基于GRNN神经网络的中厚板轧机宽展预测模型

为了实现中厚板轧机在轧制过程中的宽展变化自动预测和控制,分析了轧制过程中宽展变化的影响因素,在神经网络技术和现场实测数据的基础上,利用Matlab人工神经网络工具葙,应用GRNN广义回归神经网络建立宽展变化预测模型,来提高轧制宽展变化预测的精度.结果表明,该方法建立的模型可以实现对宽展变化的预测,预测精度有较大提高.     

广义回归神经网络在非线性系统建模中的应用

广义回归神经网络具有设计简单、收敛快等优势,因此在复杂非线性系统建模中得到了广泛应用;在简要介绍了广义回归神经网络的结构和算法的基础上,基于广义回归神经网络和均匀设计理论,提出了一种新的非线性系统稳健建模方法,并给出了仿真算例;仿真结果表明,用文中提出的方法建立非线性系统预测模型,具有预测结果稳定、模型稳健性好等优点.     

A GRNN based frame work to test the influence of nano zinc additive biodiesel blends on CI engine performance and emissions

Environmental friendly fuels have increasing demand in the field of automotive industry. Hybridizing with biodiesel and addition of nanoparticles are two evident techniques to improve the performance and to control the hazardous emissions. Combination of appropriate weight fractions of nanoparticles, biodiesel, and diesel further optimize the engine outcomes. The present work aims at investigating the effect of zinc oxide (ZnO) nanoparticles suspensions in diesel and Mahua biodiesel blended fuel on single cylinder diesel engine performance characteristics. Experimental tests are performed with neat diesel fuel, biodiesel blends and ZnO added biodiesel blends. The results indicate that ZnO particulate addition yields favorable performance and emission control of the engine. A generalized regression neural network (GRNN) is implemented for predicting the performance and emissions of the engine at various operating conditions based on the experimental results. The neural network predictions are corroborated with the experimental results and are found in good agreement. A classical differential evolution algorithm (DEA) is further used on the network model to find out optimal combination of nanoparticles, biodiesel and diesel and proven through experimental validation.     

人工神经网络在近红外光谱建模中的应用及研究现状

对国内外近十年来人工神经网络在近红外光谱建模中的应用和研究进行了详细的综述,包括误差反向传播网络、径向基网络、支持向量机、自组织特征映射网、广义回归神经网络、概率神经网络、小波神经网络、模糊神经网络以及集成神经网络等的应用和研究。概括了这些网络的基本工作原理及优缺点。最后根据神经网络的发展方向和工农业的发展需求,提出了今后人工神经网络在近红外建模方面的发展方向。     

基于典型气象周的GRNN光伏发电量预测模型

由于光伏发电量具有波动性,且现有的光伏发电量预测技术存在气象因素考虑不全面、特征提取不充分等问题,为提高光伏发电量预测精度,文中提出一种改进的典型气象年方法(TMY Method)生成典型气象年数据,并结合广义回归神经网络(GRNN)进行光伏发电量预测。首先,选择6种历史气象指标,利用Finkelstein-Schafer统计方法选择典型气象周,并生成典型气象年数据;然后,使用因子分析法对会影响光伏发电量的气象指标进行筛选,对筛选出的气象指标和日光伏发电量进行标准化处理后,将其作为GRNN模型的初始输入量,得到预测日的光伏发电量;最后,利用江苏省南京市的历史气象数据及日发电量数据对所设计的模型进行训练和预测。结果表明,与标准TMY Method-GRNN预测方法相比,文中所提预测方法有较好的预测性能。     

基于GRNN的输电线路覆冰厚度预测方法研究

为了保障输电网的安全运行,输电线路覆冰厚度预测极为重要。本文还将人工神经网络原理引入输电线路覆冰厚度预测中,并针对BP网络收敛速度慢、已陷入局部极小的缺陷,提出了基于广义回归神经网络(GRNN)的预测模型。实例研究证明GRNN模型相比较BP模型,能更有效地预测输电线路覆冰厚度。     

GRNN神经网络在电力系统负荷预报中的应用

为了预报电力系统负荷,采用GRNN（广义回归网络）的方法,通过GRNN神经网络和BP神经网络建立电力系统负荷预报网络模型,用MATLAB7.0仿真,达到了预测的目的。利用GRNN神经网络预测结果准确率高,避免了BP网络预测同样的数据库,算法冗长,网络预测结果不稳定的缺点,GRNN网络具有更好的预报精度。     

基于RS-GWO-GRNN的充填管道失效风险研究

为克服充填管道失效风险评判指标间的复杂性,传统方法预测精度低及适用性差等缺陷,提出基于粗糙集(RS)和灰狼优化(GWO)算法融合广义回归神经网络(GRNN)的充填管道失效风险评价模型。选取10项风险评价指标,通过属性约简提取影响充填管道失效的主要风险因素,运用GWO优化GRNN的参数,构建预测模型,以国内某具体矿山充填系统为例进行实证研究,结果表明:与其它预测模型相比,RS-GWO-GRNN模型的预测精度更高,泛化能力更强,为充填管道失效风险研究提供了新思路,具有较好的借鉴意义。     

Application of a general regression neural network for health index calculation of power transformers

A power transformer is one of the most important components in a transmission network. To assess the overall condition of this valuable asset, health index calculations are recently gaining more attention from the utility companies that operate networks. Only limited research has been conducted on health index calculations of transformers. Most of the past approaches are based on the linear combination of weighted scores of measurements following the industry standards such as IEEE, IEC and CIGRE. A few previous methods based on artificial intelligence and statistical approaches such as fuzzy logic, multivariate analysis and binary logistic regression have been published in recent years. In this paper, a General Regression Neural Network (GRNN) which has a nice nonlinear property and can work with measurements without quantization has been evaluated. The GRNN allows multi-dimensional measurements to be combined through an optimal weighting and scoring system to compute a quantitative health index of power transformers. The weighting of each test was assigned based on a smoothly interpolated continuous function. The efficacy of the model has been validated against expert classifications and data sets published in the literature. The comparative results demonstrate that, the proposed method is reliable and very effective for condition assessment of transformers through an automated health index calculation.     

基于广义回归神经网络与遗传算法的煤灰熔点优化

考虑固态和液态排渣锅炉对煤灰熔点的不同要求，采用广义回归神经网络建立了煤灰软化温
度模型。神经网络的输入变量为7个，即煤灰中SiO2、Al2O3、Fe2O3、CaO、MgO、TiO2、Na2O & K2O的质
量分数。以煤灰软化温度作为目标函数，采用遗传算法寻优计算获得当煤灰软化温度最高和最低时煤灰中
氧化物的组成。广义回归神经网络仅需30个训练样本，最大和平均相对误差分别为21.8%和1.55%。优化结
果表明，掺烧高钙煤或者向燃煤中添加石灰石等富含Ca的原料可以降低煤灰熔点；而增加Al2O3的质量分
数可以提高煤灰熔点。