4 ms·

are you certain this remains true on a grid dominated by wind and solar? it is true at the moment because the grid just deals with the vagaries of these sources

are you certain this remains true on a grid dominated by wind and solar? it is true at the moment because the grid just deals with the vagaries of these sources and other generators (and consumers) bear the cost. Will this continue to be the case?

toomuchtodo 5y ago
Renewables have zero marginal cost to generate. When the spot price goes negative, they can simply stop exporting. Other generators are paying others to take power and for the fuel they're using to generate. As renewables penetration increases, it only puts more pain on legacy generators and increases the arb opportunity for storage (which can charge when power is cheap, and discharge when demand is high). This is the entire model around Tesla's Autobidder platform [1] and their energy trading division [2], and their desire to become a virtual utility (with the value in energy orchestration between system components). [1] https://www.tesla.com/support/energy/tesla-software/autobidder https://www.tesla.com/support/energy/tesla-software/autobidd... [2] https://www.reuters.com/business/energy/tesla-plans-begin-trading-solar-wind-battery-storage-energy-2021-09-08/ https://www.reuters.com/business/energy/tesla-plans-begin-tr... EDIT: @eldaisfish: Energy storage growth in the US: https://www.eia.gov/analysis/studies/electricity/batterystorage/pdf/battery_storage_2021.pdf https://www.eia.gov/analysis/studies/electricity/batterystor... Tesla currently orchestrates 1.2Gwh of storage, and their Megapack manufacturing facility will have the capacity to build 40Gwh of storage per year: https://electrek.co/2021/03/16/tesla-autobidder-managing-over-12-gwh-energy-storage/ https://electrek.co/2021/03/16/tesla-autobidder-managing-ove..., https://electrek.co/2021/10/08/tesla-megafactory-aims-40-gwh-megapacks-per-year/ https://electrek.co/2021/10/08/tesla-megafactory-aims-40-gwh... The Hornsdale Power Reserve (built and orchestrated by Tesla) is used for grid services (frequency and voltage response, "continuous power"), and has exceeded expectations: https://pv-magazine-usa.com/2020/05/20/hornsdale-and-its-big-tesla-battery-exceed-expectations-as-neoens-storage-revenue-surgesneoen-reports-strong-revenue-increase-teslas-hornsdale-big-battery-exceeding-expectations/ https://pv-magazine-usa.com/2020/05/20/hornsdale-and-its-big... Utility scale PV can also provide grid forming services at minimal cost (firmware upgrade to inverters): https://www.nrel.gov/news/features/2020/renewables-rescue-stability-as-the-grid-loses-spin.html https://www.nrel.gov/news/features/2020/renewables-rescue-st...
- eldaisfish 5y ago
  tesla's service does not scale to gigawatts and gigawatt hours. Battery storage works for minutes of demand, not hours. At the moment, any battery tries to profit from arbitrage, not supplying continuous load. You mention zero energy prices but fail to address the major problem - reliability. The power is always on and losing power is catastrophic. The solution to the current problem is more transmission capacity. Not batteries. Wind and solar can help but they are not the solution unto themselves. Oh and the cost of wind and solar are not only in fuel. There's lots of costs to O&M and grid balancing, both of which are not included in their energy cost.
  - p_l 5y ago
    While of course oversimplified, https://model.energy https://model.energy is interesting site to simulate the issues, as it tries to model a "base load" generation using green energy. The moment the storage costs of hydrogen go beyond ridiculously low (based on assumptions of available salt caverns), or if you try to use batteries only, the numbers game is over for non-nuclear - unless you add gas turbines which defeat the 0-emission target. For a simulated replacement of 2GW power plant (2-3 reactors, enough to provide for a metropolitan region base load, assuming Poland for weather/solar data) using only batteries for storage, we look at needing to build: ~29GW of solar power ~12GW of wind power ~46.7GWh of battery storage (with ~3.5GW output) Going for hydrogen storage, the result is ~10.2GW solar ~6.5GW wind ~5.5GWh battery (with 0.98GW output) ~1.5GW hydrogen gas turbine ~760GWh hydrogen storage. All of that, to replace a 2GW power plant with non-nuclear zero emission sources. (for comparison, polish base load demand is effectively around 14GW, and we don't really have any place for more pumped storage or hydro power without some funky exclusion zones and environmental destruction)

