Dynamic transverse correction method of middle and long term energy forecasting based on statistic of forecasting errors

Based on statistic of forecasting errors, a correction model of middle and long term energy forecasting is proposed. The correction steps are also presented. The factors that influence intending load energy are classified into three categories, long-term factors, middle-factors and short-term factors. According to forecasting errors variation amplitude, high and low error sections are also divided out. So there are six combinations between the load influence factor types and high-low error sections. By the analysis and statistic of history forecasting errors, the six error correction factors value are calculated dynamically. Then assign different weights to the correction factors, the forecasting error value of forecasting model for the next forecast period is gained. The validity and practicability of the proposed method are tested with the actual data.