Design and optimization of a new large stroke micropositioner based on cricket-mimicked bistable mechanism for potential application in polishing

• Published in: International journal of advanced manufacturing technology 2023-07, Vol.127 (1-2), p.123-149
• Main Authors: Tran, Ngoc Thoai, Dang, Minh Phung, Dao, Thanh-Phong
• Format: Article
• Language: English
• Subjects: Artificial neural networks; CAD; CAE) and Design; Computer aided design; Computer-Aided Engineering (CAD; Design optimization; Design parameters; Engineering; Fused deposition modeling; Fuzzy logic; Industrial and Production Engineering; Mathematical models; Mechanical Engineering; Media Management; Micropositioning; Multiple objective analysis; Optimization; Original Article; Performance indices; Polishing; Stiffness; Three dimensional models; Three dimensional printing
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-023-11489-9

To overcome the limited stroke of existing micropositioning stages in precision engineering systems, this article proposes an optimal design of a new micropositioner based on the cricket-mimicked bistable mechanism. The suggested micropositioner is potential for polishing application. The proposed bistable mechanism is combined with a positive-stiffness mechanism to achieve a large stroke with centimeter range. The design targets of micropositioner are to deliver a large stroke, a high frequency, and a small parasitic motion but also ensure working safety. To solve three objective functions with four constraints, a hybrid optimization approach is proposed, namely fuzzy logic, teaching learning-based optimization (TLBO), and artificial neural network (ANN). The fuzzy logic is proposed to combine three objective functions into a single objective function, so-called output fuzzy. In the modeling, the TLBO is employed to determine the optimum ANN structure. Then, the multi-objective optimization problem of the micropositioner is converted into the single optimization task through the TLBO. Besides, the influences of the geometrical parameters on the performance qualities of the micropositioner are investigated. The results showed the performance indexes of TLBO-assisted ANN are reasonable and reliable. The optimum design parameters are found at l 1  = 60.93 mm, l 2  = 10.42 mm, t 1  = 2.77 mm, t 2  = 1.37 mm, and t 3  = 9.28 mm. A prototype of ABS micropositioner is manufactured by fused deposition modeling 3D printer. The displacement and the parasitic motion were experimentally measured about 14.9513 mm and 0.0061 mm, respectively. The frequency and the stress were simulated in ANSYS software about 617.9227 Hz and 38.7659 MPa, respectively. The output stroke is over 1 cm which is potential for wide applications.