Solar

#SWC2015 - Launching the Regional PV Simulation System in Daegu, Korea at the Solar World Congress

Our conference paper is entitled: Real-time Simulations of 15,000+ Distributed PV Arrays at Sub-Grid Level using the Regional PV Simulation System (RPSS)

[download it here]

It's an absolute pleasure to be writing to you from Daegu, Korea, where a few hundred solar energy scientists are gathered for the International Solar Energy Society's 2015 Solar World Congress.  I've been inspired by the work that I've seen here, and I'm quite proud to be a part of this community.  

At this conference, it is my great pleasure to officially launch my latest research outcome, which progresses my PhD thesis out of the laboratory, and into the real-world.  It's been a fun challenge to make this leap from science to technology, and I'm honestly quite enthused by the outcome.

From Science to Tech, How We Did It

RPSS stands for Regional PV Simulation System, which was developed by Nicholas Engerer during his PhD thesis, and is now undergoing further development to achieve operational deployment in Australia.

Over the past several months, I have been working with ANU Engineering student James Hansard to get the Regional PV Simulation System up and running, live, in the Canberra, Australia region.  This has been made possible through several mechanisms, the foremost of which was excellent ingenuity and determination from James (you can download his thesis here). 

 

There are 100+ sites reporting their power output into the RPSS as of November 2015

There are 100+ sites reporting their power output into the RPSS as of November 2015

Using an allocation on the National Computational Infrastructure, the RPSS is now running in near real-time via a series of three virtual machines.  The first machine is gathering PV power output data in near real-time (about 10-15 minute delay) from 100+ sites via PVOutput.org, a webpage where users publicly report their PV system's power output.  The second machine runs the RPSS algorithms and produce the output files/graphics.  The latter of these three VMs, is a web server, which hosts the latest simulations at http://rpss.info.  At this webpage, you can see the RPSS output for the day, mapped to ActewAGL's distribution network.  

RPSS version 2: 15,000+ PV Systems Simulated by Transformer Node

In the latest version of the RPSS, I am now simulating 15,000+ PV systems, grouped by transformer.  This is a significant advancement from the 12,000+ PV systems mapped by suburb in my PhD thesis, which progresses the RPSS towards industry relevant outcomes.  If we can quantify the contributions of PV to the grid at a high resolution, and forecast any significant changes ahead of time, then the intermittency of distributed PV can be mitigated.  Getting the RPSS running in near real-time is the first step in this process.

Quantifying PV Power Output by Distribution Node (Transformer)

When we break down our simulations by transformer, we start to observe very interesting behaviours from the collectives of distributed PV installed behind them that have not previously been observed in Australia.  Let me show you what I mean!  First, let's watch a simulation (RPSS output) from 5 March 2014:

Now, if you don't think that video is pretty cool, then you must be much less nerdy than I am.  I think its fantastic how, using only input PV data from a 100+ sites, we can observe cloud-shadow like features in the simulation! (That's the power of KPV!)

Now, let's look in detail, at the simulated power output grouped by transformer node (below)..

The three different coloured lines above, each correspond to the simulated power output at a given transformer (node).  These are shown on the right side of the simulation video above.  What is very interesting, is the observed differences in ramp event timing that occur between the three nodes during the strong negative collective ramp event that occurs around 11AM.  From this, we can see that the timing of these events are very important to predict precisely, in both space and time, as the strong ramping can occur at different times for each of them.  In the strong positive ramp that follows, there is a similar mismatch in the timing of the ramp peaks.  For a scientist, this is fascinating.  For a distribution company, its a clear example of the challenge of PV intermittency.  

And all of it motivates further development and deployment of this tool

If you found this post interesting, you can dig into the science a bit deeper by downloading our paper below, or by viewing the slides from the presentation.  There is also a tag cloud below, to help you navigate through other content in my webpage.

[You can download our ISES SWC 2015 paper here]

[Or you can download a PDF of the talk]

Or view it in the slideshow below:

SWC2015 Action:

Tag Cloud (Find More Stuff)


PhD Final Seminar - City-wide Simulations of Distributed Photovoltaic Power Production

I've just successfully completed my PhD defence seminar, and am excited to share with you the results of my thesis project.

