Study on Localization of High Radiation Dose Regions Based on Dose Gradient Characteristics and the Least Squares Method

Abstract: In nuclear decommissioning scenarios, determining the distribution of radiation doses in the spatial domain, especially the location of high-dose areas, plays an important role in monitoring nuclear decommissioning facilities. The conventional method is to perform interpolation between measurement points and judge the location of high-dose areas based on the interpolation results. However, the positioning deviation of results obtained by currently popular interpolation methods is relatively large. Therefore, this paper proposes a method for locating high-dose radiation areas based on dose gradient characteristics and the least squares method, aiming to improve the accuracy of high-dose area localization. In this study, the dose gradient characteristics during the radiation propagation process are used to determine the direction of high-dose areas, combined with the least squares optimization method to predict the radiation source location, and the high-dose area location results are obtained through the radiation dose attenuation formula. Comparative experiments with currently popular interpolation methods show that the proposed method reduces the mean squared error (MSE) of dose calculations by 82.1% compared to the Kriging interpolation method and by 47.9% compared to the inverse distance weighting method, significantly improving localization accuracy. This method is suitable for radiation scenarios composed of gamma ray (γ-ray) detection point data and can provide technical support for radiation risk assessment and emergency response to contamination, improving the nuclear safety monitoring system.