抄録
The genetic algorithm (GA) has been already effectively applied by the present authors to the particle pairing problem of the PTV (Particle Tracking Velocimetry) system software. However, from the viewpoint of practicability, the formerly reported algorithm still leaves some room for improvement, in particular with respect to the allowed number of particles to be tracked, the treatment of the unpaired particles between the frames and, finally, the computing time. This time, a number of new ideas have been introduced in the genetic encoding of the particle pairing problem as well as in the relating genetic operations, both with a view to reducing the computing time up to convergence and dealing with the unpaired particles more appropriately. At the same time, the formerly proposed ideas have been used in a more refined form in order to improve the performance of particle pairing.
