Differential evolution (DE) is an efficient population based algorithm used to solve real-valued optimization problems It has the advantage of incorporating relatively simple and efficient mutation and crossover operators However, the DE operator is based on floating-point representation only, and is difficult to use when solving combinatorial optimization problems In this paper, a modified binary differential evolution (MBDE) based on a binary bit-string framework with a simple and new binary mutation mechanism is proposed Two test functions are applied to verify the MBDE framework with the new binary mutation mechanism, and four structural topology optimization problems are used to study the performance of the proposed MBDE algorithm The experimental studies show that the proposed MBDE algorithm is not only suitable for structural topology optimization, but also has high viability in terms of solving numerical optimization problems

/pdf/topology-optimization-of-structures-using-modified-binary-56us0aqvz6.pdf

Topology optimization of structures using modified binary differential evolution

Design of Variable Stiffness Actuator Based on Modified Gear–Rack Mechanism

Topology optimization of hyperelastic structure based on a directly coupled finite element and element-free Galerkin method

Variable stiffness can improve the capability of human–robot interacting. Based on the mechanism of a flexible rack and gear, a rotational joint actuator named vsaFGR is proposed to regulate the joint stiffness. The flexible gear rack can be regarded as a combination of a non-linear elastic element and a linear adjusting mechanism, providing benefits of compactness. The joint stiffness is in the range of 217–3527 N.m/rad, and it is inversely proportional to the 4th-order of the gear displacement, and nearly independent from the joint angular deflection, providing benefits of quick stiffness regulation in a short displacement of 20 mm. The gear displacement with respect to the flexible gear rack is perpendicular to the joint loading force, so the power required for stiffness regulating is as low as 14.4 W, providing benefits of energy saving. The working principles of vsaFGR are elaborated, followed by presentation on the mechanics model and the prototype. The high compactness, great stiffness range and low power cost of vsaFGR are proved by simulations and experiments.

Design and implementation of a variable stiffness actuator based on flexible gear rack mechanism

Differential evolution (DE) is a heuristic optimization method used to solve many optimization problems in real-value search space. It has the advantage of incorporating a relatively simple and efficient form of mutation and crossover. However, the operator of DE is based on floating-point representation only, and is difficult to use in solving combinatorial optimization problems. In this article, a modified binary DE is developed using binary bit-string frameworks with a logical operation binary mutation mechanism. Further, a new stress-based binary mutation mechanism is also proposed to drive the binary DE search towards the optimal topology of the structure with higher performance and fewer objective function evaluations. The numerical results show that the performance of the proposed algorithm using stress-based binary mutation has high capability and efficiency in topology optimization of the structure.

http://pic.sagepub.com/content/224/2/443.full.pdf

Stress-based binary differential evolution for topology optimization of structures

A new topology optimization method for designing thermomechanical compliant actuators using meshless methods is presented. The meshless method is employed to discretize the design domain and the bulk density field, which can avoid the undesirable mesh distortion when the standard finite element method (FEM) is applied. The method for imposing boundary conditions in the meshless method is discussed in details. An interpolation scheme is applied to indicate the dependence of material mechanical and thermal properties on pseudo-densities which are distributed to the related integration points. Furthermore, the design sensitivity analysis is performed by incorporating the adjoint approach into the meshless method. The optimization problem is then solved by the method of moving asymptotes (MMA). The availability of the proposed method is demonstrated by using a widely studied compliant structure.

Zifan Fang

Papers

Thermomechanical compliant actuator design using meshless topology optimization