Multi-objective optimization algorithm innovates multipass cell design, boosting LITES sensor sensitivity

Traditional multi-pass cells (MPCs) rely on grid search algorithms for design, which suffer from problems such as long design time and high computational cost. Additionally, these conventional methods mostly focus on a single objective—maximizing the optical path length—while neglecting the impact of ratio of optical path length to volume on system integration.

In this study, the PNSGA-II algorithm was introduced for the first time, enabling the simultaneous optimization of both the optical path length and ratio of optical path length to volume of multi-pass cells. A single run of the algorithm can generate multiple sets of high-performance multi-pass cell design parameters that meet the requirements of different application scenarios.

The multi-pass cell with a fifteen-ring spot pattern designed by PNSGA-II exhibited an optical path length of over 80 m, far exceeding the performance of multi-pass cells reported in existing literature.

Commercial QTFs have a resonant frequency of approximately 32.768 kHz, which results in a short energy accumulation time and limits the improvement of signal amplitude. The round-head QTF independently designed by the team has a reduced resonant frequency of 9.5 kHz; the lower resonant frequency extends the energy accumulation time and increases the generated signal amplitude. Experiments showed that at an acetylene concentration of 100 ppm, the peak value of the second harmonic signal detected by the round-head QTF was 4.17 times that of the commercial QTF, and the minimum detection limit was improved by 3.45 times.