烷基硫醇在不同固定相上的QSRR研究

计算出体系中烷基硫醇化合物的各种结构参数,以优选出的分子连接性指数和量化参数为结构描述符,首次采用反向传播算法(BP)人工神经网络、径向基函数网络(RBF)2种非线性方法建立了参数少且精度高的定量结构-色谱保留相关(QSRR)模型,预测了烷基硫醇在4种极性固定上的气相色谱保留指数(RJ)。结果表明:在4种固定相上建立的BP模型均优于RBF模型且非线性方法(BP、RBF)优于文献中多元线性回归(MLR)方法,所建定量结构保留关系(QSRR)模型具有良好的稳定性和预测能力。     

BP网络与RBF网络在感官评估中应用比较

介绍BP网络与RBF网络的方法原理以及二者在感官评估中的应用.利用企业提供的单料烟感官数据设计建立BP网络和RBF网络模型,并利用模型预测单料烟感官质量指标,然后通过行业专家提供的符合度公式对建立的模型进行评估,评估结果以百分比的形式展示.实验结果表明:在感官评估应用中,RBF网络模型预测性能优于BP网络模型.     

有机化合物吸着系数(Koc)估算模型比较

基于592种化合物的Koc实测数据,分别采用片段常数法和分子连接性指数法建立了其估算模型.根据模型残差分析了两种模型的预测误差,并运用2种方式的jackknife检验比较了它们的稳健性.片段常数模型涉及较多变量,其应用受到有限的已知片段常数以及烦琐的片段划分过程的限制,相比之下,分子连接性指数模型变量数较少,其应用时只需了解化合物的结构式.片段常数模型的估算精度略高于分子连接性指数模型,两者的差别主要体现在logKoc<1和logKoc>4的化合物上.分子连接性指数模型的稳健性比片段常数模型高.在依次删除一类特定类别化合物时,其估算误差的变异幅度均小于片段常数模型.     

基于ESN的锂电池SOC评估方法与仿真研究

以新能源车载锂电池为研究对象,建立基于回声状态网络（ESN）预测锂电池的荷电状态（SOC）评估模型。采用交叉验证方法优选回声状态网络参数,以此解决网络模型的参数选择困难。通过带遗忘因子的递归最小二乘法训练建立的回声状态网络模型,实时更新输出权值矩阵以此提高网络的适应性和精度。通过模型仿真分析验证了预测算法的可行性,进一步对比分析了所建立的ESN预测模型与BP神经网络算法、径向基（RBF）网络算法在UDDS、US06和NYCC工况条件下的锂电池SOC评估预测效果,结果表明所建立的回声状态网络模型方法用于锂电池SOC评估预测的性能和效果优于BP算法和RBF算法,具有较好的应用前景,可以为锂电池SOC长期长效预测评估提供参考。     

人工神经网络预测取代芳烃类化合物对黑呆头鱼的毒性

运用Gsussian98程序的密度泛函(DFT-B3LYP)方法,在6-311G**水平上几何全优化下对92个取代芳烃类化合物的量子化学参数进行计算,将计算得到的12个量子化学参数和1个结构参数作为描述符引入QSAR研究,最终选取了4个参数用于建模,运用误差反向传播网络(BP)和径向基函数网络(RBF)建立了取代芳烃化合物对黑呆头鱼毒性的预测模型,平均相关系数R分别为0.959和0.967,模型具有很好的预测能力。并应用模型对其中7个未知毒性的化合物进行了预测。将所得结果与多元线性回归结果进行对比,发现BP网络和RBF网络结果均优于多元回归预测模型,表明毒性与量化参数之间存在着非线性关系。     

基于混沌理论的网络数据流RBF神经网络预测

应用相空间重构理论，研究了网络数据流的混沌特性，计算了实际网络数据流的关维数、Lyapunov指数，证实网络数据流存在混沌现象；据此建立了基于径向基函数(RBF)预测模型，对实际网络数据流进行预测。仿真实验表明，相对于反向传播(BP)神经网络预测，基于混沌理论的RBF神经网络预测方法学习速度快，预测精度高。     

多位取代喹诺酮药物的人工神经网络构效关系研究

利用人工神经网络研究69个喹诺酮类药物的构效关系,分别使用BP(Back Propagate)、LM(Levenberg-Marquardt)和RBF(径向基,Radial Basis Function)三种网络方法研究量子化学方法(MNDO法和AM1法)计算所得到的参数。BP网络结果不收敛,RBF网络收敛速度最快,不超过100次迭代就可得到较精确的迭代结果。LM和RBF两种网络预测训练集的正确率均达到100%。其中使用RBF网络和AM1法所得到的参数预测23个预测集的正确率82.6%,MNDO法预测预测集的正确率78.3%,体现了喹诺酮药物的氢键作用成分,结合多元回归分析和相关的耐药机理研究,提出喹诺酮药物的双氢键作用机理。人工神经网络方法可作为研究药物构效关系和研发新药物的有力工具。     

基于IRBF的草原环境污染物预测模型

综合考虑传统RBF模型在污染物预测中容易出现参数学习不足、泛化能力低下等问题,构建基于IRBF网络的草原环境污染物预测模型.利用均值思想,通过最近邻聚类之后每个类的平均值作为新的聚类中心,对中心参数进行改进,利用共轭梯度下降法优化RBF网络的权值参数.实验结果表明,IRBF网络模型相比传统的RBF和BP网络模型,预测误差得到大幅度降低,在检测性能上有明显提高,对神经网络应用于草原污染物预测进行了有效探索.     

