Fluid-Improved Particle Swarm Optimization for Parameter Optimization of XRD-Based Os Draconis Identification Model.

During the X-ray Diffraction (XRD) identification of the traditional Chinese medicine Os Draconis, the identification model often suffers from limited classification accuracy due to the difficulty in determining optimal parameters. To address this issue, this paper proposes a Hydrodynamic Improved Particle Swarm Optimization (HIIPSO) algorithm for the deep optimization of model parameters. In practical identification scenarios, the high complexity of XRD data poses severe challenges to the convergence speed and global search capability of optimization algorithms. To enhance model performance, this study introduces the interaction mechanism from fluid dynamics into the particle swarm optimization process. Specifically, HIIPSO incorporates a Voronoi neighbor topology to enhance population diversity and spatial distribution rationality. Concurrently, a hydrodynamic interaction mechanism is constructed to simulate the cooperative behavior of particles in a fluid environment, thereby effectively preventing the algorithm from falling into local optima. A theoretical analysis of the computational complexity of the HIIPSO algorithm in the parameter search task for XRD identification models was conducted, confirming that it falls within an ideal range for engineering applications. Statistical analysis of the experimental results demonstrates that, in the parameter optimization task for the Os Draconis identification model, the HIIPSO algorithm significantly outperforms traditional and other baseline algorithms across key metrics, including the optimal value, mean, standard deviation, and median of the objective function. The experimental data indicates that the HIIPSO algorithm can substantially improve the robustness and identification accuracy of the XRD-based Os Draconis identification model, making it an optimal solution for parameter optimization problems in the digital identification of complex mineral-based traditional Chinese medicines.