Short-term Power System Load Data Forecasting Based On Subset Entropy Attribute Recursive Decision Tree Algorithm
In order to ensure the safe, stable and economical operation of the power system, it is necessary to forecast the short-term power load. This helps to coordinate supply and demand balance among grid companies and planned maintenance of grid equipment, thereby improving the economics of grid operatio...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tamkang University Press
2022-11-01
|
| Series: | Journal of Applied Science and Engineering |
| Subjects: | |
| Online Access: | http://jase.tku.edu.tw/articles/jase-202306-26-6-0001 |
| Summary: | In order to ensure the safe, stable and economical operation of the power system, it is necessary to forecast the short-term power load. This helps to coordinate supply and demand balance among grid companies and planned maintenance of grid equipment, thereby improving the economics of grid operation. In the process of constructing the decision tree model, this paper makes full use of the feature that the decision tree recursively constructs new nodes from top to bottom. We divide the principal component space data set based on the attribute selection method of information entropy, and restore the divided data subsets to the original data space. This paper selects the real power load data in a certain area for experimental simulation, determines the values of important parameters through multiple experiments, uses k-means to calculate the Euclidean distance of each component to obtain the corresponding weights, and conducts experiments according to the model prediction process. The prediction results and prediction accuracy of each modal component and total prediction curve under several different models are compared. Through the mapping conversion and restoration of the original data space and the principal component space between the parent node and the child node, the information loss in the process of dimensionality reduction is avoided, and the influence of the information loss on the prediction accuracy of the algorithm is reduced. In addition, in order to verify the prediction effect of the decomposition algorithm and weighted reconstruction, a model comparison was carried out for whether the weighted sum was decomposed or not. The experimental results show that the Subset Entropy Attribute Recursive Decision Tree Algorithm (SEARDTA) has higher accuracy and exhibits excellent performance. |
|---|---|
| ISSN: | 2708-9967 2708-9975 |