Virtual Inertia Estimation of New Energy Power System Based on WRARMAX Model and LSTM

In order to precisely monitor the inertia of power systems, it is crucial to fulfill two key goals. Firstly, the inertia identification model should have a minimal computational time. Secondly, the accuracy of the outcomes must be enhanced. This study introduces the WRARMAX model, which is an Autore...

Full description

Saved in:
Bibliographic Details
Published in2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) pp. 626 - 631
Main Authors Lv, Longbiao, Wang, Bao, Yang, Yunhao, Jia, Jianxiong, Ma, Yanru, Jiang, Xiaotian
Format Conference Proceeding
LanguageEnglish
Published IEEE 21.07.2024
Subjects
Autoregressive processes
Computational modeling
Inertia identification
Long short term memory
LSTM
Monitoring
Power system stability
Real-time systems
Reliability engineering
Renewable energy
Renewable energy sources
Stability analysis
Systems simulation
WRARMAX
Online AccessGet full text

Cover

More Information
Summary:In order to precisely monitor the inertia of power systems, it is crucial to fulfill two key goals. Firstly, the inertia identification model should have a minimal computational time. Secondly, the accuracy of the outcomes must be enhanced. This study introduces the WRARMAX model, which is an Autoregressive Moving Average model with exogenous input solved using Weighted Recursive Least Squares, to expedite the initial phase of model identification. Additionally, the incorporation of multithreaded parallel processing in the code further accelerates the identification process. For the subsequent phase, the Long Short-Term Memory (LSTM) network is utilized to refine the residuals from the initial results, thus enhancing the correlation between the derived inertia values and their actual values. The integration of these two strategies leads to a more rapid and precise inertia identification process, surpassing the efficacy of prior techniques.
DOI:10.1109/CCSSTA62096.2024.10691830