Lecture 9/ 10 - Cascading failure Flashcards
(15 cards)
What is a “Common” Cause of Failure?
Multiple failures that result directly from a single common or shared root cause
Cascading Failure definition
DOMINO effect
several components shared a common load –> 1 component failure may lead to increase load on the remains ones –> increased likelihood of failure –> repeat
Detail the steps of the cascading failure model using the “Local propagation of a fixed amount of load”.
What type of results does it provide?
What kind of conclusions can be drawn from them? What are the limitations of this model?
- the initial load of each component is sampled from a Uniform random distribution
- a disturbance is applied to all components
- the load of each component is compared to the failure load
- if the load of the component is greater then the failure load the component fails and a FIXED amount of load is transferred to the neighbouring working nodes (Spreading rule)
The model provides insights into the robustness of the system against cascading failures under varying initial conditions. It outputs the number of failed components and the overall system stability after the cascade stops
LIMITATIONS:
- The initial loading conditions are randomly sampled and do not necessarily result from the actual network structure or operational conditions.
- ## model operates in discrete steps, which might not capture the continuous and dynamic nature of real-world cascading failures
Detail the steps of the cascading failure model using the “Flow- based failure propagation”.
What type of results does it provide?
What kind of conclusions can be drawn from them? What are the limitations of this model?
- initial load: The initial load of each node is determined by the number of shortest paths that pass through that node
- Node capacity definition: C = (1+ alpha)*L
alpha is the tolerance parameter representing the operating margin that allows safe operations under possible load increments - The model introduces a disturbance by removing the most congested component, which is typically the node with the highest load relative to its capacity.
- After a node fails, the shortest paths are recalculated, and the loads are redistributed across the network. This can lead to changes in the load on other nodes.
- Failure Propagation: The process of recalculating shortest paths and redistributing loads continues iteratively. If any node’s load exceeds its capacity, it fails and is removed from the network.
This model provides insights into how the failure of critical nodes can lead to widespread cascading failures in the network. It highlights the importance of node connectivity and the impact of load redistribution on network stability.
Limitations:
the initial loading condition is dictated by the
network structure (topology)
Define the concept of cascade-safe operations of network systems.
How can we use the results of cascading failure models to identify cascade-safe operating margins?
s. This concept involves ensuring that the network is resilient to initial disturbances and can contain or mitigate the propagation of failures, thereby maintaining overall system stability and functionality.
Cascading failure models can be used to identify cascade-safe operating margins by simulating various scenarios and analyzing the results to determine the conditions under which the network remains stable
Discuss strategies to prevent or mitigate the propagation of cascading failures in single network systems
- redundant components and path
- maintenance and testing
- understand which notes are the most important ones
What is the difference between “network static structure analysis” and “Dynamic modeling of cascading failures propagation in network systems”.
What different types of results are given by these two analyses?
Network Static Structure Analysis examines the fixed properties of a network, such as its layout and connections. It provides insights into the network’s topology, critical components, and potential structural vulnerabilities.
Dynamic Modeling of Cascading Failures Propagation simulates how failures spread through a network over time. It offers insights into failure pathways, system resilience, safe operating limits, and strategies to mitigate failure propagation.
What is “abstract” cascading failure ?
Sequence of failures in interconnected systems
- Triggered by an initial event and spreading over the system according to the connectivity pattern (structure) and a spreading rule (dynamics)
speaking rule eg: fixed load (5%) transferred after a failure to neighboring nodes
What are the main steps of a physics-based model for cascading failure analyses of a power system?
- initialise
- introduce failures - monte Carlo, Markov chains…
- check frequency
- check voltage
- implement voltage and frequency actions if needed (connect / disconnect generators or demand)
- analyse the results
Why do we introduce contingencies (failures)?
How do we generate the contingencies?
to simulate disruptions or failures
We generate the contingencies using simulation such as Monte Carlo, or using historical data, or Markov models
What does the frequency control do?
It checks the frequency of the system
What does the under-frequency load shedding do?
disconnects demand
What does the under-voltage load shedding do?
protects equipment, disconnects load, detects voltage drop
What type of results a cascading failure model provides?
Risk assessment and Impact analysis !!
What are the main difference between topological and physics-based models for cascading failure analyses? Mention some advantages and disadvantages?
Topological:
Advantages
▪ Require little data
▪ Easy to implement
▪ Work on different networks
Disadvantages
▪ Disregard physics
▪ Qualitative results
Physics-based
▪ Advantages
▪ Consider the physics
▪ Quantitative results
▪ In-depth analyses
Disadvantages
▪ Large amount of data
▪ High modeling effort
▪ Computationally demanding
