There are different technologies available, not all suitable for all scenarios, which may require differing modells. The technologies include Pumped Storage, Chemical Battery (Li-Ion, Lead Acid, Fe, NiCD etc) UPS', Flywheel UPS, Flow Batteries, Compressed Air Energy Storage (CAES) and Liquid Air Energy Storage (LAES) I think one may have to seperate the generation (PV, Wind etc), storage- (battery, flywheel, dam) and the converter elements (synch mach, inverters) and model independently and then optimize the models by itteration..

Sizing should not be a problem, as one can make reasonable predictions about the expected renewable generation and size the plant and the storage accordingly. Take into account that, in South Africa, our best solar and wind generation sites do not necessarily coincide with adequate transmission capability. The latter may yield even the best model unfeasable.

To model this should not be a problem in any mathematical package such as Matlab, Scilab, Python etc.., once the properties of the individual elements are known.

Commercial packages are available with this functionality built in, e.g. pvsys: http://www.pvsyst.com/en/software/functionalities , for solar PV with storage.

Hello Gerhard, you can evaluate some papers as "A Simple Sizing Algorithm for Stand-Alone PV Wind Battery Hybrid microgrids" or "Size optimization of a PV-wind hybrid energy conversion system". These are research paper you can find on the web. I'm study a similiar problem as well, actually I investigate the customer loads and a PV system only; adding a WindGenerator decreases the pragmatically of the study if you consider grid-connected customer. In the case of islanded mode, both PV and WG are welcome and realistic.

If you want more papers, I can send you a folder were I save many pdf files on this issue; let me know.

Alessandro

[email protected]

Hello Gerhard,

from the technical point of view, batteries are very important and mostly use to storage of surplus energy or provides energy when sources don't work. But from the economical point of view it is the weakest part of system due to their life time. Ofcourse we can use supercapacitor - but it will be definitely not economical, not yet. In my opinon It will be better to use a energy storage as a helpful to control of quality energy. Some kind of buffer for safe work of inverters.

Best regards

Marek

Thanks for the input Alessandro. A link to the papers will be useful.

I agree with you Marek, but traditionally people have used Lead-Acid batteries. Li-ion has made huge advances and Metal-Halide batteries (e.g. Na-S) are being used at utility-scale - both with better performance than lead-acid. Li-ion has a 7-10 year life and can be used with greater depth-of-discharge. This means they end up being not as expensive as one would imagine. There are very large Li-ion renewable implementations being put up in China and the USA. They have become economical, especially at Asian prices for batteries.

Hello Gerhard, batteries (Lead-Acid or other chemical storage) is one option. Other more environmentally friendly options can be phase change materials (or water) to store solar energy as latent heat. Mechanical energy storage systems are also under development for wind turbines.

In any case, you need to take into consideration the uncertainties of supply and demand, together with the system availability requirements. You may need to use a convensional baseload source to meet these requirements without wasting too much on energy storage.

Gerhard, when considering large-scale energy storage of intermittent renewable energy over extended periods of time (in particular to balance seasonal production variations), as an addition to batteries, storage in the form of chemical energy carriers as a complementary solution should not be forgotten. Archetypes of such energy carriers are e.g. hydrogen produced by electrolysis or synthetic natural gas made from CO2 and H2. But there are of course a number of further options available since also liquid fuels could be produced when "green" H2 is available. You therefore might be interested in the attached paper.

Article Fuel cell electric vehicles and hydrogen infrastructure: Status 2012

And coming back to your original question related to battery real-life studies, please check the related link describing an ABB-GM collaboration on Chevy Volt battery re-use for stationary energy storage applications (R&D status).

http://www.greencarcongress.com/2012/11/gm-abb-20121115.html

There are different technologies available, not all suitable for all scenarios, which may require differing modells. The technologies include Pumped Storage, Chemical Battery (Li-Ion, Lead Acid, Fe, NiCD etc) UPS', Flywheel UPS, Flow Batteries, Compressed Air Energy Storage (CAES) and Liquid Air Energy Storage (LAES) I think one may have to seperate the generation (PV, Wind etc), storage- (battery, flywheel, dam) and the converter elements (synch mach, inverters) and model independently and then optimize the models by itteration..

