Systemy SHM opracowane w AGH w oparciu o niskie częstotliwości Akademia Górniczo Hutnicza Katedra Robotyki i Mechatroniki Al. Mickiewicza 3, 3-59 Kraków
Plan prezentacji. Monitorowanie stanu konstrukcji w oparciu o pomiary wizyjne. Implementacja filtru modalnego
Monitorowanie stanu konstrukcji w oparciu o pomiary wizyjne Pomiary stanów statycznych Pomiary procesów dynamicznych Zespół: Prof. dr hab. inż. Uhl Tadeusz, Dr inż. Kohut Piotr, Mgr inż. Holak Krzysztof, Krupiński Krzysztof Dr inż. Szwedo Mariusz,
Ugięcie [mm] Ugięcie [mm] Wizyjne pomiary stanów statycznych Tani system wykorzystujący ogólnie dostępne aparaty cyfrowe (lustrzanki) do pomiaru odkształceń konstrukcji Segment wiaduktu tramwajowego Obraz krzywej ugięcia Krzywa ugięcia wiaduktu [mm] 8 7 6 7 Ugięcie Krzywa ugięcia wiaduktu [mm] 5 4 6 3 5 4 3.5.5 3 Położenie [mm] x 4.5.5.5 Położenie [mm] x 4 Monitorowane obiekty: mosty, kładki, wiadukty kominy, przęsła, dźwigary, hale, maszty, turbiny wiatrowe, budynki, rurociągi, maszyny i urządzenia
PositionZ [mm] Trans [mm] Wizyjne pomiary procesów dynamicznych Maszyna wibracyjna Robot RV-AJ kamera 8 Components of Motion(R+T) vector in WORLD Trójwymiarowe przebiegi drgań TRANSx-component TRANSy-component TRANSz-component Trajektoria ruchu 3D 3D Motion Trajectory (.3Hz) 6 4 3D Trajektoria ruchu chwytaka robota.5 -.5 -.8.6.4 -. -. -4 -.4 PostionY [mm] -.6 -.8 - -5-4 -3 - - PositionX [mm] 3-64 -8 5 Trójwymiarowe przebiegi drgań 4 6 8 4 6 8 Time [sec] przednia Maska samochodu Monitorowane obiekty: mosty, kładki, wiadukty kominy, przęsła, dźwigary, hale, maszty, turbiny wiatrowe, budynki, rurociągi, maszyny i urządzenia
Lockin signal mixing for modes visualisation based on camera measurement
Acquired frame Lockin signal mixing
resonance _ frequency camera _ frame _ rate Electromagnetic exciter Tested object Camera fixed :: Frame rate Signal generator unit Image acquisition unit PC
Wizualizacja postaci drgań analizowanych konstrukcji Procesy szybkozmienne Rejestracja pracy absorberów klatek/sek Kamera cyfrowa Rejestracja pracy wyrobów medycznych - nakłuwaczy ~ klatek/sek Utwierdzona belka 4.375 Hz 88.4375 Hz
Algorithms modal filters The method was introduced in 98 by Baruh and Meirovitch The modal filter extracts modal coordinates from measurements of the system response. It can be applied for: dealing with the control spillover problem, the vibration control of flexible structures, correlation analysis for experimental and analytical modal vectors, identification of operational forces from the system response, damage detection Filtering of environmental changes (temperature changes) influence on structural dynamic properties of a structure
Modal filters - reciprocal modal vectors To construct the discrete modal filter from experimental data, a new type of modal parameters has to be introduced reciprocal modal vectors r. They should be orthogonal with respect to the modal vectors, and thanks to that, they are applied to decomposition of the system responses to the modal coordinates r. the assumption that the modal residue R rpp is in the imaginary form: R rpp j Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions r H T p x the DOF frequency response function H pp ( ) pp j R rpp r j r T j r R * rpp the experimental frequency response functions k x N matrix: H kn H H H k H H H k H H H N N N k r * r T * r j * T r * r reciprocal modal vectors matrix p determination: p H kn H pp
Algorithms modal filters Damage Environmental change Local change of stiffness M ( K u ) K K K d u K t K u t and t ( K u t t t M ) Conclusion: any method based on the modal vectors is resistant for environmental changes
Algorithms modal filters damage localization ifferences between mode shapes of amaged and undamaged beam with % crack Difference 5 x -6 between mode shapes no. for undamaged and damaged structure Difference 3 x -6 between mode shapes no. for undamaged and damaged structure 4 3 - - - - -3-3 5 5 5 Number of node DI 4 f s x i f s x ref x ref d d -4 5 5 5 Number of node Procedure Preliminary calculations performed for an object in the reference (undamaged) state modal analysis, calculation of Reciprocal Modal Vectors for all sensors and for sensors from consecutive areas. Measurements of object characteristics in their current state and their filtration with the use of a global modal filter. In the case of damage detection in Point of the procedure, filtration of groups of characteristics from consecutive object areas with the use of local modal filters. Calculation of the damage index DI for each of the sensor regions.
Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Numerical verification - model Damage 5% stiffness drop in spring no. Temperature change 5% stiffness drop in all springs m 6 m 7 m 6 m 7 m 5 m 5 m m 3 m 4 m m 3 m 4 m m Modal model (natural frequencies, modal damping coeff., modal vectors) FRF synthesis Modal filter (reciprocal modal vectors) Modal filtering
Amplitude Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Numerical verification - results Outputs of modal filter set to MS no. 7 theoretical FRF - no damage theoretical FRF - 5% damage theoretical FRF - 5% temp. change - - -3 5 5 5 3 Frequency [rad/sec]
Amplitude Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Numerical verification - results Outputs of modal filter set to MS no. 7 (with noise) theoretical FRF - no damage theoretical FRF - 5% damage theoretical FRF - 5% temp. change - - -3 5 5 5 3 Frequency [rad/sec]
Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Model of the rail viaduct CAD model was based on documentation of real structure. Bridge is 7 m long and consists of steel (beams, barriers, rails, reinforcement of main plate), concrete (main plate, pavements), soil (sub-crust) and wood (sleepers) elements. FEM model was built. Is consists of approximately 85 elements and 35 nodes. Solid, shell and beam elements were used in the model
Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Model of the rail viaduct There were three cracks introduced in the model. First and second are vertical cracks in web, third is flange crack. Crack localization was based on linear-static stress analysis. To take into account influence of moisture, different material densities were used The following normal modes analysis scenarios were considered: Without crack (dry) Without crack (moist) Without crack (wet) Crack Crack Crack 3 Crack 3
Damage Index Damage Index Magnitude Damage Index Damage Index Magnitude Damage Index Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Humidity - results 3 x 4 Values of Damage Index for the damaged bridge - p 7 6 5 4 FRFs for selected nodes filtered with modal filter set to MS no. undamaged dry system undamaged wet system undamaged moist system DI 4 f s x i f x ref x ref d d.5.5.5 3 s 8 x 7 Values of Damage Index for the damaged bridge - p3 3 4 5 6 7 8 9 Mode Shape Number 8 x 4 Values of Damage Index for the moist bridge 7 6 -.5..5..5.3.35.4 Frequency [Hz] FRFs for selected nodes filtered with modal filter set to MS no. 6 5 4 3 undamaged dry system damaged dry system - p damaged dry system - p damaged dry system - p3 6 5 4 3 3 4 5 6 7 8 9 Mode Shape Number 8 x 9 Values of Damage Index for the wet briddge 6 4 8 6 4 8 7 3 4 5 6 7 8 9 Mode Shape Number Values of Damage Index for the damaged bridge - p 4 6 5 8 4 6 4 3 -.5..5..5.3.35.4 Frequency [Hz] 3 4 5 6 7 8 9 Mode Shape Number 3 4 5 6 7 8 9 Mode Shape Number
Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Damage localization The idea for extension of the method by adding damage localization, bases on two facts:. Effective modal filter can be formed for just a few sensors (in practice about 4-5),. That damage, in most of the cases, disturbs the mode shapes (modal vectors) only locally.
Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Experimental verification laboratory stand Three modal tests were performed, first on the undamaged beam. Next the cut between measuring points and 3 was introduced. The magnitude of the cut amounted consecutively % and 3 % of the beam cross-section area.
Damage index Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Experimental verification results DI 4 f s x i f x ref x ref d d s damage location.4. Damage index no. % damage 3 % damage.8.6.4. -5 4-8 7- Group of sensors
Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Case studies modal filter for footbridge Experimental modal analysis with impact excitation Nr CzDW CzDW [Hz] WTM [%].35.36.59.9 3.6.3 4.8.5 5 3..77 6 3.66.3 7 4.99. 8 5.5.6 9 5.6. 6.9.33
Magnitude Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Case studies modal filter for footbridge Ambient excitation.9.8 Output of Modal Filter set to MS no. Theoretical FRF undamaged system damaged system (added mass).7.6.5.4.3. Added mass - about kg. 3 4 5 6 7 Frequency [Hz]
Damage Index Magnitude Magnitude Magnitude Introduction Research at UST on SHM system development Designed SHM systems Vibrothermography Guided Waves Modal filtering Rotating machinery Conclusions Case studies modal filter for footbridge Modal Filter output for selected points (Group no. ) filtered with filter set to MS no. Modal Filter output for selected points (Group no. ) filtered with filter set to MS no. Modal Filter output for selected points (Group no. 3) filtered with filter set to MS no.. Theoretical FRF undamaged system damaged system (added mass). Theoretical FRF undamaged system damaged system (added mass). Theoretical FRF undamaged system damaged system (added mass).8.8.8.6.6.6.4.4.4....5.5.5 3 Frequency [Hz].5.5.5 3 Frequency [Hz].5.5.5 3 Frequency [Hz] 4 Damage Index Calculated for the Consecutive Modal Filters (Groups of Sensors) 35 3 5 5 Added mass - about kg 5 3 Group no.
Badania kładki dla pieszych nad ulicą Opolską dla Zarządu Infrastruktury Komunalnej i Transportu w Krakowie Obiekt badań: kładka dla pieszych nad ulicą Opolską. Rodzaj kładki: podwieszona, dla pieszych; Ilość przęseł: podwieszone + część podparta; Rozpiętość: 39, m + 39, m + 4, m; Pylon: stalowy, typu A; Wysokość pylonu: 34,5 m; Układ nośny: układ podwieszony, promienisty, zbieżny; Przeprowadzono dwa testy modalne, Wymuszenie impulsowe Sieć 4 punktów pomiarowych Pomiar przyspieszeń drgań. Pasmo 5 Hz
Damage Index Magnitude Badania kładki dla pieszych nad ulicą Opolską dla Zarządu Infrastruktury Komunalnej i Transportu w Krakowie.9.8.7.6.5.4.3.. Output of Modal Filter set to MS no. Theoretical FRF undamaged system damaged system (added mass) Uszkodzenie symulowane było poprzez dodanie masy około kg zlokalizowanej w punkcie pomiarowym Praw:. (około 5% masy obiektu w najbliższym otoczeniu). 4 35 Damage Index Calculated for the Consecutive Modal Filters (Groups of Sensors) 3 4 5 6 7 Frequency [Hz] 3 5 Zasymulowane uszkodzenie zostało poprawnie wykryte i zlokalizowane 5 5 3 Group no. Rejon uszkodzenia
Conclusions SHM systems improve safety of operation of critical structures Long term monitoring is required Local methods are more sensitive then global one SHM system are expensive (local system) Detection and localization, assessment are relatively easy Prediction is very difficult task Interdisciplinary approach is required SHM systems commercial use in civil structures, aviation, space, railway,power plants.