You are on page 1of 35

FUELL

CELLS

1
Contents

What are they

how do they work

types

Why FC

2
Types of FC

PEMF
SOFC AFC MCFC
C
low weight and volume
compared with other fuel cells. closely related to conventional PEM
very high temperatures—as high as fuel cells, except that they use an
high temperatures of 650°C
1,000°C alkaline membrane instead of an acid
membrane.

low temperatures, around 80°C


(quick start)

MCFCs do not require an external


reformer to convert fuels such as
development of low-cost materials the small amount of CO2 in the air natural gas and biogas to hydrogen.
with high durability at cell operating can dramatically affect cell At the high temperatures at which
temperatures is the key technical performance and durability due to MCFCs operate, methane and other
better durability challenge carbonate formation light hydrocarbons in these fuels are
converted to hydrogen within the
fuel cell itself

3
Your footer goes in here 4
components

PEM ANODE CATHODE FLOW PLATES

› The PEM is a thin, solid, organic › The anode is the electrode at › The cathode is the electrode at › They channel hydrogen and
compound, typically the which oxidation (loss of which reduction (gaining of oxygen to the electrodes
consistency of plastic wrap and electrons) takes place. In a fuel electrons) takes place. In a fuel
about as thick as 2 to 7 sheets of cell, the anode is electrically cell, the cathode is electrically › They channel water and heat
paper. This membrane functions negative. The anode is composed positive. The cathode is away from the fuel cell
as an electrolyte: a substance that of platinum particles uniformly composed of platinum particles
conducts charged ions (in this supported on carbon particles. uniformly supported on carbon
› They conduct electrons
case protons), but does not The platinum acts as a catalyst, particles. The platinum acts as a from the anode to the
conduct electrons. This allows increasing the rate of the catalyst, increasing the rate of the electrical circuit and from
the solution to conduct oxidation process. The anode is reduction process. The cathode is the circuit back to the
electricity. This membrane must porous so that hydrogen can pass porous so that oxygen can pass cathode.
be kept moist to conduct particles through it. through it.
through it.

5
6
PEM
positively charged protons pass through membrane, The negatively
charged electrons must flow through an external circuit. This flow
of electrons forms an electrical current.

The
Chemical
Process

H2 and Anode
catalyst separates the hydrogen’s charges

Cathode
negatively charged electrons and positively charged hydrogen ions
(protons) combine with oxygen to form water (H 2O) and heat.
Your footer goes in here 7
System Components
Therefore, individual fuel cells are typically
combined in series into a fuel cell stack

FUEL CELL
STACK
FUEL
Power conditioning includes controlling P.
current (amperes), voltage, frequency, and POWER PROCESSOR
other characteristics of the electrical current CONDITIONER
to meet the needs of the application. Fuel cells
produce electricity in the form of direct
current (DC).

AIR
HUMIDIFIER
COMPRESOR
Fuel cell performance improves as the
pressure of the reactant gases increases. It Membrane must be constantly humid
raises the pressure to 2–4 times the ambient
pressure.

8
Artificial Intelligence in Aviation
AI applications across the lifecycle of an
aircraft
Complex optimizations
AI Benefits

Real time decision making

Predictive Analytics

Rapid deployment on large scale possible

Compatible with current and future fleet


Why Aircraft
Operations?

Environmental benefits + cost savings

Possibility of keeping human in loop -> less resistance from


unions and regulatory bodies
Flights safer with less delays -> more
customer satisfaction

9
Artificial Intelligence in Operations

Flight Route
Optimization
Air Traffic
Management

Predictive
Maintenance

10
AI for Flight
Route Optimization

11
About Flight Route Optimization
Determining the most efficient route for an aircraft to travel from one point to another, taking into
account various factors such as weather conditions, air traffic congestion, and fuel consumption

What is What is
the best the wind
altitude? today?

Is there any
military
training?

“You have a tab open for the Weather Channel, a tab for CNN. So it’s just click, click, click. If you look at a dispatcher,
they have, like, 19 tabs open that they’re flipping [through].” - Alaska Airlines’ Head of Corporate Development, Pasha
Saleh

12
Case Study: Alaska Airlines Flyways
What is it? What did it do?
Utilizes machine learning for diverse data Miles and fuel consumption reduction
analysis, including weather and air traffic. for 64% of mainline flights.