It sure is tough to fit a description of everything that I've accomplished over the past four years into an hour long seminar.  Well, in fact, it's impossible!  So much goes into a PhD thesis (dissertation), and only the author themselves will truly understand all the work that went into it.

But I think I've managed to do a good job getting the best of the best information into a manageable format - and I'll be able to share a video of the final seminar with you in a few weeks.  But for now, I'd like to get the slides up on the web, and post some of the simulation videos.

In the shortest summary possible: I've created a simulation system for Canberra PV installations, which uses a subset of monitored PV systems (approximately 80-160 systems, depending on the site availability/date) to simulate 12,500+ PV arrays.  The installed PV arrays are based on data provided by ActewAGL (local utility/distributor) about the rated capacity and suburb-level location of each install through December 2012. 

The heart of the simulation system is an application of the KPV methodology, which uses the power output from one PV system to simulate the performance of another nearby PV system. I've published a paper on the topic in Solar Energy  for those of you who are interested in the more technical side of things.

For now, I'm going to stop writing, post the slides and simulation videos, and let them speak for themselves.  I hope you enjoy them:

[download presentation]

Presentation Slides:

Presentation Videos: Simulations

Video of a a clear sky day for all PV systems installed in Canberra. The images you see are the result of using ~70 PV systems to simulate the remaining 12,500. 

Video of a high variability day where broad convective clouds moved through Canberra. 

Video of a positive ramp event resulting from fog dissipation in Canberra. The images you see are the result of using ~80 PV systems to simulate the remaining 12,500. 

Video of a negative ramp event resulting from convection moving into Canberra. The images you see are the result of using ~140 PV systems to simulate the remaining 12,500. 


Solar 2014: Estimating Hourly Energy Generation of Distributed Photovoltaic Arrays

 

If you were tasked with estimating the energy generation from an entire city of PV systems - how would you do it?

A simulation probably jumps into your mind right away.  Scale up a model of PV system performance and that must get you close right?  Well that's a step in the right direction, and you could do that very accurately if you knew the amount of radiation arriving at the surface of all those PV arrays.

But that's a bit trickier than it sounds!  First, where are we going to get an estimate of the available solar radiation at a given location in the city?

The most common answer I get to this is: a pyranometer.  And that's a great start - you'd get a measurement of global horizontal irradiance (GHI) at a point location, which is very helpful.  But you're left with two major problems:

1. How representative is that pyranometer of the rest of the city's radiation resource?  Those clouds are tricky!

2. How do you estimate the amount of radiation arriving on all of those various tilted surfaces around the city?

So, OK, now we need multiple pyranometer sites around the city and at each site we need to tilt and orient them in various directions in order to get a representative sample.

Well, the bad news is that pyranometers cost a few thousand dollars each, need regular cleaning/calibration/maintenance and it's actually pretty difficult to find appropriate sites for them.  If you'd like to find out just how difficult it is to install scientific equipment on buildings - be my guest! (Hint: paperwork, approvals, PITAs galore!).

But I think I've got a better idea...

But I think I've got a better idea - what if we used the photovoltaic arrays that are already installed in a given region as our primary input to our city-scale modelling project?

They're pointed in many different directions, there are many of them already reporting data publicly in real-time, someone else has paid for the equipment AND they're representative of the systems we are trying to estimate in the first place. Sounds like a pretty sweet deal to me!

But, they are subject to shading, soiling and wiring inefficiencies, not to mention that they are not really the most scientific form of equipment.  Still - they are inherently a type of radiation sensor.  And we can probably deal with a lot of those things with some fancy machine-learning algorithms.

So, here is where I introduce our paper:

Estimating Hourly Energy Generation of Distributed Photovoltaic Arrays: a Comparison of Two Methods

J. Tan, N. A. Engerer and F. P. Mills

[download it]

 In it, we compare a two methods for estimating the energy generation of distributed PV arrays.  

The first uses pyranometers, radiation models and PV system modelling for the estimation.

Method 1: Based off of my Masters Thesis (see the 'Publications' page')

Method 1: Based off of my Masters Thesis (see the 'Publications' page')


The second uses a monitored PV system and my KPV methodology to make the estimation.  

