TY - JOUR
T1 - Towards a lifecycle oriented design of infrastructure by mathematical optimization
AU - Kufner, T.
AU - Leugering, G.
AU - Martin, A.
AU - Medgenberg, J.
AU - Schelbert, J.
AU - Schewe, L.
AU - Stingl, M.
AU - Strohmeyer, C.
AU - Walther, M.
PY - 2019/3/15
Y1 - 2019/3/15
N2 - Today’s infrastructures are mainly designed heuristically using state-of-the-art simulation software and engineering approaches. However, due to complexity, only part of the restrictions and costs that show up during the lifecycle can be taken into account. In this paper, we focus on a typical and important class of infrastructure problems, the design of high-pressure steam pipes in power plants, and describe a holistic approach taking all design, physical, and technical constraints and the costs over the full lifecycle into account. The problem leads to a large-scale mixed-integer optimization problem with partial differential equation (PDE) constraints which will be addressed hierarchically. The hierarchy consists of a combinatorial and a PDE-constrained optimization problem. The final design is evaluated with respect to damage, using beam models that are nonlinear with respect to kinematics as well as constitutive law. We demonstrate the success of our approach on a real-world instance from our industrial partner Bilfinger SE.
AB - Today’s infrastructures are mainly designed heuristically using state-of-the-art simulation software and engineering approaches. However, due to complexity, only part of the restrictions and costs that show up during the lifecycle can be taken into account. In this paper, we focus on a typical and important class of infrastructure problems, the design of high-pressure steam pipes in power plants, and describe a holistic approach taking all design, physical, and technical constraints and the costs over the full lifecycle into account. The problem leads to a large-scale mixed-integer optimization problem with partial differential equation (PDE) constraints which will be addressed hierarchically. The hierarchy consists of a combinatorial and a PDE-constrained optimization problem. The final design is evaluated with respect to damage, using beam models that are nonlinear with respect to kinematics as well as constitutive law. We demonstrate the success of our approach on a real-world instance from our industrial partner Bilfinger SE.
KW - Lifecycle oriented design
KW - Mixed integer nonlinear programming
KW - Structural optimization
UR - http://www.scopus.com/inward/record.url?scp=85056138599&partnerID=8YFLogxK
U2 - 10.1007/s11081-018-9406-5
DO - 10.1007/s11081-018-9406-5
M3 - Article
AN - SCOPUS:85056138599
VL - 20
SP - 215
EP - 249
JO - Optimization and engineering
JF - Optimization and engineering
SN - 1389-4420
IS - 1
ER -