Prof. Abdennaceur Karoui

Details

Last Updated: Friday, 09 November 2018 05:50

Is Very Large Scale Storing of Harvested Solar Energy in Batteries a Reasonable Approach?

Will it Hinder the Dream of "Free Energy" Era?

Some factors distracting from (or holding) a rapid and conscious development of efficient Renewable Energy (RE) technologies are discussed here. This article is a feedback on a comment of a recent publication on the use of batteries for large scale transportation [2]. The commentator noticed a “scary” development (not the first) related to large scale use of batteries for storing energy from a large photovoltaic (PV) power plant [3].

Picture from [1]

The pilot project employs huge batteries to store excess of harvested solar electricity. The battery system would work as an “intermittent energy source”. However, batteries are, in principle, incorporated in a PV power station to just homogenize the energy and to stabilize short fluctuations of both energy production (rapid change of the overall intensity of incident solar radiations, distribution across the PV field that lead to hot spots, localized failure of regions in the PV source, PV field reconfiguration, islanding PV modules,...) and consumption (loading, unloading, disjunction, switching spikes,...), but never, for feasibility reasons (the economics and the used materials) to solve the problem of the shifted usage peak, where domestic consumption is mainly at night. Compensating for the fluctuation requires a relatively small battery system. However, the shear size of the reported system makes me feel that... every other project is tiny; and that's just a portion of the complete dream project of that Utility Company. Phase 1 will employ batteries that have a whopping capacity of 7.2 MWh, which is enough to power simultaneously2,000 homes for 3.6 hours (say, evening time) at an average consumption rate of 1kW per home. It seems very good, and most people would agree on that !

The question is: is this technology realistic?

Also, will it have a chance to survive, or to produce derivative technologies? What is the impact of success or failure of such investment on the renewable energy technologies?

What is the overall cost? Is this approach economically viable, now and in the future?

What is the mean time between failures (MTBF) of these batteries when operated in such huge and complex system that is source of continuous strains for the battery cells? What is the cost of the maintenance? Is the owner ready to change these batteries periodically (possibly each 5 years, but maybe earlier!)?

Is it reasonable to push such technology above its capability? What if the chosen option is inapt of supporting technologies for RE harvesting?...

I think most of the engineers know the risks of this large scale test, but a well-founded quantitative study is expect to show that the situation is even worse than the easy answers to these questions.

Quickly, it is useless to power at night time homes with RE, with a consumption rate similar to rates when energy source is oil or gas. So, one could say that the whole technology (relying on such huge storage system) does not have much chance to prove itself! For the investors, it will be awful to discover that at the end of the day.

It is understandable that the owner (a local utility company) has money to afford large scale tests (at least this one). But a fundamental question remains: Why confusing testing phase and production phase? The former requires small pilot and a “constrained” and limited budget, while the latter needs large scale machinery that provides energy at the application scale? Since this is a test, why this scale and this spending (not necessarily investing in the future technology)? Instead, this experiment should have been done at small scale, tested, then scaled-up according to the viable solution?

Despite these many questions, let's focus on the problem only from the technical and functionality view points and relate that to the scope of current debate on “usefulness of batteries for industrial RE systems”.

First, let me state the following important RE technical considerations.  When I started research on photovoltaics, I learned in the lab and in the field some fundamental concepts for making a useful, efficient, and economic renewable energy system. These considerations are (hopefully they will become part of RE system design rules):

1) Reduce the number of energy conversion stages when designing a RE system. For instance, always reduce the role of batteries and when possible, avoid them. Also, when possible, do the most direct coupling of PV modules-Electric loads, reduce the length of cables,... Likewise, for solar water heating systems, shorten the transport of the hot fluids,...

2) Harvested energy from RE sources must be consumed locally, right where it is produced (or very close). There are some exceptions to this consideration, though. This consideration stems from the nature of RE, which is diluted, unlike, for instance, the chemical energy stored in hydrocarbons that we burn to produce energy, or the more obvious case of nuclear energy,...

This essential consideration has two main implications: i) most likely centralization of RE harvesting is unnecessary for there will be lot of energy losses, ii) energy harvested from RE sources should not be transported but consumed locally, as local as possible (very close to the source). To clarify this idea, let’s compare a PV power plant and a mid-size PV module (powering a pump, for instance). The solar electricity plant requires [generation][transformation][transport][back‑transformation][distribution][consumption]. Each step has its own efficiency. Ultimately, only 25% of the harvested energy will be actually utilized (at the application level), as compared to the smaller system directly coupled to the PV generator. A total of 75% will be lost in extra and unnecessary transformations and transport! The efficiency here is only related to the energy distribution system. It excludes the efficiency of the PV generators, considered equal for both the large system (PV power plant) and the optimized PV system (small PV generator).

