دانلود رایگان مقاله انگلیسی تجزیه و تحلیل قابلیت اطمینان در سیستم های شبکه هوشمند وابسته به یکدیگر - الزویر 2018

Abstract:

Complex network theory is a useful way to study many real complex systems. In this paper, a reliability analysis model based on complex network theory is introduced in interdependent smart grid systems. In this paper, we focus on understanding the structure of smart grid systems and studying the underlying network model, their interactions, and relationships and how cascading failures occur in the interdependent smart grid systems. We propose a practical model for interdependent smart grid systems using complex theory. Besides, based on percolation theory, we also study the effect of cascading failures effect and reveal detailed mathematical analysis of failure propagation in such systems. We analyze the reliability of our proposed model caused by random attacks or failures by calculating the size of giant functioning components in interdependent smart grid systems. Our simulation results also show that there exists a threshold for the proportion of faulty nodes, beyond which the smart gird systems collapse. Also we determine the critical values for different system parameters. In this way, the reliability analysis model based on complex network theory can be effectively utilized for anti-attack and protection purposes in interdependent smart grid systems.

5 Conclusions

We study the reliability and cascading risk of a smart gird system in which a cyber network overlays a grid network. To check the network reliability against random nodes failures, we estimate the fraction of nodes that still remain functional after the cascading failures process stops, and then we can obtain the correct results by simulation experiment. Our findings show that there is always a critical threshold value. If the percentage of failing nodes is greater than the critical value, the interdependent smart gird systems will collapse. Our theory analysis and simulation experiment also show that, if both networks satisfy the same degree distribution, the system reliability does not have the direct connection with the system size. However, our proposed analysis model still has some limitations which could be our future work. For instance, we consider both networks are ER networks while the realistic settings are scale-free. Meanwhile, the giant components could not always work in reality. It is also of interest to study models that are more realistic than the existing ones in this paper. Clearly, there are still many open questions about interdependent smart grid systems. We are currently investigating related work along this avenue.