The application of nonlinear least-squares estimation algorithms in atmospheric density model calibration

Abstract

Purpose The purpose of this paper is to focus on the performance of three typical nonlinear least-squares estimation algorithms in atmospheric density model calibration. Design/methodology/approach The error of Jacchia-Roberts atmospheric density model is expressed as an objective function about temperature parameters. The estimation of parameter corrections is a typical nonlinear least-squares problem. Three algorithms for nonlinear least-squares problems, Gauss–Newton (G-N), damped Gauss–Newton (damped G-N) and Levenberg–Marquardt (L-M) algorithms, are adopted to estimate temperature parameter corrections of Jacchia-Roberts for model calibration. Findings The results show that G-N algorithm is not convergent at some sampling points. The main reason is the nonlinear relationship between Jacchia-Roberts and its temperature parameters. Damped G-N and L-M algorithms are both convergent at all sampling points. G-N, damped G-N and L-M algorithms reduce the root mean square error of Jacchia-Roberts from 20.4% to 9.3%, 9.4% and 9.4%, respectively. The average iterations of G-N, damped G-N and L-M algorithms are 3.0, 2.8 and 2.9, respectively. Practical implications This study is expected to provide a guidance for the selection of nonlinear least-squares estimation methods in atmospheric density model calibration. Originality/value The study analyses the performance of three typical nonlinear least-squares estimation methods in the calibration of atmospheric density model. The non-convergent phenomenon of G-N algorithm is discovered and explained. Damped G-N and L-M algorithms are more suitable for the nonlinear least-squares problems in model calibration than G-N algorithm and the first two algorithms have slightly fewer iterations.

Radiation levels outside a patient undergoing 177Lu-PSMA radioligand therapy

“…where A is the activity of 177 Lu in GBq, R is the distance from the source point in cm, θ is the plane angle in radians (−π/2 to π/2), a, b and c are the constant terms to be fitted. In this study, the MATLAB (MathWorks, R2021a, USA) was used to explore the constant terms of a, b and c. The computation processes involved the utilization of the Levenberg-Marquardt optimization algorithm, which combines the merits of gradient descent and Gauss-Newton algorithms, thereby offering enhanced convergence performance during the search for the least square solution (Gavin 2020, Zhang et al 2020. To compare the function values with the simulated results, the index of relative root mean square deviation (RRMSD) was used to test the accuracy of fitting method.…”

A Consensus-Based Diffusion Levenberg-Marquardt Method for Collaborative Localization With Extension to Distributed Optimization

“…We also report our numerical results to illustrate the effectiveness of Algorithm 1 by applying to the decentralized least squares problem. Least squares model plays an important role in many real applications including image recognition [33], spectrum sensing [34] and atmospheric density model calibration [35]. Typically, least squares can be formulated as a global minimization problem over a multi-agents system such (33) and has the following local cost function for each…”