Research On Parameter Dimension Reduction And Partition Method Of Generating Coincident Design Days

Outdoor calculation parameters for air conditioning design mainly include hourly dry temperature and wet bulb temperature and solar irradiance of the day,which are important basic parameters for air conditioning s ystem design.There are some limitations in the current outdoor calculation parameters for air conditioning design in China: the simultaneous occurrence of meteorological parameters is ignored in the process of parameter selection;the heat transfer process is simpli fied in the process of parameter selection;the variation of dry and wet bulb temperature and the influence of solar radiation are ignored in the process of design day construction.The above problems lead to the design load of air conditioning is too large,the coincident design day can solve the above problems to a certain extent.However,in order to meet the design accuracy requirements of air-conditioning system at this stage,it is difficult to generate general coincident design day for air-conditioning design for different rooms or buildings.To meet the needs of different rooms or buildings,it must has a large amount of calculation,does not have flexible use.In order to reduce the amount of calculation of coincident design day,a near extreme method of parameter dimension reduction and partition is presented to generate coincident design days.The near extreme combination can approximately represent the most extreme case of the influence degree of characteristic parameters on the design day selection.Partitioning under the near extreme combination can ensure that the results are applicable to most cases.For the characteristic parameters of room,The method is based on the radiation time series to determine the near extreme combination of each room’s characteristic parameters,the each characteristic parameters of room was reduced and partitioned under the near extreme combination.Based on the Monte Carlo method,the design cooling load determined by the long term calculation was used to verify the parameter dimension reduction and partition method of the room characteristic parameters.The results show that the accuracy of room design cooling load calculation is improved compared with the design day determined by the traditional method.The results of dimensionality reduction and partition using Hong Kong long term data have good applicability in Hong Kong and Changsha,and also have good applicability in Beijing after dimensionality reduction and partition using Beijing data as data source.Considering the building characteristic parameters,based on the dimension reduction and partition of the room characteristic parameters,the near extreme method was used to reduce and partition the building characteristic parameters and verified.The results show that the accuracy of building design cooling load calculation is improved compared with the design day determined by the tradition al method.The results of dimensionality reduction and partition based on long term data of Hong Kong have good applicability in Hong Kong,Changsha and Beijing.A case study was conducted to compare the design day obtained by the traditional method and the coincident design day.In the case,the room and building design cooling load calculated on the selected coincident design day were reduced by21.63% and 10.92%.The results show that the design day parameters obtained by traditional methods are overestimated,which leads to large cooling load of air conditioning design.