In Figure 2-2, the position of the shear walls, the dimensions, and the sizes are specified. Figure 2-1 displays the building’s aerial image in which the mentioned instruments can be observed. The position of the shear walls and building’s orientation relative to the north is also demonstrated in Figure 2-2. The building is placed on a concrete mat foundation with concrete pile caps.

Despite the existence of doors, windows, entrances, on the shear walls up to the roof level, the building still could be considered as symmetric structure.

Figure 2-2 Aerial image of Building 54 and plan of shear walls

2-2 Site Condition

The Building 54 is near the Charles river. The structure was placed on the alluvial deposits of the river. The geotechnical studies of site condition have been completely performed, and the soil properties especially shear wave velocity of the soil were determined. Figure 2-3 represents the structure position near Charles river. The site was 6.1 meters above mean sea level and according to the geotechnical studies; the shear wave velocity was variable between 100-600 meter per second. The depth of bed-rock was estimated around 30.55 to 34 meters from the ground surface. Figure 2-4 displays the condition and profile of the shear wave velocity which various investigations have been conducted on the amount of site damping and resonant frequency, as well.

 

 

 

 

In 2001, Hyles et al. have investigated the natural frequency (resonant frequency) of the Building 54, and measured it to 0.7 Hz.

 

 

 

 

 

 

 

 

 

 

In addition to geotechnical data, the site seismicity data also were available. In 2012, a relatively mild earthquake was occurred in Maine State at a distance of 170 kilometers from Massachusetts, in which the recorded data with the accelerometers were available. They were used as input for the dynamic analysis and structure evaluation. The acceleration time history of this earthquake is depicted in Figure 2-5.

Figure 2-3 A view of Building 54 beside the Charles river

Figure 2-4 The condition and profile of shear wave velocity

Figure 2-5 The accelerogram of the earthquake occurred near the Building 54

2-3 Modeling and Dynamic Analysis

In the research studies, it is essential that a reference model be available in order to evaluate the earthquake characterization and the soil/site properties. Previously, I used to utilize a library building in Caltech, called Millikan, as the reference model. However, it seems to be more advantageous to consider the Building 54 as the reference because it includes more precise structural data. In addition, a supplementary evaluation of this structure has been performed by Professor Oral Buyukozturk (faculty member of civil and environment department, MIT, Cambridge) and Professor Kausel (faculty member of civil engineering) from MIT, Cambridge.

On this structure, 36 points were detected to install the accelerometers as every of which was connected to a 36-ports data logger. When the Maine earthquake happened, the quake acceleration at different points on the structure, the displacements, and other information related to structure’s behavior were recorded via the installed sensors. The simulation and dynamic analysis of the building under seismic loads were compared to the real measured values. Figure 2-6 shows the position of installed accelerometers on the Building 54.

Every of accelerometers had the capability to save the 200 records per second. It is noteworthy to realize that the natural frequency of the structure, using the sensors, was also measured equal to 0.7 Hz. The weight of each story was about 422 tons, and the total weight of the building containing the instruments was estimated as 8862 tons.

My motivation for modeling and dynamic analysis of this building was the similarity of geometry and height match of this structure to the Zurich silo, which was described in section 1. The both structures are nearly 100 meters in height; both are thin and have prismatic section. The thin structures normally are soft on one side and rather tough on the other side. Moreover, the two buildings are based on extended foundation and placed on the piles. As the alluvial bed situation of Switzerland site was similar to the site of Building 54 in Cambridge, I came to the decision to compare the seismic behavior of these two structures. Note that the dynamic analysis of the Zurich silo was calculated considering a weaker earthquake rather than the one assumed for the Maine State because Zurich silo is located in a much lower risky zone. However, due to the higher weight of the Zurich silo, the base shears were close to each other and thus the results were comparable.

The modeling of Building 54 has been thoroughly conducted; it means that the structure was simulated along with the surrounding soil. Although the structure data were clear and definite, there was considerable uncertainty in soil properties which caused complex challenging in analysis.

The cone model was utitized for the soil simulation. This method contains some approximations in comparison with Finite Element (FE) and Boundary Element (BE) methods. Figure 2-7 demonstrates the concept of Soil-Structure Interaction (SSI) using CONE method.

In my opinion, the most specialized scientists in the field of dynamic foundation response and dynamic soil-pile interaction, who have presented reliable models, are as following:

Professor Kausel from MIT, Cambridge; Dr Wolf from ETH; Professor Gazetas from Athens University.

Figure 2-6 The accelerometer’s positions on the Building 54

Figure 2-7 The general concept of CONE method in soil-structure interaction

2-4 Hints

Regarding the recorded data of earthquake and the information obtained from the accelerometers and installed sensors, the structure was simulated; then the dynamic load was applied using the earthquake, and the dynamic analysis was conducted considering the soil-structure interaction based on CONE model. The values of measured accelerations and displacements were compared at different points of the building, particularly at the roof. The results indicated a difference of about 30-35%. It should be mentioned that the CONE model is not as precise as the other methods with the approximation value of about 25%. Hence, the obtained error of 30-35% would be acceptable.

It should be noted that in such analyses a proper range should be considered to select the mechanical, dynamical, and soil seismic characteristics. Considering the soil structure interaction typically cause dealing with several uncertainties and vague points. The results of such dynamic analyses may require determining the limitation of materials properties and models. In the engineering and research modeling, the limitation ranges are determined in such a way that one could benefit a part of this information domain based on his technical consideration, which mostly are conservative choices. The uncertainties have remarkable impact on structure response and hence it is essential to be cautious in this matter.

In the current research, students and colleagues from the University of Tehran; Amirkabir University of Technology; Tabriz, Shiraz, Isfahan, and Azad Qazvin Universities,  also have cooperated in the simulation process and performing the other steps, and I would gladly appreciate that.

Finally, I would like to present a summary of the Professor Oral Buyukozturk’s supplementary investigations which could be classified in the Non-Destructive Testing (NDT) category. In this method, the super speed cameras were installed on the roof of the building adjacent to the MIT building, with a distance of 175 meters, in which the cameras have recorded every displacement of more than 0.07 millimeters.

The mentioned characterization and structure evaluation method is normally used for vibration analysis of civil infrastructures as it was also applied to the Building 54. Figure 2-8 shows the required instruments for this experiment.

Based on adapting the obtained images with the results of laser vibration meter, the resonant frequency of the structure was calculated to 0.7 Hz which sufficiently agrees well with the other methods result. In the meeting session with Professor Oral Buyukozturk on 28^th May 2016, it was approved that the hybrid of these methods with performing the dynamic modeling explains how the dynamic properties and dynamic responses may get fluctuated under the influence of some parameters. The examples of these factors are elapsed time, structure dilapidation, crack appearance, displacements, and some other similar cases. In fact, using the mentioned methods would evaluate the behavior variations of structure over the life cycle.