Sizing should not be a problem, as one can make reasonable predictions about the expected renewable generation and size the plant and the storage accordingly. Take into account that, in South Africa, our best solar and wind generation sites do not necessarily coincide with adequate transmission capability. The latter may yield even the best model unfeasable.

To model this should not be a problem in any mathematical package such as Matlab, Scilab, Python etc.., once the properties of the individual elements are known.

Commercial packages are available with this functionality built in, e.g. pvsys: http://www.pvsyst.com/en/software/functionalities , for solar PV with storage.

Concentrated solar thermal power plants (solar energy, a renewable technology) can include a large thermal energy storage system that provides 'dispatchability' to the plant. Biomass-based power plants, or hybrid solar-biomass power plants, are other renewable energy options that can provide stability and controlability to the grid.

Hi everyone,

I really agree with David Johnson : not every technology fits with every need.

Energy storage systems are usually classified by the energy they can store (large for hydraulics and CAES, small for flywheels and SMES), and the ratio power/energy they can deliver (large for flywheels, sizable for hydraulics, small for Lead batteries, around 1kW/kWh for Li-ion).

But to integrate renewables on a grid, the use of storage systems is not always necessary: it depends on the size of your network and the amount of renewable "fatal" energy, ie not predictable (excluding for example dam hydraulic, biogas, ...). In Europe, it is generally admitted that 30% of the power needs can be fed by fatal sources without any problem on the grid, but many people assume this could easily be more.

You can find almost all the information you need about energy storage on the ESA website, the Energy storage association : http://www.electricitystorage.org/

Enjoy your research!

Matthias.

Hi Gerhard. I thought I heard of a system in Fairbanks;http://www.greenchipstocks.com/articles/utility-scale-power-storage-innovations/804 . This is a little old now, hopefully helpful.

Thank you all for your very meaningful inputs! This is a rapidly developing field with many technology options. It is clear that there are regional and local issues that need to be considered when deciding on storage type and size - certainly no "universal answer".

I think that in the short term, the South African utility Eskom, will still be relying mostly on Hydro, but in a dry country this is limited to specific locations and requires moving vast amounts of energy long distances. In the long term I think that Thermal Batteries (or phase change storage if CSP is the electricity source) may make better sense for large installations. For small, non-grid-connected installations perhaps Li-ion batteries (from Asia, with lower pricing).

Thanks again to you all !!

Hi Gerhard,

you can find a very useful database from DOE on grid-connected energy storage projects around the world. Although it's more focused on US, you can find other projects and see some commercial applications of energy storage combined with renewables.

Antonis

sorry, here's the link:

http://www.energystorageexchange.org/

Antonis

Within an economically developing country, more warm water is needed by many households. To avoid peak loads on the grid, home installed electric boilers can be heated first with PV if no SWH is available. This requires the development of small controllers and proper thermal insulation of the same. Since water has a very high heatstorage capacity and can be heated up to 100 degrees this results in massive storage capacity.

An other development is Hydrogen production for transport vehicles. It is wise however to wait untill the storage technology for small vehicles is commercially developed. In both cases (water heating and Hydrogen) there is hardly eny energy loss in the reconversion, while reconversion from almost all battery types is rather large.

There is no reason to concentrate on batteries for storage. Commercial buildings have tremendous potential storage because of thermal inertia. Projects are underway at the university of Florida to provide ancillary service automatically from HVAC systems, without impacting climate quality, http://humdoi.mae.ufl.edu/~prabirbarooah/Research/PBResearch_buildings.html

http://plaza.ufl.edu/hehao/papers/ACC13_building.pdf

Low frequency variability can be addressed through flexible demand - ALCOA, for example, is a great source of flexibility. Here are two tutorials,

Cambridge UK, 2013 http://www.slideshare.net/spmeyn/control-of-the-grid-in-2020-and-how-economics-can-help-us-21616248

Montreal Canada, 2012 http://www.slideshare.net/spmeyn/2012-tutorial-markets-for-differentiated-electric-power-products

I would rather say that grid connected renewables normally do NOT require storage, but a fast adjustment of the main (traditional) power suppliers to the grid.

