Research On Intelligent Control Decision Method For Flexible Circuit Board Roll-to-roll Processing

With the rapid development of flexible material processing technology,the preparation process of flexible thin film materials has also been continuously improved.New flexible film materials,such as flexible circuit boards and solar cell films,are increasingly used in various industries.Roll to Roll(R2R)flexible material is one of the main production equipments for thin film materials,and is widely used in flexible circuit boards(FPC),organic light emitting diodes(OLED),solar cell films,dry adhesives In the preparation and production of flexible film materials.The process of roll-to-roll equipment has the characteristics of high speed,high precision and continuous production.When the performance of its core components declines,the processing accuracy will be greatly reduced,causing the deformation of the flexible film products,which will reduce the yield and cause economic losses.It is very necessary to perform predictive maintenance on roll-to-roll equipment.The flexible circuit board R2 R continuous manufacturing system works fast.If a quality problem occurs in a certain station,the final product will be scr apped and the life of related equipment will be affected.At present,most of the roll-to-roll processing control is based on the operator's production experience,and these empirical knowledge are obtained through simple statistical production data.Due t o the complex site conditions and the uneven technical quality of the operators,the stability and accuracy of roll-to-roll processing control are lower than the production requirements.Data-driven research on intelligent control decisions for roll-to-roll processing needs to be on the agenda.This thesis is titled "Intelligent Decision-making Method for Roll-to-Roll Processing Control Based on Equipment Health",combined with artificial intelligence methods to explore the main influencing factors of roll shaft performance degradation,especially the relationship with roll shaft vibration characteristics.R2 R processing system and quality influencing factors of flexible circuit board analysis,focusing on the research of flexible circuit board processing roll performance degradation prediction modeling method and its system processing control RS decision method.The main problems solved by the thesis include:? The process flow of the flexible circuit board R2 R manufacturing system is analyzed,and the mechanical structure of the flexible circuit board processing system is established.The mechanical analysis of the roller during the processing of the flexible circuit board,and the finite element simulation of the driving roller module.Infer the possible performance degradation of the roller shaft and the possible impact on the flexible circuit board processing process.? Aiming at the problem of roller performance degradation,a neural network prediction model combining LSTM and SVM is proposed to predict the health status of roll-to-roll equipment.The historical data is used to train the model,the roller vibration characteristics are used as input,and the performance level of the sample is classified by adjusting the parameters of the neurons in the neural network model to output the device health status.The experiment proves the superiority of LSTM-SVM model in prediction accuracy and speed.? According to the requirements of flexible circuit board roll-to-roll processing equipment processing control decision-making,combined with equipment health status,a flexible circuit board processing control RS decision method is proposed.According to the influencing factors of the processing quality of the flexible circuit board processing system,through the historical health status data,the processing control RS decision model is established.