Let me add this example. What I call the "energy generation and distribution legacy" started around 1875 by T. Edison, C. Brush, and W. von Siemens. At that time, the powered electric machines were huge and heavy. But not every factory was capable of owning its electric generator. That justified the idea of central power generation. The concept was then understandable, and worked very well for about a century and a half, and will still work for few more decades. Recall that in 1947, the landmark ENEAC computer was built. Though the machine needed very stable power, it was not possible to use batteries (or other small generator). The ENEAC was a huge machine made of thousands of electronic vacuum tubes, akin of light bulbs; they were truly electricity gluttons. Today’s powerful computers need not large power. Your laptop could run with just a small battery and/or two solar cells.  

3)  For a given application, there is an optimum size for the chosen RE system (regardless of the chosen technology).  To a certain limit (depends on the application), the smaller the system size, the higher the overall efficiency and the more economical the system will be. In addition, the smaller a RE system is, the easier to maintain and repair, and the more agile and flexible it will be. The smaller a PV field, the less probable a single failure will occur, and the avalanche effect will be reduced.

4)  Energy mix, is an important concept, where optimization targets best technology combination to fulfill a given need. Often time one would combine at least two complementary RE technologies.

5)  For optimal RE systems, batteries are used for the regulation of fluctuations due to rapid variations in the energy generation and consumption. For such systems, the regulation must focus on protecting the equipment, rather than providing energy when the energy source is absent. The last problem can only be solved (at the production scale) with the energy-mix approach, where various RE sources are combined. Therefore, batteries cannot be regarded as a long term storage device (for instance, to offset the energy to nighttime utilization, in case of solar energy generation, or several days in case of wind energy). The reasons are both economic and materials sustainability. Hence, batteries as “intermittent energy source” is a wrong concept that, if used, will fail and will damage the progress of RE technologies.

RE technologies developed with disregard of these basic concepts will cause a lot of energy waste and might be fatal to the technology development, in particular when the large scale deployment starts. Furthermore, to develop robust and efficient RE technologies, we must separate ourselves from legacy technologies, including i) Power Plants (exceptions do exist), and ii) traditional grids. Electric grids are large networks that are costly and heavy to operate, such as a National Grids (stretches over entire country) or International Grids (crossing boundaries) [4], with the exception of the ambitious MENA-Europe Solar Energy project [5]. Like pipelines, highways… national grid is one of the heaviest and the costliest infrastructure a nation has, and constitutes one of the corner stones of the economy. But grids need tremendous maintenance and supporting infrastructure (poles, lines, land corridors, transformers, breakers,… maintenance trucks) and technical and administrative personal. Will we still need the grid and the supporting huge infrastructure to run RE? The answer is: Absolutely not due to the nature of the new energy technologies (both RE generators and modern electric receivers)! Imagine Utility Companies that mange mini-sized or micro-sized RE power plants, each plant powers one subdivision! The utility companie(s), which could be many small businesses, will do better jobs when they operate small power plants that are not connected to nation-wide grid and provides electricity directly to homes,…

Isn’t it clear that is the correct way to handle harvested RE and the most energetically efficient, too? The micro-sized plant would be made of a series of modules powering one building, or a row of townhomes (for instance, 10 homes in one row),… The mini-sized plant could be one that powers an entire subdivision (250 homes). But we do need small and simple ones that allow connection to backup systems, for instance to a sufficiently big windmill or a diesel, a gas turbine, or a hydropower turbine that power the back-up electric generator.

Consideration 2) is due to the energy diluted nature of RE. It dictates the need for producing energy form RE source at the Utilization Site, or else, losses will severely degrade the efficiency. Thus centralized production and therefore grid distribution of the harvested energy are not all adapted for RE. However, micro-grids, local grids, subdivision grids, neighbors' grid,... no-grid, are conceivable for RE. To have the most economic system, the optimum grid topology, grid size,... can be calculated.

Consideration 2) implies that there is no need to design large systems that harvest a massive RE amount. It is rather better to distribute RE generators than to have them centralized. Thus, there is no need to develop large Solar Energy Plants (large plants are old fashion), and transporting the produced energy. Transport of solar electricity over a grid is not going to be efficient, and has no future, except in few cases such as the MENA-Europe project [6], that I intend to write about it.

