Correlation Mining for Reconstruction Measures and Performance Indexes of Distribution Network Planning Based on BP Neural Network

Since the traditional correlation analysis based on complex power flow calculation cannot meet the requirements of performance evaluation of current distribution network planning, the BP neural network (Back Propagation Neural Network, the BPNN) based correlation mining is proposed in this paper. Wi...

Full description

Saved in:
Bibliographic Details
Published in2018 International Conference on Power System Technology (POWERCON) pp. 4370 - 4376
Main Authors Chen, Bin, Liu, Junyong, Wang, Gang, Xiang, Yue, Chai, Yanxin, Ju, Li
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2018
Subjects
BP neural network
Correlation
Correlation mining
Data mining
Distribution network planning
Genetic algorithm
Genetic algorithms
Investment
Neural networks
Performance analysis
Performance indexes
Reconstruction measures
Training
Online AccessGet full text

Cover

More Information
Summary:Since the traditional correlation analysis based on complex power flow calculation cannot meet the requirements of performance evaluation of current distribution network planning, the BP neural network (Back Propagation Neural Network, the BPNN) based correlation mining is proposed in this paper. With the reconstruction measures and performance indexes as the training sample sets, the corresponding correlation model through the offline learning of sample data can be obtained. As a result, when given reconstruction measures in practical application, the neural network training can give the results of performance indexes quickly and accurately. In addition, in order to improve the generalization mapping capability of BPNN, the genetic algorithm is used to optimize the weights and thresholds of the BPNN. Experimental result based on the IEEE 33 node network shows the accuracy and effectiveness of the presented methodology.
DOI:10.1109/POWERCON.2018.8601579