Structure optimization of a monolithic high-shock three-axis piezoresistive accelerometer using Ant Colony Optimization

Abstract

The monolithic high-shock three-axis piezoresistive accelerometer has the advantages of smaller bulk and less axis-alignment error and so on. However, because of the complexity of the accelerometer design such as multi-dimension coupling, the longitudinal and transverse overload, natural frequency, maximum deflection and other constraints for high-shock applications, it's useful to use optimization method to improve the performance. For those sensing elements without appropriate mathematic models, a feasible way to optimize the structure is using intelligent algorithms such as ACO (Ant Colony Optimization), genetic algorithm and so on. In this paper, in order to adapt the basic ACO algorithm for continuous application such as structure optimization, the correlation between the continuous value of the design variable and the path chosen by an ant is established. The sensing element structure and the principle of our designed monolithic high-shock three-axis piezoresistive accelerometer are given in succession. Then the mathematic model is established, where, the sensitivity in Z direction is chosen as the characteristic value to improve the accelerometer's sensitivities. The procedure of the optimization contains the ACO module and the FEA (Finite Element Analysis) module. The ACO module, which runs on the MATLAB platform, is used to generate the design variables, update the pheromone and control the optimization process. The FEA module, which runs on the ANSYS platform as a subroutine, is used to conduct the finite element analysis. The compassion between the optimized and un-optimized module shows the efficiency of the optimization method. This method can also be widely used in the design of other accelerometer with complex structure.