The US electrical infrastructure is extremely vulnerable to deliberate disruption (this isn't only a US problem). This is disturbing given that the vast majority of the US economy is tightly linked to the availability of electricity. The potential of wide-scale, sustained disruption is extremely likely given structural factors in the power industry and the increasing amount of intelligence data that indicates terrorists are targeting infrastructure. Massoud Amin of EPRI (Electric Power Research Institute), concludes in his presentation "Electricity Infrastructure Vulnerabilities" (PDF) and his article "Security Challenges for the Electricity Infrastructure" that these structural problems are:
- Underinvestment. The power industry ranks near the bottom of all US industries in R&D spending as a percentage of sales. Further, transmission and generation capacity has fallen well behind demand for over 30 years. The result is that the current power infrastructure is at a critical level of usage, which makes it easy to disrupt. The lack of R&D spending means that there aren't any solutions ready for immediate deployment.
- Over centralization. In the pursuit of efficiency, power infrastructure has become increasingly centralized. Additionally, the power system control system was built without security (a cost) in mind and commonly uses low cost public communications networks. The combination of these factors has resulted in a system that can be easily comandeered and driven to fail.
- Too much complexity. The push to maximize the utilization of existing infrastructure to meet rapidly growing demand and the increasing use of power system integration to share resources has radically increased the complexity of the power network. This complexity has made it extremely difficult to exercise rational control over the network when faced with disruption. Further, the complexity of the network itself can lead to large scale disruptions when faced with perturbations.
The result of these factors leaves us with a power system that is ripe for exploitation. The result of an attack of this type would be catastrophic to our economy. To mitigate these dangers, the power industry recommends the following changes should be made to our power grid:
- Wide area sensing and control. A new sensor infrastucture that greatly improves the quality and timeliness of data from the network must be built. Addtionally, a new command system should be installed that allow more robust vulnerability assessment and failure analysis for operators.
- Intelligent adaptive islanding. A major area of vulnerability for the current US system is its interdependence. A failure in one area can spill over into other areas due to tight coupling. A system that enables local problems to remain local is needed.
- Adaptive self healing. The most ambitious proposal is to build intelligence into the all aspects of the grid such that the system is able to heal itself when assaulted. Intelligence at the periphery mitigates the vulnerabilities of centralized control, reduces complexity (if designed right), and improves responsiveness.
The likelihood of these improvements to the power system being made are slim. They cost too much, particularly during a period of austerity. As terrorists increasingly move towards attacks on systems, the likely result will be:
- Substantial economic damage. The US power system will likely suffer a major sustained outage due to terrorist assult in the next ten years. The resulting economic damage will be measured in the tens of billions. Multiple attacks could push this economic damage to a point or two of the US GNP (enough to qualify as a severe recession).
- Increasing Self-reliance. US industries and individual home owners can mitigate losses by investing in local power infrastructure. Alternatives such as solar power become increasingly cost effective when potential disruptions to the grid are factored in.
- Cascading system failures. The central role of the US power system to our computerized economy cannot be overstated. A disruption of the power grid will cause major failures in coupled networks. Addtional attacks on these ancillary networks would in turn slow repair efforts on the power grid and extend times of general failure.