论文部分内容阅读
由于具有能以任意精度逼近任意复杂非线性函数的优良性能,神经网络在灰色系统预测中得到了较大的应用。在已有的研究基础上,针对灰色神经网络进化时容易陷入局部最优,参数修正受阻的问题,建立基于遗传粒子群混合算法优化的新型灰色神经网络模型。首先将灰色神经网络进行数学建模,以便于优化算法的应用;其次,综合遗传算法与粒子群算法的优点,构造一种混合算法,运用混合算法对灰色神经网络进行优化;最后通过日本入华游客数量预测的算例研究,比较新型灰色神经网络与灰色神经网络、单一算法优化的灰色神经网络的预测精度。所得结果表明,混合算法优化的新灰色神经网络具有更好的预测性能,在社会经济领域有着广泛的应用前景。
Due to its excellent performance of approximating any complex nonlinear function with arbitrary precision, neural network has been widely used in gray system prediction. Based on the existing research, aiming at the problem that the gray neural network is apt to fall into the local optimum and the parameter correction is blocked, a new gray neural network model based on the genetic particle swarm optimization algorithm is established. Firstly, the gray neural network is mathematically modeled to facilitate the application of the optimization algorithm. Secondly, by combining the advantages of the genetic algorithm and the PSO, a hybrid algorithm is constructed and the gray neural network is optimized by the hybrid algorithm. Finally, A case study on the prediction of the number of tourists is carried out to compare the prediction accuracy of the new gray neural network with the gray neural network and the single algorithm to optimize the gray neural network. The results show that the new gray neural network optimized by the hybrid algorithm has better prediction performance and has a wide range of applications in the social economy.