High-Speed Vision Based Control System For Nanoparticle Micro-Chemical Manufacturing Processes

Since microreators have numerous advantages, the technology of nanomaterials manufacturing that using microreactors is becoming one of the most promising scientific research fields in recent decades. In practical applications, pipeline jam often occurs when using traditional closed microreactors, which makes impinging-jet micromixers have obtained more research attention. However, impinging-jet micromixers have some certain problems, such as, instable impingement affects the mixture quality, which directly influences the quality of the reaction product.This thesis independently designs and develops a high-speed vision based control system for nanoparticle manufacturing, and use the device for the manufacturing of nano barium sulfate particles. The designed system can monitor and control the micro channel impinging process in real-time, trying to solve the harmful effect caused by unstable impingement of two liquid reactants, to improve the quality of the reaction products.The main research work and innovations in this thesis are as follows:Firstly, as the impinging stream cannot be detected by the traditional industrial inspection tools, this thesis uses image processing technology such as contour extraction, etc. to detect the impingement flow. And an improved contour local-search algorithm is proposed regarding to the system requirements in processing time and accuracy.Secondly, this thesis develops a novel flow actuator based on high speed on/off valve in micro pipes, due to the high difficulties and cost of the flow control in micro pipes. The actuator utilizes the ultra high speed on/off valve and a valve driver based on microcotroller minimum system. Moreover, the actuator can respond the control signal in millisecond level.Thirdly, we identify the static model of the system in the time series. In consideration of the system characteristics such as the flow characteristics of the actuator, mico pipes’characteristics, history experimental data, etc., a feedforward-feedback compound controller is proposed. The feedforward controller uses the look-up table method, and the feedback controller is designed based on fuzzy control theory.Finally, this thesis uses the designed system to produce nano barium sulfate sample, and the transmission electron microscopy experiment is introduced as characterization techniques. The experiment result firstly verifies the effects on nanoparticles caused by impingement shape, and then the practical TEM result indicates that using the novel control system can eliminate the agglomerate phenomenon and improve the nanopaticle’s quality obviously.