There are some exceptions to consideration 2), for instance, Concentrated Solar Power (CSP) [7], solar ovens [8] and high concentration solar cells [9]. A side from these exceptions and few others, in general, it does not make sense to concentrate the originally diluted RE, store it, transport it, then dilute-it-back, and distribute it and use it in order to power electric devices at home or in a farm or in a manufacturing plant. In that case, four to five additional (often unnecessary) steps are involved, each has a certain efficiency (at best between 60 and 85%). Together they slay the overall efficiency of the system down to 17-24% (average efficiency: 0.7⁵=0.17 up to 0.7⁴=0.24). It is not unusual that people complain of weakness of harvested solar energy. But then why engineers overload the technology with extra steps and extra-devices, which result in bringing down the overall efficiency to 25% at best; this means 3/4 of the harvested energy is wasted through system. It is important to note that the overall efficiency can go lower, if the efficiency of the batteries is added. Each step, at a converting device, is costly in terms of energy losses.

Are these steps really necessary?

Absolutely not If we correct our view about RE, and as a consequence, we change the current design concepts! In that case, 80% of harvested energy can be utilized by the user application (the 80% is the efficiency of the PV system fitted to the application, which excludes the efficiency of the solar cell). Therefore, engineers must "rewind" and start designing truly efficient RE systems, and avoid every bit of energy that goes to waste. We can no longer afford wasting energy!

Back to the large battery pack utilized to store RE, which is the main subject of this discussion. Think about it, with such huge experiment, we are determining the future of PV technologies. Everything we do wrong will negatively impact the future of these technologies and the acceptance by the public and the investors. An error led by an ill-designed small experiment (in the lab) does not hurt at all, we learn from it without effecting our budget. An error in a small pilot is forgiving,... But, the reported "Trial and Error of Large Scale Battery Storage" is just delusional. Of course experimenting is always the way to go for testing new technologies. But when playing with a mountain of batteries, any error will not forgive, and will not be forgotten. It will have a lasting impact, at best it eradicates that specific RE technology and will delay others. Therefore, we must be careful since the technology needs improvement. We must first push research and development towards providing adequate solutions to RE. The best and most efficient design must be researched in the Lab, not in the field! How many large scale failures can RE withstand? How about public opinion and the investors?

It is remarkable that industry has neglected/omitted the above five simple fundamental concepts. Some of the reasons are known, and this is one of them: it has been easier for the engineers to just adhere to the existing energy infrastructure legacy. They could not break away from the old system concepts to implement the brilliant and young ideas brought by PV, Solar Smart Windows, PV Skin,... Also, companies abused the enthusiasm of the populations towards RE and alternative energy sources, they drew large investments for their ill-approaches. Many mistakes have been done; remember Solyndra scandal. As a consequence, the ill-informed business has been pushed towards making large RE plants. Unknowingly, they are hurting the nascent RE technologies. I have been telling this to decision makers, but they appeared powerless. Unfortunately, only business people and utility companies have, practically, the keys to the true RE solutions; if they just listen progress will come quicker.

Some people think that what played against the development of true RE technologies and delayed the large adoption of RE are: 1) the fact that energy industry is keeping the old ideas of "Power Plants", power grid,..., and 2) the big question of who should control the energy sources; some believe that, it cannot be given to citizens. Consider for instance the case of utility companies when they do not allow, after a hurricane, using PV backup systems installed on home roofs. I understand the technical issue; but, it is a simple one, however, it remains unsolved. Why? The true reason is of course not technical, it is just because the utility companies do not support PV on the roofs, while they do support large PV plants (again to preserve their legacy to save the old infrastructure). Maintaining the legacy and the business-as-usual, led to designing and constructing huge PV plants around the world, instead of adopting rooftop PV, portable PV,...

Really, it does not make sense to build a 100 MW PV plant to power homes. All what it does is just concentrate solar energy in high power and high voltage electricity (the old way), and thereafter transport it, dilute-it-back, distribute it in the city, and then  use it. It is exactly that old-fashion technology path that is now pushing towards storage and looking for large scale batteries, to solving new issues for the problem they created. It is exactly that path that is holding the progress of solar energy. It is not allowing development of true RE solutions. It is now about to generate new environmental problems that will be fatal for both RE and Batteries.

Isn't it smarter to just put these same PV panels on the roofs of homes to power an entire suburb, while leaving the concentrated hydrocarbon-produced energy to downtown? Besides, why do we have to destroy fertile lands to install ugly plants and to extend electric lines? That is another environmental catastrophe being caused by some misled companies. It could be avoided, for instance, by integrating PV in buildings, and creating new architectural designs. By the way, Buildings Integrated PV (BIPV) could be very artistic, and pleasant to the eye, while producing energy in towns.

