Enhancing Wavelength Modulation Spectroscopy for Gas Detection via Chaotic Map-Based Particle Swarm Optimization

This study proposes a Chaotic Map-based Particle Swarm Optimization (CM-PSO) algorithm to enhance the performance of Wavelength Modulation Spectroscopy in gas concentration retrieval. By introducing the chaotic map-driven inertia weights and the last elimination mechanism, CM-PSO significantly enhances the accuracy and stability of gas concentration retrieval. This method leverages the dynamic nonlinear characteristics of chaotic map to optimize the particle search process, effectively overcoming the tendency of traditional algorithms to become trapped in local optima. Simultaneously, by periodically eliminating particles with poor fitness and introducing new random particles, it substantially enhances population diversity and increases the probability of finding the optimal solution. Experiments demonstrate that compared to traditional PSO and the Levenberg-Marquardt algorithm, CM-PSO reduces the average relative error by 1.05% and 2.3%, respectively, and shortens the average single retrieval time by 29.2% and 73.6%, respectively. This technique provides a high-precision, high-robustness solution for calibration-free gas detection.