Data driven process modeling and simulation: an applied case study

The yarn production is a very complex industrial process, and the relation between the spinning variables and the yarn properties has not been established conclusively so far. However, the existing process cases which were recorded to ensure the ability to trace production steps can also be used to control the process itself. This paper presents a novel process simulation model with data driven approaches such as case based reasoning and support vector machine hybrid algorithms for optimization of complex spinning parameters. The process simulaiton model is able to predict the yarn properties through continually self-learning, which can help the yarn producer to make the best process decision The applied cases are demonstrated that the intelligent system for the solving the hard problem of complex process control is promising.