Abstract
In order to solve the problem of vulnerability assessment of complex power systems facing complex structures and large sizes, a novel data driven method based on random matrix theory is proposed in this paper. Firstly, with the use of phasor measurement units (PMUs) big data, evaluation matrices are constructed to extract statistical characteristics of power systems operation. Then, with the combination of random matrix theory and entropy theory, vulnerability evaluation index are constructed considering the degree of influence of some faults in power systems. With full use of big data, the model-free method is more accurate and comprehensive. Simulation results in IEEE 39-bus test system and a real-world power grid in China verify the effectiveness of the method.
Highlights
Due to the investment of new power electronic devices and the implementation of power market policy, a power system is more and more prone to disturbance, resulting in economic losses and serious social impact [1]–[4]
In order to consider the reaction of both single bus and overall system, s1 has been improved in this paper, a higher identification accuracy is achieved and complex modeling of the system is avoided.Section II presents the mathematical formula of random matrix theory and its corresponding laws; Section III shows the methodology of identifying critical buses using random matrix theory; Simulation results and analysis in IEEE 39-bus test system and real-world case are given in Section IV;and Section V is the conclusions and future work
The voltage of each bus remains constant when in normal operation, only with some fluctuations caused by measuring errors, noises or small disturbance, and the voltages of all buses are independent and in normal distribution, the data shows a statistical random characteristic, the empirical spectral distribution converges to M-P law and ring law, and Mean Spectral Radius (MSR) is larger than the inner circle
Summary
Due to the investment of new power electronic devices and the implementation of power market policy, a power system is more and more prone to disturbance, resulting in economic losses and serious social impact [1]–[4]. Paper [44]–[49] argues calculating the vulnerability of each bus based on random matrix theory and entropy theory, but the index proposed is not comprehensive and may cause misjudgment under some circumstances; Our previous work [50]–[53] proposes a data-driven method to evaluate vulnerability of power network, the index is not comprehensive and accurate enough. In order to consider the reaction of both single bus and overall system, s1 has been improved in this paper, a higher identification accuracy is achieved and complex modeling of the system is avoided.Section II presents the mathematical formula of random matrix theory and its corresponding laws; Section III shows the methodology of identifying critical buses using random matrix theory; Simulation results and analysis in IEEE 39-bus test system and real-world case are given in Section IV;and Section V is the conclusions and future work. Without knowing much about the model, the use of raw data can help to implement vulnerability optimization in power systems
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
