Abstract
Magnetic target state estimation is a widely applied technology, but it also faces many challenges in practical applications. One of the most critical challenges is the issue of estimation accuracy. The Grey Wolf Optimizer (GWO) is one of the more successful swarm intelligence algorithms in recent years, but its shortcomings have also been exposed when facing increasingly complex problems. Therefore, a Multi-Strategy Improved Grey Wolf Optimizer (MSIGWO) algorithm has been proposed to enhance the accuracy of magnetic target state estimation. In the initialization phase, Tent chaos mapping is introduced to enhance population diversity, prevent falling into local optima, and improve convergence speed. Multi-population fusion evolution strategies enhance population diversity, convergence accuracy, and global search ability. Nonlinear convergence factors better balance exploration and exploitation behaviors. Dynamic weight strategies increase the diversity of search samples and reduce the likelihood of falling into local optima. Adaptive dimensional learning better balances local and global searches, enhancing population diversity. Adaptive Levy flight enhances the ability to jump out of local optima and ensures convergence speed. In the CEC2018 benchmark function set of 29 benchmark function problems and magnetic target state estimation problems, the proposed MSIGWO was tested, and statistical indicators and Friedman test results show that compared with GWO and its advanced variants, the MSIGWO algorithm has superior performance. The application of this algorithm in magnetic target state estimation problems has proven its effectiveness and applicability.