4.2. Routing procedure of optimization.
In order to maximize the
performance of the MRTOF-MS, it is crucial that the injection optics
provide a well-focused beam. To this end, numerical simulations are
employed to track the trajectories of ions from the preparation trap FLT
into the MRTOF analyzer. While in the MRTOF system, there are numerous
electrodes that can influence the ion motion, an automatic search
algorithm can greatly assist in finding the optimal voltage
configurations [17,
18]. It should be noted that the NMS
algorithm is not an efficient solver for high-dimensional optimization.
Arranging the variables into groups and iteratively searching within a
realistic interval is a feasible and effective approach. The procedure
is outlined graphically in Fig. 5. Eight variables, including three
accelerators (AC), three drift tubes (DT), and two lenses in the MRTOF
analyzer (Lens), are grouped according to the laws of electrostatic
lenses, e.g., A1 + A2 + A3, A3 + DT1 + DT2 + DT3, DT3 + Left-Lens +
Right-Lens, and other combinations like A2 + DT1 + Left-Lens +
Right-Lens, or A3 + DT1 + DT3 + Left-Lens + Right-Lens. The voltage
configurations are iteratively optimized until the feedback function
value is satisfactory.