Kinetic Energy Recovery from a landing aircraft: Evaluating Onboard Energy Solutions

This paper compares onboard Energy Storage Solutions (ESSs) for a Kinetic Energy Recovery System (KERS) from a landing aircraft. Energy is stored temporarily and reused so that it enables engine-less taxiing. This paper evaluates the choice of onboard Energy Storage Solutions (ESSs) (flywheels, batteries and supercapacitors) for recovering energy during the landing roll and storing it in the device. A design of an ESS with each of the three technologies was made, using commercially available products. The resulting devices are compared on the basis of weight, charging time, discharging time and complexity in retrofitting to existing systems. Results shows that while batteries have the highest energy density and will have the lowest weight, they are unable to charge/discharge quickly enough to satisfy this application. Conversely, supercapacitors have this ability but their low energy density make them heavy which in turn would offer penalty to the aircraft in flight. Flywheels emerge as the most interesting proposition due to their high energy density and fast charging ability, which satisfy the requirements for application.


Introduction
With the ever-increasing pressure from climate change, the European Union has set targets to reduce emissions from air travel, while maintaining its social and economic benefits.In the last decade, R&D targeted the airborne phase.However, there is now the intent to extend the requirement for greener aviation within the airport boundaries.This requires all aircraft movements on the ground to be emission-free by 2050 [1].Airport carbon footprint analysis and accreditation demonstrate that aircraft ground movement accounts between 5-20% of all airport emissions [2].Conventionally, engine thrust is used during taxiing to provide forward propulsion.A typical 10-minute taxiing process for a narrow body aircraft consumes approximately 100 kg of fuel with a considerable amount of carbon and NOx pollutants released at ground level.The reduction of emissions on the ground is important as it has strong links with respiratory illnesses, amongst others.As airports and cities continue to grow, these get in closer proximity to each other, heightening the effects of the problem.
The environmental impact of the various taxiing techniques with reference to the baseline standard taxiing method was studied in [3].Onboard technologies were shown to have lower taxiing emissions than the use of external taxiing such as tow trucks.Onboard solutions also offer fewer logistical challenges to implement and allow aircraft to maintain their autonomy in airport operations.This characteristic is preferred by airlines which are keen to remove dependencies.However, studies on electrical taxiing showed that while motor technology was viable, the auxiliary power unit (APU) had to be redesigned such that the generator would be able to supply sufficient electrical power to the inwheel motors [4].Changing the APU would result in a costly retrofit.A new concept was hence developed which proposes that a portion of the kinetic energy of the aircraft is recovered during the landing rollout.The energy recovered is stored temporarily into an onboard energy storage device, and is then transferred back to the wheels for engineless taxiing.A schematic of the system components and the energy flow between them is shown in Figure 1.We consider a slow-speed, high torque traction motor (TMs) installed at each wheel of the main landing gear (MLG).The TMs are mechanically coupled to the wheels and electrically coupled via a power electronic converter to the energy storage device.In our earlier work [5] we demonstrate that there is enough energy from a landing aircraft to enable a taxiing process of around 5 minutes.In Low-cost carriers, particularly those operating from secondary airports show a very promising potential to benefit from such technology.These tend to operate short flights with a high turn-around time allowing them to operate the aircraft up to 4 flights daily.Thus, they provide a unique opportunity to turn an existing emission problem into a green solution with multiplier effects on environment and health.The economic study showed that the due to the fuel savings during taxiing, green technology does not need to be a financial burden but presents economic incentives to airline operators.The study presented here investigates the type of energy storage device that would be most adequate for this application.The paper is therefore organised in the following manner.Section 2 provides an overview of the system requirements, highlighting the energy that needs to be stored and the response time required.Sections 3, 4 and 5 present the sizing models developed for each of the energy storage technology: flywheels, batteries and supercapacitors respectively.Section 6 provides a discussion, comparing the technologies and draws a conclusion.

System requirements
Having described the concept, this section focuses on the overall system requirements.The energy storage device would be required to store 50 MJ of energy [5], which will be then used for taxiing.The energy storage device needs to be charged during the landing roll of the aircraft, whose duration is typically 30 s.The power and energy flow between the main landing gear and the energy storage device was presented in [6].The results showed that the most used setting for normal operation, with LO (low) autobrake deceleration setting, spoilers (for the A320) and no reverse thrust, recovers enough energy to provide sufficient energy for the typical taxi time for both taxi-in and taxi-out for majority of airports.The main challenge for energy capture is the short time available during the aircraft roll out.It is therefore crucial to size the energy storage component adequately.The design process for the ESS is shown in Figure 2. In the coming sections, the design of a flywheel energy storage, battery energy storage and supercapacitors for this application are produced.

