Large-scale distribution network load data cleaning research based on  improved random forest

Min Li; Jing Tan; Yigang Tao; Song Wan; Qiju Zhang

doi:10.1117/12.2679593

18 July 2023 Large-scale distribution network load data cleaning research based on improved random forest

Min Li, Jing Tan, Yigang Tao, Song Wan, Qiju Zhang

Proceedings Volume 12722, Third International Conference on Mechanical, Electronics, and Electrical and Automation Control (METMS 2023); 127221C (2023) https://doi.org/10.1117/12.2679593
Event: International Conference on Mechanical, Electronics, and Electrical and Automation Control (METMS 2023), 2023, Hangzhou, China

Abstract

Some large-scale distribution network load data cleaning methods have the problem of excessive relative error. A load data cleaning method for large-scale distribution network based on improved random forest is designed. Based on the release characteristics of electrical loads, large-scale distribution network open source data is collected, and an anomaly data processing model is constructed based on improved random forests to describe the processing process of numerical error data, using the original sample set as a classifier and optimising the load data cleaning model. Test results: The mean relative errors of the large-scale distribution network load data cleaning method in this paper and the other two large-scale distribution network load data cleaning methods are 3.618%, 5.162% and 5.007% respectively, indicating that the performance of the designed large-scale distribution network load data cleaning method is better after incorporating the improved random forest algorithm.

Citation Download Citation

Min Li, Jing Tan, Yigang Tao, Song Wan, and Qiju Zhang "Large-scale distribution network load data cleaning research based on improved random forest", Proc. SPIE 12722, Third International Conference on Mechanical, Electronics, and Electrical and Automation Control (METMS 2023), 127221C (18 July 2023); https://doi.org/10.1117/12.2679593

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available