| Title |
A Deep Learning Based Optimal Power Flow Model considering Binding Constraints and a Case Study of Large-Scale System Application 503 |
| Authors |
윤상민(Sangmin Yun) ; 신한솔(Hansol Shin) ; 곽규형(Kyuhyeong Kwag) ; 오효빈(Hyobin Oh) ; 윤형석(Hyeongseok Yun) ; 김욱(Wook Kim) |
| DOI |
https://doi.org/10.5370/KIEE.2023.72.4.503 |
| Keywords |
Optimal Power Flow; AC OPF; Deep Learning; Deep Neural Network; Binding constraints; Merit Order Stack Model |
| Abstract |
Recently, research using machine learning and deep learning is being conducted to quickly solve the AC optimal power flow (AC OPF) problem. The problem with the existing research is that, unlike the actual system, it did not consider the generator start-up status according to demand change by using a test system in which the minimum power generation of the generator is zero. This paper proposes a deep learning-based AC OPF model and post-processing for a simulated large-scale system in Korea. It is a difficult problem to predict in Korea's large-scale simulation system as the generator start-up status changes as the demand changes.
Accordingly, the accuracy of the deep neural network model, which is a deep learning algorithm, was increased by inserting the binding constraints and generator start-up status as variables. In addition, we want to satisfy the constraints of AC OPF with post-processing including the merit order stack model and AC power flow. |