| CPC G06F 30/23 (2020.01) [G06N 3/126 (2013.01); G06F 2111/06 (2020.01); G06F 2113/22 (2020.01); G06F 2113/26 (2020.01); G06F 2119/08 (2020.01)] | 7 Claims |
|
1. A method for molding a part made of a fiber-reinforced polymer composite, comprising:
heating the fiber-reinforced polymer composite to a predetermined temperature at a predetermined heating rate;
keeping the fiber-reinforced polymer composite at the predetermined temperature for a predetermined duration; and
cooling the fiber-reinforced polymer composite at a predetermined cooling rate to produce the part with reduced manufacturing cost and improved molding part quality;
wherein the predetermined temperature, the predetermined heating rate, the predetermined duration and the predetermined cooling rate are determined through the following steps:
(S1) establishing a Python library required for calculation; wherein the Python library comprises an Abaqus library and a genetic algorithm library, and the Abaqus library comprises an abaquesConstants library, a caeModules library, and an odbAccess library;
(S2) setting simultaneous minimization of the predetermined duration, temperature gradient, and residual stress as an optimization goal;
(S3) setting a variable interval for the predetermined temperature, the predetermined heating rate, and the predetermined cooling rate among molding process parameters followed by assigning values to the predetermined temperature, the predetermined heating rate, and the predetermined cooling rate using a random function;
(S4) establishing a macroscopic thermochemical model of the part according to dimensions and molding environment of the part by using a finite element method, and assigning a material property to the macroscopic thermochemical model; analyzing evolution of temperature and curing degree of the fiber-reinforced polymer composite during the molding process of the part by using a heat transfer analysis module; rewriting the macroscopic thermochemical model into a thermochemical model function of the molding process parameters, wherein a temperature boundary is applied by transmitting curing process parameters to a first subroutine, and the curing process parameters are randomly generated; importing the first subroutine followed by job submission to output calculation results; extracting a minimum curing degree in the macroscopic thermochemical model from the calculation results; calculating a maximum temperature gradient during the molding process; and exporting temperature-time data at a center of the part;
(S5) establishing a two-dimensional representative volume element (RVE) model by using a Python script from the Python library based on a proportion of individual components and size of a fiber in the fiber-reinforced polymer composite; analyzing an elastic modulus change of a resin of the fiber-reinforced polymer composite during the molding process by using a cure hardening instantaneous linear elastic (CHILE) model; analyzing stress and strain transformations during the molding process by using a static general module; rewriting a microscopic thermomechanics model in the two-dimensional RVE model into a thermomechanics model function of the curing process parameters and simultaneously incorporating the temperature-time data exported in step (S4) as a predefined field into the two-dimensional RVE model; importing a second subroutine followed by job submission, wherein the second subroutine is an Abaqus user subroutine; and extracting the residual stress from calculation results of Abaqus;
(S6) calculating a total time of the molding process; wherein the total time comprises a heating time, the predetermined duration, and a cooling time;
(S7) writing a dynamic penalty function, and evaluating a violation-constrained molding process parameter among the molding process parameters using the dynamic penalty function;
(S8) writing a main function of a genetic algorithm, defining a population size, the number of generations, a crossover distribution index, and a mutation distribution index; and calling the thermochemical model function, the thermomechanics model function, and the dynamic penalty function to evaluate individual fitness; and
(S9) plotting a multi-objective optimized Pareto optimal solution set according to the individual fitness, and selecting a set of molding process parameters satisfying actual needs from multiple sets of molding process parameters provided by the multi-objective optimized Pareto optimal solution set, wherein the set of molding process parameters comprises the predetermined temperature, the predetermined heating rate, the predetermined duration and the predetermined cooling rate.
|