Decision Trees Using Synchronized Phasor Measurements for Wide-Area Response-Based Control

This paper proposes using phasor measurement unit (PMU) data to trigger one-shot control in response to loss of synchronism detection. Decision trees (DTs) are trained to associate voltage magnitude and angle measurements with loss of synchronism. Different approaches to the DT construction process are illustrated in an early section of the paper by restricting the input vector to only two elements. Response-based control is obtained by triggering one-shot control the first time the DT classifies measurements as “Unstable”. The same combination of one-shot controls is applied to any event the first time the DT outputs “Unstable”. The method is demonstrated on a 176-bus reduced-order model of the western U.S. interconnection. An effective one-shot control consists of a combination of fast power changes on four buses. This control is similar to 500-MW fast power increases on two HVDC interties and it reduces angle differences in the AC network.

[1]  N. D. Hatziargyriou,et al.  On-Line Preventive Dynamic Security of Isolated Power Systems Using Decision Trees , 2002, IEEE Power Engineering Review.

[2]  Vijay Vittal,et al.  An Online Dynamic Security Assessment Scheme Using Phasor Measurements and Decision Trees , 2007 .

[3]  Ruisheng Diao,et al.  Decision Tree-Based Online Voltage Security Assessment Using PMU Measurements , 2009, IEEE Transactions on Power Systems.

[4]  S.M. Rovnyak,et al.  Response-based decision trees to trigger one-shot stabilizing control , 2004, IEEE Transactions on Power Systems.

[5]  N. Senroy,et al.  A conceptual framework for the controlled islanding of interconnected power systems , 2006, IEEE Transactions on Power Systems.

[6]  A. A. Girgis,et al.  A new method for power system transient instability detection , 1997 .

[7]  Louis Wehenkel,et al.  Decision tree based transient stability method a case study , 1994 .

[8]  C. W. Taylor,et al.  Decision trees using apparent resistance to detect impending loss of synchronism , 2000 .

[9]  S.M. Rovnyak,et al.  Integral square generator angle index for stability ranking and control , 2005, IEEE Transactions on Power Systems.

[10]  James S. Thorp,et al.  Decision trees for real-time transient stability prediction , 1994 .

[11]  Ruisheng Diao,et al.  Decision tree assisted controlled islanding for preventing cascading events , 2009, 2009 IEEE/PES Power Systems Conference and Exposition.

[12]  S.M. Rovnyak,et al.  Clustering-Based Dynamic Event Location Using Wide-Area Phasor Measurements , 2008, IEEE Transactions on Power Systems.

[13]  N.D. Hatziargyriou,et al.  Decision trees for dynamic security assessment and load shedding scheme , 2006, 2006 IEEE Power Engineering Society General Meeting.

[14]  N. Senroy,et al.  Decision Tree Assisted Controlled Islanding , 2006, IEEE Transactions on Power Systems.

[15]  S. P. Teeuwsen,et al.  Genetic algorithm and decision tree based oscillatory stability assessment , 2006, 2005 IEEE Russia Power Tech.

[16]  Jin Lu,et al.  Predicting future behavior of transient events rapidly enough to evaluate remedial control options in real-time , 1995 .

[17]  I Kamwa,et al.  Development of rule-based classifiers for rapid stability assessment of wide-area post disturbance records , 2009, IEEE PES General Meeting.