ELECTRICITY STORAGE TO COVER WIND-SOLAR LULLS AND SEASONAL VARIATIONS

Daily, Weekly and Seasonal Electricity Storage

 

MIT Professor Steven Chu, former US-DOE Secretary, stated:

 

- Whereas, the turnkey capital costs of battery systems likely would be about 50% less over the next decade, that storage approach would never be cheap enough to accommodate the big seasonal shifts in renewable power production.

- Battery systems could prove viable for storing solar electricity produced during midday hours for use during late afternoon/early evening hours, and “maybe” for up to a week later, but not over seasonal time­frames.

- New technologies are needed to convert “low-cost” renewable energy into chemical fuel whenever excess solar and wind electricity is available. Low-cost hydrogen from renewables, stored underground, may become an economically feasible approach. See note.

- Fuel cells hold more promise for urban power storage, particularly those based on liquid hydrocarbons. However, whereas technically feasible, they are not economical at present.

http://www.powerengineeringint.com/articles/2018/01/former-obama-en...

 

NOTE: One approach would be to produce electricity with base-loaded nuclear plants, and produce hydrogen with electricity and store it in underground caverns all over the world. This would need to apply to at least 70 - 80% of total primary energy, not just the 40% used for producing electricity.

 

Examples of Electricity Storage to Cover Seasonal Variations

 

For those of you who may have illusions/fantasies of 100% RE for electricity*, here are some examples of storage capacities needed to ensure 24/7/365 electricity service, year after year.

* Energy for services other than electricity, such as transportation, space heating/cooling, industrial processes, etc., is a separate issue.

 

There cannot be heavy penetration of wind and solar electricity without storage and standby generation because the sum of wind and solar may be simultaneously near zero, especially in winter with snow and ice on the PV panels, i.e., PFB functions by the other generators are needed 24/7/365. Those functions:

 

- Cannot be performed by wind and solar.

- Can be performed by large-scale hydro storage, which is the lowest cost option, by far.

- Can be performed by battery systems, but that would be very expensive, even if battery system cost became 50% less. See above Dr. Chu’s comments and URLs.

 

The euanmearns URLs show the type of analysis required to determine seasonal storage capacities:

California

1) The seasonal storage would need to be capable to provide about 20 TWh delivered as AC.

http://euanmearns.com/how-californias-electricity-sector-can-go-100...

 

2) The seasonal storage systems, with the Jacobson, et al, renewables generation mix in place, (5% hydro + geothermal, 37% wind, 58% solar), would need to be capable to provide about about 25 TWh delivered as AC.

- Such a high solar percentage would generate a large surplus of electricity during midday hours that would need to be stored for later use.

- Battery storage would provide mainly diurnal storage, which agrees with Dr. Chu’s comments, and stored gas, such as hydrogen produced with renewable energy, would provide the rest of the storage. See Appendix 5.

- US and CA electricity is about 37.5% greater in 2050 than in 2015, due to (heat pumps + plug-ins - increased efficiency). See table 1.

 

http://euanmearns.com/battery-storage-in-perspective-solving-1-of-t...

http://web.stanford.edu/group/efmh/jacobson/Articles/I/CountriesWWS...

https://www.ecowatch.com/100-renewable-energy-by-2050-2519335518.html

https://www.sciencedirect.com/science/article/pii/S0960148118301526

https://news.stanford.edu/2018/02/08/avoiding-blackouts-100-renewab...

http://www.energy.ca.gov/almanac/electricity_data/total_system_powe...

 

Jacobson		Present	2050	Storage 2050
Population	billion	7.3	9.7
Electricity	TWh	24310	48800	16000
North America	TWh
US	TWh	4000	5500	1803
California	TWh	291	400	131

Chile

The seasonal storage would need to be capable to provide about 2.7 TWh delivered as AC

http://euanmearns.com/how-chiles-electricity-sector-can-go-100-rene...

 

Texas

The seasonal storage would need to be capable to provide about 50 TWh delivered as AC to balance ERCOT’s wind and solar generation against ERCOT’s demand over the three-year period considered in this 100% renewable scenario.

http://euanmearns.com/can-texas-go-100-renewable/

 

Germany

The seasonal storage systems would need to to be capable to provide about 80 TWh delivered as AC, at a turnkey capital cost of about $20 trillion, at $250/kWh delivered as AC.

