consumption, 50% comes from heating, ventilation and air conditioning system (HVAC, hereinafter collectively referred to as central air conditioning system). If it is possible to detect the thermal comfort level of the human body in real time, adjust indoor parameters (temperature, humidity, airflow, etc.) in a targeted manner, or perform local heating / cooling, it can meet the needs of individual thermal co...

AI Technical Summary

Problems solved by technology

This method can reflect the user's thermal comfort feeling, but has obvious disadvantages

On the one hand, the temperature control has a strong hysteresis; on the other hand, it needs continuous intervention from the user, especially when resting at night, it cannot be controlled

For the heating system, currently users cannot know the specific temperature, and can only switch gears based on experience

Images