Research On Robust Optimization Of Sizing Process Based On Soft Measurement

Sizing process is an important part in textile industry and the sizing qualities directly affect the next step. Sizing process should be accurately controlled as much as possible, which ensure the quality of sizing in the acceptable range. In the sizing process, sizing percentage is one of the most important indicator of sizing quality. So it often adopts controlling the sizing rate to ensure their quality. In the sizing process, there are many factors affect sizing rate. Then the system and the external load have un-certainties. The process variable or restricted parameter range will continue to change in the sizing process. It inevitably exist all kinds of uncertainty in the sizing process. So it is necessary to research sizing process in order to ensure control sizing percentage and improve the sizing quality.The robust optimization method based on soft measurement of sizing process is presented in this paper. First, it identified 10 factors that affect the sizing process, through the analysis and understand of the sizing process. Due to the complexity of the process, the prediction model of sizing is established using soft measuring data-driven method. Second, the optimization theory and the characteristics of the sizing process should be understood. Then the objective optimization model is established on the basis of the soft measurement prediction model. Further, because of the sizing process is complex and its operating parameters often contain disturbances, the robust optimization model of sizing processes established according to the concept of robust optimization. Finally, the hybrid particle swarm optimization and genetic algorithm (HPSO-GA) is proposed combining the advantages of particle swarm optimization and genetic algorithm. The common test functions are used to checkout the HPSO-GA has high efficiency and accuracy. The algorithm used to solve the robust optimization model of sizing process and find out the optimal parameters. Then it can stabilize the sizing percentage and improve the quality of sizing. And the downstream operations of sizing can smoothly run in textile industry. Further, the resource waste and environmental pollution will reduce, which reduce the production cost of enterprises.