Design of a Flywheel Energy Storage Technology
A flywheel stores energy through its rotation.The energy stored in a flywheel is described by: where   is the kinetic energy stored in the flywheel. is its rotational speed and I is the moment of inertia of the flywheel.This is described by: where m is the flywheel mass and r the radius.The weight efficiency of a flywheel is defined by its energy density e, which is described as: where K is the shape factor of the flywheel. is the material tensile stress and the  is the material density.By applying eqn.( 3) to various shapes and materials, it can be shown in Figure 3 that the laval disc offers the highest energy density.In his work, [7] shows that a material combination of M46Jepoxy for the constant thickness material and T1000-epoxy for the constant stress material provides the highest energy density.This is therefore chosen as the base model.The geometry of a simplified laval disc is shown in Figure 4.The disc was therefore sized so that it satisfies the energy requirement of 50MJ for this application.In his work, [7] shows that the mass of the disc can be defined as: where ha is the height of the rim, ho is the height of the disk.ra is the radius of the disc.rb is the radius of the disc with rim.  ,   and   are the material density, poisons ratio and Youngs Modulus of elasticity for the disc.Likewise,   ,   and   are the material density, poisons ratio and Youngs Modulus of elasticity for the rim.B is a ratio of height defined as: [8] found that the optimal design can be achieved when the ratio of ho/ ha is fixed to 2. The moment of inertia of the disc can be defined as: (6) By varying the dimension of the disc (and thus varying the mass and moment of inertia) and using eqn. 1, the required rotational velocity of the disc for each design to store 50 MJ could be found.This is presented through the carpet plot shown in Figure 5. Due to limitation in bearing technology, a limit to the flywheel speed was set to 60,000 rpm.Likewise, a limit to the mass of the flywheel was arbitrarily set to 50 kg.The geometric configuration of the selected flywheel (marked with a dot on Figure 5) is shown in the insert.Flywheels can be mechanically coupled or electrically coupled.In this application, a mechanically coupled flywheel will require a direct connection to the aircraft landing gear.A reduction gear will be required to increase the speed of the flywheel to 60,000 rpm.Packaging such a device close to the landing gear might be very challenging.On the other hand, an electrical flywheel will be coupled to a high-speed electrical machine which will convert the kinetic rotational energy to electrical energy and vice versa.Likewise, a slow speed electrical machine will be fit to the main landing gear to convert the electrical energy into rotation of the aircraft wheels, and vice versa.The electrically coupled flywheel will also provide greater flexibility in packaging the flywheel onboard the aircraft.In their work, Lian et al. [9] showed that a high-speed electrical machine for the flywheel would weigh just under 5 kg.The system would operate at 400 V and independently to other aircraft electrical subsystems.Considering an added 10 kg which are arbitrary taken to adjust for the coupling of the flywheel to the electrical motor, additional housing and accessories, the total mass of the energy storage system using an electrically coupled flywheel would be 65 kg.Having designed the dimensions of the flywheel, the focus of the next section is now turned to the design of an energy storage comprising of battery technology.

