Data Compression and Sampling Period of Daily Energy Consumption

Data compression will be extremely necessary in order to solve the problem of large data volume generated by smart meters. The aim of this work, is to investigate the source coding theory, which establishes the entropy as a fundamental limit on the performance of conventional data compression algorithms applied to daily load curve of a typical resident consumer. We have proposed the applying of Pulse Code Modulation+Huffman and Differential Pulse Code Modulation+Huffman as methods to represent the consumption as few bits are possible, and we have explored the performance of compression when the typical sampling period of 15 minutes is reduced. Furthermore we have determined the second and third order entropies as a limit to data source compression schemes.