A problem in SA is that ever more people install airconditioners which require massive amounts of power per household. Obviously when the sun shines more cooling is needed and more power is required. In such a scenario no excess power will be produced by PV. A simple survey will indicate that if you thermally insulate the houses and apply reflective outside coatings the power demand will significantly deminish, in some countries up to 30%.

When houses are well insulated you will get a power surplus with strong sunshine.

During the winter season wind may be available , but than you also need space heating. All heat storage facilities will function as energy storage. Also here the point is to develop the right controll equipment rather than installing batteries.

Hi Sjoerd

In SA our peak power consumption is in the early evening, primarily from cooking, hot water and heating (we don't have piped gas). During this period we use diesel-fired turbines to supplement the coal and nuclear base load. The first priority from both an economic and CO2 perspective is to use renewables to reduce this peak.

Although I agree that building insulation will make a big difference, it is not going to be enough. We need a way of storing the renewable power until this peak period and use it then.

The coal and nuclear generation cannot respond to changes in load very quickly and is actually providing the base-load only.

Totara Valley in New Zealand and some Islands in Scotland. These are off-the-grid places where there is an integrated energy system (renewables and fossil) with storage. You may find more detailed information in this OECD report http://www.oecd.org/regional/linkingrenewableenergytoruraldevelopment.htm . I would add that in Scotland, the presence of a reliable energy system created some business opportunities for tourism.

Hi Gerhard, Hi everybody,

I designed an optimization problem to size a PV and a batteries energy storage (BES) system for a building of my university of calabria, Italy.

Input data are: the real energy consumptions each 15 minutes, the irradiance and temperature at the rooftop, the electricity prices and taxes in the Italian deregulated market for medium voltage utility.

In 2011, the energy consumption of the considered building was about 133MWh, the 29% was for internal lighting whereas the remaining 71% was for plug loads; the energy measurements cover the entire year.. The electric bill for the 2011 was about € 23 000.00 (21% VAT not included).

The optimization problem returns the optimal solution: a PV plant of 30kWp, a BES system of 25kW (2h is the discharge time at the rated power, the deep of discharge is 20%, the deep of charge 100%).

Such a solution requires a subsidy of about 13k euro.

Installments were calculated using the well-known French amortization system, characterized by equal annual payments; the cash flows is positive, the life prediction returns 13 years.

Hi Alessandro, Raffaele

These are interesting case studies, thank you. Alessandro, I presume that the system was grid-connected so that the storage was discharged during the peak (high cost) time of the day only?

Gerard, The peakload for cooking needs to be shifted to the midday by introduction of high tariff for early evening load and the stimulation of solar cooking during midday in combination with heat retention box cooking. The food will still be hot in the evening, see HRBox on my webpage ICS www.nienhuys.info

With house insulation the heating can also largely be done durng the sun hours. The problem is that houses are not at all insulated and especially concrete and cement buildings require massive amounts of energy to heat up. In such a situation the electricity companis have an interest in thermal insulation for the rooms that are most heated and advertisements should go together with the sale of electric heating equipment and tariff management.

Hot water is the same as cooking. With sufficient solar insolation, also during the winter, SWH are very effective, but the in-house heat storage boilers should be well insulated and be fitted with a time device that avoids power consumption during peak hours.

The above is approaching the problem of the peakhours from another side, buit in the end teh country will be far cheaper off and more durable.

Hi Gerhard, my system is grid-connected, you are right.

I do not use the batteries for mitigating peaks of the load profile, absolutely not.

Peak shaving is an interesting issue but it must be evaluated for the specific case because the benefit due to the peak shaving has not a corresponding dollar/euro benefit.

A further example, I calculated that charging night-time for discharging day-time is a good idea, it is also an economic solution; but it is a bad solution by the financial point of view because the pay back period is too long, very very long, about 20 years when batteries with significant rated power are adopted.

Money has a cost that people generally do not take into account.

Worth noting that in my previous ansewer I said that cash flow was positive....

Indeed, my experience says that any storage system must be investigated in term of financial and cash flows; where not, we have fairy tale.

