Page 159 - AI for Good-Innovate for Impact

AI for Good-Innovate for Impact

36�2� Use case description

36�2�1 Description

Use case status: The use case is part of a larger research project use case description: The 36-US Sussex
shift towards renewable energy resources for environmental sustainability and achieving net-
zero has given rise to some challenges like instability and fluctuations in power supply, and
massive costs incurring in completely changing the infrastructure of electricity generation.
In this scenario, small scale distributed renewable energy sources are one of the low-cost
solutions to meet the increasing energy demands in an environment-friendly manner. This
Use case proposes the backup battery of Base Stations (BSs) as a distributed energy source
to be supplied in case of high demand. It explores the concept that BSs can take part in
achieving net-zero through bidirectional energy exchange with gird, i.e., Telecom-to-Grid (T2G)
and Electric Vehicles (EVs) i.e., Telecom-to-Vehicle (T2V). Well-established and thoroughly
analyzed solutions of bidirectional Vehicle-to-Grid (V2G) energy exchange are in practice but
T2G and T2V are amongst relatively emerging approaches. It can be further evaluated with
the help of energy-efficient AI approaches. The feasibility of the use case will be investigated
with accurate energy consumption modeling of BSs and EVs to forecast their demand and
supply, and optimization algorithms to optimally match buyer and seller BSs and EVs. The
existing energy consumption and optimization models of 5G BSs are limited. The AI based
models produce accurate results but are high power consuming themselves. We aim to work
on energy-efficiency of AI based energy consumption and optimization solutions. The solutions
to reduce power consumption of AI models such as quantization and Tiny Machine Learning
can produce models with low carbon footprints but may trade-off accuracy. This use case aims
to achieve optimal accuracy with energy-efficient AI solutions designed to achieve net-zero in
an electrified transportation network assisted by telecommunications.

Is it publicly available?: Yes

is it privately available?: No

Repository url: link
UN Goals:

• SDG 7: Affordable and Clean Energy,
• SDG 8: Decent Work and Economic Growth,
• SDG 9: Industry, Innovation and Infrastructure,
• SDG 11: Sustainable Cities and Communities

Justify UN Goals selection: Firstly, this use case encourages distributed peer-to-peer energy
exchange. EVs and BSs can sell their energy at times of high demand and later charge themselves
at reduced off-peak prices. The concept will help in achieving net-zero, specifically when the
supply from renewable sources is low and therefore it is aligned with Goal 7 of Affordable
and Clean Energy. Secondly, the use case brings a novel concept in telecommunications
industry and framework of energy exchange and matches with Goal 9 of Industry, Innovation
and Infrastructure. Unique approaches to propose energy-efficient AI solutions will further
contribute to innovation. Thirdly, the use case aims to achieve net-zero in EV transportation.
The access to energy for EVs will be easier as they can either travel to a charging station
or a BS. Less travel will further lead towards environmental sustainability. Therefore, the use

143