Method 2: Based off of my KPV Methodology, Read the Solar Energy journal publication at the 'Publications' page

Method 2: Based off of my KPV Methodology, Read the Solar Energy journal publication at the 'Publications' page

 

I'll let you read the paper to get the details, as that's not the point of blogging (all the boring stuff is for the papers - Ok, I really do actually think that stuff is fun too, #supernerd).  But I will let you know that we've found a few interesting things:

1.  The pyranometer methods does tend to do a bit better (RMSE  15-20% versus 15-25%)

2. BUT when we start to leverage the prolific availability of the PV systems (there are many more of them out there!), we find the KPV method actually does best! (for distances less than 5km) 

3. We actually detected a calibration error in one of the pyranometers using the PV systems - so much for pyranometers being the pinnacle of scientific monitoring!

Overall, I find this result very encouraging.  If we can use PV systems as our primary input to our city-wide modelling idea, then we are one step closer to making the estimate we need.  And we can do it on the cheap - which is really good for solar! 

Now it's time to scale it up, test it on different time scales and handle all those pesky quality control issues.  But don't worry, you can count on me to bring you the results soon! 

Until then, enjoy my new webpage!

-Nick-

 

 

 

AMOS 2014 Talk: Categorising meteorological events as inputs to machine learning based solar forecasts

hobart_clouds

So I’m in Hobart, Tasmania.  A beautiful city I might add, with some picturesque upper level cirrus arranged in awesome gravity wave bands.  But I’m not here as a cloud tourist, but in fact to present some interesting research I am completing with an undergraduate student (Sonya Wellby) at the Australian Meteorological and Oceanographic Society's Annual Conference (AMOS 2014).


In the course of producing solar forecasts for our ARENA USASEC grant project (I really do need to actually write up a blog post on that at some point), we’ve discovered something interesting about the inclusion of weather data in our machine learning algorithms:  it doesn’t seem to help at all.


This is quite strange when you consider that the entire challenge of solar energy prediction is related to the clouds and those are driven by recognizable weather phenomena.  So including information like the temperature, wind speed/direction, surface pressure, etc should in theory, help the forecast improve.


But this is not what we’ve found.  In our single-site, Support Vector Machine (SVM with linear loss function) model estimates for 10 minute interval data/forecasts, we see an increase in Mean Bias and Root Mean Squared Errors and a decrease in the correlation between predictions/observations:


For hourly data/forecast intervals, we see no improvement in the forecasts with the inclusion of weather data.

10_min_forecasts
10 minute interval forecasts/data (1). 60 minute interval forecasts/data at (2).    Red line is persistence, black line SVM without weather data, green line SVM with weather data.  All forecasts were produced using data from one PV site.

10 minute interval forecasts/data (1). 60 minute interval forecasts/data at (2).  

Red line is persistence, black line SVM without weather data, green line SVM with weather data.

All forecasts were produced using data from one PV site.

So what is going on here?

 

My hypothesis is that the weather data is full of too many small fluctuations and seemingly random signals for the machine-learning algorithm to see the “Big Picture" (thanks to Sonya for that phrase).  This is to say, it doesn’t recognize the overall synoptic or mesoscale event, which us meteorologists are trained to interpret.  It has no physical understanding of the data it is seeing – just a lack of direct relationships and therefor a diminishing weight to that data. 

So what can we do?

Well, our approach is to remove the individual feature vectors of weather data and replace them with a feature vector that signals weather or not a "significant" weather event is going to occur

 

“Significant” is defined here as the types of weather events that result in large scale ramp events for collectives of PV systems.  Currently, we are using data from 30 PV systems in Canberra to ID large ramp events.

 

Major events so far are fog clearing, morning cloud dissipation after easterly surges, the departure of a low pressure system/cold front and thunderstorm events.  We’re identifying more, but at this stage, I’ll refer you to our talk – which I’m posting here with audio (how cool are we?).

Check out our #AMOS2014 talk here, complete with Audio Transcript

[Download Talk]

See the slides:

Listen to the audio:

/*