One must note that the urbanization plans in the US South East region, where tens of subdivisions are around downtown of a city, is ideal for the above suggested energy-mix model. Policymakers and the business can seriously think about it, and push towards adopting it; it has tremendous benefits. It would be the most optimal energy production-consumption model. It does not need transport of produced energy from the RE source to the consumers.

Energy conservation: This is a must for our future. For RE to work on large scale and to work better, we need to re-educate ourselves and learn about energy conservation and conservation of nature, in general. It is vital that we turn-off the lights and the AC immediately after usage, reduce energy consumption, use better construction materials to build homes with higher thermal mass, use LED lamps in homes, do the chores during day time, watch TV not too late (big screens use a lot of energy), improve the management of the daylight timing, use public transportation when possible and always smaller cars … These plausible ways of conscious usage of energy represent a large part of the solution, they are doable !  That way, the need for excessive energy storage (as it appears in the commented article) will become pointless; as a result, battery requirement will be reduced to minimum.

Conclusion:

A drastic change in the energy production means and the modification of electric equipment have been witnessed. The immense miniaturization of computers has been done similarly for all electric hardware that we use every day. The main purposes were: i) energy saving, and ii) increased efficiency of the electric hardware. Today, it has become pointless to bring, electricity from tens to hundreds of kilometers to power a small iphone, or even a TV set. So, we can safely say that the "power generation and transport legacy" has become unnecessary as it is readily inefficient and costly. 

RE technologies do not need large scale batteries, large power plants, and grids for transporting electricity nation-wide. RE storage and transport are pointless and are misguiding future development of appropriate RE technologies. In most regions of the world, the future is not for megawatt power plants, not for “intermittent energy sources” (batteries), not for PV grids,... The future is for distributed RE systems, such as rooftop PV (panels or tiles), PV skin, BIPV, Solar water heaters, small PV powered cars,… While electric stoves are wasteful, solar oven will be economic efficient, and a sort of life saver in may regions of the world. As for batteries, we would better leave them for the billions of portable electronic systems, planes, computers, smart phones, autonomous machines,… and medical devices.

Finally, the progress of the new alternative energy sources and related new technologies must not be hindered by that legacy, and more research on RE must be duly supported.

A. Karoui

June 30, 2018

For this debate on "usefulness of batteries for large scale applications and for renewable energy" some counter-productive experiences and wrong concepts were used to illustrate the irrationality of large scale storage of RE. The large scale utilization of batteries is a serious problem that was induced by ill-designed applications, due to a deep misunderstanding of renewable energy and the correct technology pathways.

Please leave comments, or contact me for any question. Whether you are a supporter of renewable energy technologies or not, please share this paper with your friends in social media, we want to learn from both sides.

References 

[1]  https://www.utilities-me.com/power/11500-dubai-to-trial-large-scale-battery-storage

[2]  https://www.linkedin.com/pulse/powering-trucks-cars-batteries-good-idea-abdennaceur-karoui/?published=t.

[3]  https://www.utilities-me.com/power/11500-dubai-to-trial-large-scale-battery-storage

[4]  Abolition of international grids should not alter international cooperation. It is only related to the specific case of exchanging RE.

[5]  International grid (built up on existing grids) are being thought out to import RE from hot countries to colder where RE is not available.

[6] Because of geographical context, where the utilization is naturally separated (but not too far) from the RE source, few cases are forced to spend “extras” for transporting energy from the RE source to the consumer; these are: MENA-Europe, USA-Canada, CentralAsia-Russia, Mexico-USA, ChinaWest-ChinaEast.

[7] CSP does require concentration of solar radiations to melt the salt, the fluid used for energy transfer. Melting the salt needs reaching a minimum temperature. Also, a critical mass is required to have an efficient system. Note that CSP is a useful technology for its unique feature of heat storage, for at least 6 hours, depending on the design. That makes the technology viable, even though it require heavy infrastructure. It will survive if further developed as a complementary technology to PV.

[8] A solar oven does require a trough to concentrate the radiation. Solar ovens are extremely useful devices that need further development for many obvious reasons. Among many benefits, the solar oven has excellent societal impacts. Due to the critical size needed to attain an efficient oven, and for economic reasons, it is better if such contraption is purchased and used by a family or a small group of families. Due to that, the oven keeps the family together around the meal, and brings other users. The way of life around the solar oven is highly applicable to Africa, MENA, Latin America, India, China and Central Asian countries.

[9] There are solar cells that are designed to efficiently work under concentrated light.