Assessing survivability of smart grid distribution network designs accounting for multiple failures

Abstract

distribution automation (DA). The automation of the smart grid brings novel challenges to the power grid engineers, such as the assessment of the tradeoffs involved to accurately engineer the power distribution reliability.The introduction of automation (intelligence) to power distribution networks has created a need for the holistic assessment of the distribution network. The distribution network reliability is a function of the correct operation of several architecture artifacts such as electrical power components, telecommunications, distribution network topology, failure detection isolation and restoration, demand response, and distributed generation (DG) and storage. The automation of power distribution requires a more integrated perspective across these domains. However, in the current mode of operation, they are still being engineered separately.Traditionally, the reliability of power systems has been quantified using average metrics, such as the system average interruption duration index (SAIDI). SAIDI is used by public service commissions in the USA to assess utilities' compliance with the commission rules. It was developed to track manual restoration times, and according to standard 166-1998, the median value for North American utilities is roughly one and a half hours. In smart grid networks, power failure and restoration events will have a finer level of granularity, because of the deployment of reclosers, which isolate faulty sections, and demand side management system activities, such as distributed generators and demand response application systems. Therefore, there is a need to extend the SAIDI metric and to develop new models and tools for the accurate computation of customer interruption indexes after power failure events occur, even if the occurrence of such events is rare. The survivability of a missioncritical application is the ability of the system to continue functioning during and after a failure or disturbance [1].In [2], we presented a proposal for a common analysis framework to support the survivability analysis of DA using extensions of the International Electrotechnical Commission-standardized common information model. The paper presented a case study of the application of the proposed method to the survivability analysis of a simple DA network that was derived from a real power distribution network. In [3], we have evaluated the impact of available active and reactive power supply after a section failure on the distributed automation survivability metric, and we derived closed-form expressions for certain survivability-related metrics. In [2][3][4], the survivability model accounted for single failures.In this paper, we present an analytical model to assess the survivability of distributed automation power grids and to predict SAIDI and related metrics as a function of different system parameters related to communications, DG, demand response, and other smart grid features, accounting for multiple failures. We use a performability model to capture how the system recovers from a fail...

Survivability Evaluation of Gas, Water and Electricity Infrastructures

“…Additionally, the model presented here supports a single faulty section only. In ongoing work [46], we are extending this to multiple failures.…”

“…In the following, we show a typical case study of applying our method taken from [46]. The load per section in the topology (cf.…”

“…Parameters of our model are p, the probability that the communication network is still operational after a section failure; q, the probability that there is sufficient backup power to supply energy for sections i+; and r, the probability that load can be successfully reduced in case of insufficient backup power. The time-related parameters in our model are α, the time to restore the upstream sections, β, the time to call for demand/response; δ, the time to manually repair the faulty section, Table 2 Choice of reward rates: lower bound on energy not supplied per hour (ENS/h) (from [46]). and γ, the time to restore communication.…”

“…Our recently introduced approach [8,46] targets assessment of transient properties of the power systems accounting for the implications of electro-mechanical and computer-based strategies to address failures in an integrated manner. In this approach, we quantify the effect of FDIR behaviour and demand/response functionality on survivability metrics, based on extended SAIDI metrics.…”

Survivability as a generalization of recovery

“…This is often referred to as escalated levels of recovery, where the recovery period is modeled in phases [12], [13]. Each of the recovery phases represents a new system/service restoration attempt.…”