Computer Science

CLC number: TP399

Study on the Effect of Spot Size and Non-Linearity on PSD Positioning Accuracy

光斑尺寸及非线性度对PSD定位精度影响研究

¹ School of Electronic and Electrical Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
² School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

^† Corresponding author. E-mail: zhenhuazhang1985@163.com

Received: 2 December 2024

Abstract

Position-sensitive detector (PSD) is widely used in precision measurement fields such as flatness detection, auto-collimator systems, and degrees of freedom testing. However, due to factors such as uneven surface resistance and differences in electrode structures, the nonlinearity of PSD becomes increasingly severe as the photosensitive surface moves from the center toward the edges of the four electrodes. To address this issue, a PSD nonlinearity correction algorithm is proposed. The algorithm utilizes the particle swarm optimization (PSO) algorithm to determine the optimal weights and thresholds, providing better initial parameters for the back propagation (BP) neural network. The BP neural network then iterates continuously until the error conditions are met, completing the correction process. Furthermore, a PSD nonlinearity correction system was developed, and the influence of different spot sizes on PSD positioning accuracy was simulated based on the current equation under the Gaussian spot model. This validated the robustness of the correction algorithm under varying spot sizes. The results demonstrate that the overall optimized error is reduced by 84.51%, and for spot sizes smaller than 1 mm, the error reduction exceeds 93.89%. This method not only meets the measurement accuracy requirements but also extends the measurement range of PSD.

摘要

位置敏感探测器(PSD)被广泛应用于平整度检测、自准直仪系统及自由度测试等精密测量领域。然而，由于表面电阻不均匀和电极结构差异等因素，PSD的非线性问题在光敏表面从中心向四个电极的边缘靠近时变得越来越严重。针对这一问题，提出了一种PSD非线性校正算法，利用粒子群(PSO)算法计算出最佳权值和阈值，为BP神经网络提供较好的初始参数，再通过BP神经网络不断迭代，直到满足误差条件，完成校正过程。此外，构建了PSD非线性校正系统，根据高斯光斑模型条件下的电流方程，仿真不同的光斑尺寸对PSD定位精度的影响，验证了校正算法在不同光斑尺寸下的鲁棒性。结果表明，优化后的整体误差减少了84.51%，当光斑尺寸小于1mm时，误差减少超过93.89%。该方法不仅满足了测量精度要求，还扩展了PSD的测量范围。