New England

The seasonal storage system would need to be capable to provide about of 8 TWh delivered as AC, at a turnkey capital cost of about $2 trillion, at $250/kWh delivered as AC.

http://www.windtaskforce.org/profiles/blogs/seasonal-pumped-hydro-s...

 

Examples of Electricity Storage to Cover Wind/Solar Lulls

 

The windtaskforce URLs show the type of analysis required to determine wind/solar lull storage capacities:

 

Germany

The storage systems would need to be capable to provide about (100-h lull + 50-h lull) x 62.0 GW = 9300 GWh = 9.3 TWh delivered as AC. Imports and exports would be minimal, as nearby countries also would have wind and solar lulls. See note.

http://www.windtaskforce.org/profiles/blogs/wind-and-solar-energy-l...

 

New England

The storage system would need to be capable to provide about 1.04 TWh delivered as AC, to cover 2 consecutive 4-day wind/solar lulls in winter. See URL, table 3.

http://www.windtaskforce.org/profiles/blogs/wind-plus-solar-plus-st...

 

NOTE: Regarding battery systems for utilities, it is entirely reasonable to assume some percent of the installed capacity would be out of service due to scheduled and unscheduled outages.

APPENDIX 1

Fossil Fraction of World Primary Energy Consumption Unchanged for 43 Years

 

In the 1970s, the big worry was fossil fuels would soon run out, and so we should “use them wisely”. But in the 1980s, the risk changed to one of an overheating planet, and so we should not use them at all. This article shows unchanged fossil energy use from 1970 to 2013, a period of 43 years.

http://notrickszone.com/2018/01/12/green-energy-revolution-a-flop-f... 

 

Here is a table of world primary energy consumption percentages (fuels, electricity, etc.) during the 2011 - 2015 period, which, indicates hardly any progress towards RE, despite worldwide investments in renewables of $250 - $300 billion in each of these 5 years. The fossil fuel percentage likely remained about the same in 2016 and 2017. Table 1 shows the data for the years 2011 - 2015, a 5-y period.

http://www.windtaskforce.org/profiles/blogs/the-world-making-almost...

 

And then COP21 comes along, with much rah-rah and fueled by champagne and 5-star restaurant food in Paris, and that 43-year trend is going to change?

 

It could, but that would require an RE investment level of at least 5 times existing, or $1.5 trillion/y for decades, as outlined in this article.

http://www.windtaskforce.org/profiles/blogs/cop21-ipcc-co2-emission...

 

APPENDIX 2

GMP, a Vermont utility, buys Hydro Quebec electricity at 5.549 c/kWh, under a 20-y contract, per GMP spreadsheet titled “GMP Test Year Power Supply Costs filed as VPSB Docket No: Attachment D, Schedule 2, April 14, 2017”. GMP refuses to buy more Quebec electricity. See URL.

http://www.windtaskforce.org/profiles/blogs/gmp-refusing-to-buy-add...

 

Vermont could have much more hydro electricity from Hydro-Quebec. All we need is more transmission lines. 
- Subsidies would NOT be required. 
- The hydro electricity would be about 6 c/kWh.

 

NE generated wind electricity on ridgelines would be 9 c/kWh, heavily subsidized, and NE generated utility-scale, field-mounted solar would be about 13.5 c/kWh, heavily subsidized, plus the cost of:

 
- Peaking, filling-in and balancing by the other generators 
- Transmission to connect all the wind turbines and solar systems
- Subsidies

 

APPENDIX 3

Vermont Energy Independent: Vermont uses about 6,100,000 MWh/365 = 16,712 MWh/d.

If at some future date, Vermont were to use 75% homegrown and out of state wind and solar, with the rest from home-grown and imported hydro, etc., two consecutive, 4-day wind/solar lulls, a common occurrence in winter, would require 2 x 4 x 0.75 x 16,712 = 100,272 MWh delivered as AC from storage, at a turnkey capital cost of 100272 x 1000 x 250 = $25 billion, at $250/kWh delivered as AC.

That is just for 40% of the primary energy used for generating electricity. The other 60% of PE is a separate issue.

Some 100% RE folks want Vermont to be "energy independent" (disconnect from the NE grid?) and to stop using fossil fuels (coal, oil, gas) and nuclear by 2050.

http://www.windtaskforce.org/profiles/blogs/vermont-energy-independent

 

APPENDIX 4

A Very Large Battery System in Australia: The Tesla Powerpack system in Australia, the largest in the world, has a rated capacity of 100 MW/129 MWh delivered as AC. The system:

 

1) Smoothens the variable output of a nearby 315 MW French-owned wind turbine system,

2) Prevents load-shedding blackouts and

3) Provides stability to the grid, during times other generators are started in the event of sudden drops in wind or other network issues.

