An integration framework of topology method, enhanced adaptive neuro-fuzzy inference system, water cycle algorithm with evaporation rate for design optimization for a flexure gripper

Design and analysis for flexure-based mechanisms are a challenging task thanks to their movements relied on elastic linkages. Hence, this paper presents a new optimization framework to provide a systematic design method for a flexure gripper. The optimization strategy includes the topology, modeling, and size optimization phases. In the first phase, the topology scheme for the gripper is proposed via the solids isotropic material with penalization method in terms of a full consideration of stress constraint and equal forces of both hands. In the next phase, modeling of the performances is implemented via an enhanced adaptive neuro-fuzzy inference system (EANFIS). The EANFIS's architectures are optimized by the Taguchi. The EANFIS's optimization is aimed to search the best parameters and improve the modeling accuracy. It showed that the EANFIS models have a good precision with root mean square error and standard deviation being close to zero, and coefficient of determination around one. In the last phase, the size optimization is performed by the evaporate rate-based water cycle algorithm. Two cases of the flexure gripper are considered in this phase. The results of case 1 found the hand's stroke of 0.0078 mm, the strain energy of 0.0354 mJ, the stress of 65.332 MPa, and the safety factor of 3.169. The results of case 2 identified the hand's stroke of 0.0075 mm, the stress of 66.208 MPa, and the safety factor of 3.795. Additionally, the optimized values are close to the finite element verifications. In comparison with the other methods, the results showed that the proposed framework is a best optimizer for the flexure gripper.