Forecasting the Intrinsic Viscosity of Polyester Based on Improved Extreme Learning Machine
Zhang‐qi Yin, Kuangrong Hao, Lei Chen, Xin Cai, Xiuli Zhu · 2019
Polymerization process plays an important role in the entire production process of polyester fibers. Polyester, the product of polymerization process, directly determines the quality of the final fiber. Since intrinsic viscosity is a key performance index of polyester, its prediction is necessary and meaningful. In this paper, a forecasting model of intrinsic viscosity of polyester based on extreme learning machine is proposed. The model inputs are the technological parameters of the polymerization process, including the temperatures and pressures of the esterification reactor, the prepolymerization reactor, the final polycondensation reactor, and the additive amount of titanium dioxide. The model output is intrinsic viscosity of polyester. In order to enhance the prediction accuracy of the model, the combination function of ReLu and Softplus is used as the activation function of extreme learning machine (ELM). And particle swarm optimization (PSO) is used to optimize the weights of the input layer and thresholds of the hidden layer of ELM. The experimental results show that the proposed prediction model has better performance than ELM, Support Vector Machine (SVM) , back propagation neural network (BPNN) and ELM based on PSO (PSO-ELM).