Investment decision optimization for distribution network planning with correlation constraint
Summary With the increasing access of distributed generation (DG), the investment decision of distribution network (DN) has become a large‐scale portfolio optimization problem with various reconstruction strategies, which reduces the applicability of the traditional investment decision optimization...
Saved in:
Published in | International transactions on electrical energy systems Vol. 30; no. 7 |
---|---|
Main Authors | , , , , , |
Format | Journal Article |
Language | English |
Published |
Hoboken
Hindawi Limited
01.07.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Summary
With the increasing access of distributed generation (DG), the investment decision of distribution network (DN) has become a large‐scale portfolio optimization problem with various reconstruction strategies, which reduces the applicability of the traditional investment decision optimization model. Therefore, aiming at reliability of DN, a novel deep belief networks (DBN)‐based correlation constraint‐integrated investment decision model is proposed in this paper. With the DBN‐based correlation constraint replacing the nonlinear and nonconvex constraints in the traditional model, a new investment decision model is established aiming at maximizing the reliability index and minimizing the total investment cost. In this way, the effects of different reconstruction strategies can be analysed, from which the optimal investment reconstruction plans are identified. Finally, an example of a regional distribution network in a city is provided to verify the rapidity, feasibility, and effectiveness of the investment decision model. |
---|---|
Bibliography: | Funding information International Visiting Program for Excellent Young Scholars of Sichuan University; Natural Science Foundation of China, Grant/Award Number: 51807127; Young Elite Scientists Sponsorship Program by Chinese Society of Electrical Engineering, Grant/Award Number: CSEE‐YESS‐2018006 |
ISSN: | 2050-7038 2050-7038 |
DOI: | 10.1002/2050-7038.12323 |