Here is an aerial photo of the system on a 10-acre site. The installed cost of the Australian Powerpack system was about $66 million, or 66 million/129,000 = $512/kWh; this is a low price, because Tesla was eager to obtain the contract.

 

NOTE: The battery does not serve to store electricity for delivery at a later hour on the same day, or a week later, or several months later. For those purposes, much larger batteries would be required.

APPENDIX 5

California has access to hydro storage and concentrated solar power, CSP, storage, which have a much lower cost, $/kWh, than battery storage at about $250/kWh delivered as AC.

 

The above analysis is based on extrapolating from existing conditions. The rate of change of those conditions, on an absolute basis, is very slow, even in California. Somewhat like turning around a loaded supertanker.

 

- The fact a growing duck curve exists in California indicates a much greater level of RE efforts are required to offset it.

- The high costs of the present level of RE efforts has caused California residential rates to increase, as happened in other RE countries, such as Denmark and Germany.

- A similar duck curve is starting to emerge in New England, albeit only on rare sunny days.

 

California residential rates increased from 14.2 c/kWh, at start 2008 to 15.4 c/kWh, at end 2015, up 12% since 2008

US residential rates decreased from 10.8 c/kWh, at start 2008 to 10.4 c/kWh, at end 2015, down 3% since 2008

http://www.latimes.com/projects/la-fi-electricity-capacity/

APPENDIX 6

Energy-Surplus Buildings and Plug-in Vehicles: Turning around the US building stock to energy-surplus buildings, and the vehicle stock towards plug-ins would take some decades.

http://www.windtaskforce.org/profiles/blogs/evs-and-plug-in-hybrids...

 

Ideally, all residential and other buildings should be energy-surplus buildings, with:

 

1) Highly insulated and sealed.

2) Energy efficient systems and lighting.

3) Heat pumps for space heating and cooling and hot water.

4) Battery systems; store PV solar during midday, use in the evening

5) Thermal storage systems; store PV solar, use as needed

6) PV solar systems. 

 

Such buildings would:

 

1) Take a long time to warm up or cool down, i.e., the internal temperature would vary just a few degrees during a day, even though the outside temperature would vary 30 degrees or more.

2) Use minimal energy for heating, cooling and electricity (Btu/ft2/y)

3) Have enough electricity left over to charge plug-in vehicles at night.

 

Such a setup would greatly reduce the daily variation in electrical demand, and thus reduce the need for generating capacity, MW.

 

However, almost all of the building stock is very far from highly insulated and sealed, etc.: they are energy-hog buildings. Placing solar panels on the roofs of such buildings makes for good visuals, but creates grid disturbing duck curves, especially in California.

APPENDIX 7

Norway Electricity Generating System

 

Norway electricity generation was about 149 TWh in 2016, about 143.4 TWh, or 96.3% from reservoir hydro plants and run of river hydro plants. Norway total hydro reservoir capacity is 84.3 TWh.

https://www.ssb.no/en/energi-og-industri/statistikker/elektrisitet/aar

http://wwwdynamic.nordpoolspot.com/marketinfo/rescontent/norway/res...

 

All those plants are connected to the high voltage network. Modulating the water flows through the turbines performs the peaking, filling-in and balancing, PFB, functions.

 

The combined storage reservoirs act as a giant battery that is continuously charged with rainwater and runoff and discharges water through turbines, as needed to meet electricity demand 24/7/365, year after year.

 

The HV networks of Denmark, Germany and the Netherlands are connected to the HV network of Norway with HVDC lines. During periods of higher winds, electricity (usually at near-zero or negative wholesale prices) is sent via these lines to Norway, which merely reduces the water flow through the turbines.

 

Norway uses the “saved” water to generate electricity when wholesale prices are high, such as during wind and solar lulls and higher demands. Germany would like to send its excess electricity from North Germany to South Germany but HVDC lines, already 15 years overdue, do not exist, mostly due to NIMBY concerns.

 

Quebec Electricity Generating System

 

Quebec electricity, generated and purchased, was about 217 TWh in 2016, of which about 172 TWh/y from 36911 MW of 63 large hydro plants.

