Now is the time to be sure about the details. Virtually all the SmartGrid products aimed at the electric utilities measure electrical parameters somewhere in the grid, process the data and send it along to another system to be acted upon or stored. Common sense has it that all this new data is only as good as the metrology creating it but entrepreneurial SmartGrid companies often don’t give this basic truth the attention it deserves.
One of the best examples of this occurs in products designed to give electric utilities a deeper view into their medium voltage networks. These devices may not have access to the low voltage side of the network so preserving ability for ubiquitous deployment means they must be powered creatively via inductive energy harvesting, solar cells, batteries, RFID methods, etc. If operating power is in short supply these devices revert to low power modes with limited functionality or “sleeping” most of the time waiting for notable power system events. All is fine until the potential utility customer asks the vendor to add a staple “bread and butter” measurement like 10-minute average RMS current or voltage to the deliverables list and the marketing/sales guy eagerly agrees in order to close the deal.
Averaging anything over time implies taking an equally-spaced time series of measurements over the interval. Computing any RMS value itself implies doing yet another integration over time, generally with smaller intervals. If the device cannot be ON for the entire duration to do this properly, some compromise is required and there have been some remarkably bad implementation choices. For example, one vendor’s clever instrument would actually go to sleep for 9 of the 10 minutes, take 10 measurements in the last minute, average those and present the result as the “10 minute average.” The problem is that the first 9 minutes are vastly undersampled and the reported result contains no information about what actually went on during that time. The last 1 minute is highly oversampled for normal grid variations and the precious power invested in doing that metrology is largely wasted. What’s worse is that the utility didn’t ask nor did the vendor tell them exactly what the “10 minute RMS” number meant. Things got messy later when the vendors’ gear did not agree with other conventional instruments the utility had come to trust over the years.
If you need to get the right answer, details matter. DJA Engineering can help protect your SmartGrid project from unrealistic expectations, misunderstandings and maybe even legal bills later on.