NLR and Wake Turbulence Separation related … and Wake Turbulence Separation related research, June 2016 8 • Generalized EDR –Based on PIREP Turbulence Intensity scale –In combination

NLR and Wake Turbulence Separation related research

FSS Wake workshop, Braunschweig, June 2016 І Gerben van Baren, [email protected]

NLR and Wake Turbulence Separation related research, June 2016 2

NLR’s involvement in European research on reduced separation concepts (2000 onwards)

Concept Development

System support

Safety metrics

Safety criteria

Safety Assessment

Safety Case

Flight / RTS

Simulations


Highlights of recent research

• Universal Atmospheric Hazard Criteria

• Wake Impact Flight Simulation campaign

• Optimised Runway Delivery Study

• Aircraft performance Model development

• Data analytics & performance monitoring

• FSS P4 Total System Risk Assessment









What are good metrics and criteria to measure

and monitor WVE safety?

How can we support ATC to accurately

deliver separation?


UFO – Ultra Fast Wind Sensors

Universal Atmospheric Hazard Criteria

• Wide range of new Lidar/Radar wind sensors for detecting/monitoring/alerting of atmospheric hazards:

– Wake vortices

– Wind shear

– Turbulence


• UFO system requires consistent assessment of the various hazards:

For each atmospheric hazard ideally one specific metric with:

Good discriminative power (strong correlation between metric value and severity)

Aircraft independent

Meaningful

Computable (without need for access to proprietary data)

Absolute

Link to common risk matrix, defining the severity (e.g. minor, major, hazardous, catastrophic) and probability thresholds


Why universal criteria?


• F-Factor

– Based on TSO C-117 alerting boundaries

0 5 10 150

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Exposure Time (sec)

Ffa

cto

r (-

)

2 kt/s

1.4 kt/s

0.78 kt/s

20 ktspeed loss

7 ktspeed loss

Hazardous

MajorMinor

No safety effect

Catastrophic

M a y A L E R T

M u s t A L E R T

M u s t NOT A L E R T

3 kt/s

30 kt speedloss


Wind shear

0 5 10 150

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Exposure Time (sec)

Avg.

Ffa

cto

r (-

)

M a y A L E R T

M u s t NOT A L E R T

M u s t A L E R T

20 kt

windspeed change

wind gradient 2 kt/s

wind gradient 0.78 kt/s


• Generalized EDR

– Based on PIREP Turbulence Intensity scale

– In combination with ICAO Annex 3

– But for Medium sized A/C only

– Made A/C independent with A/C chord

1 2 3 4 5 6 7 80

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9relation in situ EDR and Pilot reported tubulence intensity

PIREP Turbulence intensity scale

Pea

k E

DR

[m

2/3

/s]

ICAO Annex 3, Severe

PIREP, Severe

ICAO Annex 3, Moderate

PIREP, ModerateICAO Annex 3, Light

PIREP, Light


Turbulence

1 2 3 4 5 6 7 80

0.2

0.4

0.6

0.8

1

1.2

1.4


Pea

k E

DR

[m

2/3

/s]

Boeing 737

Boeing 747

Small Businessjet

1 2 3 4 5 6 7 80

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5


Pea

k G

en

era

lise

d E

DR

[m

1/6

/s]

Relation Generalised EDR and Pilot reported tubulence intensity

Boeing 737

Boeing 747

Small Businessjet



Wake vortex -> Dimensionless Equivalent Roll Rate

– Equivalent to RMC

– Meaningful to pilots as it can be interpreted as the angle of attack at the wingtip; wingtip stall criteria can be used for severity rating

– Only few parameters needed: wingspan, airspeed, circulation

– Easy to assess operationally in UFO concept

– Need for ‘real-time’ prediction of airspeed and circulation

– Applied in RECAT-EU

10

• Further validate that the RMC as a metric scales conservatively with increasing aircraft size: If a given RMC is acceptable for an aircraft, the same RMC will also be acceptable for a larger aircraft

• In view of RECAT PWS, collect additional evidence for acceptability of WT severity alignment of aircraft types of various size

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Objectives

11

B744 A332 A320 F100 C550

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA)

NLR GRACE Reconfigurable Research Flight Simulator

Mixed cock-pit configuration • Fly-By-Wire A320/A330 • Classic Controls B747/F100/C550

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA)

NLR GRACE Reconfigurable Research Flight Simulator

• Two scenario’s

– Level flight at 3,000 ft altitude

– ILS approach to EHAM RWY 06 (3500 x 45m)

• Light background turbulence to create realistic environment

• Fixed assumed interception angle of 10 degrees

13

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) / Experiment set-up

– The “F65” is a scaled version of the F100 to represent an A/C with a wing span of 20 meters like the E145 which is used as pivot in the Safety Case.

