Alaska Village Daily Load Profile

Replies to This Discussion

Permalink Reply by Alan Fetters on March 10, 2010 at 3:29pm

David

We have a limited amount of load profiles for some of the communities we have assisted. Keep in mind the information is centric around the period we worked in the community. These do not include AVTEC, AP&T or North Slope. Contact me with your interests I will see what we have available.

I am actively working on a project to add historical trending to the Supervisory Control And Data Acquisition (SCADA) system we have at some of the powerhouses. We have a goal to make this information readily available to the general public. Keep in mind the project does have severe funding limitations. Everyone wants the information few want to fund it long term. Our conditions for success are challenging but very achievable.

We use LM 5400 Enetics power analyzers. http://www.enetics.com/prod-LM5400.html

I am toying with the idea to start a working group for rural power SCADA.

Alan

Permalink Reply by David Karabelnikoff on March 10, 2010 at 4:14pm

I'm sure that I am no different the the rest in that we all would prefer actual load profile data rather than generalized approximation. SCADA technology offers functionality which seems indispensable with high levels of intermittent RES penetration.

But how to make if funded, perhaps starting a group would be a way to coordinate such an effort. I'll try to find out more about the funding mechanisms they are using here with the smart grids. my professor in smart girds: João Peças Lopes, University of Porto, Portugal / INESC is demonstrating 3000 smart meters on demo microgrid basis (I'll try to find some links on the project in english)... I have a portegues IP they give me porteguese hits on google ... I have some presentation slides on the project if your interested.

The demo projects in the EU offer models on how to innovate, develop human capacity and commercialize technologies which I am looking forward to learning more about. If there is anything I should inquire about please feel free to advise.

Attachments:

• Smart Grids and Mobility Lopes.pdf, 1.8 MB

Permalink Reply by Don Eller on March 10, 2010 at 5:42pm

Hi David,

If the addage, "you cannot manage what you do not measure" holds true, we should worry about rural electrical production because no ones at the helm.

Attached is a screen shot of the load today in Tanana. More information is available if desired. The Load in a village varies dramatically based on season, day of the week and whether a major user like the school has something going on.

Attachments:

• TPC Load.pdf, 32 KB

Permalink Reply by David Karabelnikoff on March 12, 2010 at 3:40am

I know this is a technical discussion but for give me while I get a little sentential - Beautiful!
(These data points get my imagination fired.)

I am curious to know more about the fluctuations in load curves both daily, seasonally and annually how accurate are forecasts based upon historical trends ? Has this been studied in micro grids / stand alone power systems ? I am working with the folks at MIT-Portugal sustainable energy systems project, in the area of smart grids they are developing dynamic models in forecasting wind production for 72 hours. I'm curious if the same type of approach could be used on the demand side, with sufficient information such as the one you posted I imagine that this could be done perhaps by an engineering grad student, this is a bigger project than I am able to include in my thesis but a needed component of research in developing and adapting smart grid technology for the dynamics in rural Alaska SAPS.

electrical production in the 21st century: this isn't your grandfathers electrical utility.

Permalink Reply by Don Eller on March 12, 2010 at 9:40am

Hi David,

I will try and make an attempt to answer some of your questions:

1) fluctuations in load curves both daily, seasonally and annually how accurate are forecasts based upon historical trends ? In general village load curves are very consistent and easy to forecast based on historical data. However because the user group is so small the actions of each user can significantly impact the daily load curve. If a user has a welding project going on, a potlatch or temperature drops significantly the typical forcast load curve for that day would be significantly off. Attached is another 24hour load curve for comparision with previous curve supplied, the curves a very similar.

2) Has this been studied in micro grids / stand alone power systems ? Not to my knowledge. The varibles which drive load which are human related are hard to forecast. TPC's approach has been to focus on efficiency through demand dispatch (matching the most efficient supply for the given load) because it is impossible to forecast and control the demand side.

3) Smart grids are almost a necessity in rural power production if IPPs are to contribute any significant energy on the supply side.

Attachments:

• TPCdemand31210.pdf, 32 KB

Permalink Reply by Rorik Peterson on March 15, 2010 at 1:29pm

What is a ballpark price for the LM5400?

