Data-based Method for Diagnosing Multiple Blockage Locations in a Microreactor with Parallelized MicroChannels

摘要

Microreactors with parallelized microchannels are widely used in chemical industries because a single microchannel processes only a small amount of raw material. Any blockage in parallelized microchannels causes poor uniformity of the residence time distribution among microchannels and thus leads to degraded product quality. To address this issue, we developed a data-based identification method that can cope with blockages in multiple microchannels. The method identifies multiple blockage locations by comparing measured pressure-distribution data with pressure-distribution data calculated by computational fluid dynamic (CFD) simulation when blockage occurs. To avoid a combinatorial explosion of CFD simulations for database construction, the proposed method identifies blocked locations using only pressure-distribution data obtained when a single channel is blocked. The results of CFD simulations showed that the method can accurately identify three blocked locations using fewer pressure sensors than there are microchannels.