1 / 84

NGGM ASSESSMENT STUDY Requirements Review Meeting TAS-I, Torino, 19 November 2009

NGGM ASSESSMENT STUDY Requirements Review Meeting TAS-I, Torino, 19 November 2009. 9:00 Welcome and Introduction 9:15 WP1100: Requirements Analysis (ULUX) current status of satellite gravimetry wrt detection of mass transport in the Earth System

crown
Download Presentation

NGGM ASSESSMENT STUDY Requirements Review Meeting TAS-I, Torino, 19 November 2009

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. NGGM ASSESSMENT STUDYRequirements Review MeetingTAS-I, Torino, 19 November 2009

  2. 9:00 Welcome and Introduction 9:15 WP1100: Requirements Analysis (ULUX) current status of satellite gravimetry wrt detection of mass transport in the Earth System scientific questions potentially addressed by improvements in spatial / temporal resolution analysis of current limitations with respect to aliasing and background models list of priorities for the NGGM 13:00 Lunch break 14:00 WP1200: System drivers (TAS-I) Payload engineering requirements and constraints Satellite platform drivers Mission architecture options Cost/performance models 15:30 WP2310: E2E Simulator progress report (TAS-I) 15:45 WP2320: Variable Gravity Model progress report (IAPG) 16:00 Discussion, work plan 17:00 End of meeting Agenda

  3. WP1200: System Drivers • Payload engineering requirements and constraints

  4. Earth gravity measurement techniques • Satellite gradiometry (GOCE) • Low-low satellite-to-satellite tracking (GRACE) Fundamental observable:distance variation between two satellites produced by the gravity acceleration: dG, obtained as d - dD d: distance variation between the two satellites produces by any source, measured by a laser metrology system. dD: distance variation between the satellites produced only by drag forces, measured by accelerometers. Fundamental observables: components of the gravity gradient (Vxx, Vyy, Vzz, Vxy, Vxz, Vyz) measured by the gradiometer as difference between the accelerations at the location of any accelerometer pair divided by the pair baseline.

  5. Gradiometry vs LLSST • Why LLSST is more suitable than gradiometry to measure temporal variations of the Earth gravity field • LLSST, thanks to the much larger separation between the “proof masses” (the satellites themselves) is intrinsically more sensitive than gradiometry. Therefore good measurement accuracy can be achieved even at the relatively high altitudes (>300 km) needed to ensure long lifetime with an affordable amount of propellant.

  6. Measurement instruments • Instruments involved in the measurement of the fundamental observable (dG)in a LLSST mission.

  7. Distance measurement scheme • Distance measurement objective: Measurement of the distance variation (d)between the satellite COMs. Involved P/L items:Distance metrology, Angle metrology, Lateral metrology Distance metrology (laser based) function: • Measurement of the distance variation between the retro-reflectors (points A, C) along the optical path followed by the laser beam (A-B-C): L = d1 + d2. Angle metrology function: • Measurement of rotation angles of Satellite1, 2 w.r.t. the line joining the satellite COMs ( laser beam): 1, 2, 1, 2 • Lateral displacement metrology function: • Measurement of RR2Y-Z offsets from laser beam axis, for beam pointing.

  8. Distance metrology concepts • Retro-reflection concept • Optical transponder concept Proposed/developed in US for GRACE-FO Michelson-type heterodyne interferometer based on transceiver scheme with master laser on satellite 1 and slave laser on satellite 2 phase-locked to the first one in “frequency-offset”. Pros: suitable for very long distances (>100 km) thanks to the signal “re-generation” scheme. Cons: two lasers and two interferometers must operate simultaneously (+complexity, -reliability) Proposed/developed by TAS-I for NGGM Michelson-type heterodyne interferometer with passive retro-reflection and chopped laser beam for long-distance (> ~1 km) operation. Pros: single laser sufficient to perform the measurement (-complexity, +reliability). Two lasers provides the single-failure tolerance. Cons: unsuitable for very long distances (> 100 km) due to the weak return optical power.