Suggests flight routes considering factors Avoided 4,600 tons of carbon emissions
like wind and altitude. lowering fuel costs

Learns from human decisions, providing Saved 480,000 gallons of fuel in six months
real-time alerts and recommendations.

Reduces fuel usage, shortens flight times, Dispatchers evaluated and accepted 32% of
and lowers carbon emissions. AI recommendations for optimized flight
routes
Data taken from: https://aerospaceamerica.aiaa.org/features/ai-how-its-delivering-sharper-route-planning/ 13
Industry Leaders in AI-Driven Flight Route Optimization

14
How is it being done ?

15
AI Route Planning Techniques

16
AI for Predictive
Maintenance

17
Statistical Overview: Impact of PM
Predictive maintenance refers to the use of data-driven and predictive analytics methods that are designed to
analyze the condition of equipment and help predict when maintenance should be performed, at the most convenient
and most cost-efficient moment, allowing equipment’s lifespan to be optimized to its fullest.

Airlines face the challenge of enhancing the availability of their fleet by avoiding flight delays
and cancellations, consequentially reducing costs to be able to support the forecasted growth of
38000 aircraft by 2025 (Lufthansa Technik, 2020).

Could generate about $3 billion of savings for the MRO industry.

IATA estimates the global cost of irregular airline operations (delays, cancellations, in-flight turn
backs, etc.) is $28B.

Past studies reported by the US Department of Energy have estimated that a predictive maintenance program
could realize an 8% to 12% savings over a preventative only program.

[13] [14] 18
AI-Powered Visual Inspections
Environmental effects

Smart image processing programs


Extended equipment life. By detecting the
Trained AI to recognize physical first signs of equipment weakness, allows
signs of repair needs experts to restore it using often minimal
repairs and to avoid premature aging.
Quality assurance
Automatized use of computer algorithms
Defect detection reduces energy-intensive
• Get signs of potential maintenance procedures
maintenance problems
• Damage assessment
• Pre-flight and post-flight checks
• Detect issues before they occur

19
Data analysis for predictive maintenance
Environmental effects
Takes insights from performance and maintenance
data
Data analysis and pattern recognition through Material recycling according to waste
continuous monitoring hierarchy. It can prolong products' lives.
Capture information to identify anomalies, anticipate Decreases the occurrence of corrective
breakdowns maintenance.
Example: Mechanical problems caused by repairment parts Reduces energy-intensive activities to do
fast repairs
The malfunction of parts leads to
overconsumption of energy.
Preventive maintenance helps design
optimized parts to reduce the ecological
impact.

[11]

20
Data analysis Examples
Given that with today’s aircraft, including thousands of sensors — the Airbus A350 has nearly 250,000 of
them, generating about 2.5 TB of data per day (Airbus, 2020) — sifting manually through all that data and
getting actionable information would be overwhelming.

Continuous monitoring (performance


& pattern recognition);

Specific occurrence in time (predict


based material longevity, condition
monitoring eq. (usual weather
conditions ac flights under):
How much can it flight still safely
still?).
18% of assets have an age-related failure pattern,
while a full 82% of asset failures occur randomly
[13]

[13] [14] 21
Systems Health Monitoring

Propulsion Avionics System Airframe Fuel Hydraulic


System Structure System Systems
Flow Rate Pressure &
Vibration Sensor Health Monitoring Temperature Landing Gear
Monitoring Strain Tension (issues w/ fuel delivery Monitoring
Monitoring
(unusual ones) & amount of fuel for Shock Absorber
opt. performance) Sensors
Environment
Communication (amount of force
Oil analysis Thermal Imaging Control System experienced during
(levels of System Diagnosis Fuel Quality landings)
(variation in surface Temperature &
contamination or wear (signal strength, clarity, (contamination)
particles) temperature) Humidity
and data integrity)
Sensors

22
Monte Carlo Predictive Maintenace System
Monte Carlo simulations in aircraft predictive maintenance provide a probabilistic approach to understanding
and managing uncertainties and risks associated with various factors affecting the health and performance of
aircraft components.

