Intelligent Networks for Chaotic Fractional-Order Nonlinear Financial Model

The purpose of this paper is to present a numerical approach based on the artificial neural networks (ANNs) for solving a novel fractional chaotic financial model that represents the effect of memory and chaos in the presented system. The method is constructed with the combination of the ANNs along with the Levenberg-Marquardt backpropagation (LMB), named the ANNs-LMB. This technique is tested for solving the novel problem for three cases of the fractional-order values and the obtained results are compared with the reference solution. Fifteen numbers neurons have been used to solve the fractional-order chaotic financial model. The selection of the data to solve the fractional-order chaotic financial model are selected as 75% for training, 10% for testing, and 15% for certification. The results indicate that the presented approximate solutions fit exactly with the reference solution and the method is effective and precise. The obtained results are testified to reduce the mean square error (MSE) for solving the fractional model and verified through the various measures including correlation, MSE, regression histogram of the errors, and state transition (ST).     

Artificial Neural Network Methods for the Solution of Second Order Boundary Value Problems

We present a method for solving partial differential equations using artificial neural networks and an adaptive collocation strategy. In this procedure, a coarse grid of training points is used at the initial training stages, while more points are added at later stages based on the value of the residual at a larger set of evaluation points. This method increases the robustness of the neural network approximation and can result in significant computational savings, particularly when the solution is non-smooth. Numerical results are presented for benchmark problems for scalar-valued PDEs, namely Poisson and Helmholtz equations, as well as for an inverse acoustics problem.     

混合神经网络及其在非线性系统控制中的应用

针对一类非线性动态系统模型的特点,提出一种非线笥和线性神经网络的并联神经网络－－混合神经网络,克服了以非线性函数逼近线性函数引起的复杂性和不精确性问题,实现了对模型的线性和非线性部分的分别逼近。仿真例子说明了混合神经网络用于这类非线性动态系统控制的可行性。     

应用神经网络模型对大学毕业生就业质量的预测

目的:尝试应用BP人工神经网络模型预测大学毕业生的就业质量。方法:以476名大学毕业生为研究对象,将毕业生的各种心理特质作为预测因子,利用Clementine数据挖掘软件构建大学毕业生就业质量的BP人工神经网络模型。结果:神经网络模型的拟合精度达到99.98%,对就业质量的平均绝对预测误差值为0.083。结论:该神经网络模型可用来预测和诊断大学毕业生找到工作后的就业质量,为就业指导工作提供技术与理论依据。     

应用概率神经网络诊断自行火炮发动机的故障

目的　研究概率神经网络模型 ,并应用于故障诊断 .方法　对基于概率统计思想和 Bayes分类规则的概率神经网络模型、网络结构、算法及其特点进行分析 ,利用其进行故障诊断 ,并提出一种优化估计平滑因子的方法 .结果　概率神经网络可很好地诊断自行火炮发动机进行中油路和气路的故障 .结论　概率神经网络在模式识别和故障诊断领域中可取得良好地应用效果