  9. Distance measurement errors • Distance measurement error tree (values from the previous study)

  10. Distance measurement limit • Ultimate limiting factor: laser frequency stability • L = distance measured by the laser •  = laser frequency (~282 THz) Best frequency stability achieved in laboratory using a reference cavity (not easy). Ref. Gerhard Heinzel, “LISA technology for gravity-field missions”, Graz Workshop, 30/9 2009. Envelope Error contribution of the laser frequency stability only

  11. Distance measurement requirement • Top-level requirement considered in the previous study • Assumed best performance on the COM-COM distance d overall relative error achievable in orbit. • Corresponding requirement on distance measurement error (d), for d = 10 km. 5 nm/Hz

  12. Ancillary metrology requirements • Lateral displacement metrology Measurement of the laser beam axis lateral displacements relative to the retro-reflector on S2, such as to ensure a laser beam pointing control with: • max. error 10-5 rad (0.1 m at 10 km). • pointing stability spectral density: • Angle metrology • Measurement of the S1, S2 rotation angles w.r.t. the line joining the satellite COMs ( laser beam): i, i • measurement range: 1° • maximum measurement error 10-4 rad • measurement error spectral density 10-7rad/Hz 1.510-7 rad/Hz Lateral displacement measurement noise 10-3 m/Hz at 10 km).

  13. Laser power impact on distance metrology Output optical power of the laser source set = 0.75 W in the previous NGGM study. Case 1: same distance measurement error (d = 5 nm/Hz) for any distance. Case 2: same distance relative measurement error d/d = 510-13 1/Hz) for any distance. Optical power requirement fulfilled up to ~85 km in Case 1 and up to >100 km in Case 2. Conclusion: the distance metrology based on the retro-reflector is still limited by the laser frequency stability (/ = L/L) and not by the optical power. Optical power received by the laser interferometer (after the retro-reflection) vs inter-satellite distance, compared with minimum power requirement in Case 1, 2.

  14. Distance measurement: required resources • S/C resources/services (preliminary) required by the distance measurement system (Distance, Angle, Lateral metrology). • Mass: 60 kg (including optical bench) • Power demand: 100 W • Telemetry generation rate: 1.5 kbps (distance + angle metrology; data output at 10 Hz) • Inter-satellite telemetry rate: 3 kbps (lateral metrology on S2: data output at 100 Hz) • On board processing power: ~20000 flops (mainly on S1 for distance metrology + beam pointing management). • Large amount of the computation on the distance metrology and angle/lateral metrology shall be performed by a dedicated FPGA/ASIC. • ~10000 flops are required by the pointing control. • Absolute / Relative Pointing Error: not very demanding if the laser beam pointing function is performed by a dedicated device and not by the S/C. • Relative velocity control: <15.9 m/s (limited by the heterodyne frequency) • Relative acceleration control: <120 m/s2 (limited by the laser beam chopping scheme) • Other: stable thermal control on the optical bench and on for the optical cavity utilized for the laser frequency stability

  15. Distance measurement system: Alternative architectures • Alternative 1: distance metrology without laser frequency stabilization • Distance variation measurement error computed for d = 10 km. Ref. M. Tröbs et al., Laser development for LISA, Class. Quantum Grav. 23 (2006) S151–S158 • The frequency stabilization system has a significant impact in cost (~2x) and complexity (frequency reference, closed loop control) on the distance measurement system, but is unavoidable to improve the GRACE performance.

  16. Distance measurement system: Alternative architectures • Alternative 2: laser beam pointing performed by the S/C attitude control • Advantages of a dedicated laser beam pointing device: • More relaxed attitude control requirements. • Satellite attitude control decoupled from the relative motion of the satellites. • Drawbacks of a dedicated laser beam pointing device: • More complexity and less reliability (mechanism continuously operating) • Induced disturbances on distance measurement and potentially on accelerometers. • Conditions for transferring the laser beam pointing task to the S/C attitude control: • Relaxation (~10x, TBC) of the current requirements on laser beam pointing and stability (10-5 rad, 10-7 rad/Hz). • Attitude motion induced by the tracking of Satellite 2 compatibility with attitude stability requirements

