Parametric hybrid modular timber construction

Parametric hybrid modular timber construction: Computational design approach for modular construction to discover different building typologies based on global and local structural requirements

Hamelijnck, Max (TU Delft Civil Engineering and Geosciences)

Hendriks, M.A.N. (mentor)
Ravenshorst, G.J.P. (graduation committee)
van der Linden, L.P.L. (graduation committee)
van Telgen, M.V. (graduation committee)

Delft University of Technology

Civil Engineering | Building Engineering - Structural Design

2021-08-17

This master thesis' main objective is to develop a parametric design tool which improves the design process and explores the building typologies of a hybrid modular timber construction based on global and local structural requirements. The hybrid modular timber construction is core-less and combines load-bearing and corner-supported modules. The parametric design tool, which is named Habitat21, consists of an automatic placement function (Grasshopper), a manual placement function (Rhinoceros) and a structural analysis program (Karamba). The hybrid modular timber construction concept is tested by comparing the displacements in the x- and y-direction of a replicate of Hotel Jakarta, generated in Habitat21, with the original displacements in the x- and y-direction of Hotel Jakarta, generated by the case study of Gijzen. It appears that the replicated hybrid modular timber construction of Hotel Jakarta in Habitat 21 is able to approximately reproduce the original displacement in the y-direction of Hotel Jakarta calculated by Gijzen (difference of 0.642 mm). However, the displacement in the x-direction could not be reproduced in the same order of magnitude as the original displacement in the x-direction (difference of 13.081 mm). This could be caused by the fact the assumed inter-module connections, connection 1 and 4, of the hybrid modular timber construction concept were modelled as line elements with two hinges (”pendelstaven”), this could have influenced the displacement behaviour of the CLT shear wall mechanism. To examine the usability of the manual placement of the modules by Rhinoceros within Habitat21 a workshop with 5 participants (2 architects and 3 engineers) and two cases is performed. In the first case each participant created 3 buildings, each building was generated within 8 minutes, resulting in a total of fifteen different building typologies of the hybrid modular timber construction concept. Ten out of the fifteen building typologies satisfied the global structural requirements, five did not satisfy the global structural requirements. Two buildings were not stable in one or more directions, two buildings did not satisfy the unit check for the displacement in the x-, y- or z-direction and for one building an error occurred. To explore the diversity of the typologies from the fifteen buildings a typology score was given for each building. The highest score was 102 points and the lowest score was 10. This indicates a relatively large diversity in typologies. In the second case each participant worked on two buildings that both did not satisfy the global and local structural requirements but the second building had 2 extra design restrictions. The participants had 5 minutes per building to fix this. For the building with extra design restrictions three participants (all engineers) were not able to meet the local structural requirements. This suggests that the feedback from Habitat21 provided to the user in this case was not adequate. To examine the automatic placement of the modules by Grasshopper within Habitat21 the calculation time of the automatic placement of one extra CLT shear wall is analysed for 6 different buildings by using an iterative process. These buildings were all stable in the y- and z-direction but unstable in the x-direction. The calculation time for the placement of the extra CLT shear wall at a building with 4 modules and 8 possible positions was 6 seconds and the calculation time for the placement of the extra CLT shear wall at a building with 24 modules and 48 possible positions was 360 seconds. Combining the calculation times of the 6 different buildings it appeared that the calculation time increased exponentially when the amount of modules increased linearly. In conclusion, it was possible to discretize a hybrid modular timber construction in a parametric design tool. Furthermore, both engineers and architects were able to work with the tool and to independently create diverse typologies of a hybrid modular timber construction. The boundaries of the hybrid modular timber construction concept in Habitat 21 were determined by the inter-module connections, as found in the comparison study with Hotel Jakarta. The applicability of the parametric design tool and the hybrid modular timber construction concept could be increased by more research into inter-module connections and point supported cross-laminated timber slabs. Also, feedback about how the structural systems works could be implemented in the parametric design tool, where the user gets necessary feedback to fulfill the global or local structural requirements instead of using a trial and error design method.

Parametric Design
Parametric design tool
Modular construction
Timber
Karamba

http://resolver.tudelft.nl/uuid:63fd5c41-77a6-44ca-8a73-4dd5449a2f22

Student theses

master thesis

© 2021 Max Hamelijnck