Electrical power system short-term load forecasting method based on big data technology

Abstract

The present invention provides a electrical power system short-term load forecasting method based on a big data technology. The user-level load forecasting is realized by utilizing a data mining technology, and user-level loads are accumulated to form a system load. The electrical power system short-term load forecasting method comprises the steps of: conducting load curve clustering analysis, and classifying load curves with similar shape features into a class; determining key influence factors, and achieving the purposes of reducing classifying rules and simplifying a forecasting model; establishing classifying rules, and adopting a CART decision tree algorithm to obtain condensation level clustering analysis results; classifying days to be forecast; training the forecasting model and forecasting, and selecting a corresponding support vector machine model to complete the forecasting according to the obtained classification results of the days to be forecast; and completing the step of calculating the system load on a Hadoop big data calculating platform. The electrical power system short-term load forecasting method based on the big data technology studies a user-level load forecasting framework, discovers electricity utilization behavior rules of a user by utilizing the data mining technology, and increases the precision of load forecasting.

Description

technical field [0001] The invention relates to the technical field of power system engineering, in particular to a short-term load forecasting method of a power system based on big data technology. Background technique [0002] The results of power system short-term load forecasting are related to the formulation of power system dispatching and production plans. Accurate short-term load forecasting results can help improve system security and stability, and can reduce power generation costs. With the massive access of distributed energy (solar energy, wind energy, energy storage, etc.) in the power system, it is more difficult to grasp the change law of the load, and this uncertainty will increase the difficulty of load forecasting in the power system. Therefore, there is an urgent need for a forecasting method that can better grasp the law of load variation. [0003] Users are the most basic part of the power grid, and they are also the source of load fluctuations in the ...

AI Technical Summary

Problems solved by technology

[0004] The present invention proposes a short-term load forecasting method for power systems based on big data technology, which effectively solves the problem of low load forecasting accuracy caused by complex power consumption laws of users

Images