Refined classification, prediction and analysis of electricity consumption law 

taking into account the big data of user carbon emissions

Research services on refined classification, prediction and analysis of electricity consumption law include following items:

(1)On the basis of the detailed classification of users, the correlation between the electricity consumption of various users and natural factors such as time, season, holidays, meteorology and socio-economic factors such as per capita wage, gross domestic product and built-up area is studied. A prediction model for the classification of electricity consumption from long-term load to short-term load is constructed, including natural factors and socio-economic factors.

(2)The dynamic estimation and early warning technology of carbon emission intensity and cumulative carbon emission intensity are used to study the dynamic change law of carbon emission factors of users in different industries, and on this basis, the dynamic prediction method of carbon emission intensity of users considering the time series characteristics and carbon emission characteristics of electricity consumption is studied. Using spatial measurement tools, the correlation analysis model between load electricity and carbon was constructed, and the cumulative estimation method of carbon emissions in different spatial and temporal domains was studied. This paper studies the risk and early warning methods and theories of carbon emission quota in key industries under the trend of carbon emission policy, and designs the prediction of carbon emission intensity and accumulation and the early warning model of overrun.

(3)Based on the correlation analysis technology of carbon emission situation based on big data analysis, the statistical characteristics and spatiotemporal distribution characteristics of electricity consumption data and carbon emission data are studied according to the production and operation data of relevant industries. The intelligent analysis technology of big data is used to study the correlation and causal relationship of carbon emissions in time and space. Combined with the carbon flow calculation method and carbon emission prediction results, a wide-area carbon emission heat map was constructed, and the prediction and systematic display method of carbon emission trend on multiple time scales were studied.