A distributed model predictive control with machine learning for automated shot peening machine in remanufacturing processes

In practical peening operation, the values of inlet air pressure and media flow rate are manually preset to acquire desired intensity requirements. The operator often needs perform intensive experimental trials determine a set operational inputs for actual production. Obtaining these parameters is time-consuming labor-intensive. Thus, in this study, we propose an optimal distributed model predictive control multiple input/multiple output system address issues. newly developed system, actions voltage optimally obtained with anticipation future states plant models, while reference at nozzle flowrate determined using proxy model. dynamical models include model, which based on measurement data physics-based knowledge sparse identification nonlinear dynamics algorithm. from intensity, pressure, deep machine-learning performance demonstrated on-site controls physical machine different scenarios. results exhibit favorable stability, robustness, accuracy. consistent target setting value; difference smaller than industrial threshold ± 0.01mmA all random tests. another word, can be achieved without need performing parameters. It also suggests that deployed