Comfort level prediction method based on big data, intelligent terminal and storing device

Abstract

The invention provides a comfort level prediction method based on big data, an intelligent terminal and a storing device. The method comprises the following steps that S101, data related to comfort level is acquired, and the data is clustered to form at least one type cluster; S102, a first comfort level index is acquired according to the feature data of the at least one type cluster, and the at least one type cluster is subjected to dimension reduction treatment; S103, a demand mode corresponding to the data in the least one type cluster is determined according to a user use behavior in the data, and a machine learning algorithm is adopted to carry out model training on each type cluster and each demand mode to form a comfort level model; and S104, one comfort level model matched with theattribute variable of an actual control object is selected according to the attribute variable of the actual control object, and a comfort level index of a user is predicted according to the comfortlevel model. According to the method, the personalized comfort level models can be obtained through distinguishing the use scenes and adding the use behaviors of the user, so that the actual control effect can better conform to the differentiated preference of the user, and the guarantee of the accuracy of the models is provided for improving the temperature control effect.

Description

technical field [0001] The present invention relates to the field of air conditioning management, in particular to a method for predicting comfort level based on big data, an intelligent terminal, and a storage device. Background technique [0002] With the continuous improvement of living standards, people's requirements for indoor thermal environment are also getting higher and higher, and the corresponding energy consumption due to indoor temperature regulation is getting higher and higher. According to statistics, building energy consumption accounts for about 40% of the world's total energy consumption, half of which is used for comfort air conditioning systems. [0003] In order to meet people's temperature control needs and reduce energy consumption, it is necessary to predict people's subjective thermal sensation, and reduce energy consumption according to the subjective thermal sensation. Among them, the predictive mean vote PMV (predictive mean vote) proposed by t...

AI Technical Summary

Problems solved by technology

[0004] However, many scholars in Asia believe that because Fanger takes young people from European and American countries as the research object, Asians and Europeans have different metabolisms, different regional climates, differences in age and gender, and different adaptability to thermal environments and psychological expectations. The theory proposed by Professor Fanger inevitably has certain limitations and deviations. Therefore, it cannot meet the actual temperature control needs of Asians. Moreover, the theory is formed based on the judgment of most people, and cannot meet the user's differentiation in practical applications. preference

Images