Process optimization of carbon fiber precursor spinning coagulation process based on surrogate model

High-quality carbon fiber precursor is the key to the manufacture of high-performance carbon fiber, and the coagulation process is one of the key steps in the preparation of the precursor. This coagulation process is an industrial process with complex mechanism and strict control accuracy requirements. Therefore, iterative optimization based on its strict mechanism model will be much time-consuming. In order to reduce the computational cost, a process optimization method for the coagulation process of carbon liber precursor spinning based on surrogate model was proposed. Using meta-modeling technology, this paper constructs the surrogate model which describes the relationship between the key operating parameters and the main output variables in the coagulation process. In the process of constructing the surrogate model, an evolutionary optimal sample point insertion criterion is adopted to improve the accuracy of the model. And the quantum-behaved particle swarm optimization (QPSO) algorithm is applied for searching operating parameters in the design space. The simulation results show that the proposed method not only provides a group of satisfactory operating parameters, but also greatly reduces the calculation cost.