  17. Distance measurement system: Alternative architectures • Alternative 3: angle metrology replaced by S/C equipments (GPS, star trackers) • The satellite rotation angles w.r.t. the line joining the satellite can be obtained in principle using: • The satellite absolute and relative position provided by the GPS, from which the inertial orientation of the satellite-to-satellite line can be reconstructed. • The inertial attitude of each satellite provided by the star trackers. • Condition for replacing the angle metrology with GPS and star trackers: • Relaxation by at least one order of magnitude (TBC) the measurement error spectral density (set to 1.510-7 rad/Hz in the previous NGGM study). • Possible increase (TBC) of the laser beam divergence (currently 10-4 rad). • Note: in case the beam pointing device will be removed (and therefore there are no bending of the laser beam outgoing/incoming from/to S1), the nearly constant rotation angles of S1 relative to the beam ( satellite-to-satellite line) could be performed using a quadrant photodiode in the laser interferometer (same concept of GRACE-FO metrology).

  18. Non-gravitational acc. measurement scheme • Non-gravitational acceleration measurement objective: Measurement of the non-gravitational differential accelerationof the two satellitesalong the line joingin the COMs: Involved P/L items:Accelerometers, Angle metrology Accelerometers function: • Measurement of the non-gravitational, linear acceleration of the COM of each satellite, in the Satellite Reference Frame: D1,D2. Angle metrology function: • Measurement of rotation angles of Satellite1, 2 Reference Frames w.r.t. the line joining the satellite COMs: 1, 2, 1, 2 dD =

  19. Non-gravitational acc. measurement • Non-gravitational differential acceleration measurement error tree (values from the previous study)

  20. Non-gravitational acc. measurement • Ultimate limiting factor: accelerometer intrinsic noise • GOCE accelerometer noise along the ultra-sensitive axes considered the ultimate performance limit (note GRACE acc. noise = 10-10 m/s2/Hz). • Non-gravitational acceleration noise increase at high frequency matched to the double time derivative of the distance variation measurement noise. Top-level requirement considered in the previous study

  21. Angle Metrology Requirement Derived requirements from Transformation Errors Satellite misalignment in the SSRF: P, P 1° Rotation angle measurement error: P, P 10-4 rad Envolope of satellite pointing requirements from distance variation measurement and non-gravitational acceleration measurement: ,   1°, ,   10-4 rad • Satellite rotations stability requirements • Satellite rotation meas. error requirements October 21 2008

  22. Acceleration measurement: required resources • S/C resources/services (preliminary) required by the acceleration measurement system (two GOCE-like accelerometers per S/C assumed). • Mass: 22 kg • Power demand: 35 W • Telemetry generation rate: 3 kbps (3 linear + 3 angular accelerations per accelerometer; data output at 10 Hz) • Drag free control in linear and angular accelerations: Angular acceleration max. value: 110-6 rad/s2 Linear acceleration max. value 110-6 m/s2

  23. Distance measurement system: Alternative architectures • S/C without drag-free control (just orbit maintenance) GOCE drag acceleration environment at ~255 km, low solar activity. • With a factor ~10 relaxation of the residual linear acceleration requirements, the drag control in radial and cross-track directions could be avoided ( benefit on thruster range), while it seems still necessary in-track (TBC by disturbance and performance analysis). • An attitude control finer than in GOCE (34º in yaw) is necessary to guarantee the optical link, even if a dedicated device is used for the laser beam pointing  electrical thrusters needed since reaction wheels are too noisy for the accelerometers.

  24. WP1200: System Drivers Satellite platform drivers

  25. System drivers vs. practical limits • Launcher type and number of launches • Physical configuration (mass, volume, shape) • Electrical power system design (solar array efficiency, power system density) • Thermal Control (complexity) • On board data handling (science data rates, ancillary data rates, mass memory size, on board processing power) • Telecommunications (inter-satellite data exchange, telemetry rates to ground station, frequency of contacts with ground station) • Orbit control (orbit maintenance needs, frequency, propellant budgets) • Attitude & drag free control (pointing accuracy, laser pointing control, linear and angular acceleration control) • Relative satellite motion (relative range control accuracy, formation keeping)

