If electricity was the general purpose technology that provided a new lease of life to mankind in the 20th century, AI is doing that in the 21st century. AI is evolving and scaling everyday, with its utility finding relevance in almost every sector and every industry. Interestingly, it is now poised to revolutionise the power sector as well. 

Conventionally, coal and other fossil fuels have been the primary source of power for nations to carry out their economic activities. Owing to carbon emissions and the detrimental effect these fossil fuels have on the environment, there has been a systemic shift to renewable energy. Now, most governments are leading a mixed approach to energy consumption - completely weaning off fossilized fuel sources isn't entirely feasible in the short run, and setting up the infrastructure for sustained renewable energy generation needs time and money. Depending on the region and policies that support clean power generation, a plethora of opportunities are available for nations to become leaders in renewable energy. For instance, Middle East is understandably at an advantage to harness solar, Norway and surrounding Nordic nations can capitalise wind and hydro energy, while geothermal is another high potential power source for Scandinavian nations. In addition to these countries, US, UK, China, the Philippines, India are gradually diversifying their power generation modes, away from fossil fuels. However, in many of these countries, renewable energy generation is still not a publicly available utility and are not connected to the nationalised power grids. On the contrary, nationalised power grids are quite archaic and antiquated in terms of their infrastructure. This massive gap in power generation and supply is posing a considerable challenge, and limiting the possibility for countries to become a circular economy faster and with efficiency. 

This is where AI is poised to make a sizeable impact, and smart grids are the answer to energy resiliency and sustaining the power distribution network of the future. 

Varying sources of power supply cause fluctuations and possible discrepancies in supply, but grid software can play a vital role in managing this demand and supply if enabled with AI. Grid software covers transmission, distribution and market management - functions that can be planned meticulously using AI and analytics. In addition, AI can also help power distributors maintain the health and efficiency of batteries - by themselves, batteries cannot regulate their own usage and deploying manual resources isn't feasible or even necessary, if AI can do the job of remotely managing sensors and gaining critical data from transmission or distribution towers. AI is also vital for collecting reams of data points across millions of towers installed across locations, and all this data can be crunched and synthesized to extract real, tangible value-driven insights. Deep learning algorithms can process these massive datasets, and detect anomalies with ease, perhaps in a matter of seconds if the training datasets are big enough. 

This is a primer on how AI can revolutionise power grids in developed and fast developing economies. Some efforts are already being made in various parts of the world to hasten the digitisation process. For instance, USA's Department of Energy (DoE) has made it a national policy to support smart grid mobilisation. Since 2010, the DoE has invested $4.5 billion in smart grid infrastructure and installed over 15 million smart meters to monitor energy usage per device. While the country's peak power demand could increase 25% by 2050, the program will limit peak electricity load on the grid by one percent only. India launched the National Smart Grid Mission in May 2015 with an outlay of Rs. 980 crore and has managed to execute a pilot project in Puducherry too, which has largely managed the area's energy needs using renewable power sources. Eventually, this can pave the way for AI-enabled power management methods. Pune-based IndiGrid is already collaborating with companies like IBM and ClimaCell to deploy AI for improved management of power assets. (Read the story here)

A coordinated effort to manage power grids, with technologies like AI, is a sure-shot way of managing the environmental fallout of conventional power generation and to allow renewable energy generation to become more reliable.