Multi-objective optimization of a single component and a two component Load cell

Sourabh Patle, Pankaj Priyadarshi (Indian Institute of Space Science and Technology)

CHALLENGE - The main aim of this project is to carry out single objective and multi-objective design optimization of single component and two component load cells. For measuring the loads on vehicles, force measurements are typically carried out. In a previous study, it has been seen that a parallelogram beam type load cell is sensitive only to the normal loading and insensitive to all other loads and moments. The best performing design needs to be explored using formal optimization. 

SOLUTION -  Effect of cavity size and fillet radius has been studied through FE analysis. We have studied the FEA on parallelogram type load cell and later optimized in single and multi-objective optimization in modeFRONTIER using ANSYS 14.5. The design variables of a single load cell which define the cavity are optimized such that the Wheatstone bridge output is maximized and mass is minimized. In the optimization of the two component load cell, the design variables define the external size, and shapes of cavities are optimized to obtain optimal configuration under constraints of factor of safety. 










BENEFITS - Multi-objective optimization studies have been carried out using the genetic algorithm NSGA-II in modeFRONTIER, with the aim of increasing the bridge output and reducing the mass. Results of the optimization of a load cell whose outer dimensions are constrained show a large discontinuity of mass domain between 0.441Kg to 0.451 Kg and domain of bridge output from 50μϵ to 500μϵ. The results for a load cell in which the external size along with cavity related variables are being optimized together show a domain of bridge output from 900μϵ to 1400μϵ.