Mark Dunger: The Meter Lives Too Long - Innovate

This posted to smartgrid@ostp.gov by Mark Dunger, ESCO Technologies.

Response to Q1 -- The smart meter should NOT serve as the primary gateway. Technological advancements in electronic products within the home and the evolution of communications protocols (which might also address evolving security concerns) embedded within such products will likely render as obsolete any gateway embedded within a meter within a matter of years.

Considering that there is an expectation of meter longevity (perhaps 10+ years) for meters placed in the field, and the cost that would be required to upgrade / update hardware within a meter once installed in the field, the inclusion of a gateway within the meter does not (intuitively) seem to be an economical approach to achieving state of the art Smart Grid communications over the long run.

Response to Q2-- Yes, a data gateway other than the smart meter should be used for all signals that require two-way communications. Again, it should be expected that a gateway within a meter will become outdated prior to the end of life expectancy of the meter asset.

Response to Q3 -- Bandwidth requirements will vary depending upon the applications that the utility hopes to employ via that smart grid, and so true needs / infrastructure cost over the planning term should be taken into account in a decision as to the use of alternative architectures.

AMI systems that utilize private networks, such as powerline or radio frequency technologies, may adequately serve the purpose of providing a communications architecture that links an in-home gateway to the utility head-end while providing real-time or near real-time data.

Public networks such as cellular or satellite might also be used. Again, an economic analysis of the communication path needs to be considered in reaching a formal conclusion as to a proper architecture. Intuitively, consumers that incur the largest energy load are also likely to have a broadband internet connection to their premise, and so the use of ISPs as a medium for interaction with consumer gateways makes some sense. Use of an AMI system as a backup would have virtues in this scenario, and so the ability to engage in IP connectivity with the primary AMI system has merit. Real-time communications with low-income consumers is probably not a pressing matter, as they are unlikely to be introducing significant loads to the grid. Therefore, some latency in the system design / architecture may be acceptable. Therefore, an alternative architecture that consists of high bandwidth internet coupled with a lower bandwidth AMI network is likely the optimal solution.

Mark Dunger