As the wind plant are built after a long-term wind measurements (2 years almost), as the storage systyem must be studied in close correlation with a long-term measurement of the load profile (1 years almost).

best regards

Thank you for touching on this issue Sjoerd. The macro energy problem needs a very broad solution that comes from a combination of interventions: energy pricing, housing building regulations, user behaviour (education) and technology. Unfortunately the SA government does not yet have a holistic approach, and most of us work in one small segment of the problem!

I am always impressed by the progress in many European countries where the is a well-coordinated and well-informed approach to these strategic issues.

Thanks for the clarification Alessandro. I agree that at the end of the day the economics must work. The paying customer needs to be convinced, and this normally means that it must make financial sense. People generally don't want to pay to resolve the world's GHG or environmental issues ;).

Found an interesting article that covers this topic:

http://www.greentechmedia.com/articles/read/EPRIs-cool-new-grid-scale-energy-storage-tool

Maybe the government does not have a holistic approach, but the powercompanies can play an important role in their tariff strategy as well as in faclitating the installation of other domestic power equipment. It is less costly to invest in a SWH subsidy than to increase power output for only a few hours per day; for that you need massive investments and these have a long pay-back period, increasing the price of electricity. The underlaying problem is that there is little awareness amoung the power users about the possibility to spread or minimise consumption. In some countries (such as in he Netherlands electricity companies have become energy companies. They address their clients with specific information about water heating and illumination and give them discounted purchase possibilities. In fact they invest in power consumption diversion rather than in increasing their peak power capacity. By the way, I hear that Netherlands' power companies have some coal fired installations for sale.

Dear Gerhard, please Check

a) Kythnos Island

b) Alaska in GVEA

c) Islands of REAo and Eigg reaching above 90% RES penetration

d) Some deliverables of the Stories project www.storiesproject.eu

solid oxide fuel cells (SOFCs)

Battery bank can be used for storage but the energy wastage is high in battery. Energy may be store in other modes like heat, air, water etc.

For grid-connected RES, there are various options for storage depending on contectual constraints..

The issue is trickier when one has RES plus storage and has to support island operation.

Hi Gerhard,

I think one of the thing that has not been mentioned as a "passive" storage are the flexibility of the power generation portfolio or the capability of the other power plants to ramp up and down really fast in order to follow the demand and the renewable generation output (e.g. OCGT are really flexible plants). However, this lead to the fact the the owner of these plants will have less revenues and they will not want to build more plants (flexibility subventions needed). Moreover, this solution will not be able to bring the penetration of renewables to 100% obviously.

The other really important factor to increase the penetration of renewable (the same aim as storage) to an higher extent are the international interconnections in order to be able to export the excess power and import it when renwable output is low.

Then demand management, physical storage, energy efficiency etc.. are part of the game..

Hi Gerhard,

In principle and in Europe (sorry for this) one needs storage capacity (electric) for durable electricity production when one goes off-grid, not when one produces grid-connected.

This said. I have just invested in an off-grid solar production facility (small sized) to operate garage doors, electric equipment in a garden and somewhat later I will buy an electric car, which will be charged by the same PV (and wind power) installation. That will be the end of my fossil fuel goggler. Believe it or not but the batteries of a fully electric car increase the storage capacity of the off-grid system drastically. When you don't use your car, its batteries join the off-grid storage capacity. A pure profit. Even better. You charge your car at your employers place and take the power home to be used for other purposes than driving! Great! Free power and additional storage capacity. The world is for the bold.

The company with which I am working to get the job done is Redénergie, a small Belgian company, but one which has a lot of off-grid experience, It has small projects in black Africal to provide elecric power for small schools especially in remote villages on the countryside in several countries in the Sahel belt.

In Europe, most people (and that crowd is growing every day) produce electric power in a grid-connected mode. If one does that, always pay attention not to overdimension power production, because every kW you don't consume within a certain time frame, is lost to the electricity comapnies. They will sell it, and the small producer gets nothin. That's why I prepare to go off-grid, at last in part, as a try-out for a city dweller.