  26. Eurokot: 1250 kg in SSO (GOCE) Vega Launcher performance

  27. Flat-cylinder-shaped satellite [cartwheel / pendulum] Top surface = solar panel area: 3.5 m² (Rockot), 4.4 m² (VEGA) Height @ 1.1 m² lateral cross section: 0.52m (Rockot), 0.46m (VEGA) Volume: 1.8 m³ (Rockot), 2 m³ (VEGA) Prism-shaped satellite accommodation in VEGA Front cross section: 1.1 m² Max lateral cross section: 7.2 m² Min lateral cross section: 3.2 m² Min lateral/front cross section ratio: 2.9 Max lateral/front cross section ratio: 6.5 Volume: 4.48 m³ Launcher volume & satellite shape

  28. Pendulum drag force scans 45° angular sector around satellite, once per orbit cylindrical satellite with long axis aligned to local vertical always offers same cross section to drag Minimum of 2 thrusters needed for drag compensation Cartwheel drag force turns around satellite body once per orbit cylindrical satellite with long axis aligned to orbit normal always offers same cross section to drag Minimum of 4 thrusters needed for drag compensation Satellite shape in cross-line formations • Drag force rotating around the body suggests shape of a flat cylinder

  29. Prismatic shape for pendulum formation Prismatic shape for N-S cartwheel formation Satellite shape in cross-line formations • In high-i orbits, air density changes by factor of 3 from equator to poles → prismatic sat (variable cross section facing drag) Smaller cross section facing drag at equator (higher density) Larger cross section facing drag at poles (lower density)

  30. Structure & Thermal Control • GOCE & GRACE: common drivers • External shell as load-carrying structure • Symmetry, aerodynamic shape • Accurate CoM trim • Thermoelastic stability • Comfortable mass margins throughout implementation phase • GOCE design parameters (reference) • Structure mass ratio: 33% [including 60kg balance mass] • Thermal control mass ratio: 3% • GRACE (reference) • Structure mass ratio: 50% [including 35kg balance mass] • Thermal control mass ratio: 3.5% • Implications for GRACE+ • will continue to be driven by high structure mass ratio, aerodynamics, CoM, thermoelastics • mass margin likely to become a driver too (twin launch on Eurokot)

  31. GOCE design parameters (reference) SSO Max eclipse < 28 min hibernation during long eclipse Max power demand  1 kW incl. IPA @ 19 mN Max sun incidence angle on solar array  30 Solar array efficiency (all included)  150 W/m² [GaAs cells] Power system density  7.5 W/kg Power system mass fraction at launch  13% GRACE (reference) Non-SSO Max eclipse < 36 min Max power demand  160W Solar array efficiency (all included)  32 W/m² [Si cells] Power system density  1.5 W/kg Power system mass fraction at launch  28% Implications for GRACE+ Large increase of power demand (OM by ion propulsion + laser system) Mass increase (solar array & battery) Trade-off: solar array accommodation vs. operational limits Electrical power system

  32. Electrical power system • GOCE IPA Power Demand (reference)

  33. Power implications of non sun-syncronous orbit • Sun revolves around orbit plane →longer eclipses & eclipse seasons

  34. Power implications of non sun-syncronous orbit • Electrical power from fixed solar panels insufficient for science+drag control →steerable solar array (cost) or seasonal degradation of performance • example: GRACE-like formation, ~8 m² side panel (GOCE) + 1m² panel on top & bottom / average active solar panel area decreases from 8 m² (1600 W) to 3 m² (590 W) Season 1: sun normal to orbit, no eclipse Season 2: sun in orbit plane, eclipse

  35. GOCE (reference) S band 1 ground station 1 pass per orbit 14 kbit/s telemetry generation rate 1.2 Mbit/s transmit rate 250 mW RF transmit power GRACE (reference) S band 2 ground stations, each satellite operated individually, different up/down frequencies used 4-5 passes per day (6-7 min) 16 kbit/s telemetry generation rate 1 Mbit/s telemetry data rate 2 W RF transmit power Implications for GRACE+ Increased telemetry rate likely non-polar orbit Ion propulsion & laser link h/k data Operations trade-off: individual sat operation vs. inter-satellite link Intersatellite link Needed anyway for laser operation Frequency separation, polarization TBD Telecommunications

