The paper addresses the question of how to deal with non-stationary power signals. The first part of the solution is at a fundamental level: the recognition that the thing being measured is known by some kind of label in a model. The label is attached to a some parameter in an equation, and is often identifiable by its position in the equation. The paper presents measurement as the act of solving the equation to find the value of the parameter. In other words, the equation is what metrologists term the measurand, and the measurement equipment must be designed around it. To measure a time-varying signal, in a world of digital measurements, one of the first questions that must be addressed is the relationship between the sampling window of the measurement system and the rate at which the signal is varying. A goodness of fit metric is identified. Several changing-frequency cases are examined.

10aAA14-0011 aKirkham, Harold1 aRiepnieks, Artis uhttps://certs.lbl.gov/publications/dealing-non-stationary-signals01428nas a2200169 4500008003900000020002200039245004300061210003900104260003400143300001000177520094800187653001301135653000901148653001001157100002001167856007101187 2013 d a978-1-4673-4894-200aThe digital revolution in measurements0 adigital revolution in measurements aWashington, DCbIEEEc02/2013 a1 - 63 aThis paper considers what it means to make a measurement, and the changes in measurement technology over the years. The impact of the latest changes, which have resulted in most electrical measurements being done digitally, is explored. It is argued that the process of measurement can be considered equivalent to one of data compression. The smart grid will certainly result in many more signals being made available, and therefore a great deal of data compression will be taking place. Measurements will be made in parts of the power system presently unmonitored, as well as parts that are already well covered by instrumentation. The smart grid engineer must decide what it means to have “useful” information. Unless care is taken, the signal processing may furnish information that is not useful, and may not even make sense. The paper concludes by examining the possibilities of data compression from multiple separate signals.

10aAA14-00110aAARD10aCERTS1 aKirkham, Harold uhttps://certs.lbl.gov/publications/digital-revolution-measurements