Its More Than a Big Grey Box – The Basics of Data

In the first chapter of the series about DAS and SCADA, we learned why these systems
are not only necessary, but required for utility scale PV generation systems. Its all
about knowledge, the power to know what a site is or is not doing. In this next article,
we dig a little deeper into what information we are seeking and why it is important. So
here goes…
As in many things of life in the Information Age, data, and access to collect it and then
investigate it allows us to make good decisions. Energy generation is no different as
having either small packets of data or ream of information, provide the owner and
operator of the site increased insight into the current and historical operation of the
site. The choice of data collected is critical to the successful operation of the systems,
and a large site requires much more data acquisition than a smaller commercial
project. But the need remains the same, only the amount of actual data collected
and what is done with the data.
Here at REIG, data is more than ones and zeros, or bits and bytes of code. Data tells a
story; it makes us aware of challenges and opens our eyes to what has happened or
is happening within the system, good, bad, or indifferent. And with the right analytical
tools, it also allows forecasts to be made, plans to be developed and actions to be
undertaken before a failure has occurred. Information gives insight, understanding
and vision into what was before a mystery of failure.
But what data needs to be collected? The honest answer…everything that can be
collected…it’s all important. The real-world answer…just the basic information that
allows the site to be monitored, for all data and particularly good data is not cheap to
collect and impacts the bottom line and ROI of the generation system. So, guess
which way most systems are built?
Is there a middle ground between the two extremes? I think so and as I see it, data
needs to be acquired holistically. In other words, it needs to be collected as a
package, and the critical data is viewed more often than the operational data that
does not impact the operation, but that could affect the bottom line. Until it does
make a difference and then the questions are asked…why weren’t those points being
monitored to begin with?

An example…inverters. Its absolutely necessary to monitor key operation parameters
such as voltages, currents, power generation and overall energy delivery. But why
collect data on cooling fan speeds and heat sink temperatures? They are not
important enough but they sure are when an large central inverter shuts down
because one of the IGBT’s fails and the entire inverter ceases operation. How cost
effective was that?
But the human operator cannot detect trends in the minutia of thousands or tens of
thousands of data points. More and more DAS and SCADA software platforms are
utilizing AI to look for those trends and anomalies that could signal a possible
hardware failure and then bring that trend to the attention of the operator. With
previously missed data in hand, then good decisions can be made to be proactive
and make a service call to the site well before the inverter ceases operation.
Thats it for this post…I hope the information has been helpful and causes you to think
of your own systems, and what data should be collected.
Next time, we start looking at how this task is accomplished through hardware and
communications. This is where he rubber meets the road.