Research For Power System Short-term Load For Casting

Great attention is always paid to power system short-term load forecasting (STLF), which is an important task of power utilities. STLF is concerned to the dispatching work and production scheme of power system, and accurate load forecasting is helpful to the security and stability of power system as well as saving its generation costs. With the establishment and development of the power market, STLF will play more and more important role in power system.After discussing the constituents and characteristics of the electric load, analyzing and comparing the merits and shortcomings of some forecasting methods, the paper chooses the hybrid method, artificial neural network (ANN) and fuzzy logic, to build the load forecasting model, which separates the forecasting job into two parts: one is the base load component, and the other is the revision for temperature and holidays. The based load (BL) component is forecasted in ANN, no considering the effect of temperature or holidays, which reduces the work for ANN and simplifies its structure; the BL revision is finished in Fuzzy, only considering the influence of temperature and holidays, where uses the expert experiences by introducing fuzzy language variable and gets the final forecasted load.The paper presents a simple algorithm for STLF: the learning speed of ANN is improved by using smoothing factor and forgetting factor, that is the fast ANN; the subjective experiences is well used in Fuzzy Logic, where the uneven membership function is introduced, reflecting load sensitivity to temperature.In the end, the paper discusses the improvement of STLF for the new situation under power market, puts forward the ready work to better power market.