Victor? Meyer? Schoenberg once pointed out in the book "The Age of Big Data: Great Changes in Life, Work and Thinking" that the information storm brought by big data is changing our life, work and thinking, and big data has opened a major era transformation. The concept of energy big data is related technologies and ideas that comprehensively collect, process, analyze and apply data in energy fields such as electricity and oil and gas, and data in other fields such as population, geography and meteorology. Energy big data is not only the deep application of big data technology in the energy field, but also the deep integration of energy production, consumption and related technological revolution with big data concept, which will accelerate the development of energy industry and business model innovation. 20 13 In March, the Information Committee of China Electrical Engineering Society released the White Paper on the Development of Big Data in China, which designated 20 13 as the "first year of big data in China", which set off a research upsurge of electric power big data. As a traditional power industry that is transforming to the energy Internet, the arrival of the era of big data and cloud computing will inject new vitality into the development of the traditional power industry that is about to undergo revolutionary changes.
Power big data mainly comes from power generation, transmission, substation, distribution, power consumption and dispatching of power production and power use, which can be roughly divided into three categories: first, power grid operation and equipment testing or monitoring data; The second is the marketing data of power enterprises, such as transaction price, electricity sales, electricity customers and other data; The third is the management data of power enterprises. Electric power big data has four characteristics: (1) large amount of data: PB level; The traditional dispatching automation system contains hundreds of thousands of acquisition points; Power distribution and data centers will reach 10 million levels; (2) There are many types of data: real-time data, historical data, text data, multimedia data, time series data and other structured, semi-structured data and unstructured data; (3) Low-value density: most of the collected data are normal data, and only a few are abnormal data, which is the most important basis for maintenance according to the situation; (4) Fast processing speed: analyze a large amount of data in a fraction of a second to support decision-making.