Research And Application Of Algorithm In Green Building Energy Saving Design Platform

With the increasingly serious resource crisis and global warming,the building energy consumption as an important incentive has also attracted more and more attention.Green building has become an inevitable trend of architectural design.Green building design needs to consider the influence of the local climate characteristics,so in the green building energy saving design platform,it is very necessary to build the basic database of building energy conservation and to query the climate characteristics of each station and building design parameters.Over the past few decades,the meteorological department has accumulated a large amount of data,provides the basis for climate analysis and architecture design,however,due to software mistakes or acquisition record collector,there are abnormal or missing values in meteorological data.Obviously,it is impossible to extract valuable information from large amounts of data by artificial ways,but thanks to the help of the computer,we can use the data mining technology to deal with large-scale data and get useful information.According to the above questions,this paper studies the related data mining algorithms and compression perception theory and its application to meteorological data processing,and then green building energy saving design platform was implemented.In this paper,the source and characteristics of meteorological data are explained,and the data is preprocessed,including the elimination of non-use stations and the deletion of redundant information,which is to be prepared for the implementation of the algorithm.And then the basic concept of clustering algorithm are described,and the advantages and disadvantages of common clustering algorithms are compared and analyzed.According to the characteristics of the meteorological data and the purpose of the algorithm in this paper,this paper mainly studied the algorithm of clustering by fast search and find of density peaks,and its application to the meteorological data,including all of the attributes and part of the attributes in the meteorological data as the research object,clustering analysis was carried out on the meteorological stations,and then the clustering results were visualized,and analyzed the clustering results,found the climate of the country's global distribution patterns,which makes up for the shortcomings of traditional climate partition according to the geographical location.In the data preprocessing stage,in order to solve the problems of filling the missing data in meteorological data,based on the study of compressed sensing principle and signal reconstruction algorithm,this paper proposed a meteorological data repair method based on compressed sensing theory,and verified the feasibility of the method in the cases of random and continuous lack of data,compared with the damaged data,the repaired data can more accurately reflect the raw data's information,And compared with traditional methods,we found that in a random lack of measured data,both the cubic spline interpolation method and the method in this paper can realize the recovery of the original data,and the method in this paper is more suitable for large data loss situation,and in the continuous lack of measured data,the method can restore the original data.Finally,this paper designed and implemented the green building energy-saving design platform,and introduced each function module of the platform,with the application of algorithm research in the process of platform design and data processing,and realized the goal of providing a good guidance for green building design.