Algorithms Seen as the Primary Defense Mechanism for Grid Cyber Attacks
A team of electrical engineers from Princeton University have developed a series of algorithms designed to prevent hackers from exploiting internet-connected devices to bring down the power grid. Specifically, the algorithms help mitigate coordinated attacks on Internet of things (IoT) devices that are meant to overload the power grid and take it offline.
If this sounds far-fetched, it’s not, and in fact several similar types of attacks have already happened. For example, in 2016 a botnet called “Mirai” coordinated an attack on over 500,000 IoT devices globally, for the specific purpose of overloading computer networks. While this particular attack did not impact the power grid, it did provide a blueprint of what is possible. According to the Princeton researchers, a botnet that is able to coordinate an attack on 600,000 high-wattage devices or appliances could instantaneously gain control of 3,000 MW of power, equivalent to the output of a large nuclear reactor, potentially triggering cascading blackouts.
How the Algorithms Work
The algorithms developed by the Princeton team aim to optimize responses to a demand spike. This includes facilitating responses such as power balancing at the plant level to prevent a line from getting overloaded, and increasing the speed of identifying and reacting to a spike.
Simply put, the algorithms help regain control of supply and demand when systems are compromised. They factor in variables such as transmission line capacity and power generation capacity to redirect power flows and adjust generator processes.
Although this technology adds some cost, the incremental improvement in grid fortification will likely justify the added expense. Of course, the proof will be in the proverbial pudding, so a key next step for the Princeton team is to prove the theory in the real world by partnering with a handful of utility companies to test the effectiveness of the algorithms. Until that happens, all of the benefits are theoretical, but hey, it’s a start!
