Performances evaluation of the optical techniques developed and used to map the velocities vectors of radioactive dust

Radioactive dust mobilization is a risk that can occur in many nuclear plants and, in order to reduce the risk related to this event, it is necessary map the velocity vectors of dust during its mobilization. The authors have designed and used a chain of measurements for air pressure and velocity, temperature, and dust velocity used on the experimental facility STARDUST-Upgrade that can replicate the thermos-fluidodynamic conditions of the loss of vacuum accidents with a pressurization rate in a range of 10–1000 Pa/s and a temperature in a range of 20–140 ℃. In this work, the authors present the optical experimental setups and software used to track dust velocities. These techniques are based on the particle tracking velocimetry and flow motion algorithms. Two different experimental setups are used to take into account the different optical properties of dust, each image obtained during the experiments has been analysed with customized software. Three different of algorithms are analysed and criticaly compared in this work: Lucas-Kanade, feature matching and Horn-Schunck. The authors will evaluate the performances of these optical techniques developed and used to map the velocities vectors of radioactive dust.