Parameter Variability: material strength, wear rates, and manufacturing tolerances.

Time to Failure Modeling: considering uncertainty in input parameters.

Supply chain variability + part availability.

False Positive/Negative rates.

Optimal Allocation of Recourses


Most Cost-effective Strategies
Optimal Maintenance Timing

[16] [17] 23
Advantages of AI in Predictive Maintenance

Cost-effectiveness Efficiency Safety

› Reduce › On resources › Aircraft equipment


unscheduled maintenance, › Processes. No need to liability. Passengers and crew
reducing aircraft downtime access inaccessible part safety
› Optimization and management › Detects problems › Lessen de workload of
of fleet. Better life expectances before they get worse maintenance engineers
› Prevents maintenance check › More time to do essential › More accuracy on
› Predict repairs determination of condition of
maintenance procedures, the aircraft
› Continuous monitoring
optimizing inventory
› Quicker decision making
management.

24
Disadvantages & Future for PM
Data Quality & Availability: Insufficient Historical Data & Data Accuracy;
Problematics & Challenges

Interconnected Systems: Deep Understanding of interactions between various systems;


System Integration: may require great engineering effort; Impact on Environment:
1) Higher Energy
Regulatory Challenges: new technologies require approval from certification entities; Consumption;
2) Exploitation of resources
Initial Investment in equipment & training and skill development; (innovation & adaptability
of infrastructure)
Unkown Failure Modes;
Cybersecurity Concerns: increased connectivity increases threats, unauthorized
access & integrity of data. Long
term investment!
Future

New sophisticated technologies: augmented reality in aviation maintenance!

[15] 25
AI for Air Traffic
Management

26
Projects and areas of implementation

ATM can make use of AI and higher levels of automation to improve the efficiency of their operations in many ways
and allow human operators to focus on safety-critical tasks: cutting costs (and delays), help maintain situational
awareness and plots a safe routes.

EUROPE Main SESAR ATM reseach


areas:
ATM Operations,
deliver a fully digital Enabling Aviation
SESAR Architecture,
scalable TMS transformation Infrastructure
(partnership Performance, and
capable of of the essential role (EAI)
between private Validation (OAPV)
handling underlying of AI
and public
growing air infrastructure
sector) [19]
traffic system High-Performing Advanced Air
Airport Operations Traffic Services
(HPAO) (AATS)

27
About ongoing projects EUROCONTROL
fully committed to support the acceleration of AI adoption in European aviation, and
more specifically in air traffic management [19]

HEATHROW AIRPORT
Extensive array of cameras to detect aircraft and other airfield objects
and provide a detailed, unified view on controllers' digital monitors [21]

THALES HONEYWELL
Efficient approach, landing and taxi operations in
Uses AI to predict traffic flow, optimal routings, and estimated take-
conditions of low visibility.
off and arrival times.
Services in over 85 locations
[20]

ARTIMATION
Algorithms lab-tested by Mälardalen University and ENAC. The result was an XAI
prototype which increased user understanding of the stages leading up to a critical
decision, alongside the underlying logic driving that process ,

AIRBUS
optimize the performance and efficiency of airlines and airports and
automatic landing and take-off,

28
AI in Air Traffic Management for Contrail Reduction

Contrail Impact Current Research


• Capable of trapping • Adoption of engines that
considerable amounts produce fewer emissions
of heat, leading to net and the use of alternative
warming effect fuels, current focus to
rerouting
• 35% of the industry's
global warming • Lateral and vertical
contribution changes in flight
trajectories using genetic
• Worse than jet fuel's algorithms
impact
• Analysis is not automated
and requires an expert
integration

29
Proposed Study Over European Airspace
Proposed Study Over European Airspace

Use a multi-objective genetic algorithm

Simulate one day of Europeans Air Traffic

Assessing impact of rerouting for contrail


avoidance

Assessing additional parameters like airline costs,


ATC workload and other environmental impacts

[12]

AI-powered flight rerouting offers an immediate and impactful solution to reduce aviation's climate impact

30
Challenges of the Implementation
Trade-offs

Balancing the trade-off between flight duration and contrail distance is a challenging task. While it's
possible to significantly reduce contrail distance, this often leads to longer flight times.