Permalink Reply by Alan Fetters on March 15, 2010 at 1:59pm

The Enetics analyzer is about $2500 with test leads but no CT's. The flexible CT's are about $250 each and you need 4 ea for 3 phase wye. Email William Bush [wbush@enetics.com] for a quote. You can connect to a analog phone line and download the data. Use a serial to Ethernet convertor and they are network capable. There are more sophisticated meters like Dranetz BMI but these are rugged, reliable and relatively easy to hook up. Three green lights and you have it right. We have about a dozen of these. I have attached a few sample kW reports from a couple of communities. We use SATEC PM172 power meters integrated into the switchgear for the permanent station service and bus feeder meters.

SATEC PM172 http://www.satec-global.com/eng/products.aspx?product=47

Alan

Attachments:

• Modified Graphs.pdf, 1008 KB
• ChitinaData1.pdf, 231 KB
• FYU 11-24 to 12-14-09.pdf, 140 KB

Permalink Reply by David Karabelnikoff on March 15, 2010 at 2:09pm

Hey Alan,

I am considering including a load forecasting algorithm/ component into the tool/program we are building here at MIT-Portugal. If this gets a green light can I get some data sets from you to assist in this endeavor? The folks whom are putting this together are PhD's in Electrical Engineering - I presume they will want as much into as they can get.

I've also contacted the folks whom put together the AK Village Load Calculator and hopefully will get in touch with TDX and AVEC if you have any inroads with them I sure could use some help. Being half way around the world doesn't help for data set requests.

There also may be a wind forecasting module as well which would make for having data with wind production ideal.

I'm on skype: AlaskanWarrior

Permalink Reply by Ian Graham on March 15, 2010 at 3:51pm

Alan and others,

You might want to check out the Shark power meters, they pretty inexspensive rugged little buggers. The communication protocol is real simple to integrate and can help from a cost standpoint if you are doing a ton of load monitoring.

http://www.electroind.com/dm.html

What about the creation of a centra data base for village load data? My buddies at Fat Spaniel (used to be the power monitoring solution for skystream and small solar), essentially they have moved into facilitiating database managment for anyone to backup, organize, and use as a central server. Pretty cost effective if you can get enough users. You just have your SCADA system send the data to them and they do all the backend work. Essential for any major smart grid activity when you are able to query different data sets with mulitple inputs. Also having it web based means its availiable to everyone (security concerns aside).

Rorik Peterson said:

What is a ballpark price for the LM5400?

Permalink Reply by Ian Graham on March 15, 2010 at 3:55pm

I am assuming that you have messed around with the Homer simulation models? Are you expanding upon that type of platform that includes several modular components to the RES? It seems to be fairly accurate and is incredibly usefull in early stages. But doesnt account for power quality, voltage or frequency imbalance issues. Shear energy to energy simulator.

David Karabelnikoff said:

Hey Alan,

I am considering including a load forecasting algorithm/ component into the tool/program we are building here at MIT-Portugal. If this gets a green light can I get some data sets from you to assist in this endeavor? The folks whom are putting this together are PhD's in Electrical Engineering - I presume they will want as much into as they can get.

I've also contacted the folks whom put together the AK Village Load Calculator and hopefully will get in touch with TDX and AVEC if you have any inroads with them I sure could use some help. Being half way around the world doesn't help for data set requests.

There also may be a wind forecasting module as well which would make for having data with wind production ideal.

I'm on skype: AlaskanWarrior

Permalink Reply by David Karabelnikoff on March 16, 2010 at 1:35am

Sure I've used homer, I think that the models and computational abilities have gone way beyond it - there are about 50 tools for the integration of renewable energy/dg systems all the way from national planning to micro grids I've attached a paper on the review of integration tools.

The ability to do wind/ res production forecasting, load shifting and load deferment are all things that homer is not able to model while the technology is able to deliver.

I'm hoping to produce a tool that is able to be use in renewable energy islands/ micro grids which is an evolutionary step forward in accelerating the evaluation and implementation of renewable energy systems.

I spoke with the folks at NREL and mentioned the abilities which would be the next evolutionary step in integration tools and wind and load precision models appear to be what no tools include in the integration algorithm.

Attachments:

• A review of computer tools Connolly.pdf, 432 KB

Permalink Reply by Walter H Rose on July 13, 2010 at 9:25am

Here is monthly data for a village of 150. I'll see if I can dig up more detailed daily or weekly data. I do not guarantee the accuracy of this data or its suitability for any purpose.

Attachments:

• VillageData.xlsx, 11 KB