Analysis On Differentiation Residential Electricity Consumption Characteristic Based On Power Load Decomposition

With the technology development and smart power grid construction, advanced measurement system based on smart meter can extend to normal users, and residential electricity consumption characteristics are involved in massive account of real time power data collected by smart meter. In view of the need of accurately grasping the characteristics of residential electricity consumption in the process of the design and construction of smart grid, by mining this data and researching user type, personalized and differential demand of power consumer can be understood, and the data is provided for deciding polices for future demand response. This paper takes non-production load as research objects, sets goals of environmental protection, energy saving and safety utilization of electric power. Then residential electricity consumption characteristic is analyzed from time dimension, generic dimension, and influence dimension, it may provide support for demand response, such as making electricity price of peak load, power load staggering management, load control. The main contents of the paper are as follows.(1) The indexes of load characteristic are classified, then calculation method and meaning of those indexes are introduced in detail, it can lay the foundation for calculation of load characteristics and drawing load curve. Then factors which influence region load are analyzed, it includes electrovalence policy, power availability, measures of DSM, weather and so on. In the meantime, the analysis methods are introduced, these methods have shortcomings according to instance analysis.(2) Residential load is decomposed to basic load and cooling load which are classified based on adaptive fuzzy C-means algorithm. Daily electricity consumption and load curve characteristics are taken into account. Then typical customers are chosen based on weighted centroid model. According to contrastive analysis with grey correlativity analysis, the result shows that the method of choosing typical customer is feasible, reasonable and effective.(3) The indexes of typical residential load characteristics which include daily load factor, consumption rata of peak period, consumption rata of slack period, consumption rata of valley period and so on are calculated. Basic load and cooling load of typical customers are analyzed based on load curve, and then the potential of load control is also analyzed, based on this, a new classified method is proposed. Electricity consumption characteristic of power consumer are analyzed from electricity type, load characteristic and potential of load control. Finally, in view of electricity consumption characteristic of different power consumer, corresponding measures of potential of load control are formulated.