基于遗传神经网络的纱线质量预测

Combining the Genetic Algorithm with Artificial Neural Networks for Yarn Quality Forecasting

针对基于单一BP神经网络的纱线质量预测模型的不足,提出了一种基于遗传算法优化的神经网络的纱线质量预测模型,采用遗传算法完成对神经网络权值和阈值空间的寻优搜索,以提高神经网络的收敛速度和获得全局最优解的能力.通过试验表明,基于遗传算法优化的神经网络可以提高纱线质量预测模型的精度和稳定性,其性能优于基于单一BP神经网络模型的纱线质量预测.

In order to solve the shortcomings of yarn quality forecasting model based on a single BP neural network, a yarn quality forecasting model based on genetic algorithm and neural network is proposed. The model adopts the genetic algorithm to complete the optimal search of the network weights and threshold space, which enhances the neural network convergence speed and the ability to obtain a global optimal solution. The experiments indicate that the neural network based on genetic algorithms can improve the accuracy and stability of the yarn quality forecasting model, and is superior to that based on a single BP neural network.