heal.abstract |
Η παρούσα διπλωματική εργασία πραγματεύεται την σεισμική συμπεριφορά γέφυρας σε ρευστοποιήσιμο έδαφος. Εφαλτήριο της εργασίας αποτελεί το ερευνητικό πρόγραμμα με επικεφαλής τον Γεώργιο Μπουκοβάλα, Καθηγητή Ε.Μ.Π, με τίτλο “Πρότυπος σχεδιασμός βάθρων γεφυρών σε ρευστοποιήσιμο έδαφος με χρήση φυσικής σεισμικής μόνωσης”. Η ερευνητική ομάδα του Ιωάννη Ψυχάρη, Αναπληρωτή Καθηγητή Ε.Μ.Π, ασχολήθηκε με την μελέτη μιας τυπικής ισοστατικής οδικής γέφυρας από σκυρόδεμα δύο ανοιγμάτων και θεμελιωμένης σε ρευστοποιήσιμο έδαφος. Κεντρική ιδέα της έρευνας είναι η αντικατάσταση της πασσαλοθεμελίωσης του μεσοβάθρου, δηλαδή της λύσης θεμελίωσης του συμβατικού σχεδιασμού τέτοιου τύπου γεφυρών, με επιφανειακή θεμελίωση με την χρήση πεδίλου επί μιας βελτιωμένης εδαφικής κρούστας, η οποία στην ουσία είναι το άνω μέρος της ρευστοποιήσιμης ζώνης του εδάφους. Η εργασία έδωσε έμφαση στις απαιτούμενες μη γραμμικές αναλύσεις του νέου σχεδιασμού. Έγινε χρήση του προγράμματος ανοικτού κώδικα Open System for Earthquake Engineering Simulation ή εν συντομία Opensees, ανεπτυγμένο από το Pacific Earthquake Engineering Research Center. Αναλυτικότερα, πραγματοποιήθηκε η ανάλυση ευαισθησίας του φορέα σε στατικούς καταναγκασμούς στην θεμελίωση, δηλαδή στην βάση του μεσοβάθρου. Αυτοί οι στατικοί καταναγκασμοί είναι οι επιβαλλόμενες από την ρευστοποίηση μετακινήσεις, δηλαδή καθιζήσεις και ευθέως ανάλογες στροφές. Στην συνέχεια, αναζητήθηκε γι’ αυτήν την ανάλυση το κρισιμότερο στοιχείο του φορέα, αυτό που θα αστοχήσει πρώτο, μεταξύ του βάθρου, των εφεδράνων και του καταστρώματος της γέφυρας. Ακολούθως, έλαβε χώρα η διερεύνηση της σεισμικής απόκρισης του φορέα υπό την μη γραμμική δυναμική ανάλυση χρονοϊστοριών. Επιβλήθηκαν οι ίδιες χρονοϊστορίες στα ακρόβαθρα, οι οποίες εξέφραζαν την διέγερση του αναδυόμενου βραχώδους υποβάθρου και διαφορετική χρονοϊστορία στο μεσόβαθρο, η οποία αντικατόπτριζε την διέγερση του ρευστοποιημένου εδάφους. Η γεωτεχνική ομάδα παρείχε τα κατάλληλα αρχεία χρονοϊστοριών. Αποτέλεσμα, αυτών των διαφορετικών διεγέρσεων ήταν η ανάπτυξη σχετικής οριζόντιας μετακίνησης στο βάθρο της γέφυρας. Τέλος, οι χρησιμοποιούμενοι σεισμοί τριπλασιάστηκαν. Αφού, μορφώθηκαν οι καμπύλες ικανότητας του βάθρου με τις στάθμες επιτελεστικότητάς του για τις μέγιστες σχετικές οριζόντιες μετακινήσεις του βάθρου κατά την διάρκεια τόσο των απλών όσο και των τριπλασιασμένων σεισμών, έγινε μια προσπάθεια ερμηνείας της απόκρισης του βάθρου, ύστερα από την ολοκλήρωση των μεγεθυμένων σεισμικών γεγονότων και την άμεσα επακόλουθη εκδήλωση του φαινομένου της ρευστοποίησης. |
el |
heal.abstract |
The present dissertation has as starting point the main subject of the research program, entitled:
THALIS-NTUA
A New Bridge Pier Design on Liquefiable Soil with Natural Seismic Isolation
with George Bouckovalas, Professor at NTUA, as coordinator.
The aim of the research team of Ioannis Psycharis, Assistant Professor at NTUA, was to explore the feasibility of the proposed new design methodology, and the resulting advantages over conventional design methods, in the case of statically determinate concrete bridges which probably constitute the most common type of bridge in our country.
