Abstract:Weather information is an important factor in short-term load forecasting (STLF). However, for a long time, more importance has always been attached to forecasting models instead of other processes such as the introduction of weather factors or feature selection for STLF. The main aim of this paper is to develop a novel methodology based on Fisher information for meteorological variables introduction and variable selection in STLF. Fisher information computation for one-dimensional and multidimensional weather variables is first described, and then the introduction of meteorological factors and variables selection for STLF models are discussed in detail. On this basis, different forecasting models with the proposed methodology are established. The proposed methodology is implemented on real data obtained from Electric Power Utility of Zhenjiang, Jiangsu Province, in southeast China. The results show the advantages of the proposed methodology in comparison with other traditional ones regarding prediction accuracy, and it has very good practical significance. Therefore, it can be used as a unified method for introducing weather variables into STLF models, and selecting their features.

No abstract

The dynamics structure of many complex system there often happens a suddenly change due to the effect of outside force in nature. This abrupt change is closely related to human life. In order to make an accurate prediction of abrupt change and take corresponding measures, abrupt change should be detected effectively. Fisher information (FI) can keenly catch and characterize a small change in probability density distribution (PDD) of a system variable. While there is a change in dynamic structure of a system, the PDD of the system variable will have some changes correspondingly. In view of this, in this paper we describe in detail that FI is used to detect and recognize the abrupt change of system dynamics structure, and it is a new method to solve the detection of abrupt change in dynamic structure of a system. First of all, this method is used to present the ability to detect abrupt change lying in ideal datasets of different types of ideal signals. Next, this method is used to analyze daily datasets of ground climate from national meteorological information center of China meteorological administration between 1960 and 2008 in Lanzhou meteorological observation station in Northwest area. The results show that the daily datasets are consistent with historical records, which further verifies that the method is effective and practical.