Particle swarm-based structural optimization of laminated composite hydrokinetic turbine blades

2014 ◽  
Vol 47 (9) ◽  
pp. 1191-1207 ◽  
Author(s):  
H. Li ◽  
K. Chandrashekhara
2021 ◽  
pp. 1-31
Author(s):  
Lorenzo Pinelli ◽  
Andrea Amedei ◽  
Enrico Meli ◽  
Federico Vanti ◽  
Benedetta Romani ◽  
...  

Abstract The need for high performances is pushing the complexity of mechanical design at very high levels, especially for turbomachinery components. Structural topology optimization methods together with additive manufacturing techniques for high resistant alloys are considered very promising tools, but their potentialities have not been deeply investigated yet for critical rotating components like new-generation turbine blades. This research work proposes a methodology for the design, the optimization and the additive manufacturing of extremely stressed turbomachinery components like turbine blade-rows. The presented procedure pays particular attention to important aspects of the problems as fluid-structure interactions and fatigue of materials, going beyond the standard structural optimization approaches found in the literature. The numerical procedure shows robustness and efficiency, making the proposed methodology a good tool for rapid design and prototyping, and for reducing the design costs and the time-to-market typical of these mechanical elements. The procedure has been applied to a low-pressure turbine rotor to improve the aeromechanical behavior while keeping the aerodynamic performance. From the original geometry, mode-shapes, forcing functions and aerodynamic damping have been numerically evaluated and are used as input data for the following topological optimization. Finally, the optimized geometry has been verified in order to confirm the improved aeromechanical design. After the structural topology optimization, the final geometries provided by the procedure have been then properly rendered to make them suitable for additive manufacturing. Some prototypes of the new optimized turbine blade have been manufactured to be tested in terms of fatigue.


2018 ◽  
Vol 52 (24) ◽  
pp. 3381-3398 ◽  
Author(s):  
Oscar Castro ◽  
Kim Branner ◽  
Nikolay Dimitrov

A probabilistic model for estimating the fatigue life of laminated composite materials considering the uncertainty in their mechanical properties is developed. The uncertainty in the material properties is determined from fatigue coupon tests. Based on this uncertainty, probabilistic constant life diagrams are developed which can efficiently estimate probabilistic ɛ-N curves at any load level and stress ratio. The probabilistic ɛ-N curve information is used in a reliability analysis for fatigue limit state proposed for estimating the probability of failure of composite laminates under variable amplitude loading cycles. Fatigue life predictions of unidirectional and multi-directional glass/epoxy laminates are carried out to validate the proposed model against experimental data. The probabilistic fatigue behavior of laminates is analyzed under constant amplitude loading conditions as well as under both repeated block tests and spectral fatigue using the WISPER, WISPERX, and NEW WISPER load sequences for wind turbine blades.


Author(s):  
Andrea Amedei ◽  
Enrico Meli ◽  
Andrea Rindi ◽  
Benedetta Romani ◽  
Lorenzo Pinelli ◽  
...  

Abstract The need for high performances is pushing the complexity of mechanical design at very high levels, especially for turbomachinery components. In this field, structural topology optimization methods together with additive manufacturing techniques for high resistant alloys are considered very promising tools, but their potentialities have not been deeply investigated yet for critical rotating components like new-generation turbine blades. In this framework, this research work proposes a methodology for the design, the optimization and the additive manufacturing of extremely stressed turbomachinery components like turbine blade-rows. The presented procedure pays particular attention to important aspects of the problems as fluid-structure interactions (forced response and flutter phenomena) and fatigue of materials, going beyond the standard structural optimization approaches found in the literature. The new design strategy enables a substantial reduction of the component mass, limiting the maximum stress and improving the vibrational behaviour of the system in terms of eigenfrequencies, modal shapes and fatigue life. Furthermore, the numerical procedure shows robustness and efficiency, making the proposed methodology a good tool for rapid design and prototyping, and for reducing the design costs and the time-to-market typical of this kind of mechanical elements. The procedure has been applied to a low-pressure turbine rotor to improve the aeromechanical behavior while keeping the aerodynamic performance. From the original geometry, mode-shapes, forcing functions (due to rotor/stator interactions) and aerodynamic damping have been numerically evaluated and are used as input data for the following topological optimization. Finally, the optimized geometry has been verified in order to confirm the improved aeromechanical design. After the structural topology optimization, the final geometries provided by the procedure have been then properly rendered to make them suitable for additive manufacturing. Some prototypes of the new optimized turbine blade have been manufactured from aluminum to be tested mechanically and in terms of fatigue.


2007 ◽  
Vol 348-349 ◽  
pp. 725-728 ◽  
Author(s):  
Omer Soykasap ◽  
Şükrü Karakaya

In this study, the structural optimization of laminated composite plates for maximum buckling load capacity is performed by using genetic algorithm. The composite plate under consideration is a 64-ply laminate made of graphite/epoxy, is simply supported on four sides, and subject to in-plane compressive static loads. The critical buckling loads are determined for several load cases and different plate aspect ratios using 2-ply stacks of 02, ±45, 902. The problem has multiple global solutions, the results of which are compared with previously published results.


Sign in / Sign up

Export Citation Format

Share Document