Title :
An adaptive importance sampling method for probabilistic optimal power flow
Author :
Huang, Jie ; Xue, Yusheng ; Dong, Z.Y. ; Wong, K.P.
Author_Institution :
State Grid Electr. Power Res. Inst., Nanjing, China
Abstract :
A new probabilistic optimal power flow method is developed to manage electricity market price risk. The proposed method is an adaptive importance sampling method which based on the principle of importance sampling and reinforcement learning. The estimation result of conventional Monte Carlo simulation is taken as benchmark. Case study is conducted on IEEE 39-Bus system to compare the proposed method with point estimate method, which shows the feasibility and efficiency of the method.
Keywords :
Monte Carlo methods; learning (artificial intelligence); load flow; power engineering computing; power markets; IEEE 39-bus system; Monte Carlo simulation; adaptive importance sampling; electricity market price risk; probabilistic optimal power flow; reinforcement learning; Accuracy; Benchmark testing; Computational modeling; Electricity supply industry; Estimation; Monte Carlo methods; Uncertainty; Importance Sampling; Point Estimate Method; Probabilistic Optimal Power Flow; Reinforcement Learning; Risk Management;
Conference_Titel :
Power and Energy Society General Meeting, 2011 IEEE
Conference_Location :
San Diego, CA
Print_ISBN :
978-1-4577-1000-1
Electronic_ISBN :
1944-9925
DOI :
10.1109/PES.2011.6039167