  36. GOCE (reference) Functions: DHS + DFAC + Thermal Complexity/ Performances: high for current standard (ERC32, 17 MIPS) 10 Hz DFAC control 4Mbyte RAM load at 80% ERC32 CPU Load above 80% Autonomy Requirements: high 8 days Complex FDIR in SW + OBCP Data Storage & routing: moderate < 20 kbits/s average 2x4Gbit storage Environment (Rad + EMC) Low dose < 1krad Demanding EMC L1-L2 Bands Interfaces Mil-1553 Integrated RTU large number of I/O AOCS I/F: STR (serial) / MTR Drivers / RTU I/O for CESS-Sun Sensor , MGM GRACE (reference) 1750 processor 1.4 Gbit data storage Clock stability 10 ms per 30 min Implications for GRACE+ Processor upgrade if DFAC needed Formation keeping TBD Moderate to high autonomy Undemanding altitude Formation keeping TBD Undemanding data storage On board data handling

  37. Propellant mass @ Isp = 3000s for 500-kg satellite: 2.14 kg/yr/mN Propellant mass for 6 years: Propellant mass is a small percentage (3% - 4%) of satellite mass from 350 km up Orbit & formation acquisition and maintenance1. orbit maintenance A/M = 10-3 m²/kg 2010 to 2020 time span With electric propulsion!

  38. Heading towards (at least) 1 full half-cycle with F10.7 < 100 Next solar maximum is now predicted to occur in 2013 2011 in mid-2007 forecasts 2012 in mid-2008 forecasts … Orbit & formation acquisition and maintenance2. solar flux prediction

  39. Repeat orbit: β revolutions in α days Repeat cycle α depends on altitude and inclination 15 < β/α < 16 in the altitude range of interest For well-behaved monthly gravity field solutions, low-period orbit resonances must be avoided Grace lesson learnt: β > 2L (L= maximum degree of L x L solution) Implies α > 15 days if L=120 Implies rather strict orbit maintenance (average interval between α < 15 resonances  3.3 km) Orbit & formation acquisition and maintenance3. repeat cycles

  40. GRACE ground track examples Jan. 2009 Sept. 2004 (475km, 61/4 resonance) June 2009 http://www.csr.utexas.edu/grace/operations/gtrk_mons

  41. N-S Cartwheel line of apsides of the two orbits must be kept in the N-S inertial direction For a polar orbit, h = 312 km, argument of perigee drifts by -4.2°/day compensation V = 0.28 m/s/day propellant mass over 6-yr mission: [Isp = 3000s; 500-kg satellite] = 10.5 kg = 2% of satellite mass Formation-maintenance propellant mass is comparable to orbit maintenance Orbit & formation acquisition and maintenance4. formation maintenance cost (example)

  42. Orbit & formation acquisition and maintenance5. formation acquisition (example) • Cartwheel • Several orbits are required to achieve the final cartwheel geometry • To avoid collision risk, sequence of orbit control accelerations must be carefully planned • The keep the thrust control authority bounded, and limit propellant consumption, the boundaries of formation control must be defined around a nominal trajectory generated by including the spherical geopotential and possibly J2 • Relative position evolution [S2 in the S1 LORF] during cartwheel formation acquisition by a low-thrust propulsion system

  43. Control systems • In order to fully exploit the potentiality of a gravity mission based on the ll-SST technique and realized with a laser interferometer and ultra sensitive accelerometers, several controllers must be put in place: • linear drag-free; • orbit; • formation; • attitude and angular drag-free; • laser beam pointing. • The implementation of all these controllers, which must work in a synergetic way, contributes to the overall system complexity (and costs). • The possibility of simplifying the system by dropping some of these controllers will be one of the tasks addressed during the study (drag-free, laser beam pointing).

