Method for determining forecast precision of multiple levels of loads

Abstract

The invention relates to a method for determining the forecast precision of multiple levels of loads, and belongs to the technical field of load forecast in the electric system. The method for determining the forecast precision of multiple levels of loads includes the steps of calculating a (Percentage Error) PE index of a load curve of the loads of each level and in each area in each time period, and carrying out exponential transformation on each PE index to obtain an (Exponent of Percentage Error) EPE index in each time period; obtaining geometrical means from the EPE indexes to obtain a (Geometric Mean Exponent of Percentage Error) GMEPE; calculating the precision datum line of the loads of each level and in each area by adopting a stability forecasting method; diving the GMEPE by the precision datum line to obtain reference precision; determining the forecast results of the loads of each level and in each area based on the reference precision. The method for determining the forecast precision of multiple levels of loads adopts the GMEPE to measure the forecast precision, and determines the forecast precision based on the reference precision, thereby improving the forecast precision of the overall loads of a power grid and lowering the operational risks of the power grid.

Description

technical field [0001] The invention belongs to the technical field of power system load forecasting, and in particular relates to a precision analysis and evaluation method for multi-level load forecasting. Background technique [0002] Power systems are among the largest and most complex systems in the world. Judging from the actual situation of our country, the development of my country's electric power industry has naturally formed a national, regional and provincial three-level power grid today. There is an obvious hierarchical relationship between the loads of various levels of power grids, and the load characteristics of different areas of the same level of power grids are also different. [0003] From a vertical perspective, the load forecasting of different levels of power grids currently has the problem of "different accuracy measurement methods". At the level of system load, the average absolute relative error index——MAPE (Mean Absolute Percent Error)—with relat...

AI Technical Summary

Problems solved by technology

There are objective differences in the load regularity of different areas that cannot be changed in a short period of time. For example, the load regularity of areas containing impact loads such as steel mills is usually poorer than that of general areas, and it is difficult to obtain better prediction accuracy

At present, the power grid adopts a unified accuracy judgment standard to require different regions, ignoring the difference in load predictability in different regions, which is not conducive to the improvement of the overall load prediction accuracy of the power grid, increases the risk of power grid operation, and reduces the economy of power grid dispatching

Images