I’m excited that the solar power field is moving towards mass scale adoption. The cost for solar power is falling rapidly, and the installation rate is rising quickly. The quicker this happens the more likely it is to change the energy mix enough to change the path of climate change. However, at the Intersolar 2015 conference last week I learned of something which could limit solar power adoption – the administrative overhead in monitoring performance of solar arrays, and detecting problems to fix.
One aspect of the issue is that investors and business managers need proof their solar array is performing as expected, and is producing the economic/business benefit it’s supposed to produce. A business installing solar power on the roof or over the parking lot needs to know the benefit being produced – the amount of electricity generated, the cost savings, and whether it’s really paying for itself or whether it’s a boondoggle.
The facilities manager needs to know additional information, such as if one/some of the panels has died, or whether it’s time to clean the solar array.
In other words you don’t just put up the solar array and watch the dollars roll in. You have to track its performance, perform maintenance, check things against business projections, and weigh whether it’s worth sinking more capital into solar panels or other business pursuits.
What’s needed is a monitoring system to measure enough aspects of the system for proper management, and reporting results to business management. It must provide performance data (electricity generated) and also warn whether the system isn’t performing up to snuff. For bonus points the solar power monitoring system might be part of a larger system, one whose scope includes managing all the power on the site including grid energy storage systems.
With the help of discussion in a pair of sessions at Intersolar, let’s take a deeper look at what’s involved.
At the top I said the administrative overhead to monitor and report on performance could limit adoption of solar power. The problem statement in the meetings I attended at Intersolar was that existing solar power management systems don’t have commonality in reporting. One system will report one way, another will report another way, and they won’t agree on a common definition to the values being reported. A given company might own several solar power systems, each installed at different times with different components and different management systems. The different reports mean extra administrator time spent to interpolate the reports to a common set of metrics, so they can present unified reporting to upper management.
Additionally, each site might have different conditions in its contract and therefore require unique reports for that site.
The first bit is data collection, which means installing “data logger” gizmos into and around the solar array. Data loggers of all kinds exist, and generally speaking they’ll have some kind of sensor(s) to take measurements at regular interval, sending a data stream over (often) a serial cable. The monitoring system will require a subsystem to receive these data streams and record them in a database.
- The useful lifetime of a solar array is 20-30+ years
- Facilities managers will want to track performance degradation over the entire lifetime of the system
Therefore the data storage system will need to hold historical data over that period of time. Someone in 2040 will be tasked by their management to evaluate if/when it’s time to replace that creaking old solar array on the roof with a new system. They’ll need the historical data with which to decide.
What kind of data is required?
- Volts / Amps (hence Watts) produced at each panel, and for the entire array
- Temperature of each panel
- Operational status
- Ambient light and temperature conditions, as well as weather (cloudiness, rain, snow, etc)
- Temperature (at least) of the inverter(s) and other electrical system equipment
- If there’s a grid energy storage system, its state of charge and other performance metrics – e.g. if the BMS provides information about cell health, that will need to be recorded
This data must be recorded at fairly granular intervals at least for short term reporting. For long term historical data, I’d imagine, the system doesn’t have to keep the five-minute-interval measurements for 30+ years.
With the data collection in hand, there’s a few things one can do with it:-
- Put a pretty display in the building lobby showing the solar power generated on the roof
- Detect problems with the solar array, and automatically dispatch technicians to fix the problem, giving them initial data about the problem
- Generate performance reports for management
- Detect the effects of long term aging
Earlier I said participants in the meetings at Intersolar complained of incompatible reporting from one system to another. They also discussed reporting and definition standards being developed by the industry, through the SunSpec Alliance.
That’s an industry standards consortium working to develop best practices, software tools and specifications meant to accelerate adoption of solar power. The SunSpec suite of standards includes several data exchange API’s for use between different sorts of organizations.
Given that the sessions I attended were sponsored by SunSpec, the answer to all the problems above was to adopt the SunSpec standards and practices. I don’t know enough about the specifications to say “yes they’re the cats pajamas” but the SunSpec website says all the right things.
- Tesla plans to keep more stores open, re-increase prices, backtracking on previous plan - March 11, 2019
- Imperial Beach CA backs off from “managed retreat” from the ocean that was their plan for sea level rise response - March 7, 2019
- Tesla closes stores to raise cash to pay off convertible bonds? - March 5, 2019
- Tesla does end-run around states prohibiting direct car sales, and starts shipping $35k Model 3 - February 28, 2019
- Trump Administration cuts off “negotiations” with California over clean air rules - February 21, 2019
- Amazon investing in Rivian, GM possibly to follow - February 15, 2019
- Pres. Trump selling Paradise to coal baron who donated money to his campaign - February 13, 2019
- GM, Amazon, looking to invest in Rivian, no sign of GM/Tesla EV truck joint venture - February 12, 2019
- Tesla’s goals may require rumored joint venture with GM on electric pickup truck - February 12, 2019
- Green New Deal outline proposed in US Congress - February 7, 2019