Digitization of the power grid is the current buzz. Modern automation and protection systems are predominantly digital and many benefits have been derived from this evolution. But the underlying systems are fundamentally analog and the digital model of that system is an approximation that always has at least a few rough edges. There is a tendency to ignore this and assume that the digital representation is perfect and infallible.
This presentation considers some of the assumptions and shortcomings of digital representations and the ways that systems handle data. A clear understanding of how these processes are performed allow designers to avoid errors and allow users to better interpret information.
The acquisition of both binary (e.g. breaker position) and analog (e.g. feeder current) are considered and the implications of different measurement techniques are considered. Resolution, accuracy, monotonicity and time domain behavior of analog signal measurements are reviewed. The impact of different applications’ requirements for time synchronization and time stamping are also considered. Some applications, especially in the power system protection domain, have very tight time constraints for the acquisition and analysis of signals. There is often a trade-off between speed and accuracy. Signal measurement is often performed in electrically harsh environments and design techniques required to survive that environment can impose limits on speed and accuracy and impact hardware cost.
In conjunction with this, the characteristics of different data messaging systems and protocols are also discussed. Some data cannot be losslessly translated between different representations or different protocols.