 

- Reservoir, 28 plants, capacity 26843 MW, production about 124.7 TWh in 2016

- Run of river, 35 plants, capacity 10068 MW, production about 47.3 TWh in 2016

 

The active water storage of the reservoir plants is about 1/3 of total storage. The active storage is equivalent to about 176 TWh/y, about equal to Quebec’s annual electricity requirements.

 

The active storage could generate another 176 - 124.7 = 51.3 TWh/y, however, this is dependent on environmental impacts, recreation, water levels, flow rates, navigation, flood control, and the weather.

 

All plants are connected to the high voltage network. Modulating the water flows through the turbines performs PFB functions.

The combined storage reservoirs act as a giant battery that is continuously charged with rainwater and runoff and discharges water through turbines, as needed to meet electricity demand 24/7/365, year after year.

 

http://www.hydroquebec.com/learning/hydroelectricite/gestion-eau.html

https://en.wikipedia.org/wiki/Hydro-Québec

 

In 2016, Quebec total net exports were 32.6 TWh, of which 20.8 TWh, directly or indirectly, to New England (per ISO-NE) and 11.8 TWh to other states, such as New York State. Revenues were $1.568 billion; average sales prices 4.8 c/kWh.

http://www.hydroquebec.com/sustainable-development/energy-environme...

 

New England Electricity Generating System

 

NE generated about 105.6 TWh in 2016, and imported about 20.8 TWh (Quebec 12.3 TWh; New Brunswick 4.8 TWh; NY 3.7 TWh). Fuel sources for generation were: gas 49.3%; nuclear 31%; hydro 7.1%; refuse 3.1%; wood 3%; coal 2.4%; wind 2.4%; solar 0.6%; oil 0.5%.

 

Modulating the outputs of the gas turbines, which requires more Btu/kWh and emits more CO2/kWh, performs most of the PFB functions. The modulating would increase as more wind and solar is added to the system.

 

NE wholesale prices have averaged about 4.5 to 5 c/kWh since 2009. Closing down gas, nuclear, coal (all of which generate at less than 5 c/kWh), and oil plants, would create a huge generation deficiency that likely would need to be offset mostly by increased build outs of wind, solar, refuse, hydro, plus increased imports via tie lines, plus large-scale storage systems. The PFB functions would be performed by storage systems to accommodate the variability and intermittency of wind and solar.

 

- Wind; ridgeline about 9 c/kWh, heavily subsidized; offshore about 19 c/kWh, heavily subsidized.

- Solar; large-scale, field-mounted about 13.5 c/kWh, heavily subsidized; residential rooftop about 19 c/kWh, heavily subsidized.

APPENDIX 8

California has made major efforts to implement electricity generation from renewables, such as wind and solar. Wind and solar electricity is more expensive than fossil and hydro electricity, due to the costs of:

 

- Generating wind and solar

- Subsidies for wind and solar

- Additional battery storage to mitigate DUCK curve effects, due to midday solar.

- Increased ramping up by traditional generators. As solar disappears in the late afternoon and demand is ramping up, generation by traditional sources has to ramp up as well; the more solar, the steeper the ramp up, MW/h.

- Increased demand management to shift demand from late afternoon/early evening to after 11 PM.

- Additional grid investments to connect wind and solar systems.

- Increased peaking, filling-in and balancing services provided by the traditional generators to accommodate the variable, intermittent wind and solar, 24/7/365.

 

As a result of those costs, California electric rates have increased much faster than US rates.

 

In 2011, California rates were 35% higher than US rates. In 2017, they were 60% higher than US rates, a clear sign increased RE leads to increased electric rates. The same is true in Germany and Denmark, which have high RE percentages. Denying it is beyond rational. Source EIA website. See table.

 

Year	California	US	CA higher than US
	c/kWh	c/kWh	%
2011	13.1	9.7	35
2012	13.5	9.6	41
2013	14.3	9.8	46
2014	15.2	10.1	50
2015	15.4	10.0	54
2016	15.3	9.9	55
2017	16.2	10.1	60

California, and other states, should concentrate on energy surplus housing and other buildings.

Over a 24-h period, their indoor temperature would vary a few degrees F, with outdoor temperatures varying 30F or more.

They would have solar systems and batteries and have enough electricity to charge plug-in hybrids at night.

With lower solar and battery prices, that would be the most rational approach to make disappear most of the DUCK curves.