The Research Of Error Analysis And Correction Method About Distributed Photometer Measurement System

The distributed photometric measurement system is an important means to measure the spatial photometric distribution characteristics of lamps or light sources.However,the limitation of measuring space and extinction materials absorption rate results in a large number of stray light in the measurement environment,which reduces the accuracy of measuring the spatial distribution photometric characteristics of light sources.Therefore,based on the comprehensive analysis of stray light error,a correction method combining physical elimination structure and artificial neural network error compensation algorithm is proposed to improve the accuracy of photometric measurement.In view of the influence of stray light on the photometric measurement system,the generation and transmission characteristics of stray light in the photometric measurement system are analyzed by the geometrical optical analysis method and the stray light radiation theory.According to the characteristics of stray light,combined with the theory of approximate absolute blackbody structure and the theory of absorbance summation,a concave narrow aperture array with high refraction effect is designed.According to the influence of different aperture intervals on the reflection and absorption of stray light,the Adaptive Inertia Weight Particle Swarm Optimization(AIPSO)is designed to search for the optimal spacing of aperture,so as to build the optimal concave narrow aperture array model and effectively suppress stray light.On the basis,aiming at the inherent error of the residual stray light and the time-varying error caused by the aging of the optical measurement system after the stray light is suppressed,the Point Source Transmission(PST)method is used to analyze the stray light distribution characteristics that affect the photometric measurement,instead of further optimizing the physical structure of the system.According to PST's approximate estimation of the distribution characteristics of the system error,and considering the approximate linear relationship between the measured value and the expected value of the photometric data and the nonlinear relationship between the stray light error characteristics,a correction algorithm of the photometric system error based on Radial Basis Function Neural Network(RBF)is designed,and the correction effect is based on the larger outgoing angle of the measured light In order to improve the accuracy of photometric data,the traditional RBF algorithm is improved.Through the simulation and comparison of the optimal model of concave narrow aperture array structure and the analysis of the correction results and performance of RBF error compensation algorithm for the secondary error after clutter suppression,the results show that the measurement error can be reduced to less than 0.24%.The method which combined the optimal physical clutter elimination structure of concave narrow aperture array with RBF error compensation network is the measurement error caused by system stray light The solution of the poor problem provides an effective means.