Spatial Load Forecasting Based On Cloud Theory And Cellular Automation Theory

Spatial load forecasting has two notable features. First, the relevant factors are more uncertain. And second, the method covered the time, space and property characteristics of geographical entity. According to the existing researches, Cellular Automata has strong ability in simulated space dynamic evolution of complex system. Cloud theory can fully illustrate the uncertainties of concepts. Introduced the above theoretical characteristics, this paper proposed a spatial load forecasting model, based on Cloud theory and the Cellular Automata theory, to simulate the evolution of residential land use types during the planning level years.The core of the Cellular Automata is to acquire conversion rules. This paper applied Cloud theory to train CA conversion rules, considering several factors affecting load development, formulate some conversion rules. The establishment of cloud model and the uncertainty calculation, taking the fuzziness and randomness of factors into account, fully demonstrated the uncertainty of the conversion rules.The core question of spatial load forecasting is the load density selection. This article proposed load index optimization model based on Interval AHP theory and TOPSIS theory. The model calculated the weights of factors making use of Interval AHP theory, computed optimization load index in ideal solution by using TOPSIS theory, according to the similarity in the categories for correction, obtained optimal load index.Finally, the model proposed in this paper was used in a region of the actual data. The results showed that this prediction model had certain accuracy and reliability, and the certain superiority in spatial load forecasting.