  44. Control systems DRAG FREE CONTROL • Linear drag-free control has the task of reducing the dynamic range of the measured acceleration. This is necessary: • for feasibility of the instrument (sensor and electronics accuracy versus range); • to reduce the impacts of accelerometer nonlinearities and misalignments on overall instrument accuracy after in-flight calibration. • According to current mission architecture, the residual linear acceleration shall be lower than 10-6 m/s² (max. value) and 10-8 m/s²/Hz in the [0.001,0.01] Hz band => each axis shall be controlled. • Drag-free control requires throttlable thrusters with dynamic range compatible with the different environmental conditions (operating altitude, solar activity). • It is very expensive in terms of fuel mass. • Since the scientific requirements, and consequently the requirements of the fundamental observables, will be reviewed during the study, even the conclusions on the drag-free control might change.

  45. Control systems ORBIT CONTROL • Orbit control shall keep the average formation altitude. • Autonomous orbit control will be traded vs. ground-based control. FORMATION CONTROL • Formation control shall provide the following capabilities: formation acquisition and keeping, anti-collision strategy. • Formation keeping control shall maintain the formation geometry during the observation phase in a box with sizes (10km average satellite distance): • along track: < 500m • across track: < 50m • radial : < 50m • The challenge of the formation control for this mission consists in keeping the relative motion within these boundaries without interfering with the scientific measurements, operating in synergy with the drag-free control and minimizing the thrusters use (in terms of dynamic range, propellant consumption).

  46. Control systems ATTITUDE CONTROL • According to the current architecture, attitude control shall: • maintain the attitude errors with respect to LORF compatible with laser beam pointing range; • constrain the angular accelerations and the angular rates to be very stable < 10-8 rad/s²/Hz and < 10-6 rad/s/Hz respectively in [0.001,0.01]Hz band because their coupling with the accelerometer displacement from the satellite COM produces a linear acceleration which contributes to the measurement error of the non-gravitational accelerations of the satellite COM. • Above requirements impact on the attitude control closed loop and on the reference attitude trajectory to be tracked.

  47. Control systems LASER BEAM POINTING CONTROL • Laser beam pointing control is used to guarantee the optical link between the satellites (acquisition and tracking). • The orientation of the laser beam towards the retro-reflector on the other satellite must be very precise and stable too < 10-5 rad and < 10-7 rad/Hz respectively • The presence of a continuously operating Beam Steering Mechanism (BSM) on a long-duration mission is certainly a weak point, besides being a potential source of disturbance for the accelerometer measurements. • The BSM can be removed if all its tasks can be handed over to the satellite attitude control. These tasks can be facilitated if the pointing requirements can be relaxed and if the laser beam divergence can be increased. • Tracking this oscillation by changing the attitude of the satellite could conflict with the attitude stability requirements on angles, angular rates and angular accelerations. • These issues will be addressed during the study.

  48. Drag control:implications of cross-line satellite formation • The main drag force rotates around the satellite →satellite must have small lateral cross section and drag control thrusters must be distributed all around it “Pendulum” formation “Cartwheel” formation

  49. Relative satellite motion • The relative movement of two satellites in orbit with low eccentricity is described by the well known Hill-Clohessy-Wiltshire (HCW) equation (chief motion almost circular, differential effects related to J2, drag, etc. are not taken into account). • The structure of the solution of the homogeneous differential equations reads: X : along track Y : radial Z : across track n : orbital rate A0,B0, , , xoff, zoff depends on init value for relative position and velocity. Bounded relative motion for zoff =0, i.e.

  50. Relative satellite motion • Considering the HCW equation, the unforced and stable formation geometries (solution of the homogeneous differential equations with particular initialization) are: • In-line (named also GRACE, trailing); • Pendulum; • Cartwheel (with or without along track offset xoff); • LISA like (with or without along track offset xoff). • The eigenvalues of the HCW equation are: • Along-track and radial (coupled axis): 0,0, ± jn (n is the orbital rate) • Across-track: ± jn • The system is not asymptotically stable • Drift from stable trajectories are due to: • Differential perturbation due to gravity field (non-spherical components like J2, etc.), drag, solar pressure, actuators activity (due to bias and drift on accelerometers replied by drag-free control); • Not exact initialization of the state variables (relative position and/or velocity) due to GNC sub-system.

More Related