High-accuracy and efficient method for calibrating spatial laser beam based on optimized PSD

Nonlinear error compensation for 2D PSD based on SSA-BP neural network

“…n1 is the number of hidden layer nodes, n is the number of input layer nodes, m is the number of output layer nodes, and is a constant between [1,10]. According to the empirical formula, the range of hidden layer nodes in this model is between [3,12]. To determine the optimal BP neural network parameters, each hidden layer node is trained 10 times and the average value is taken, with mean square error (MSE) as the standard.…”

“…They can be widely applied in displacement measurement, attitude detection, beam positioning, tracking, and calibration. [1][2][3][4] Due to the working environment and inherent characteristics, the output current of PSD devices often produces nonlinear errors, leading to a decrease in the overall accuracy of the measurement system. [5] To compensatethe nonlinear errors and improve the accuracy of PSD devices, two major types of nonlinear error compensation methods based on hardware system and software system compensation have been proposed.…”