Open AccessEmergency control strategy of power system transient instability based on DBN
Author(s) -
Ziyue Qiang,
Junyong Wu,
Baoqin Li,
Ruoyu Zhang,
Liuyun Qin
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/645/1/012016
Subject(s) - tripping , transient (computer programming) , control theory (sociology) , electric power system , generator (circuit theory) , instability , electronic stability control , computer science , stability (learning theory) , power (physics) , blackout , engineering , control (management) , control engineering , automotive engineering , artificial intelligence , machine learning , physics , circuit breaker , electrical engineering , quantum mechanics , mechanics , operating system
In order to satisfy the real-time emergency control decision-making after power system transient instability, an emergency control strategy based on deep belief network is proposed in this paper. Firstly, the DBN instability degree prediction model is established to fit the mapping relationship between generator power angle characteristics and transient stability coefficient; secondly, according to the fitting instability degree index, the sensitivity of generator tripping is solved to determine the generator tripping control action bus; finally, according to the emergency control optimization model and transient stability constraints, the optimal generator tripping strategy is solved and verified in New England 10-machine and 39-node system. The results show that the proposed method has high prediction accuracy andefficiency, which can make the unstable system quickly resume stable operation.