An off-grid system is quite flexible. When you have wind and solar, there is not much than can go wrong with the continuity of power supply. Batteries will for 95% of the time by fully charged. That depends of course as well on your power consumption pattern. If you would happen to experience some 'dead' periods, you can always use a conventional fossil fuel generator to take over in emergencies. These fossil fule generators are on the market everywhere, even in versions which take over automatically, when power supply drops to zero. Actually the same principle applied by computer based companies in thier case for for safety reasons. If the mains fail, the generator takes over. (Hospitals do this as well well evidently). Of course one needs to have a fairly good idea of ones actual and future power consumption needs (electric cars, electric devices, heating, cooling, you name it.

Technically speaking, this type of more or less domestic power supply is long acquired technology in Belgium and some other countries as well. It is something else when we look at the number of people going off-grid. In Belgium that's close to nobody. In black Africa that's everyone, because the grid is is not reliable or is just not there. By the way in India they face exactly the same unstable power supply as in black Africa. I don't have to tell you that for the moment Redénergie is making money like hell.

Strange world we liven in no?

Frank

PS.: look at http://www.redenergie.be

It's in French, but your PC will translate without a doubt.

Thank you for the valuable perspectives.

I think all the power utilities would like some storage to relieve the need for "instant" adjustment of generation to follow the grid loading. This demands great complexity that is made worse by large-scale PV, CSP or wind.

In South Africa this is made worse by our peak power use being in the evening, and not having other countries on other time zones to shift the consumption/generation to, this sets the generation capacity needed. Adding wind/solar does then not reduce the PEAK demand, which you would be able to do if there was energy storage, saving a significant amount of money.

In the web http://www1.eere.energy.gov/hydrogenandfuelcells/pdfs/business_case_fuel_cells_2012.pdf,

You can find profile several recent sites that made plants of renewable energy with storage in fuel cells (of 1 kW to 6 MW) during the past year 2012.

Personally, I think that in the next future is it in the smart cities, without High Power Grids and with an electrical transport with electro-stations located near to the Wind farms or the PV farms.

The actual technology is trained to undertake this target successfully.

To add on this discussion, have a look to our paper with V Rious published in RSER on "Review of supporting scheme for island power system storage".

Hi Félix, I agree the Fuel Cells will have a role to play. Thank you for the very useful reference. I think believe that in the long term we will have a great deal of regional diversification in energy sources (and yes, also in energy storage and transport). However, the problem with Hydrogen remains its inefficiency when compared to battery storage. This may prevent it from being the dominant energy carrier except where there are specific regional reasons that support its use.

One such case is Germany, where the gas network is so vast and well integrated, that it makes a gas-based technology very attractive for other reasons.

Hi Yannick, thank you also for your input on Island Storage. I agree with the writers that it is important to ensure that technology is mature enough to be useful before we start pushing it too hard.

http://www.offgridenergy.com.au/index.php/grid-battery-back-up.html

This is a good resource to start. I did hear of a town that has a backup battery system of unusually high capacity, but I'm not sure about the data they have available. The hydrogen revolution has not happened, but some engineers are already working around the storage of hydrogen for energy cell use by storing hydrogen at 120 psi in multiple tanks. It would be inefficient for a large storage system however, but should be examined next to the cost and life of traditional battery systems. (High pressure underground storage of the specialized tanks can be examined against the cost and life of a more traditional battery system)

If we are talking about the storage it does not battery only. We can store the energy in different form like in kinetic, heat, water, air etc.

FWIW the Graciosa Project is one real-life implementation that I am aware of; it supports 3MW Peak Load, capable of 100% coverage by renewable generation, covering more than 70% of the island's total annual consumption, and, of course, it has a rather large (2.5 MWh) battery system.

Your question includes assumptions which are not generally correct.

- Wind and solar power can be predicted. Such predictions are used operationally by grid operators in Europe. Research in the field of energy meteorology is continuously improving the accuracy of these predictions.

- In most grids at least twenty percent of electricity can by provided by wind/solar without the need for storage or extra back-up capacity. At what level storage or extra back-up (gas power plants or hydro ) become necessary depends on the spatial size on the grid, the transmission capacity, what other type of of power plants are on the grid etc.

@Jethro

Thanks. I agree with your observations, but as you add wind & solar to the grid, there comes a point where it is difficult to add more without also adding some storage to help match the demand with the supply. My concern is particular smaller grids that do not run across multiple time zones.