Design of a Battery Energy Storage Solution
Batteries convert electrical energy and store it as a chemical energy.Batteries can be divided into the following two types: • Primary batteries are designed to be used until their energy is exhausted.Their chemical reactions are generally not reversible and hence they cannot be recharged.• Secondary batteries can be recharged which means that their chemical reactions can be reversed by applying electric current.
This research focuses on Li-ion secondary batteries.Large battery packs are built of a number of battery cells, put together in series and parallel.The chemical energy E stored in a battery is described as: where V is the overall voltage of the battery pack, I is the current and t is the time.
To compare the mass for the various energy storage technologies, the energy density for each has to be found.The energy density for a battery is defined as: where m is the mass of the battery.Conversely, the power density for a battery is defined as: Using a library of current state of the art battery cells, the cell with highest energy density was selected.Its properties are shown in Table 1.To establish the battery pack geometry that would provide with the required energy (50 MJ equivalent to 14 kWh), a sweep of a number of cells stacked in rows and columns was done.This allows us to find the various configurations that would satisfy the energy requirement, and identify the one with the highest energy density.The result is shown in Figure 6 (left).The geometry of the optimal configuration is shown in Figure 6 (right).Using a library of current state of the art battery cells, the cell w ith highest energy density w as selected.Its properties are show n in Table 1.To establish the battery pack geometry that w ould provide w ith the required energy (50 M J equivalent to 14 kWh), a sw eep of a number of cells stacked in row s and columns w as done.This allow s us to find the various configurations that w ould satisfy the energy requirement, and identify the one w ith the highest energy density.The result is show n in Figure 7 (left).The geometry of the optimal configuration is show n in Figure 7 (right).It can be seen that the battery pack w ill be made of 1116 battery cells and w ill have an overall dimension of 1.1m x 0.27m.The w eight of the pack is 52 kg.Considering a 1 mm aluminium protective housing around the battery pack w ould add approximately another 2 kg.In order to choose the series/parallel electrical configuration, the charging time should be taken into consideration.Figure 8 (left) show s the analysis the time taken to charge he pack w ith the maximum charging rate limitations imposed by each cell.Likew ise, Figure 8 (right) show s the potential configurations for this pack, highlighting the one satisfying the charging time.It can be seen that for a charging of about 60 s, a configuration of 1116p, 1s is required.A t this configuration the battery pack It can be seen that the battery pack will be made of 1116 battery cells and will have an overall dimension of 1.1m x 0.27m.The weight energy density is 269 Wh/kg with the mass of the pack being 52 kg.A total of 1116 cells are required.Considering a 1 mm aluminium protective housing around the battery pack would add approximately another 2 kg.An additional 10 kg were added for BMS and cooling.In order to choose the series/parallel electrical configuration, the charging time should be taken into consideration.Figure 7 (left) shows the analysis the time taken to charge he pack with the maximum charging rate limitations imposed by each cell.Likewise, Figure 7 (right) shows the potential configurations for this pack, highlighting the one satisfying the charging time.It can be seen that for a charging of about 60 s, a configuration of 1116p, 1s is required.At this configuration the battery pack will need to be 4018 V and 3.5 Ah.The process was repeated with an alternative 21700 cell which allows faster charging rates at 7.5 C.This was found to be able to charge in approximately 30 s.The pack voltage is reduced to approximately 1800 V but the mass of the pack increases to 142 kg

Design of a Super Capacitor Energy Storage System
Having designed the geometries for the flywheel and battery energy storage system, this section will focus on the design of a super capacitor energy storage system.A capacitor is a device for storing an electrical charge in close proximity to each other.Supercapacitors on the other hand combine the workings of capacitors and batteries by storing electrical energy using two mechanisms: double-layer capacitance (which is electrostatic) and pseudo-capacitance (which is electrochemical).The energy E stored in a capacitor is described as: where C is the capacitance and V is the voltage across the capacitor.Using a library of current state of the art supercapacitors, the one with highest energy density was selected.Its properties are shown in Table 2.A similar technique used in battery was used here, in which a sweep with a number of units stacked in rows and columns was done.This allows us to find the various configurations that would satisfy the energy requirement, and identify the one with the highest energy density.The result for the supercapacitor bank is shown in Figure 8 (left).The geometry of the optimal configuration is shown in Figure 8 (right).It can be seen that the super capacitor bank will be made of 3330 units and will require an overall dimension of 1.5m x 6.8m.The Weight energy density is 8.7 Wh/kg with the overall mass of the supercapacitor bank would be 1631 kg.Considering a 1 mm aluminium protective housing around the battery pack would add another 70 kg.In order to choose series/parallel electrical configuration, the charging time is also taken into consideration.Figure 9 (left) analysis the potential configurations for this pack.Likewise, Figure 9 (right) analysis the time taken to charge he pack with the maximum charging rate limitations imposed by each cell.It can be shown that a charging time under 30 s is possible using a 3s110p configuration.

Discussion and Conclusion
In the previous sections, the process to design three energy storage systems made up of a flywheel, battery pack or capacitor bank respectively was shown.The design of each of the three technologies was based on commercially available products.Each system is required to satisfy the energy demands of a landing aircraft thus enabling the aircraft to recover 50 MJ of energy in under 30 seconds, store it temporarily and use it to power a 5 engine-less taxiing.Table 3 shows a comparison between high- In order to choose the series/parallel electrical configuration, the charging time is also taken into consideration.Figure 11 (left) analysis the potential configurations for this pack.Likew ise, Figure 11 (right) analysis the time taken to charge he pack w ith the maximum charging rate limitations imposed by each cell.It can be show n that a charging time under 30 s is possible using a 3s110p configuration.

