The Design Of Load Forecasting System Based On Multi-Agent

The forecasting to mid-long term load can provide important evidence to the power planning. The accurate load forecasting can improve the economics and reliability of power system operation. The load is related with a variety of factors, there is a complex nonlinear relationship between the load and the factors. The forecasting to mid-long term load is important because it can provide important evidence to the power planning.Multi-Agent technology has the characteristies of autonomy, distribution and coordination, and it also has the ability of organization, learning and reasoning. Using the multi-agent system to solve practical problems have many superiority, such as, strong robustness and reliability, and high problem-solving efficiency. Multi-agent systems is distributed from the field of artificial intelligence research, the focus is to research the coordination acts between a group of intelligent agent. It has the ability of to solve problems under the conditions of distributed information and do the parallel asynchronous decision-making under the adverse conditions, such as the individual information is incomplete, limited the ability of individuals, data fragmentation and so on.The mid-long term electric load forecasting provide essential data for the grid planning, it is helpful to optimize the planning of the power system. Solving the problem by the way of some agents’cooperation is the focus of multi-agent systems theory and technology. It is more useful to do the forecasting by the multi-agent collaboration than individual agent do it independently.In this paper introduces a load forecasting model based on multi-Agent. This model include three diagnosis-Agents, a management-Agent and fusion-Agent.In this model the Diagnosis-Agent are established according to the trend prediction model, exponent model and GM(1,1) model. Those diagnosis-Agents are regulated and controlled by the Management-Agent, which purpose to achieve the load in consultation under all diagnosis-Agents. The fusion-Agent will give the final load under the control situation which is from the Management-Agent’s message, predict result which should be from all the Diagnosis-Agents’predict result, different probability of each diagnosis-Agent and numbers of different diagnosis-Agent runs.