Font Size: a A A

Study On Usage Characteristics Of Room Air Conditioner In Shanghai Based On Big Data Monitoring Platform

Posted on:2019-05-01Degree:MasterType:Thesis
Country:ChinaCandidate:S CaiFull Text:PDF
GTID:2382330566476954Subject:Heating, Gas Supply, Ventilation and Air Conditioning Engineering
Abstract/Summary:PDF Full Text Request
Room air conditioner is an important means to improve the residential thermal environment in the hot summer and cold winter region,as well as the major source of energy consumption for residential buildings in this region.In recent years,with the rapid development of computer technology and internet technology in China,the arrival of the big data era will bring new opportunities and challenges to all walks of life.In this paper,the room air conditioner(RAC)in Shanghai is taken as the object and the big data method is used to analyze the energy consumption mode and characteristics of room air conditioner.First of all,this chapter analyzes the running time characteristics?setting temperature and indoor temperature of room air conditioners in Shanghai by mathematical statistics.It is found that users mainly use air conditioners in summer months of July and August and winter months of January and February.The usage rate of air conditioners is the lowest in transition season.The change rules of the usage rate of new and old city are similar throughout the daytime.The air-conditioner usage rate of the new city is 3% higher than that of the old city.The air-conditioner usage rate in old city area is between 7% and 12%;The cumulative operating time of 34% of the samples during the whole year had 552 more hours than average.There was a large difference in the running time among samples throughout the year.The setting temperature variation coefficient of more than 80% of the samples was less than 0.1.The user's setting temperature of the air conditioner was fixed and generally no adjustment is made.The indoor temperature difference between the on-off switch of air conditioner in winter is obviously greater than that in summer.The indoor temperature in summer is mainly distributed from 25.8 to 27.6°C when the air conditioner is turned on.When the air conditioner is in the off state,the indoor temperature is mainly distributed from 26.5 to28.6°C.In winter,the indoor temperature is mainly distributed from 19 to 23.4°C when the air conditioner is turned on.When air conditioner is not in use,the indoor temperature is significantly reduced by about 6°C,mainly distributed from 13.6 to17.3 °C.Secondly,the overall level of energy consumption 930 air conditioners was introduced and the Mann-Whitney test method was used to test the difference among the energy consumption values of 8 types of air conditioners.In order to analyze the influencing factors of the energy consumption value of the room air conditioner,the relationship among the energy consumption value of the air conditioner ?the outdoor temperature and the indoor temperature was studied using the control variable method.The Spearman correlation coefficients of the air conditioner energy consumption values of the selected air conditioner samples with the outdoor temperature and the indoor temperature were calculated,and the linear correlations among them were pointed out.Then,an unsupervised learning method based on cluster analysis was used to explore the classification of the usage modes of room air conditioners in Shanghai and four typical usage modes of air conditioning were obtained.At the same time,the practical application background was combined to analyze and summarize the characteristics of each typical usage model.It was found that the clustering classification results have a good interpretation in the actual context.Using the supervised learning method of artificial neural network based on the classification results of unsupervised learning method to build a predictive classification model for the research results of cluster analysis,it is finally found that the predictive model has relatively high prediction accuracy under the exploratory classification results of cluster analysis.Finally,Dest-H software is used to simulate the energy consumption of the type of clustering.It is found that the energy consumption simulation results are compared with the measured values.It is found that the annual energy consumption of the bedroom is closer to the measured value,the gap is less than 50%,while the gap between the annual energy consumption and the measured value of the living room is larger.Compared with the winter,the summer simulation results are closer to the measured ones..In order to further explore the cause of the difference between the simulated and measured values,six different Dest-H energy consumption simulation schemes are formed by changing the unit,air conditioning,setting temperature and air conditioning tolerance temperature to get the energy consumption of the main bedroom,the sub bedroom and the living room air conditioning.The energy consumption of the six schemes is within the actual energy consumption range,which shows that the statistical results are universal.Changing the Huxing will obviously affect the air conditioning energy consumption,and the direction of winter and summer will not necessarily be the same.The use time of air conditioning increases(decreases),and energy consumption increases accordingly.Changing the setting temperature of air conditioning has an obvious impact on the energy consumption of air conditioning in summer,and has little effect on air conditioning energy consumption in winter.Air conditioning tolerance temperature range reduction will lead to obvious increase in air conditioning energy consumption.To sum up,the use and rest of the air conditioning,the setting temperature of air conditioning in summer and the tolerance temperature of air conditioning have great influence on the energy consumption of air conditioning,and the temperature of air conditioning in winter has little effect on the energy consumption of the bedroom air conditioning.
Keywords/Search Tags:Room Air Conditioner, Cluster Analysis, Neural Network Prediction, Differential Test, Energy Consumption Simulation
PDF Full Text Request
Related items