A bus load forecasting method based on distribution factor and support vector machine

Abstract

The invention discloses a bus load forecasting method based on a distribution factor and a support vector machine. The method comprises steps of step 1, importing load data; 2, preprocess that load data; 3, classify that busbar load and determining each object in the hierarchical structure of the system distribution model; 4, optimize that parameters of the support vector machine model; Using Support Vector Machine Algorithm to Forecast the Total Load of the System; 5, calculate that load distribution coefficient of each region; 6, calculate distribution coefficients of each type of load; Step7, according to the data of step 4, step 5 and step 6, the predicted value of the bus load is solved, and the method can be completed. The invention can better deal with the emergencies of various loads in the bus load forecasting, achieve good forecasting effect, and comprehensively improve the scientific and refined level of load forecasting. The method of the invention can fully meet the requirements of lean management of power network operation, reasonably arrange production dispatching plan and implement energy-saving power generation dispatching.

Description

technical field [0001] The invention relates to the field of power system loads, in particular to a bus load prediction method based on distribution factors and support vector machines. Background technique [0002] Electricity has become the main source of energy in people's work and life. Electricity load is the sum of the electric power that the electric energy user's electric equipment takes from the power system at a certain moment. In order to ensure the normal power supply of the region, people will predict the load of the power system . Traditional power system load forecasting includes: short-term load forecasting (in units of days), medium-term load forecasting (in units of weeks and months) and long-term load forecasting (in units of quarters and years). [0003] In general load forecasting, the research object of various forecasting methods is generally a certain area or a certain power grid system, and its historical load data is comprehensive, and it is imposs...

AI Technical Summary

Problems solved by technology

However, unlike the system load, the change law of the bus load itself is not stable, and it is also prone to sudden changes, which leads to the inability to obtain high prediction accuracy simply by referring to the system load prediction method.

Images