Designing a theoretical microgrid

Pinechen

I don't know whether this is a proper topic here...but I still want to ask you guys.Hope anyone know something about this and help me. I am working on designing a theoretical microgrid for a school building, but before I can even select theoretical components, I need to select what the structure of my microgrid will be.

Here's what I know:

•     That MPPT charge controllers work by finding the maximum power points on your solar panels to ensure you get the most bang from your solar panel (and charge your battery safely and quickly)
•     MPPT CC's can work with hybrid inverters to invert however much you need to power your AC loads while the rest gets sent to batteries (refer to the diagram at the bottom)
•     DC coupled systems are much more efficient at charging batteries than AC coupled systems
•     About 160 [kW] of PV panels would be used
•     Unlike hybrid inverters, grid-tie inverters sync with the grid frequency and therefore scale infinitely
•     A purely grid-tied system with no batteries is the most economical, but my project involves looking into a system with batteries anyway as a "what-if" scenario
•     A DC busbar would be used to connect the batteries, hybrid inverter, and output of the MPPT CC's for the model. The higher the voltage, the less losses in the lines but there are safety considerations and component selection limitations to consider as well.

Here's what I don't know (these are my questions):

•     How to make a system that uses MPPT CC's, hybrid inverters, and grid-tie inverters (to "sell back" excess generation when the batteries are full). A picture of what I was thinking is shown below. Otherwise, I feel as if I would need to have some of my PV panels only connected to a MPPT and a grid-tie inverter. Then the rest connected to the MPPT CC, battery bank, hybrid inverter, and loads. The problem with that is that there will be times when the batteries are full so the part of the PV panels that is connected to the batteries and hybrid inverter would be wasted.
•     How AC combiner boxes work. Are they manufacturer specific where they only work with a specified inverter model (and number of inverters) or are they more of a stand-alone product like an inverter can be?

Current idea:

So what I was thinking was having a switch that just switches the PV panels to the grid-tie inverter when the batteries are full. As you can see, DC-coupled system. The problem with this idea is then the school building will start draining the batteries unnecessarily (batteries cost money and using them when you don't need to lowers their lifespan).

Another possibility I was thinking was just to leave however much the building will ever need at a given time always connected to the hybrid inverter and then just switching the rest of the PV panels either to the hybrid inverter or grid-tie inverter based on need (batteries need charged or batteries are full and I need to sell the excess).

I'm not an EE, so I wanted to see what your thoughts on this was (and info on AC combiner boxes).

mcgivor

Welcome to the forum.

Wanting the best of both worlds is not always as easy at it may appear, generally grid tied (GT) inverters use higher voltage strings, MPPT controllers generally needs lower, the closest you could get in a setup where switching between the two would be to use  600VDC charge controllers,  but the limitations on output current could mean multiple controllers, limitations of usable output would be determined by the battery capacity. So how about dividing the array, one section feeling a MPPT controller (s) at a lower voltage to satisfy the battery demands including  supporting loads during excess production, the ballance using GT inverters feeding the grid, obviously during daylight hours. What's fed in during the day could be withdrawn at night from the grid using an inverter which allows grid support, this is a theory I've been thinking about when i connect to the grid, this way there is no switching.  Below is a pdf which you may find informative.

Dave Angelini

It has already been done for several years now. Outback and Schneider both have scalable micro-grids up to several hundred KW and beyond soon. The XW Pro (in test now) will go way up there and be easy to configure, as will the new gear from Outback. Look at their websites and even on U-tube. High voltage DC solar is the only way to do large installations in my opinion.

Estragon

It seems to me a better definition of loads and objectives is needed. 160kw of pv is mentioned, but why not 50kw or 300kw?. Is the battery meant to back up all loads, critical only, how big, for how long, and how often?

Also, it may be worth confirming grid connected is even possible. Depending on location and the attitude of local utility, there may be physical or regulatory issues preventing this or limiting potential size. If more than an engineering exercise (ie a cost/benefit component), the economic terms offered by the utility should also be determined.

westbranch

Nor has the poster told us the total max load to be supplied,  if GT is  allowed/legal or if there is a benefit to pushing power into the grid, feed-in tariffs, etc?

Raj174

Attached is a good PDF on micro grids with diagrams of DC and AC coupled designs. Also provides example real world installations of various sizes.

clockmanfran

Raj174 said:

Attached is a good PDF on micro grids with diagrams of DC and AC coupled designs. Also provides example real world installations of various sizes.

Interesting basic document.

Yes those systems are possible if you throw vasts amount of cash at them, but maintenance issues are always a hassle a few years down the road.

AC coupling seems the way forward as PV and GTI's becomes cost effective solutions.

Storage,...... Simple lead acid are still the most proven and cost effective.

Generator...... Most cultures on this planet have folk that can maintain simple diesel generators.  

MarkC

Inclusion of EV battery storage to stabilize micro-grids (eventually national grids?) and make them more cost effective is an area of substantial research, but seems to be mainly in Europe.  Anyone have good references for US based projects/programs?

Interesting that Princeton Power Systems - big in the micro-grids - now is offering a V2X energy controller (they call it an EV charger - need to come up with a new class of "chargers"?).  It's CHAdeMO compliant protocols only, but I'm sure CCS will not be far behind (Bolt!!). Appears many paradigms are about to be broken?  We'll see.

BTW - thanks to Raj174 for the interesting .pdf

Dave Angelini

Outback just came out with their first HV Mppt which should help micro grids. Should be in the store here in the next few weeks.
Their web page on microgrids.
http://www.outbackpower.com/global/micro-grid

Schneider has been very active worldwide in microgrids for many years.
https://www.schneider-electric.com/b2b/en/solutions/microgrids/