– Final approach speed aligned with RECAT-PWS, based on measurements

– Aircraft at 90% MLW

– Each type is verified in GRACE with type rated pilots 14

B744 A332 A320 F100 / “F65” C550

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Six aircraft types evaluated

• The rolling moment is directly imposed on the aircraft.

– Ensures a predictable and worst case encounter through the core

– Suited for the evaluation of the effect of a WV of certain strength on aircraft of different sizes.

– Avoids undesirable effects of differences in intercept angle and relative position of the aircraft to the vortex core

• A shaping function is developed to vary the wake vortex induced rolling moment with time.

• Further validated in GRACE simulator with an experienced test pilot involved in WVE flight tests

• Participating pilots confirmed realistic behaviour of WVE

15

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - WV Modelling

• Altitude 3000 ft (level flight at altitude), 200 & 100 ft (on ILS approach)

• RMC 0.04 / 0.06 / 0.08 / 0.10

– Broadly covering worst case values that aircraft are exposed to

• Variation in left and right upset

• Random sequence of the runs

• Pilots asked to continue landing and focus on WVE impact and indicate whether or not go-around would have been initiated in an operational setting afterwards

16

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Configuration of runs

17 17

Noticeable disturbance, No or negligible pilot compensation required to maintain desired flight path

1

Small disturbance Light pilot compensation required to maintain desired flight path

2

3

Minor

Major

4

5

Large disturbance Moderate pilot compensation required

to maintain desired flight path or avoid ground contact. (Safe landing possible)

Minor

Major

6

7

Severe disturbance Significant or maximum pilot compensation required

to maintain desired flight path or avoid ground contact. (Safe go-around possible)

Minor

Major

8 Extreme disturbance,

Maximum pilot control authority exceeded, inability to maintain desired flight path or avoid ground contact.

Pilot rating

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Rating scale

Topics addressed:

Preparations sufficient?

Actual experience on WVE?

Realism of the simulated encounter?

Quality of the rating scale?

Other feedback?

18

Wake Impact Severity Assessment - Post Experiment Questionnaire

1 Was the pilot briefing sufficient for you to prepare for the experiment? (What can we do to improve?)

2 How many times have you experienced a Wake Vortex Encounter in real life and what was the highest severity? (Could you relate the severity with the proposed severity rating scale?)

3 Do you consider the behaviour of the simulated aircraft type realistic enough to represent a real Wake Vortex Encounter?

4 Did you consider the Wake Vortex Encounter scenario’s realistic?

5 In the event you initiated a go-around or would have initiated a go-around, on what ‘elements’ did you base your decision? (remember we ask you always to try to land, the focus of the experiment is the assessment of the WTE severity)

6 Could you distinguish the different severity levels of the Wake Vortex Encounters?

7 Do you think the severity you rated as large disturbance/moderate compensation can be used for separation design for your aircraft type as follower?

8 Do you have any other feedback or comment? ( e.g. on the setup of the experiment …)

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Questionnaire

• Selected pilots experienced on the type including test pilots (EASA, Airbus, NLR) and airline pilots

• Pilot briefing guide (beforehand) and presentation (on site)

• Experiment runs

• Debriefing and post-experiment questionnaire

• In total 64 hours of simulation time and ~900 runs

19

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Experiment execution

• For each of the runs:

– Ratings of the severity by Pilot Flying (PF) and Pilot Not Flying (PNF)

– PF & PNF indications whether a go-around would have been initiated

– Recordings of 125 parameters and 30 derived metrics after post-processing

– Video recordings

• Pilot feedback on post-experiment questionnaire

20

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Collection of data

21

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - How does it look like?

