| Power load is one of the most important indices for urban power network planning and management. The analysis and forecasting of load is the basis of planning. The research of load characteristics and its development trend have great significance for the secure and economical operation, planning and construction of power networks.Association rules based method is applied to the urban power load analysis in this paper. The method is based on the theory of data mining, considering the characteristic of power load. Firstly, the data of power load and its correlative factors are integrated, and then data warehouse with the theme of power load analysis is built. Secondly, in order to make analysis more effectively, the clustering method is used to generalize the original data. Last, the association rules algorithm is developed to find frequent items and achieve strong association rules from the historical data set. Therefore, the effect of correlative factors to power load can be analyzed.Base on the method above, a program was composed and load data of 48 cities in china were analyzed. The relations between annual power consumption and city type, such as the center hierarchy and executive category of city etc, were discovered in this paper. The relations between power consumption growth rate and some economy factors, such as the proportion of second industry and GDP growth, were analyzed further. Those results are realistic and also have given a specific numerical interval of factors. Some of the rules can't be revealed by traditional method. In the end, the validity of the theory and method applied in this paper were verified through install rules in data beforehand.With enriching of data, the method proposed in this paper has a very broad prospect of application. This deep data mining technique can help researches to find characteristics and development rules of power load which may provide a scientific basis for decision making on urban network planning.
|
PDF Full Text Request