1 / 42

Towards a scientific basis for determining En Route capacity

Towards a scientific basis for determining En Route capacity. Alex Bayen Charles Roblin Dengfeng Sun Guoyuan Wu. University of California, Berkeley Department of Civil Engineering Systems Engineering. March16 th , 2006, Pacific Grove, CA.

nyx
Download Presentation

Towards a scientific basis for determining En Route capacity

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Towards a scientific basis for determining En Route capacity Alex Bayen Charles Roblin Dengfeng Sun Guoyuan Wu University of California, Berkeley Department of Civil Engineering Systems Engineering March16th, 2006, Pacific Grove, CA Work funded by NASA Ames under Task Order TO.048.0.BS.AF

  2. Motivation for this work Importance of en-route capacity (analysis) Safety, quality of service, delays, efficiency, performance metrics,… Establishing a scientific base for the envelope of operations Question 1: reachability in terms of delays Question 2: reachability in terms of counts or similar metrics Provide input data for estimating storage capacity Question 1: how much aircraft / delay can one portion of airspace absorb Question 2: what is the relation to WITI? Question 2: space / time definition of capacity

  3. Outline Building blocks towards scientific capacity analysis • Systematic identification of the topological features of the National Airspace System (graph theoretic) • Automated model building (aggregation procedure) • Parameter identification (travel time) • Model Analysis (storage) • Model validation • Capacity assessment (in progress) • Delays that can be absorbed • Aircraft that can be stored • Stability of the storage (backpropagation)

  4. The National Airspace System (NAS)

  5. Air Route Traffic Control Centers in the US [Menon, Sweriduk, Bilimoria 2004] [Sridhar, Soni, Sheth, Chatterji 2004] [Roy, Sridhar, Verghese 2003], [Devasia, Meyer 2002]

  6. Air Route Traffic Control Center (Oakland)

  7. Flight trajectories

  8. Conceptual goal: graph building [Histon, Hansman, 2000]

  9. A state of the art clustering algorithm

  10. A state of the art clustering algorithm

  11. Outline Building blocks towards scientific capacity analysis • Systematic identification of the topological features of the National Airspace System (graph theoretic) • Automated model building (aggregation procedure) • Parameter identification (travel time) • Model Analysis (storage) • Model validation • Capacity assessment (in progress) • Delays that can be absorbed • Aircraft that can be stored • Stability of the storage (backpropagation)

  12. Data analysis procedure Sequential (automated) algorithm • Airspace segmentation using sector boundaries • Link building using clustering techniques • Data aggregation using ASDI/ETMS information (flight plan information) • Filtering using LOAs, and observed flow patterns • Computation of the aggregate flow pattern features Output: topology of the flows

  13. Data analysis procedure

  14. Data analysis procedure

  15. Data analysis procedure

  16. Data analysis procedure

  17. Main question: what is the aggregate dynamics? How to relate inflow to outflow (MIMO)? What is the internal dynamics? How to use these models to assess capacity?

  18. Outline Building blocks towards scientific capacity analysis • Systematic identification of the topological features of the National Airspace System (graph theoretic) • Automated model building (aggregation procedure) • Parameter identification (travel time) • Model Analysis (storage) • Model validation • Capacity assessment (in progress) • Delays that can be absorbed • Aircraft that can be stored • Stability of the storage (backpropagation)

  19. System identification (example: one link)

  20. System identification (example: one link)

  21. System identification (example: one link)

  22. Outline Building blocks towards systematic capacity analysis • Systematic identification of the topological features of the National Airspace System (graph theoretic) • Automated model building (aggregation procedure) • Parameter identification (travel time) • Model Analysis (storage) • Model validation • Capacity assessment (in progress) • Delays that can be absorbed • Aircraft that can be stored • Stability of the storage (backpropagation)

  23. Delay system: Link level Link level model Cell

  24. Illustration of the model: state vector time step 1 delay system at the link level state: cell counts

  25. Illustration of the model: states transition time step 2 delay system at the link level state: cell counts

  26. Illustration of the model: states transition time step 3 delay system at the link level

  27. Illustration of the model: states transition time step 4 delay system at the link level

  28. Illustration of the model: entry input time step 5 delay system at the link level input to the link (forcing)

  29. Illustration of the model: entry input time step 6 delay system at the link level

  30. Illustration of the model: delay control time step 7 delay system at the link level one aircraft about to be held

  31. Illustration of the model: delay control time step 8 delay system at the link level control input

  32. Illustration of the model: delay control time step 9 delay system at the link level two aircraft about to be held

  33. Illustration of the model: delay control time step 10 delay system at the link level

  34. Delay system: Sector level Sector level [Robelin, Sun, Wu, Bayen 2006]

  35. Delay system: Center (ARTCC) level Center level: [Robelin, Sun, Wu, Bayen 2006]

  36. Outline Building blocks towards systematic capacity analysis • Systematic identification of the topological features of the National Airspace System (graph theoretic) • Automated model building (aggregation procedure) • Parameter identification (travel time) • Model Analysis (storage) • Model validation • Capacity assessment (in progress) • Delays that can be absorbed • Aircraft that can be stored • Stability of the storage (backpropagation)

  37. Aggregate model validation

  38. Outline Building blocks towards systematic capacity analysis • Systematic identification of the topological features of the National Airspace System (graph theoretic) • Automated model building (aggregation procedure) • Parameter identification (travel time) • Model Analysis (storage) • Model validation • Capacity assessment and control (in progress) • Delays that can be absorbed • Aircraft that can be stored • Stability of the storage (backpropagation)

  39. IP: Formulation Challenges >1M variables, >1M constrains. CPLEX: <6 minute running time (LP) [Robelin, Sun, Wu, Bayen 2006] [Borelli 2003, Feron 2000, Bertsimas 1997]

  40. Overload control

  41. Eulerian models Generic features of Eulerian models • Eulerian models scale well: complexity is independent of number of aircraft • Control volume based: appropriate for capacity analysis • Linear features make them suitable for analysis • Can rely on control theory for controllability, observability • Combinatorial optimization algorithms can be applied Features of the current model • Can take any set of ETMS/ASDI data as input • Eulerian model, validated against ETMS/ASDI data • Compared to 2 other existing models (AIAA GNC 2006) • Interface with FACET

  42. Acknowledgments UC Berkeley: Mark Hansen NASA Ames: Banavar Sridhar, Kapil Sheth, Shon Grabbe, George Meyer FAA: Dave Knorr CNA: Doug Williamson

More Related