Optimal Resource Allocation in Electrical Network Defense
Infrastructure networks supplying electricity, natural gas, water, and other commodities are at risk of disruption due to well-engineered and coordinated terrorist attacks. Countermeasures such as hardening targets, acquisition of spare critical components, and surveillance can be undertaken to detect and deter these attacks. Allocation of available countermeasures resources to sites or activities in a manner that maximizes their effectiveness is a challenging problem. This allocation must take into account the adversary's response after the countermeasure assets are in place and consequence mitigation measures the infrastructure operation can undertake after the attack. The adversary may simply switch strategies to avoid countermeasures when executing the attack. Stockpiling spares of critical energy infrastructure components has been identified as a key element of a grid infrastructure defense strategy in a recent National Academy of Sciences report [1]. Consider a scenario where an attacker attempts to interrupt the service of an electrical network by disabling some of its facilities while a defender wants to prevent or minimize the effectiveness of any attack. The interaction between the attacker and the defender can be described in three stages: (1) The defender deploys countermeasures, (2) The attacker disrupts the network, and (3) The defender responds to the attack by rerouting power to maintain service while trying to repair damage. In the first stage, the defender considers all possible attack scenarios and deploys countermeasures to defend against the worst scenarios. Countermeasures can include hardening targets, acquiring spare critical components, and installing surveillance devices. In the second stage, the attacker, with full knowledge of the deployed countermeasures, attempts to disable some nodes or links in the network to inflict the greatest loss on the defender. In the third stage, the defender re-dispatches power and restores disabled nodes or links to minimize the loss. The loss can be measured in costs, including the costs of using more expensive generators and the economic losses that can be attributed to loss of load. The defender's goal is to minimize the loss while the attacker wants to maximize it. Assuming some level of budget constraint, each side can only defend or attack a limited number of network elements. When an element is attacked, it is assumed that it will be totally disabled. It is assumed that when an element is defended it cannot be disabled, which may mean that it will be restored in a very short time after being attacked. The rest of the paper is organized as follows. Section 2 will briefly review literature related to multilevel programming and network defense. Section 3 presents a mathematical formulation of the electrical network defense problem. Section 4 describes the solution algorithms. Section 5 discusses computational results. Finally, Sec. 6 explores future research directions.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- US Department of Energy (US)
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 15009766
- Report Number(s):
- UCRL-TR-201959; TRN: US200430%%1287
- Resource Relation:
- Other Information: PBD: 15 Jan 2004
- Country of Publication:
- United States
- Language:
- English
Similar Records
Countering Cyber Sabotage: Introducing Consequence-Driven Cyber-Informed Engineering (CCE)
Architecture and Methods for Substation SCADA Cybersecurity: Best Practices