RMC correlates well with WVE severity Comparison of Level flight versus Approach: Correlation best visible at 3000 ft (level flight); Other effects, causing more variation, play a role close to ground at 200 – 100 ft

22

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Results

23

Support to EUROCONTROL in SESAR Project 6.8.1 Wake Impact Severity Assessment (WISA) - Results Safe landing possible up to RMC=0.06 for all A/C and up to RMC=0.08 for M/L.









What are good metrics and criteria to measure

and monitor WVE safety?

How can we support ATC to accurately

deliver separation?

Congested airports Reduced separation

concepts

Current practice? Understanding by data analysis

Forecast capability

Supporting tool

Support to EUROCONTROL in SESAR Project 6.8.1 Optimised Runway Delivery Study


Objective:

• Baseline the current practices for separation application throughout Europe

• Understand factors causing variability and compression

• Demonstrate there are today means for going below minima

• To put the SESAR concepts into the right perspective

• New concepts should not be more stringent than current ops

Activities/inputs:

Analysis of flight track and weather data

• Combination of radar/ADS-B data and weather data

• Focus on busy operational hours

• Analysis of:

– Ground speed

– Distance and time separation

– Spacing buffer

• Considering:

– Headwind

– Aircraft type / Wake turbulence categories

– Airport / runway

Support to EUROCONTROL in SESAR Project 6.8.1 Optimised Runway Delivery Study – Data analysis

Support to EUROCONTROL in SESAR Project 6.8.1 Optimised Runway Delivery Study - How does separation delivery work?

Speed profile follower

28

IBE652R H

015 ↓ 160

A343

IBE652R H

009 ↓ 135

A343

EZY475A M

019 ↓ 160

A319

Speed profile leader

Compression of spacing Final Target Distance

Initial Target Distance

=

+

Spacing Buffer

At 10 NM At THR

Actual separation delivery is combination of: • Separation minima -> Final Target Distance (FTD) • Anticipation of compression ->

Initial Target Distance (ITD)

Final Target Distance


012345678910

Distance to THR (NM)

0

1

2

3

4

5

Spa

cing

buf

fer

[NM

]

Evolution of spacing buffer

012345678910

Distance to THR (NM)

0

1

2

3

4

5

Spa

cing

buf

fer

[NM

]

Evolution of spacing buffer

mean

mean +/- standard deviation

5th and 95th percentile

Spacing buffer

decreases because

of compression

Close to

threshold

buffer on average

0.5 NM

Fraction below minima,

justified by e.g., visual

separation

Support to EUROCONTROL in SESAR Project 6.8.1 Optimised Runway Delivery Study – More accurate separation delivery Optimised runway throughput not only requires optimised separation minima, also accurate delivery of separation

31

• More accurate spacing? => Develop capability to predict aircraft speed and time-to-fly on final approach

Time/ distance

separation

minimum

Actual conditions:

A/C type

Airline

Wind

Traffic pressure

Predicted speed profile from data driven model

Initial and final targeted spacing

indicators to support ATCo

Support to EUROCONTROL in SESAR Project 6.8.1 Aircraft Performance Model – Prediction of airspeed / time to fly

NLR and Wake Turbulence Separation related research, June 2016

32

0246810

Distance to threshold [NM]

110

120

130

140

150

160

170

180

Ind

icat

ed A

ir S

peed

[kt

s]

IAS profiles from data

A346B772B738A320DH8D

0246810

Distance to threshold [NM]

0

50

100

150

200

250

Tim

e to

fly

[s]

Time to fly profiles from data

Time separation converted into FTD/ITD using aircraft speed/T2F profile

Support to EUROCONTROL in SESAR Project 6.8.1 Aircraft Performance Model - Prediction of airspeed / time to fly

Ground speed and Time-to-fly per segment of 0.5 NM and headwind condition at altitude

33

10 9 8 7 6 5 4 3 2 1 0

< -5-5 - 00 - 5

5 - 1010 - 1515 - 20

>20

Hea

dwin

d co

nditi

on [k

ts]

A320 GS model

207 205 200 197 192 187 184 180 177 172 167 161 154 146 148 152 140 139 140 139 139

