An excerpt from "using artificial Intelligence to accelerate Energy Transformation"

China Electronic Information Industry Development Research Institute

.

China Electronic Information Industry Development Research Institute (CCID, Sadie) is a unit directly under the Ministry of Industry and Information Technology. It is a famous think tank in the fields of industrial economy, information technology and Internet in China. It is also a public technical service organization and media communication group with extensive influence. Welcome to cooperate and communicate with us through this platform!

In September 2021, the World Economic Forum, Bloomberg New Energy Finance and the German Energy Agency jointly released a research report on "using artificial Intelligence to accelerate Energy Transformation". The report has been compiled by the Institute of Planning, a Sadie think tank, which is expected to be helpful to the relevant departments of our country.

The report analyzes the huge demand and scope of application of artificial intelligence technology in the process of energy decentralization, digitization and decarbonization, and puts forward nine principles from three aspects of design, empowerment and governance, aiming to release the potential of artificial intelligence and assist in energy transformation. The report puts forward guiding suggestions for energy companies, policy makers and other stakeholders, that is, to strengthen cooperation among all parties, optimize market operation mechanisms, and establish clearer energy data regulations.

"

The sixth assessment report issued by the Intergovernmental Panel on Climate change in August 2021 and climate changes such as heatwaves, floods and wildfires that have become increasingly apparent in recent years have attracted the attention of policy makers, businesses and investors. With the approach of the 26th United Nations Conference of the parties to Climate change (COP26), it is expected that the release of climate targets will be further accelerated. The transformation process of low-carbon economy needs to be accelerated, and the energy field is the core challenge of this process. Artificial intelligence will play an important role in promoting energy transformation.

First, energy transformation requires artificial intelligence.

At present, the global energy system is in transition, and the potential of artificial intelligence to accelerate energy transformation is constantly being stimulated.

(1) Energy systems need to use digital means to promote rapid transformation

In order to achieve deep decarbonization, carbon dioxide emissions from the energy system need to be rapidly reduced to very low levels. The decarbonization transformation of the energy system has brought about integration and electrification, and the interaction between power, transportation, industry, construction and other industries has been significantly enhanced, and the system will be composed of interdependent energy and telecommunications networks. In order to accelerate the transition to extensive, affordable, low-carbon energy supply, it is necessary to further optimize all aspects of the energy system and strengthen coordination and cooperation between each link.

(2) decarbonization in the power industry is the focus of decarbonization in the whole energy system.

The transformation of the energy system includes the rapid expansion of the supply of renewable energy and the large-scale clean electrification of heating, industry and transport. With the increase in the number of electric vehicles, the decline in battery energy storage costs and the trend towards net zero-emission electricity in construction and heavy industry, electricity is expected to account for 60% of global energy demand from 2019 to 2050. Electricity will increasingly be used for heating and cooling, transportation, and even the preparation of hydrogen.

(3) Energy transformation requires a large amount of investment

In Bloomberg New Energy Finance's New Energy Outlook 2020, there is a long-term expectation for the future transformation of the energy economy, that is, by 2050, 56% of the electricity generated will come from solar and wind energy, reaching 7.6 terawatts and 4.6 terawatts, respectively. And the assumption is based on maintaining the current policy, which reflects that solar, wind and energy storage economies have become important drivers of rapid decarbonization in the power industry, even without taking into account high coal prices or net zero emission targets.

(4) the future power system will be highly decentralized

Increasing the proportion of renewable energy generation will make the power system more powered by intermittent generators and more decentralized. At present, distributed small photovoltaic power stations account for 4% of the world's installed capacity, and medium-sized power plants have an installed capacity of 944 megawatts. According to the forecast of Bloomberg New Energy Financial Energy Transformation, the proportion of distributed small photovoltaic power stations will increase to 13% by 2050. The installed capacity of medium-sized power plants will be reduced by more than 80%, to only 158 megawatts.

(5) the complexity of power system management will increase significantly.

According to the decarbonization target and the current development trend, a large number of physical equipment will be connected to the power grid in the future, especially the distribution network. In the distribution network, the current will also become more and more dynamic and multidirectional (see figure 1). Devices such as small distributed devices may generate electricity and sell back to the grid, fast charging of electric vehicles may lead to a surge in demand, and devices such as smart homes may be connected to the grid without the knowledge of grid operators, all of which will have a significant impact on current stability.

Powered I only includes battery, electric and plug-in hybrid passenger vehicles (excluding commercial vehicles and two-wheeled or tricycles). Ii includes only solar and wind energy (excluding other renewable sources). Iii includes large power station level and energy-saving lithium-ion battery storage. Source: adapted from the German Energy Agency (2020), picture from Bloomberg New Energy Finance (2020)

(6) artificial intelligence can accelerate energy transformation.

Artificial intelligence refers to a broader concept, not a single technology or product, but a set of algorithms that can mine effective information from a large number of databases, carry out pattern recognition and predict potential results. There are already some applications of artificial intelligence in the industry, but it is necessary to deploy artificial intelligence technology on a larger scale and faster in order to get rid of the dependence on fossil fuels quickly, safely and economically.

II. Application of artificial Intelligence in Energy Transformation

Artificial intelligence is a powerful tool that can cope with the complexity of global energy transformation and improve system efficiency, thereby reducing costs and speeding up the transformation. It is mainly used in four priority areas: renewable energy generation capacity and demand forecasting, grid operation and optimization, energy demand management, and material discovery and innovation. According to the German Energy Agency's analysis of artificial intelligence applications in the energy industry in 2020 (see figure 2), artificial intelligence applications can be divided into the following categories according to the data used: market, commodity and meteorological data, images and videos, device and sensor data. These applications are described in detail in the following sections.

Source: German Energy Agency Analysis (2020)

(I) forecasting the power generation capacity and demand of renewable energy

The application of artificial intelligence in renewable energy generation capacity and demand forecasting is mainly shown as 1-7 in f