基于OWO-HWO算法的MLP网络在织物染色配色中的应用

讨论了具有非线性、不确定特性的织物染色配色过程建模与仿真问题。针对传统的织物染色配色方法效果差、精确度不高和难以达到期望结果的问题,结合MLP神经网络的特点,提出了基于OWO-HWO算法训练的MLP神经网络,同时分别优化网络输入层到隐层和隐层到输出层的权值,并利用基于OWO-HWO算法的MLP神经网络建立织物染色配色模型。针对此种模型,利用NuMap神经网络软件进行仿真实验。仿真结果表明,该配色模型收敛速度快,精确度高,在解决织物染色配色问题上取得了令人满意的配色效果。

Application of MLP Neural Network Based on OWO-HWO Algorithm in Textile Dyeing Color Matching

The modeling and simulating problems of textile color dyeing process with non-linearity and uncertainty are discussed. Considering the problem of poor effect and precision of the textile color dyeing method, a MLP neural network based on OWO-HWO algorithm is presented combining with characteristic of MLP neural network. The output weights and hidden unit weights of neural network are optimized alternately by the proposed algorithm, which is used to establish the model of textile color dyeing. The simulating process of the model is implemented in NuMap software platform for neural network. Simulation results show that the model with fast convergence speed and good approximation accuracy has showed a satisfying result for color matching in textile dyeing.