Spatial Electric Load Forecasting Using a Local Movement Approach

Abstract

An agent based model for spatial electric load forecasting using a local movement approach for the spatio- temporal allocation of the new loads in the service zone is presented. The density of electrical load for each of the major consumer classes in each sub-zone is used as the current state of the agents. The spatial growth is simulated with a walking agent who starts his path in one of the activity centers of the city and goes to the limits of the city following a radial path depending on the different load levels. A series of update rules are established to simulate the S growth behavior and the complementarity between classes. The results are presented in future load density maps. The tests in a real system from a mid-size city show a high rate of success when compared with other techniques. The most important features of this methodology are the need for few data and the simplicity of the algorithm, allowing for future scalability Index Terms—Spatial electric load forecasting, land use, knowledge extraction, distribution planning, agent based models. (http://www.dee.feis.unesp.br/lapsee). This work presents an agent-based model approach to the spatial electric load forecasting, considering the city as a dynamic landscape that interact continuously with a walking agent. In a previous work (3), the characterization of the preferences for land use in the service zone was presented, using this information as a development probability for each of the major consumer classes in a city, and complementing it with information about redevelopment probability, the city can be represented as a dynamic landscape, where each sub-zone have a different probability to change from one load level to another. A walker agent is set loose in a pseudo-random path considering the development and redevelopment probabilities. This agent has the capacity to modify the status (increase the load) of each sub-zone in the dynamic landscape according to some rules, similar to the rules used in cellular automata algorithms. The information is presented in a grid, each sub zone have a discrete state that can be modified in discrete step times according to rules involving its own state, the state of the neighborhoods and the interaction with the walker agent. The initial state of each sub-zone is the actual electric load density, and that state can change to another level of electric load density in discrete time steps. In summary, this paper deals with a new approach for spatial electric load forecasting, considering new developed zones and redevelopment of the existing ones, using concepts of knowledge extraction algorithms, and elements from evolutionary algorithms, with special emphasis on developing a simple but powerful methodology with a reasonable use of available data, taking into account the stochastic nature of new consumers, and modeling the spatio-temporal allocation of the new loads using a structured algorithm based on agent-based models.