Carbon neutrality powered by AI
As the world enhances its efforts to mitigate the impact of climate change, it will have to rely much on emerging technologies. The foundational shift of energy transition from fossil-based to renewables will have to be done at scale and rapidly, for any positive impact on the environment.
A new report by World Economic Forum (WEF) says that the transition to low-carbon energy can be accelerated and deepened by focused application of artificial intelligence.
The report Harnessing AI to Accelerate Energy Transition identifies the key shifts that will take place and underlines the role that emerging tech will play.
The supply of renewable energy is increasing in grids which have been built for fossil fuel-based power. The time for setting up renewable power generation is much less than for setting up transmission and distribution lines. As a result, existing electricity grids will have to be managed with far more efficiency to cope with rising supply and usage of renewable energy.
AI can optimise and even add to the lifecycle of existing grid infrastructure. Renewable energy is not a steady supply, since it depends on weather conditions. Solar works on sunny days and wind turbines when there is a strong breeze. Such intermittent supply of renewables poses problems for managers, who must maintain
Renewable energy is not a steady supply, but depends on the weather. Artificial intelligence can anticipate the amount of power that will reach the grid by combining weather conditions with supply parameters
stability of energy passing through the grid.
With predictive analysis, AI can anticipate the amount of power that will reach the grid by combining weather conditions with supply parameters. This will allow grid managers to be prepared rather than be impacted by unanticipated surges in supply. The report flags several problems which can hurt grids when intermittent power increases. These include power frequency imbalances, blackouts and brownouts, and significant capacity overbuild. Without real-time data, advanced analytics and automation, the increasingly complex power and energy systems of the future will become impossible to manage, the WEF says.
Research firm Bloombergnef (BNEF), which contributed to the report says, “13 per cent of all global power capacity in 2050 will comprise distributed small-scale photovoltaic (PV) energy and batteries, up from 4 per cent today. This will accelerate an ongoing trend of shrinking median power plant to shrink over 80 per cent.”
The WEF reports states that “it is clear that in the future there will be vastly more physical assets connected to the power grid and, in particular, the distribution grid, where power flows are becoming increasingly dynamic and multidirectional.”
For example, the rise of energy storage in batteries means that some of this power can be reused when required. A household can have the option of switching between battery power, on-site solar source and the grid based on the situation. Similarly, for individuals, AI can help decide when to charge electric vehicles. AI can decide the charging time and duration based on peak or off-peak rates.
Such switching at a large scale across millions of homes will require an AI platform to track usage in real-time.
The German Energy Agency, another contributor to the WEF report, says that AI can help in efficient designing and location of solar, wind and other renewable farms. According to the report, 56 per cent of power generation could be provided by solar and wind in 2050 — a massive 7.6 Tw of solar and 4.6 Tw of wind. And this would need investments of $15.1 trillion in solar, wind and batteries, and $14 trillion in power grid by 2050. According to BNEF, power system costs would be higher by 6-13 per cent in 2040 if intelligence automation systems are not used.
Countries like India, which are leading the adoption of climate change measures, will have to encourage energy companies to embrace higher levels of AI usage.