PSI - Issue 10

N.G. Pnevmatikos et al. / Procedia Structural Integrity 10 (2018) 195–202 N.G. Pnevmatikos et al. / Structural Integrity Procedia 00 (2018) 000 – 000

199

5

3. Case study: Analysis, results and discussion

Linear and non-linear analyses were executed in a two bay of 4m span and two story of 3-meter height steel moment frame. The material was of grade S235. The beams and columns sections are IPE330 and HEB240, respectively. The vertical loads cause bending moment at the beam equal to 40% of the yield moment. The model of structure and the point which the stresses were calculated is shown in Fig.3. For low cycle fatigue procedure, the Aigio, 1995, earthquake excitation with Mw=6.1 Richter scale and peak ground acceleration PGA=0.54g was considered as fatigue load. The twice of this excitation was also used. For low cycle fatigue a nonlinear time history analysis was performed. While, for high cycle fatigue a 1/3 of Aigio earth quake excitation was applied and linear time history analysis was performed. The time history record and the spectral acceleration of the Aigio earthquake are shown in Fig.4. In Table 2 the combinations of fatigue method (low or high cycle fatigue) and analysis case (linear or nonlinear) which were performed, with respect to the level of earthquake excitation, are summarized.

Fig. 3. Structural model and the connection point of calculation the stress.

Acceleration (m/sec 2 )

(m/sec 2 )

Spectral Acceleration

Time (sec)

Period (sec)

Fig. 4. The time history record and the spectral acceleration of the Aigio earthquake excitation.

For high cycle fatigue, linear dynamic analysis of frame subjected to 1/3 of Aigio ground motion was performed and the moment history at point 3 (see Fig.3) was calculated. Dividing the moment by the beam section modulus, stresses at connection vs. time are calculated and shown in Fig.5. For high cycle fatigue, applying the reservoir method in the stress- time diagram, the spectrum of variation of stresses, Δσ, at the beam -column connection were calculated and shown in Table 3. The range of stress, Δσ, is increased by a factor of 1.5 as defined by the EC. From Tables 8.1 to 8.10 of EN 1993-1-9, a connection detailed category is chosen. The detail examined is in node 3 of Fig.3 and belongs to category 90. By making a parametric analysis, category 80 is also examined and is comparatively considered. A mean number per year of small earthquakes, representing the aftershocks earthquakes, were assumed to happen in the region. Using the fatigue chart from EC3 part 1- 9 for a detail category 80 and 90 and applying the Miner’s