Environmental and Economic


Implications

The extended flight times contribute to increased fuel consumption, exacerbating global warming
and raising economic costs for airlines striving to minimize operational expenses.

Complexities in European
Airspace
European airspace, constrained by capacity limitations, struggles to optimize fuel-efficient flight
paths amid challenges like congestion, safety regulations, adverse weather, and operational
constraints. Rerouting increases controller taskload, complicating cost analysis.

31
Possible Problems of AI

Data Privacy and Security concerns

Biasing and Discrimination based on data

Transparency and Accountability Issues

Job Displacement problems

32
Conclusions

Global Emissions AI Focus in


Anticipated Challenges
Impact Aircraft Operations
› Route Optimisation › Airspace Capacity: congestion,
› Aviation contributes to
3% of global emissions › Predictive Maintenance safety regulations, increased
› Air traffic management controller, adverse weather,
› Urgent action is taskload
required to address › Artificial intelligence
environmental concerns provides a prompt and › Potential environmental
effective solution to impacts: Higher energy
› AI is a powerfull tool to
minimize aviation's consumption. Resource
provide quick solutions exploitation (innovation and
climate impact – tried
and tested! adaptability of infrastructure)

33
References
[1] https://ts2.space/en/ai-takes-flight-the-science-behind-flight-path-optimization/#gsc.tab=0
[2] https://www.airbus.com/en/newsroom/press-releases/2021-09-airbus-and-partners-target-more-energy-efficient-flights
[3] https://www.dubaiairshow.aero/artificial-intelligence-transforming-aviation-industry
[4] https://www.triplepundit.com/story/2021/artificial-intelligence-air-travel/723586
[5] https://ts2.space/en/the-greening-of-aviation-how-ai-flight-path-optimization-is-reducing-environmental-impact/#gsc.tab=0
[6] https://airwaysmag.com/ai-in-aircraft-maintenance/
[7] https://www.odysight.ai/articles/aviation/machine-learning-use-cases-for-condition-based-monitoring-and-predictive-maintenance-in-aviation/
[8] https://www.qoco.aero/blog/ai-in-aviation-maintenance-how-its-changing-the-industry
[9] https://www.amiraltechnologies.com/en/news/blog/how-does-predictive-maintenance-have-a-positive-impact-on-the-environment/
[10] https://www.clarifai.com/blog/what-is-ai-powered-visual-inspection
[11] https://aviation.stackexchange.com/questions/32630/why-do-jet-engines-smoke
[12] Yin, Feijia & Volker, Grewe & Frömming, Christine & Yamashita, Hiroshi. (2018). Impact on flight trajectory characteristics when avoiding the
formation of persistent contrails for transatlantic flights. Transportation Research Part D Transport and Environment. 65. 10.1016/j.trd.2018.09.017.
[13] Opportunities for Explainable Artificial Intelligence in Aerospace Predictive Maintenance Bibhudhendu Shukla, Ip-Shing Fan, and Ian Jennions,
Cranfield University
[14] https://www.heavy.ai/technical-glossary/predictive-maintenance
[15] https://toolsense.io/maintenance/why-predictive-maintenance-in-aviation-can-save-lives/
[16] Robust long-term aircraft heavy maintenance check scheduling optimization under uncertainty, Tim van der Weide, Qichen Deng
[17] The Monte Carlo Simulation Rustom D. Sutaria – Avia Intelligence 2016, Dubai
[18] “Building trust in air traffic management AI,” CORDIS | European Commission. Accessed: Nov. 14, 2023. [Online]. Available:
https://cordis.europa.eu/article/id/442199-building-trust-in-air-traffic-management-ai
[19]“CORDIS Results Pack on AI in air traffic management – Bringing intelligent and trustworthy automation to Europe’s aviation sector”.
[20] “How AI Makes Air Traffic Management More Predictable and More Efficient.” Accessed: Nov. 14, 2023. [Online]. Available:
//interactive.aviationtoday.com/avionicsmagazine/november-december-2022/how-ai-makes-air-traffic-management-more-predictable-and-more-efficient/

34
THANK
YOU
Questions?

35

You might also like