205 198 194 191 186 183 179 176 174 169 164 158 150 143 142 147 137 137 138 136 136

198 194 190 186 183 179 176 172 169 165 159 154 147 140 139 141 133 133 135 133 133

193 188 183 180 175 174 170 167 164 160 154 149 141 135 135 136 131 131 135 131 131

187 182 180 175 173 169 166 162 159 156 150 145 138 133 132 135 129 129 135 130 130

180 176 171 167 164 161 158 156 152 149 144 139 133 130 132 137 128 129 133 127 127

180 176 171 167 164 161 158 156 152 149 144 139 133 130 132 137 128 129 133 127 127

10 9 8 7 6 5 4 3 2 1 0Distance to threshold [NM]

< -5-5 - 00 - 5

5 - 1010 - 1515 - 20

>20

A320 T2F model

216 208 199 190 181 171 162 152 142 132 122 111 100 88 76 63 52 39 26 13 0

222 213 204 195 185 175 166 156 145 135 124 113 102 90 77 65 53 39 26 13 0

227 218 209 200 190 180 170 160 149 139 128 117 105 93 80 67 54 40 27 14 0

234 225 215 205 195 185 175 164 153 142 131 119 107 95 81 68 55 41 27 14 0

239 229 219 209 199 189 178 167 156 145 133 121 109 96 82 69 55 41 27 14 0

246 236 226 215 204 193 182 171 159 147 135 123 110 96 83 69 56 42 28 14 0

246 236 226 215 204 193 182 171 159 147 135 123 110 96 83 69 56 42 28 14 0

Support to EUROCONTROL in SESAR Project 6.8.1 Aircraft Performance Model – Headwind dependent model

34

• Model build based on historic data • Assume a specific headwind profile • Identify the associated model cells • Resulting in an average T2F profile ... • ... and uncertainty band due to variation within a cell • To account for uncertainty in the headwind profile itself ... • ... another uncertainty band may be added

• Model can learn taking into account new data, resulting in smaller variations

Support to EUROCONTROL in SESAR Project 6.8.1 Aircraft Performance Model – How to use it?

NLR own research – Data analytics & performance monitoring

35

• Potential application in view of WT research:

– To support monitoring and accurate delivery of separation after deployment of a new separation concept

• NLR cross-domain and multidisciplinary working group of performance experts, data scientists, IT engineers

• Development of own big data base, combination of ADS-B data, weather data, safety data ...

• Development of (regression, machine learning) techniques to predict indicators based on combination of historical and actual data

• Looking at performance of the ‘whole picture’: final approach + runway + taxi ways + TMA + take-off

• Real-time information for tactical monitoring & predictions

• Frequent updates for strategical monitoring


NLR own research - Data analytics & Performance monitoring, Dashboard development to visualise safety / performance data

36 NLR and Wake Turbulence Separation related research, June 2016

Future Sky Safety Project 4 – Total System Risk Assessment

37

• Development of Prototype Risk Observatory

– Combining safety data from abroad the sector

– Integration with risk models

– Potential for integration of WVE safety information


38

Conclusions

• Research has been important to build up knowledge and create the tools to deploy reduced separation concepts

• For optimizing runway throughput:

+ Optimized separation minima

+ Optimized – more accurate – delivery of separation

• Requires accurate prediction of aircraft speed & time to fly

• Future work focus on:

– Accurate prediction of aircraft performance and weather/wind

• More advanced data analytics and Machine Learning techniques

• Broader, larger, and more accurate data sets

– Extension to departures

– Integration of reduced separation in broader scope of the airport runway throughput operation

– Monitoring: separation performance, WVE metrics NLR and Wake Turbulence Separation related research, June 2016

NLR Amsterdam

Anthony Fokkerweg 2

1059 CM Amsterdam

p ) +31 88 511 3113 f ) +31 88 511 3210

e ) [email protected] i ) www.nlr.nl

Fully engaged Netherlands Aerospace Centre

Documents

NLR and Wake Turbulence Separation related … and Wake Turbulence Separation related research, June 2016 8 • Generalized EDR –Based on PIREP Turbulence Intensity scale –In combination