Peter Rabins: "Widespread Power Failures: Programmatic and Emergent Causality"

This week our featured book is The Why of Things: Causality in Science, Medicine, and Life, by Peter Rabins. This is the fifth article in a series of six articles by Peter Rabins.

And, don’t forget to enter our book giveaway for a chance to win a free copy of The Why of Things.

Widespread Power Failures: Programmatic and Emergent Causality
By Peter Rabins

In a Commentary in the11 July 2013 edition of the journal Nature, electrical engineer Massoud Amin outlines an approach to avoiding massive power failures. He advocates a “resilient” power system that is a “self-healing” power grid. His recommendations mirror the discussion of power failures in my recent book The Why of Things. He writes that the principles underlying his recommendations are the same for a range of complex systems, including fighter jets and telecommunication systems, in which sudden or emergent failure is the result of interactions among multiple units of a system.

Specifically, he recommends interventions at the local nodes of programmatic networks that make them secure and smart. This will require replacement of electromechanical switches with solid state circuits that can carry higher voltages than now possible. At the level of interconnections among systems, he recommends the installation of systems that would foster self-sufficiency at the local subsystem level. These are examples of programmatic network analysis operating at the individual and node level. At the next higher level of analysis, individual highly connected hubs (regional distribution systems, for example) will require solutions that vary by the needs and design of that hubs power resources. His example is that coastal systems have different design needs than inland systems. At a broader system level of analysis and intervention, Amin recommends flow-direction technologies that will even out differences between supply and demand and customer feedback inputs that will allow ongoing monitoring needs and improved coordination among users.

Amin’s commentary illustrates how a systems or network analysis (what I refer to as analysis at the programmatic level) can identify interventions at multiple levels. Emergent phenomena, such as widespread power failures may not have single predisposing or precipitating causal elements; rather it is interactions among elements of the system at multiple levels i.e. local, regional, and system-wide that best explain a range of network failures