Two approaches to the optimal design of composite flywheels

Abstract

In this article two approaches to the design of reinforced composite flywheels are presented. The main goal of the optimization procedure is to maximize the accumulated kinetic energy of a flywheel. The first approach is based on a discrete model of reinforcement, causing the discontinuity of static fields along reinforcement and preserving the continuity of kinematic fields. In the second approach, the material of the reinforced flywheel is subjected to the homogenization procedure using the Halpin-Tsai assumption and then the continuity of both static and kinematic fields is preserved within flywheel domain. The evolutionary algorithm was used in both cases to determine the optimal shape of reinforcements, while the finite element method was applied in order to analyze the mechanical response of a flywheel.

A bilevel game theoretic approach to optimum design of flywheels

“…A uniform cross-section for flywheel is quite uneconomical because all the material is not fully stressed. Besides the hyperbolic crosssection profile proposed by Stodola, mathematical programming techniques have been applied (Bhavikatti and Ramakrishnan 1980, Sandgren and Ragsdell 1983, Dems and Turant 2009) to optimize the cross-section of rotating discs. In some approaches (Bhavikatti and Ramakrishnan 1980), the disc is approximated by a number of rings and the optimum profile is obtained by smoothing the stepped shape.…”

An efficient reanalysis assisted optimization for variable-stiffness composite design by using path functions

“…However, as the number of control variables is usually relatively larger, the heuristic algorithms are more commonly used in the variable-stiffness composite optimization. The GA may be the most widely used heuristic algorithm in variable-stiffness composite optimization [43][44][45][46][47], and other methods are also involved, such as artificial immune algorithm [48] and evolutionary algorithm [49]. In engineering problems, the optimization objective functions are usually calculated by computational expensive simulations, which will lead to a low efficiency of optimization, especially when heuristic algorithms are used.…”

Optimum Design of Composite Structures: A Literature Survey (1969–2009)

“…886 Other than behavioral constraints, side constraints may be applied; e.g. design space may be restricted to positive values of thickness and cross-sectional area 50 584 stress interaction criterion, 511 C2 criterion, 150 Ye criterion, 964 Puck failure criterion, 882,974,975 Chang-Chang failure criterion, 937 Cheng and Lessard criterion, 299,475 Yamada failure criterion, 143 maximum distortion energy theory, 9,15,47,58,200,242,267,291,330,542,674,759,763,765,767,827,847,894,951 Huber-Mises criterion, 563,745,942,954 Treska, 5 maximum normal stress theory, 761,970 critical failure volume method, 949 point stress criterion, 112,183,519 where failure criterion is checked at a specific distance for a notch, multiscale stress criteria, 773,846 fracture mechanics, 73,299, 338 continuum damage mechanics, 501 energy based failure criteria, 438 failure-mechanism based failure criteria, 917,…”

“…In many of these studies, plain fiber-reinforced laminated composite plates were considered for optimization; besides, others types of composite structures were optimized: Structures reinforced with short or long fibers, 6,70,113,300,545,618,696,899,958 particulate fillers, 352,425,737,821 braided or woven fibers, 291,326,336,347,425,530,531,710,880,884,923,1006 or carbon nanotubes, 993 laminates with layers having variable fiber orientation, 149,173,219,243,247,271,311,325,355,359,446,564,633,709,720,730,732,771,787,796,797,814,831,877,885,900,906,910,951–954 or variable fiber density, 325,472,…”