Date Approved
8-17-2026
Graduate Degree Type
Thesis
Degree Name
Engineering (M.S.E.)
Degree Program
School of Engineering
First Advisor
Arjumand Ali
Academic Year
2025/2026
Abstract
Actively controlled truss structures are important in applications requiring efficient vibration suppression and dynamic performance beyond the capabilities of passive methods alone. This study directly compares predetermined truss structures derived from engineering intuition with topology optimization-derived trusses under identical boundary and loading conditions. To evaluate the robustness of topology optimization as a design methodology, each truss was evaluated using three controller architectures (PID, LQR, and MPC) and two optimization algorithms, one deterministic (Stress Loop Method) and one stochastic metaheuristic (Grey Wolf Optimization). Each controller, truss, and algorithm combination was subjected to noisy operating conditions with 100 replicates for analysis of variance (ANOVA). The performance metrics considered were settling time, actuator effort, structural weight, and a combined objective function incorporating all three metrics. To isolate the effect of topology optimization, boundary condition, controller type, and optimization algorithm were included as blocking factors in the analysis.
Three of the four performance metrics were statistically significant settling time, structural weight, and the combined objective function, with estimated topology effects of −0.233, −13.536, and −1.125, respectively. Although the estimated effect for actuator effort was negative (−1478.9), suggesting a directional reduction, the effect was not statistically significant (p = 0.233). Because topology optimization minimizes compliance, effectively increasing structural stiffness, topology-optimized trusses reduced structural weight and settling time while requiring comparable actuator effort to return the structure to equilibrium. These results demonstrate the inherent coupling between structural and control performance, as changes in topology alter the system matrices governing dynamic response and, consequently, controller behavior. This coupling emphasizes the importance of considering structural and control objectives together when designing actively controlled truss structures. The statistically significant reduction in the combined objective function further suggests that topology optimization provides a robust framework for improving overall performance, while the diverse selection of controller architectures and optimization algorithms considered provides insight into the potential generalizability of topology optimization across design methodologies.
ScholarWorks Citation
DiCanio, Charles Richard, "Statistical Evaluation of Topology Optimization Across Design Methodologies for Actively Controlled Truss Structures" (2026). Masters Theses. 1194.
https://scholarworks.gvsu.edu/theses/1194
