Accurately obtaining aircraft aerodynamic parameters is essential for improving flight performance, optimizing design and control strategies, and ensuring flight safety. In this study, the improved Northern Goshawk Optimization (SPNGO) algorithm is used to optimize the kernel parameters and regularization coefficients of the Kernel Extreme Learning Machine (KELM). To address the defects of the original NGO algorithm, such as insufficient global optimization ability and being prone to falling into local optimums, two improvement strategies are proposed. The enhanced SPNGO algorithm is verified by 14 benchmark test functions, and the proposed SPNGO-KELM model is evaluated using open-source F-16 nonlinear simulation data for longitudinal aerodynamic parameter identification. The results demonstrate its effectiveness under the considered simulation conditions, while further validation with real flight-test data is required before application to actual flight environments. Comparative analysis with KELM, NGO-KELM, SSA-KELM, and WOA-KELM models shows that a single KELM is difficult to achieve high-precision aerodynamic parameter identification, and other comparison models have obvious fitting deviations in non-steady-state and strong nonlinear regions. Notably, the SPNGO-KELM model achieves the best identification performance, with a determination coefficient (R²) of 0.96537 and a mean absolute percentage error (MAPE) as low as 3.1574%. Its comprehensive identification accuracy is 1.81% to 37.98% higher than that of the comparison models, and it can effectively suppress error oscillations in nonlinear regions. Experimental results show that the proposed algorithm has excellent identification accuracy, generalization ability, and anti-interference performance.

Standard Detective Behavior Algorithm (DBA) may exhibit limited exploitation in later iterations, insufficient use of individual historical information, and unstable local search on complex continuous optimization problems. To address these limitations, this paper proposes a hybrid DBA with particle swarm optimization, denoted as DBA-PSO. The method retains the search-space-centered exploration, Lévy-flight-based local search, direct-attack refinement, and opposition-based learning mechanisms of DBA, while introducing the velocity update and personal-best/global-best guidance mechanisms of particle swarm optimization (PSO). At each iteration, DBA-PSO generates one DBA candidate and one PSO candidate and selects the next position through a fixed probabilistic fusion rule. Experiments on 23 classical benchmark functions and the CEC2017 test suite, together with supplementary CEC2022 and constrained engineering-design tests, show that DBA-PSO improves the mean fitness or stability on several functions compared with DBA or PSO and remains competitive on many test cases. Overall, the results support the usefulness of PSO-based velocity memory and historical-best guidance for improving the later-stage search behavior of DBA, while the mixed CEC2022 ranks, uneven engineering feasibility rates, and fixed fusion probability show that the method does not provide uniform dominance across all problem classes.