A General Approach for Optimization of Ultrasonic Array Parameter and Setup Using Principal Component Regression.

Abstract

When an optimization approach is taken for ultrasonic array design, the relationship between imaging performance and acoustic beam characteristics (ABCs) needs to be considered. A general optimization approach for ultrasonic array design is proposed based on principal component regression (PCR), which is used to establish the relationship between imaging performance indicators (IPIs) and ABC. A linear and a circular ultrasonic array design scenarios are simulated to demonstrate the effectiveness of the proposed approach. A set of beam and reflector response simulations are implemented to collect beam characteristics and IPIs, an explicit mapping function between which is built by PCR. Then, the mapping function is applied as an objective function to form an optimization model. The global optimization algorithm is used to locate the optimal solution of array parameters and setup. Simulation results demonstrate that the acoustic beam and imaging performance are associated quantitatively by PCR mapping, making the relationship explicitly interpretable, that the beam and imaging can be optimized simultaneously, and that acoustic response modeling may be skipped during iterative optimization, speeding up the approach.