toomuchtodo 5y ago

Renewables have zero marginal cost to generate. When the spot price goes negative, they can simply stop exporting. Other generators are paying others to take power and for the fuel they're using to generate. As renewables penetration increases, it only puts more pain on legacy generators and increases the arb opportunity for storage (which can charge when power is cheap, and discharge when demand is high). This is the entire model around Tesla's Autobidder platform [1] and their energy trading division [2], and their desire to become a virtual utility (with the value in energy orchestration between system components). [1] https://www.tesla.com/support/energy/tesla-software/autobidder https://www.tesla.com/support/energy/tesla-software/autobidd... [2] https://www.reuters.com/business/energy/tesla-plans-begin-trading-solar-wind-battery-storage-energy-2021-09-08/ https://www.reuters.com/business/energy/tesla-plans-begin-tr... EDIT: @eldaisfish: Energy storage growth in the US: https://www.eia.gov/analysis/studies/electricity/batterystorage/pdf/battery_storage_2021.pdf https://www.eia.gov/analysis/studies/electricity/batterystor... Tesla currently orchestrates 1.2Gwh of storage, and their Megapack manufacturing facility will have the capacity to build 40Gwh of storage per year: https://electrek.co/2021/03/16/tesla-autobidder-managing-over-12-gwh-energy-storage/ https://electrek.co/2021/03/16/tesla-autobidder-managing-ove..., https://electrek.co/2021/10/08/tesla-megafactory-aims-40-gwh-megapacks-per-year/ https://electrek.co/2021/10/08/tesla-megafactory-aims-40-gwh... The Hornsdale Power Reserve (built and orchestrated by Tesla) is used for grid services (frequency and voltage response, "continuous power"), and has exceeded expectations: https://pv-magazine-usa.com/2020/05/20/hornsdale-and-its-big-tesla-battery-exceed-expectations-as-neoens-storage-revenue-surgesneoen-reports-strong-revenue-increase-teslas-hornsdale-big-battery-exceeding-expectations/ https://pv-magazine-usa.com/2020/05/20/hornsdale-and-its-big... Utility scale PV can also provide grid forming services at minimal cost (firmware upgrade to inverters): https://www.nrel.gov/news/features/2020/renewables-rescue-stability-as-the-grid-loses-spin.html https://www.nrel.gov/news/features/2020/renewables-rescue-st...

eldaisfish 5y ago
tesla's service does not scale to gigawatts and gigawatt hours. Battery storage works for minutes of demand, not hours. At the moment, any battery tries to profit from arbitrage, not supplying continuous load. You mention zero energy prices but fail to address the major problem - reliability. The power is always on and losing power is catastrophic. The solution to the current problem is more transmission capacity. Not batteries. Wind and solar can help but they are not the solution unto themselves. Oh and the cost of wind and solar are not only in fuel. There's lots of costs to O&M and grid balancing, both of which are not included in their energy cost.
- p_l 5y ago
  While of course oversimplified, https://model.energy https://model.energy is interesting site to simulate the issues, as it tries to model a "base load" generation using green energy. The moment the storage costs of hydrogen go beyond ridiculously low (based on assumptions of available salt caverns), or if you try to use batteries only, the numbers game is over for non-nuclear - unless you add gas turbines which defeat the 0-emission target. For a simulated replacement of 2GW power plant (2-3 reactors, enough to provide for a metropolitan region base load, assuming Poland for weather/solar data) using only batteries for storage, we look at needing to build: ~29GW of solar power ~12GW of wind power ~46.7GWh of battery storage (with ~3.5GW output) Going for hydrogen storage, the result is ~10.2GW solar ~6.5GW wind ~5.5GWh battery (with 0.98GW output) ~1.5GW hydrogen gas turbine ~760GWh hydrogen storage. All of that, to replace a 2GW power plant with non-nuclear zero emission sources. (for comparison, polish base load demand is effectively around 14GW, and we don't really have any place for more pumped storage or hydro power without some funky exclusion zones and environmental destruction)