Energy
AI4Cities challenges suppliers to come up with AI solutions that reduce CO2 emissions in the energy field. Energy – crucial for cities’ infrastructure, heating, cooling, security and light – is the biggest polluting source, accounting for 58% of GHG emissions in Europe's urban areas. In this context, AI technologies can be issued for storage needs, more efficient energy uses and the integration of renewable sources. Furthermore AI can help citizens to better understand how their energy consumption impacts their carbon footprint and give them tools that can facilitate the change.
Subchallenges
The purpose of these sub-challenges is to describe and give examples of the cities’ current needs. It is not expected that a solution will cover every issue mentioned under a sub-challenge. On the other hand, solving only one of the issues listed may not be enough for the solution to advance to the later phases of the PCP. Tenderers are encouraged to address several problem areas mentioned under one or multiple subchallenges.
The increased use of renewable energy, combined with fluctuating production patterns, increasing electrification of e.g. vehicle fleets, and heat production is a challenge to the current electrical grid. If cities are to make use of renewable energies there is a need for shifting the energy consumption patterns to accommodate for the fluctuations as well as flattening the peak loads during the day. This supports the wider grid to use renewable energy when it is abundant and to reduce consumption when electricity (or heat) is produced from fossil fuels. There is a substantial need to renovate and upgrade the current electrical grid across Europe to adapt to this new reality so it allows a flexible use of energy by turning energy consuming equipment on and off based on signals from the energy market. Solutions to reduce investments in electrical infrastructure and to better utilise fluctuating energy production can both promote the green energy transition and save energy costs.
Addressable problem areas (non-exhaustive list):
- Flexible electricity supply and usage
Locally produced renewable energy could be utilised in the smart grid, thus allowing excess energy to be fed to other parts of the city, thereby reducing the need for major energy providers to produce energy based on fossil fuels.
- Energy flexibility in buildings
Energy usage (e.g. lighting and heat) should take into account how and when a building is used. Too often buildings are illuminated and heated at a standard that does not fit the actual usage or when no one is even in the building. Flexible heat and electricity consumption are relevant instruments to avoid consumption peaks while at the same time avoiding negative effects to the comfort levels of the users.
- Flexibility in electric vehicle charging stations
The use of a flexible energy consumption can also affect electric vehicle charging infrastructure in terms of charging patterns and charging locations without compromising comfort of having the vehicles charged when needed. This also impacts the Mobility challenge of AI4Cities.
- Interaction with market stakeholders
Using predictive user behaviour market stakeholders (such as utility suppliers and assets owners) are knowledgeable as to the needed energy supply at given times, which they can then forecast in terms of probable exceedances of renewable energy production. At these times the stakeholder could incentivise the users (such as workers and citizens) to shift their usage to other times, thus limiting the production of energy through fossil fuel, while analysing the effects on comfort levels (e.g. being indoor climate) and the structural integrity of the asset (e.g. the building construction not suffering damages).
It is hard to meet cities’ current energy needs through only renewable energy sources in a timeframe given by cities' climate goals. Therefore, it is essential that cities use energy more efficiently by enabling deeper analysis for all technical systems at a city scale. The idea of energy efficiency is to make better use of energy without decreasing the quality of life by prioritizing essential energy needs and reducing the amount of energy needed to satisfy current needs. Solutions to improve energy efficiency in building portfolios and similar large energy consuming assets can lower the overall energy consumption and reduce the usage of peak power plants that produce CO2.
Addressable problem areas (non-exhaustive list):
- Energy refurbishment prioritization
Prioritising strategies for the refurbishment of buildings can be hard to navigate, as it is hard to find the most effective strategz. Using existing data on the building materials and various other data of the actual usage of buildings could provide a prioritization tool of efforts with regard to energy savings, including physical climate screening such as windows, roofs and walls.
- Optimising energy usage of buildings and assets
Every technical installation within a building uses energy. While most receive scheduled maintenance, different needs based on activity levels and actual usage are often not considered. At the same time the energy profiles of buildings are not taking various public (e.g. computers) or private (e.g. mobile phones) assets into consideration even though they produce a lot of energy consumption. The consequences of inefficient maintenance and unaccounted assets using energy are higher CO2 emissions and higher economic costs as the installations run ineffectively. Using AI, it may be possible to define indicators that can inform and engage procedures that optimise the energy usage or redefine the energy profiles of the building by taking into account the assets.
- Predictive maintenance
Every larger technical installation and building has its own Management System, which gathers large amounts of operational data. When these systems send signals that some equipment is failing to operate it can be hard to pinpoint where the problem is. It is therefore needed to create smart maintenance procedures that define indicators on specific installations for when maintenance is needed, and automatically engage a maintenance procedure thereby reducing energy needs to the planned levels.
- Lack of energy efficient construction of buildings
During the construction of new buildings (including materials used) huge amounts of energy are used. Construction projects could often be much more energy efficient.
- Citizens and other private building owners are not aware how and/or lack motivation to use energy more efficiently
Majority of the estates and houses in the cities are privately owned and have a need for light and easily maintained energy consumption optimisation solutions. Raising the awareness of citizens and engaging them in the low-carbon city transition will also enhance the social acceptance of interventions and maximize environmental impacts.
- Industries are not aware how or it is not profitable for them to use energy more efficiently
Industries use vast amounts of energy and they might not always have the incentives or the awareness to be more energy efficient.
Merely increasing energy efficiency and reducing energy consumption is not enough for cities to reach carbon neutrality. In addition, it is important to develop and integrate renewable energy sources into the grid in order to gradually replace fossil fuels.
Addressable problem areas (non-exhaustive list):
- Improving storage of renewable energy
- Sub-optimal integration of renewable energy within cities’ buildings and other assets
- High costs of renewable energy (including the investment costs of building and installing RE facilities) and long term return on investment
- Mapping of potential renewable energy sources
- Help in creating adaptive strategies for integration of renewable energy in the city
- Lack of opportunities for renewable energy trading at a small scale (e.g. buildings, private energy producers)
AI4Cities has also reserved a possibility for the suppliers to independently propose a solution for an energy challenge that we have not defined. In their bid, the suppliers must outline the specific sub-challenge that they have identified and the proposed solution. The solution must still satisfy the general criteria: CO2 and other GHG emissions reduction and the use of AI technologies.