The main work tasks were the following:
(a) Initially, for this type of bridges, the allowable foundation movements (settlements and rotations) had to be established. The relevant criteria took into account the permissible damage and serviceability levels (e.g. driving discomfort, repairable damage, non-repairable damage), as well as the anticipated seismicity level (e.g. seismic excitation with 90, 450 or 900 years return period), and were established after a joint evaluation of: an extensive literature survey of relevant codes and guidelines (e.g. Eurocode 2-Part 2, Eurocode 8-Part 2, Eurocode 7, MCEER & FHA-chapter 11.4), examples of actual bridge performance during recent earthquakes, and parametric analyses of various bridge components (e.g. piers, deck, etc.) under static and cyclic dynamic loading.
(b) Next, a typical statically determinate concrete bridge with spans of 40-60 m, was designed using the conventional foundation approach, i.e. pile groups with ground improvement between and around the piles. It was the team’s intention to select an actual (existing or in the design stage) river bridge site, where the subsoil conditions are well established by geotechnical surveys, while extensive liquefaction is expected underneath one or more of the bridge piers.
(c) Finally, the static and seismic design of this bridge was repeated with the new methodology of “natural” seismic isolation (i.e. shallow foundation and partial improvement, only on the top part of the liquefiable soil), in connection with the allowable foundation movements which were established in work task (a) above. The comparative advantages and limitations of the new design methodology, relative to the conventional ones, were consequently evaluated on the basis of technical, as well as cost criteria.
Nikoleta Psilla, Dr. Civil Engineering at NTUA, developed the bridge model using SAP2000 software and the deck model using Sofistik software.
In this dissertation the bridge is modelled in Open System for Earthquake Engineering Simulation or briefly Opensees, created by the Pacific Earthquake Engineering Research Center in programming language tcl. This program is based on an open source code, which provides the user the freedom of using it without any legal restraint. In the same time, it boosts the international collaboration between the researchers in their efforts at improving and understanding the program.
It is preferable than other programs because it offers a great variety of material models, elements, solution algorithms and many combinations of the parameters above. Moreover, the user is free to interfere in the code and to develop applications, specialized in his model, for the analysis and the data processing. Plus, Opensees is perfect for the execution of dynamic non linear analyses. Finally, the time needed for a large number of analyses is relatively short and the program’s direct interaction with other ones, such as Matlab, Excel worksheets etc, is more than effective and efficient. Its only disadvantage is the absence of a graphical environment. This lack of model response visualization is a repeller and an obstacle for the user.
This study focuses on task c of the research program. Concerning the allowable displacements of the shallow foundation caused by the liquefaction, a non linear static analysis is taking place, where the imposed settlements and rotations at the pier foot are gradually increased, until either the pier, bearings or deck fails. It is proven that the pier is the most sensitive element of the bridge. That is a totally justifiable fact, since the structure is a statically determinate bridge, where the pier yield strength should not be exceeded, otherwise a mechanism is triggered and the collapse is inevitable. The same analysis was executed by the research team in SAP2000, in which the displacements were imposed one by one and not in combination. Concerning the seismic response of the structure, the time histories describing the outcrop excitation are imposed at the abutments, and a different time history describing the surface excitation is imposed at the middle pier. In Opensees, a different time history can be defined at every node of the structure. The result of this analysis was the relative horizontal displacement of the pier. During the research, due to SAP2000’s inability to run a dynamic non linear time history analysis, a dynamic spectral analysis with a simultaneous enforcement of horizontal displacement at the pier took place. Finally, the seismic excitations are multiplied by three and, taking into account the results of liquefaction, the pier capacity curve is formed.
The structure of the dissertation is the following:
Chapter 1 covers the introduction.
Chapter 2 describes the bridge simulation in Opensees. The materials and the elements of the structure are presented. With the aid of Matlab several figures of the bridge, as it is modelled in Opensees, are showed.
In Chapter 3 a non linear static analysis is carried out in order to calculate the allowable liquefaction settlements of the bridge foundation.
In Chapter 4 the code which describes in detail the prestressed deck according to the conventional design, written by Nikoleta Psilla in Sofistik, is modified in order to insert the liquefaction displacements at the bearings. After that, the deck’s serviceability limits are checked according to the Eurocodes.
In Chapter 5 the dynamic non linear time history analyses take place. As it is mentioned above, different time histories are imposed at abutments and at the middle pier. Afterward, the given time histories are multiplied by three and the residual relative horizontal pier displacements are recorded.
Chapter 6 delineates the effect of the liquefaction on the structure. We consider that the liquefaction is triggered right after the end of the seismic excitation. We distinguish two different phases. Initially, the residual displacements caused by seismic excitation, are recorded at the pier. Then, pier settlement and rotation due to liquefaction, whose values are calculated after the geotechnical Excel worksheet, set in. Finally, the pier capacity curves, which contain the pier’s limit states, are exported in terms of moment and angle chord rotation. |
en |