Hi Gerhard,

This paper tries to "determine the optimum size of the wind generation-battery system".

Can this be usefull to you?

http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6630120

Hallo Gerhard,

Much of the earlier studies on the grid integration of wind and solar power were done on a national scale and did not consider international power transmission. These kind of studies for countries like the Netherlands, Belgium and Ireland could therefore be useful to you. However, the results of these studies were often only presented in reports (grey literature) that are not always easy to find.

Below some wild mixture of references and links that might be useful, or at least could lead you to other helpful literature:

http://digital-library.theiet.org/content/journals/10.1049/iet-rpg_20070056

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5237665&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5237665

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5237665&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5237665

Wijk van et al., 1992

A. Wijk van, N. Halberg, W. Turkenburg

Capacity credit of wind power in the Netherlands

Electric Power System Research, 23 (1992), pp. 189–200

http://ejournal.windeng.net/3/01/GGiebel-CapCredLit_WindEngEJournal_2005_right_links.pdf

http://ivem.eldoc.ub.rug.nl/FILES/ivempubs/dvrapp/EES-2012/EES-2012-158M/EES-2012-158M_HarmenSytzedeBoer.pdf

http://www.sciencedirect.com/science/article/pii/0142061594900353

http://ivem.eldoc.ub.rug.nl/FILES/ivempubs/dvrapp/EES-2012/EES-2012-136T/EES-2012-136T_MartinVelthuis.pdf

http://www.we-at-sea.org/wp-content/uploads/2013/01/RL0-1-2004-012-Implementation-scenarios.pdf

http://www.sciencedirect.com/science/article/pii/S0360544207000345

http://www.sciencedirect.com/science/article/pii/S0301421510008785

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=4275692&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D4275692

Alsema, E.A., A.J.M. van Wijk, and W.C. Turkenburg. 1983. “The Capacity

Credit of Grid-Connected Photovoltaic Systems.” In Proceedings of the 5th EC PV Solar Energy Conference. Reidel, Dordrecht, Netherlands

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=964838&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D964838

http://multi-science.metapress.com/content/91883ln284v87q0w/

@Jethro

Thank you! There are some really good reads in there.

There is a new IEA * study that finds that a very substantial amount of renewables can be integrated in most grids without much cost. So even in South Africa there is probably more possible than you expect:

http://www.modernpowersystems.com/news/newshigh-renewables-penetration-can-be-achieved-in-every-country---iea-study-4186773

http://www.iea.org/Textbase/npsum/GIVAR2014sum.pdf

http://www.iea.org/newsroomandevents/speeches/140221_GIVAR_speech.pdf

http://www.iea.org/media/presentations/PowerOfTransformation_Factsheet.pdf

The actual report is behind a paywall, but it is not overly expensive:

http://www.iea.org/w/bookshop/add.aspx?id=465

*not really known for their pro-renewables bias

Sizing an energy storage system from technical view requires two most important considerations; 1) Energy storage system capacity, and 2) rate of input and output power, for renewable energy storage systems it is proposed to use ESS with high absorbing and producing rate of energy. A good choice is SOFC/Electrolyser combination which its round trip efficiency is around 80%.

In smaller systems battery storage migt be an option but not in bigger systems as the battery cost may be enormous. A 25kW system exists in IIT Delhi that uses a battery bank. Please contact Prof. V. Dutta at [email protected]

@ A.K. Thank you. In some areas the peak consumption is much higher than average so you can reduce generating capacity by putting storage in place. You can also reduce distribution ratings if you have distributed storage as the peak power is reduced. Some people have found that compared to the potential saving on this type of infrastructure, batteries are actually cheaper per kWh throughput.

Because of the low cost of batteries as compared to increasing the peak capacity of the power factories, these are increasingly used.

In relation to the planning of a small PV station with batteries I posted a paper on the subject related to hot climate zones:

http://www.nienhuys.info/mediapool/49/493498/data/V6_Small_Photo_Voltaic_Power_Station_in_a_Hot_Climate_Zone.pdf

and

http://www.nienhuys.info/mediapool/49/493498/data/Small_Scale_PV_Energy_Storage_In_a_Hot_Climate_Zone.pdf