Big Data Analysis Of Demand Side User Power Based On Cloud Platform

The contradiction between power supply and demand is now becoming worse with the development of society.The Demand Side Management(DSM)will be an effective means of power management in the aspect of solving the contradiction between power supply and demand in the future.It can reduce the consumption and demand of electricity as much as possible and improve the efficiency of power utilization at the situation of meeting the same function of using electricity,by taking effective motivation and countermeasures for guidance and cooperating with appropriate operation modes.The development of large data will bring great changes to DSM in technology,development concept and management system.So,this paper combined large data analysis with power demand side management,deeply analyzed and studied the large electro-data of demand side users,and then solved the problem of large volume of electro-data and low data information density and unclear influence of load factors in the application of the large data of demand side,finally put forward the dimensionality reduction method and load forecasting method suitable for large data of demand side.The research work is summarized as follow:Firstly,aimed at the problem of the large volume of electro-data and low data information density,PAA and wavelet transform are used to reduce the dimension of electro-data,and an example is given to verify the effect of dimension reduction.Combined the advantages and disadvantages of two methods,a dimensionality reduction method based on PAA-wavelet transform is proposed.The results of an example showed that this method is not only fast but also can keep the characteristics of original data as much as possible.So,it is more suitable for the dimensionality reduction of the large data of cloud platform.Secondly,the large data of cloud platform are classified and sorted,and the influence of main influence factors of different power loads is analyzed quantitatively by using the principal factor analysis method.And then the influence factors are sorted according to the magnitude of the influence.Finally,the research results are built on the cloud platform database and a detailed principal factor analysis decision-flow diagram is designed for cloud platform,which lays a foundation for DSM service and decision-making.Finally,load clustering and load forecasting based on large data of demand side are studied,and then load forecasting method based on large data is proposed.