Discussion and Conclusion
In the previous sections, the process to design three energy storage systems made up of a flyw heel, battery pack or capacitor bank respectively w as show n.The design of each of the three technologies was based on commercially available products.Each system is required to satisfy the energy demands of a landing aircraft thus enabling the aircraft to recover 50 M J of energy in under 30 seconds, store it temporarily and use it to pow er a 5 engine-less taxiing.It can be seen that while the battery pack has highest weight energy density and the lowest mass at 54 kg.However, despite the extremely high operating voltages, this is still unable to recharge effectively in the time frame synonymous with the landing roll of the aircraft.Conversely, supercapacitors have a very high-power density but poor energy density.Thus, while they can be charged very quickly, a supercapacitor bank will weigh thirty times more than the battery pack at 1631 kg.Finally, the flywheel shares an energy density that is similar to battery technology but with the power density closer to that of supercapacitors.The overall mass of a flywheel energy storage would hence be approximately 65 kg with the ability to charge within the landing roll of the aircraft.The electrical coupling would require the design of a high-speed machine which can be designed to operate at acceptable voltages.The authors are hence of the opinion that in this context, a flywheel energy storage device would be better suited.This paper demonstrated a concept whereby the kinetic energy from a landing aircraft can be recovered and stored temporarily in a device to allow engineless taxiing.The paper shows that it is technically feasible to achieve this by designing an energy storage system for three different technologies.Flywheels emerge as the most interesting proposition due to their high energy density and fast charging ability, which satisfy the requirements for application.

Figure 1 .
Figure 1.Schematic showing the concept of a kinetic energy recovery system.

Figure 2 .
Figure 2. Flowchart for Sizing of the energy storage system

Figure 3 .
Figure 3.Comparison for of the energy density for various flywheel shapes and materials.

Figure 4 .
Figure 4. Schematic of the laval disc geometry.

Figure 5 .
Figure 5. Carpet plot of the resulting laval disc dimensions, mass and rotational velocity.(Insert) Schematic of the resulting laval disc geometry marked with a blue dot.

Figure 6 .
Figure 6.(left) Sweep of cells stacked in various rows and columns, to identify the configurations that satisfy the energy requirement.(right) Overall geometry of the battery pack to satisfy the energy requirement of 14 kWh.

Figure 7 :
Figure 7: (left) Sw eep of cells stacked in various row s and columns, to identify the configurations that satisfy the energy requirement.(right) Overall geometry of the battery pack to satisfy the energy requirement of 14 kWh.

Figure 7 .
Figure 7. (left) Charging time for each electrical configuration.(right) Electrical configurations for the battery.

Figure 8 .
Figure 8. (left) Sweep of super capacitors bank stacked in various rows and columns, to identify the configurations that satisfy the energy requirement.(right) Overall geometry of the super capacitor bank to satisfy the energy requirement of 14 kWh.(insert) packaging details of the super capacitors.

Figure 9 .
Figure 9. (left) Charging time for each electrical configuration.(right) Configurations for the capacitor bank.

Figure 10 :
Figure 10: (left) Sw eep of super capacitors bank stacked in various row s and columns, to identify the configurations that satisfy the energy requirement.(right) Overall geometry of the super capacitor bank to satisfy the energy requirement of 14 kWh.(insert) packaging details of the super capacitors.

Figure 11 :
Figure 11: (left) Charging time for each electrical configuration.(right) Configurations for the capacitor bank.

Table 1 .
Properties of the battery cell.

Table 2 :
Properties of the super capacitor. 7

Table 3
show s a comparison betw een the highspeed, electrically-coupled flyw heel, lithium-ion batteries pack and ultracapacitor bank.It can be seen that w hile the battery pack has the highest w eight energy density and results w ith the low est w eight at 54 kg.H ow ever, despite the extremely high operating voltages, this is still unable to recharge effectively in the time frame synonymous w ith the landing roll of the aircraft.Conversely, supercapacitors have a very high-pow er density but poor energy density.Thus, w hile they can be charged very quickly, a supercapacitor bank w ill w eight thirty times more than the battery pack at

Table 3 :
Comparison of the energy storage technologies.