China
Africa
Explore chapters and articles related to this topic
Regional resilience analysis
Published in Paolo Gardoni, Routledge Handbook of Sustainable and Resilient Infrastructure, 2018
Neetesh Sharma, Armin Tabandeh, Paolo Gardoni
For the region of interest (i.e., Shelby Country), we considered a detailed model of the electric power infrastructure that captures the variability of the initial impact and recovery of different areas of Shelby County (Sharma & Gardoni 2018c). The electric power infrastructure in Shelby County is operated by MLGW which sources its power from the Tennessee Valley Authority (TVA). TVA constitutes its own balancing authority in the eastern interconnection of the continental US power transmission grid. We modeled the power infrastructure of TVA with sufficient details to be able to run a power flow analysis (Sharma and Gardoni 2018c). Figure 28.5(a) shows the topology and service areas of the electric power infrastructure in Shelby County, as explained in Chang et al. (1996), and Figure 28.5(b) shows the topology of the infrastructure in Tennessee. The TVA operated infrastructure in Figure 28.5(b) is synthetically generated but it is representative in accordance with the data provided by Birchfield et al. (2017). To estimate the hourly power demand at different service areas, we used the MLGW annual fact sheet (MLGW 2015) and the per capita power demand provided by Birchfield et al. (2017). We also added the generators from Allen and Southaven power plant, located near Memphis, which were not included in past studies.
Cyber-resilience
Published in Stavros Shiaeles, Nicholas Kolokotronis, Internet of Things, Threats, Landscape, and Countermeasures, 2021
E. Bellini, G. Sargsyan, D. Kavallieros
In [46], an end to end cyber-physical attack-resilient framework for a Wide Area Monitoring, Protection, and Control (WAMPC) is reported. The framework is composed by the following steps: Risk Assessment, Prevention, Detection and Mitigation/Resilience.C1—Prevention: would be achieved through a combination of multiple approaches as quantitative risk assessment, attack-resilient measurement design, and moving-target inspired algorithms. Quantitative risk assessment involves modeling all the components of risk, namely, threats, vulnerabilities, and impacts, using approaches such as probabilistic or game-theoretic modeling. Attack-resilient measurement design involves algorithms that identify and recommend redundant measurement deployments that could be fed additionally into the WAMPAC applications, thereby increasing the difficulty of creating successful attacks. The optimal selection of redundant measurements is achieved by formulating a design problem that optimizes the placement of new sensors (e.g., PMUs) such that the accuracy, bad-data detection capability, and observability of the system improve while satisfying cost constraints. Moving-target inspired algorithms leverage a redundant measurement design and randomize the associated design parameters at the WAMPAC algorithm level while still ensuring that the functionality of the algorithm is maintained.C2—Detection: In the proposed framework, CPS model-based anomaly detection approaches that leverage sound mathematical tools from machine learning and related domains are critical for the detection of cyber-attacks beyond traditional IT intrusion detection techniques. Additionally, specification-based anomaly detection approaches serve a complementary role to CPS model-based approaches and enable the reduction of false-positive rate by capturing the normal behavior and existing security policies as part of a formal specification or languageC3—Mitigation: In the proposed framework, attack mitigation/resilience would be achieved using CPS model-based mitigation, and dynamic system reconfiguration and resiliency algorithms appropriately. One of the aspects of attack mitigation is the ability of the attack-resilient algorithm to recover from faults, either partially or completely. Based on the output of the anomaly detection module, if the data are considered “anomalous,” a CPS model-based mitigation method would be triggered. For example, if the SCADA measurement data used by the Automatic Generation Control (AGC) application are found to be untrustworthy, then the control signal would be calculated based on a statistical model that uses short-term load forecast information along with system generation parameters to predict its most likely value for a particular balancing authority area until a redundant, trusted source of SCADA telemetry is restored.
Estimating Spinning Reserve Capacity With Extreme Learning Machine Method in Advanced Power Systems Under Ancillary Services Instructions
Published in Electric Power Components and Systems, 2022
Two different mechanisms are used to control the frequency. Under normal circumstances, independent manufacturers ignore the system frequency; generator regulators typically have a ± 0.35 Hz dead band. Instead, each balancing authority operator focuses on balancing generation and load while monitoring system frequency. The system frequency is stable when the total system output and load are in balance. Instead of directly measuring production and load, system operators focus on maintaining the planned amount of net exchange with their neighbors. Included in the area control error (ACE) equation is a frequency bias term that requires each compensation field to increase generation when there is system frequency. When low frequency is high, it reduces production. The deviation is determined in MW/0.1 Hz and depends on the MW size of the stabilization area.Primary (fall) control = dead band blocks response under normal conditionsSecondary (AGC) control = regulation reserve, ISO computer controlledTertiary control = manual, six-hour, and hourly energy market response