The Quantitative Study Of District Energy Use Characteristics For Distributed Household Air Conditioners’ Cooling In Hot Summer And Cold Winter Zone

According to Chinese Statistical Yearbook data,the average number of air conditioners’ ownership of one urban household is 1.5,and the total number of air conditioners’ ownership is about 520 million.Therefore,air conditioner is the most important household electrical equipment in China.Especially in hot summer and cold winter zone,urban residents have close to or more than 2 air conditioners per household,and household air conditioners are the most important cooling equipment for residents in this area.Based on this,within a certain area,the energy demand of household air conditioners with a high amount of "accumulates",as well as its impact on the load of the district energy network in summer in hot summer and cold winter zone,should not be underestimated.At the same time,in the context of "carbon neutrality" and "carbon peaking",the research topic of disctrict energy use characteristics of distributed air conditioners’ cooling has important academic significance and application value.In the previous research,due to the problems of data dimension and acquisition method,there is a certain bottleneck in the research on the district energy use characteristics of distributed air conditioners.In recent years,with the development of smart air conditioners,it is possible to rely on the Internet of Things technology to obtain large-scale distributed air conditioner operation data at a district scale.Relying on this data,this study used the data-driven approach to investigate the topic of district energy use characteristics of distributed household air-conditioners’ cooling in hot summer and cold winter zone from three aspects: "input of district residential air-conditioning cooling demand modeling","diverse characteristics of district energy use for household air conditioners’ cooling",and "prediction method of district enery demand for distributed air condtioners’ cooling".Focusing on the bottom-up input of the bottom-up district air-conditioning cooling demand modeling(air-conditioning usage behavior),this research applied cluster analysis to obtain the behavior of air-conditioning users in setting temperature,setting wind speed,operation schedule,and operation duration respectively.And then,based on these diverse characteristics of user behavior,the prediction models for the air-conditioning on/off pattern in three typical cities in hot summer and cold winter zone,Shanghai,Wuhan,and Chongqing,were established repectively.And providing prediction tools to lower the threshold for relevant researchers to use the research results.Then,using the association rule method,the setting temperature adjustment behavior of distributed air conditioners in the process of "once operation" was quantified,and the diverse setting temperature adjustment behavior patterns were obtained,among which,for the wall-mounted air conditioners,a total of 25 temperature regulation behavior patterns with strong correlation were obtained,and for the floor-standing air conditioners,a total of 23 temperature regulation behavior patterns with strong correlation were obtained.Focusing on the diverse characteristics of district energy use for household air conditioners’ cooling",this study was conducted from micro to macro,focusing on the“single air conditioner” level and the “district” level in turn.A series of data-driven methods such as cluster analysis,statistical analysis,decision tree,were used to identify the electricity load patterns of distributed air conditioners at different scales.From the perspective of the single air conditioner,five typical daily electricity load patterns were obtained for wall-mounted air conditioners and floor-standing air conditioners,respectively;and based on the longitudinal variability of user load patterns,five and eight typical energy consumption users obtained for wall-mounted and floor-standing air conditioners,respectively.Then,from the district perspective,six typical district airconditioning daily electricity load patterns are obtained,and the regulation potential of district system peak load under different typical district load patterns was analyzed.The influencing factors of "typical district daily electricity load pattern for air conditioning"are discussed,and it was found that "the average outdoor temperature of the day" and the"difference between the maximum outdoor temperature of the day and the maximum outdoor temperature of the previous day" are the most important factors that determine the typical district load pattern of a natural day.And finally,according to the key influencing factors,the decision-making path of "typical district daily electricity load pattern" was established by using the decision tree model.Focusing on the problem of how to use the data-driven approach to predict the "district energy demand for distributed household air conditioners’ cooling",two district air-conditioning operation indicators that influence the energy consumption of district airconditioning were firstly proposed.Two indicators are daily air conditioner open rate and daily average power consumption per air conditioner in the district,respectively.The logistic regression model was used to quantify the relationship between the daily air conditioner open rate and the outdoor temperature,and the 4P model was used to quantify the relationship between the daily average power consumption per air conditioner and the outdoor temperature.Then,based on the research results of air-conditioning load patterns at different scales,from two perspectives: “based on the typical daily load pattern at the single air conditioner level” and “based on the typical daily load pattern at the district level”,two prediction methods aiming at " district energy demand for distributed household air conditioners’ cooling " were proposed,and verified with actual measured data.The verification results show that two proposed prediction methods have good prediction performance when the outdoor temperature in summer is higher than a certain value.In this study,the constructed air conditioner usage behavior model,the obtained typical district energy use characteristics of distributed household air conditioners,and the proposed district energy demand for distributed household air conditioners’ cooling prediction methods can provide important data and method support for reasonable planning and operation management of the district energy demand of distributed household air conditioning cooling in hot summer and cold winter zone.