Multi-model data-driven dynamic state estimation and fault diagnosis of virtual synchronous machines
Multi-model data-driven dynamic state estimation and fault diagnosis of virtual synchronous machines
As power systems transition toward Inverter-Based Resources (IBRs), grid dynamics are increasingly governed by the discrete control logic of Virtual Synchronous Generator (VSG) inverters rather than the continuous physics of synchronous machines. A critical observability gap exists during Fault Ride-Through (FRT) events, where VSG inverters instantaneously switch from voltage-regulating to current-limiting modes, introducing hybrid dynamical non-linearities that destabilize conventional estimators. While recent approaches have addressed mode identification, they often rely on sliding windows that introduce detection latencies incompatible with next-generation setting-less protection schemes. This paper proposes a recursive Interacting Multiple Model Unscented Kalman Filter (IMM-UKF) tailored for sub-cycle detection of VSG mode transitions. Unlike existing works, this framework explicitly incorporates negative-sequence dynamics to maintain observability during asymmetrical faults.
Fault Diagnosis, IMM-UKF, State Estimation, Virtual Synchronous Generator, Virtual Synchronous Machine
819-823
Qing, Hanshan
3e2f5e64-d095-495b-8c68-236d53f9c3ac
Singh, Abhinav Kumar
6df7029f-21e3-4a06-b5f7-da46f35fc8d3
Batzelis, Efstratios
2a85086e-e403-443c-81a6-e3b4ee16ae5e
22 May 2026
Qing, Hanshan
3e2f5e64-d095-495b-8c68-236d53f9c3ac
Singh, Abhinav Kumar
6df7029f-21e3-4a06-b5f7-da46f35fc8d3
Batzelis, Efstratios
2a85086e-e403-443c-81a6-e3b4ee16ae5e
Qing, Hanshan, Singh, Abhinav Kumar and Batzelis, Efstratios
(2026)
Multi-model data-driven dynamic state estimation and fault diagnosis of virtual synchronous machines.
Lie, Tek-Tjing, Dai, Ningyi and Liu, Youbo
(eds.)
In 2026 11th Asia Conference on Power and Electrical Engineering (ACPEE 2026).
IEEE.
.
(doi:10.1109/ACPEE69242.2026.11524929).
Record type:
Conference or Workshop Item
(Paper)
Abstract
As power systems transition toward Inverter-Based Resources (IBRs), grid dynamics are increasingly governed by the discrete control logic of Virtual Synchronous Generator (VSG) inverters rather than the continuous physics of synchronous machines. A critical observability gap exists during Fault Ride-Through (FRT) events, where VSG inverters instantaneously switch from voltage-regulating to current-limiting modes, introducing hybrid dynamical non-linearities that destabilize conventional estimators. While recent approaches have addressed mode identification, they often rely on sliding windows that introduce detection latencies incompatible with next-generation setting-less protection schemes. This paper proposes a recursive Interacting Multiple Model Unscented Kalman Filter (IMM-UKF) tailored for sub-cycle detection of VSG mode transitions. Unlike existing works, this framework explicitly incorporates negative-sequence dynamics to maintain observability during asymmetrical faults.
Text
Multi_Model_Data_Driven_Dynamic_State_Estimation_and_Fault_Diagnosis_of_Virtual_Synchronous_Machines
- Version of Record
Restricted to Repository staff only
Request a copy
More information
Published date: 22 May 2026
Keywords:
Fault Diagnosis, IMM-UKF, State Estimation, Virtual Synchronous Generator, Virtual Synchronous Machine
Identifiers
Local EPrints ID: 513259
URI: https://eprints.soton.ac.uk/id/eprint/513259
PURE UUID: 09b5db49-37f2-4c43-9895-80cd69811fee
Catalogue record
Date deposited: 10 Aug 2026 16:35
Last modified: 13 Sep 2026 15:50
Export record
Altmetrics
Contributors
Author:
Hanshan Qing
Author:
Abhinav Kumar Singh
Author:
Efstratios Batzelis
Editor:
Tek-Tjing Lie
Editor:
Ningyi Dai
Editor:
Youbo Liu
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics