Overload Analysis of Distribution Transformers Based on Relevance Analyse and Machine Learning
| Autor: | Jian Wang, Wang Xingzhao, Hou Zhiyuan, Ying Deng, Hanming Liu, Yalei Huang |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Apriori algorithm
Computer science business.industry 020209 energy Multivariable calculus 02 engineering and technology Distribution transformer Machine learning computer.software_genre Data modeling Electric power system Robustness (computer science) 0202 electrical engineering electronic engineering information engineering Relevance (information retrieval) Artificial intelligence Time series business computer |
| Zdroj: | 2021 IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC). |
| Popis: | Aiming at the long-term heavy overload problem in distribution network operation. In order to avoid the impact of heavy overload on the power system and prepare response measures in advance,, this paper proposes an analysis method of heavy overload influence factors based on relevance analyse and machine learning. Firstly based on distribution transformer historical load information, weather, customer information, etc., to find out the factors affecting the overload of the distribution transformer with Apriori algorithm, on this basis, determine the input variables. Then, use the long short-term memory (LSTM) network to establish the overload prediction model, and the experiment was carried out on the collected multi element overload data. The results show that the model has high prediction accuracy and strong robustness in dealing with large scale multivariable time series data. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|