Depth decision tree algorithm-based power load characteristic mining method

Abstract

The invention discloses a depth decision tree algorithm-based power load characteristic mining method. The depth decision tree algorithm-based power load characteristic mining method comprises the steps of acquiring load characteristic data of a power user and factor data affecting change of a power load by an acquisition device; pre-processing the acquired load characteristic data of the power user and the factor data affecting the change of the power load, and building a training data sample set and a test set; setting a parameter of a depth decision tree model, namely an initial forecast model parameter in a depth decision tree algorithm; training the depth decision tree model by the training data sample set; testing the trained depth decision tree model by the test set, and determiningthe depth of the depth decision tree model; obtaining the depth decision tree model after training and test are completed, and then inputting observation value data of which load characteristic is needed to be mined into the model and outputting a load characteristic forecast result; and achieving a depth mining target on the load characteristic of a power system so that a power enterprise is guided to safely dispatch and stably run.

Description

technical field [0001] The invention belongs to the field of electric load characteristic analysis, in particular to an electric load characteristic mining method based on a deep decision tree algorithm. Background technique [0002] The power system should provide safe, reliable and standard electric energy to all kinds of users, and meet the power demand of power users, that is, loads, at all times. With the rapid development of social economy, the upgrading of industrial structure, the change of global climate and environment, and the continuous improvement of people's living standards, the characteristics of power load have undergone greater changes than before. This has had an impact on the power system to ensure power balance and safe and stable operation at all times. In order to cope with such a situation, it is an effective measure to solve this problem to deeply dig the characteristics of electric load and grasp the changing law of electric load characteristics un...

AI Technical Summary

Problems solved by technology

This method is time-consuming and laborious, and cannot respond to rapidly changing electrical load characteristics in a timely manner.

Images