Online Inference of the Dynamic Security Level of Power Systems Using Fuzzy Techniques

Machine learning and digital twins: monitoring and control for dynamic security in power systems

“…They must comply with boundary definitions from grid codes or other specified thresholds. Some indices considered in DSA applications for transmission systems are given in [35,36]. We furthermore distinguish between offline and online DSA.…”

Archives of Electrical Engineering

“…The power variation in all lines of the network with respect to a power injection in bus k can be represented by (3). P i is the initial power value in line i, P k is the power injection in bus k, P (k) i is the resulting power transfer to line i after the injection in bus k, and γ ik is the PTDF of line i with respect to the power injection in bus k [15,16]:…”

“…To simplify this task, some authors prefer to remove a number of low impact contingencies and only focus on critical events to appropriately monitor power system security [1,2]. Furthermore, critical contingencies for online applications require a significant reduction in calculation time [3][4][5].…”

Techniques of power system static security assessment and improvement: A literature survey

“…SA can be broadly categorized into two: dynamic SA and static SA [ 14 , 27 , 28 ]. The purpose of dynamic security assessment (DSA) is to identify and evaluate contingencies that may cause transient (angle), frequency or voltage instabilities by evaluating the time dependent transition from the pre-contingent to post contingent steady state [ 28 , 29 ]. Methods such as continuous power flow solutions of P –V or Q-V curves, modal analysis, or computation of voltage stability margin for voltage stability [ 8 ], time-domain simulation for transient stability, and analysis of eigenvalues for small signal stability are used to carry out DSA [ 30 , 31 ].…”

“…Data-driven artificial intelligence (AI) techniques such as fuzzy inference systems [ 29 , 33 ], Artificial Neural Network ( ), support vector machine (SVM) [ 14 ], decision tree (DT), graph convolutional neural network [ 34 ], cascaded convolutional network [ 35 ] and long short-term memory network ( ) [ 30 ] have been widely employed to overcome the limitations of classical model-driven simulation approaches in DSA [ 3 ]. Due to their capacity to tackle nonlinear problems with remarkable speed and accuracy, these and other AI approaches have attracted the researchers’ attention [ 8 ].…”