基于改进RBF神经网络的GNSS高程拟合

针对传统的RBF神经网络模型在GNSS高程拟合中拟合精度较低、稳定性较差、相关因子需提前人为设置等问题,通过将改进的自适应权重粒子群优化算法与MATLAB RBF神经网络函数newrb相结合,实现RBF神经网络函数模型中隐含节点数和SPREAD值的自动优化选取,提高算法在GNSS高程拟合中的精度和稳定性。通过实例分析,该方法拟合精度高,可达到mm级精度,相对于传统的二次多项式模型精度提高17%,稳定性良好。     

���������Ż�RBF�����ڶ�̬����Ԥ���е�Ӧ��

?????????????????????????????????????????????????RBF?????????????????????CIOA????RBF??????????????????????????????CIOA?????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????RBF???????????????????????У???Ч????????????????????     

Application of Improved Multi-Objective Ant Colony Optimization Algorithm in Ship Weather Routing

This paper presents a novel intelligent and effective method based on an improved ant colony optimization(ACO)algorithm to solve the multi-objective ship weather routing optimization problem,considering the navigation safety,fuel consumption,and sailing time.Here the improvement of the ACO algorithm is mainly reflected in two aspects.First,to make the classical ACO algorithm more suitable for long-distance ship weather routing and plan a smoother route,the basic parameters of the algorithm are improved,and new control factors are introduced.Second,to improve the situation of too few Pareto non-dominated solutions generated by the algorithm for solving multi-objective problems,the related operations of crossover,recombination,and mutation in the genetic algorithm are introduced in the improved ACO algorithm.The final simulation results prove the effectiveness of the improved algorithm in solving multi-objective weather routing optimization problems.In addition,the black-box model method was used to study the ship fuel consumption during a voyage;the model was constructed based on an artificial neural network.The parameters of the neural network model were refined repeatedly through the historical navigation data of the test ship,and then the trained black-box model was used to predict the future fuel consumption of the test ship.Compared with other fuel consumption calculation methods,the black-box model method showed higher accuracy and applicability.     

۶���������ƴ��ˮ׼�澫���о�

???????????????????BP?????緽????????????????????棬?????????????????????????????Ч???????????????????????????????????????????????????????????????????????????????????仯???????????????????????????y?????ж???????????п?????θ?????????￡?PSO-BP??????????????÷???????Ч????     

基于灰色关联和相关向量机的隧道地表沉降预测

基于灰色关联算法确定与地表沉降有直接重要关联的主要影响因子,构建高斯核函数和多项式核函数的加权核函数,利用遗传算法优化模型参数,建立相关向量机地表沉降预测模型。实验结果表明,灰色关联算法能定量地反映系统影响因子与地表沉降变化的关联程度,有效处理不是完全明确的灰色系统信息;加权核函数的合理组合可较好地通过低维空间线性不可分映射变换到高维特征空间线性可分;遗传算法具有计算过程简单和自适应迭代寻优特点;相关向量机模型可极大地减少核函数的计算量,计算过程和结果均具有概率解释。该模型预测结果的多项精度指标值均优于BP神经网络和GR-SVM方法。     

遗传算法优化支持向量机在大坝变形预测中的应用

建立大坝变形预测的支持向量机模型,并用遗传算法对支持向量机模型的核函数参数、惩罚参数和损失参数进行优化。将同一优化方法不同支持向量机核函数、不同优化方法同种支持向量机核函数进行横向对比,将BP神经网络、自回归AR(p)模型、多元回归分析法和周期函数拟合法进行纵向对比。结果表明,该GA-SVM(RBF)模型不仅能较好地预测大坝的变形趋势,而且能大幅提高预测精度。     

����ѧϰ���������Ӳ�Ԥ���е�Ӧ��

?????????????????????????????Sigmoidal??Sine??Hardlim??????????????????????????????????????????????????????б??????????????????????????????????????????????????????????????????????磬?????Sigmoidal????????????????????????     

����RBF���˲�������Բο�վ�߳�ʱ�����з���

?????????IGS???????вο??????????仯???????仯????????????????????????????仯??????RBF?????????缰С??????????????????MATLAB7.0????????IGS???????????????GRNN?????????С??????????????????????????????????????????????????????д??????仯????????????????????????????仯?????????????????????     

基于实数编码遗传算法的大气颗粒物源解析

实数编码遗传算法(RCGA)是在二进制编码遗传算法的基础上提出来的,它具有精度高、执行效率快等优点,非常适用于模型参数寻优.在采用化学质量平衡法(CMB)建立了源解析方程组的基础上,应用RCGA对方程组进行参数寻优,得到各污染源对大气颗粒物的优化贡献率.RCGA法与其它方法相比,解析结果具有更高准确性.     

基于RBF的指标规范化的水安全评价模型

为了建立科学合理、计算简便和普适通用的水安全评价模型,在适当设定指标参照值cj0和指标值的规范变换式基础上,提出了基于径向基函数网络的指标规范化的水安全评价模型。采用具有全局优化的猴王遗传算法对模型中的参数进行优化,得出优化后对任意m(1≤m≤23)项水安全指标共同适用的水安全评价模型。应用模型对山东省水安全状况进行了评价分析,其评价结果与其它方法的评价结果基本一致,从而表明:指标规范值的径向基函数网络模型为水安全评价提供了一个简单实用、结果可靠的新方法。