拓扑集成指数与TIBO类衍生物构效关系的研究

4H-甲基咪唑苯二氮(?)酮(TIBO)类衍生物是抗爱滋病的一种新药,分子连接性指数是经证明应用广泛、较为成功的一种指数,本文定义并计算了TIBO类衍生物原子的特征值δi,利用量子化学计算方法,建构新的拓扑集成指数G和分子连接性指数mX,基于多元回归技术建立的对TIBO类衍生物药物的油水分配系数,作出精确估算和预测的定量结构-活性相关关系,得到的多元回归方程为:logP=0.782 G-0.1430X 0.2312X-3.829,估算的平均相对误差为2.53%。为了检验模型的稳定性和预测能力,做了留一法交互校验,预测平均相对误差为3.40%。该模型相关系数高,稳定性好,预测能力强。     

A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy

Two parameters, C and σ, must be carefully predetermined in establishing an efficient support vector machine (SVM) model. Therefore, the purpose of this study is to develop a genetic-based SVM (GA-SVM) model that can automatically determine the optimal parameters, C and σ, of SVM with the highest predictive accuracy and generalization ability simultaneously. This paper pioneered on employing a real-valued genetic algorithm (GA) to optimize the parameters of SVM for predicting bankruptcy. Additionally, the proposed GA-SVM model was tested on the prediction of financial crisis in Taiwan to compare the accuracy of the proposed GA-SVM model with that of other models in multivariate statistics (DA, logit, and probit) and artificial intelligence (NN and SVM). Experimental results show that the GA-SVM model performs the best predictive accuracy, implying that integrating the RGA with traditional SVM model is very successful.     

Statistics-based wrapper for feature selection: An implementation on financial distress identification with support vector machine

Support vector machine (SVM) is an effective tool for financial distress identification (FDI). However, a potential issue that keeps SVM from being efficiently applied in identifying financial distress is how to select features in SVM-based FDI. Although filters are commonly employed, yet this type of approach does not consider predictive capability of SVM itself when selecting features. This research devotes to constructing a statistics-based wrapper for SVM-based FDI by using statistical indices of ranking-order information from predictive performances on various parameters. This wrapper consists of four levels, i.e., data level, model level based on SVM, feature ranking-order level, and the index level of feature selection. When data is ready, predictive accuracies of a type of SVM model, i.e., linear SVM (LSVM), polynomial SVM (PSVM), Gaussian SVM (GSVM), or sigmoid SVM (SSVM), on various pairs of parameters are firstly calculated. Then, performances of SVM models on each candidate feature are transferred to be ranking-order indices. After this step, the two statistical indices of mean and standard deviation values are calculated from ranking-order information on each feature. Finally, the feature selection indices of SVM are produced by a combination of statistical indices. Each feature with its feature selection index being smaller than half of the average index is selected to compose the optimal feature set. With a dataset collected for Chinese FDI prior to 3 years, we statistically verified the performance of this statistics-based wrapper against a non-statistics-based wrapper, two filters, and non-feature selection for SVM-based FDI. Results from unseen dataset indicate that GSVM with the statistics-based wrapper significantly outperformed the other SVM models on the other feature selection methods and two wrapper-based classical statistical models.     

基于SVM的PVC汽提过程预测控制方法

针对汽提塔温度控制系统的非线性和参数时变的特性,采用支持向量机对汽提塔温度进行建模,结合非线性模型的实时线性化和广义预测控制隐式算法,提出了基于支持向量预测模型的广义预测控制算法.同时将该算法应用到聚氯乙烯汽提生产过程当中,并将模型在线校正和误差反馈校正相结合,根据实际情况进行了多种情况下的仿真,仿真结果表明了方法的有效性.     

多阶灰色支持向量机集成预测模型研究

对灰色预测模型GM(1,1)和支持向量机SVM预测模型进行分析,提出了多阶灰色支持向量机集成预测模型Dm_GM(1,1)-SVM。通过多阶缓冲算子改进灰色预测模型的预测精度,对最终预测值的各个相关指标进行预测;同时,采用粒子群优化算法对支持向量机模型进行径向基核参数和惩罚参数寻优,得到最佳参数对(c,g),从而确定支持向量机的最佳回归模型;最后将各指标预测值作为支持向量机模型的输入,依据预测模型和预测模型的输入值求得预测结果。实验实例表明,多阶灰色支持向量机集成模型和传统的预测模型相比,在本例中预测精度更高,说明多阶灰色预测模型和支持向量机模型相结合在解决实际预测问题中具有实用价值。     

基于SVM算法的碟形水下机器人姿态预测方法研究

碟形水下机器人的运动控制过程是非常复杂的,涉及到很多影响因素,并且是一个非线性控制过程,其姿态控制过程的系统模型的辨识对于实现机器人精确的控制和不同水文环境的自适应预测控制有着重要意义。因支持向量机(SVM)算法经过严格的数学推导,且在非线性等方面的良好表现,提出了将SVM算法用于碟形水下机器人模型的辨识,并设计了一组基于SVM的多输入多输出系统辨识器,可针对控制量进行姿态变化预测。通过在水池中测试的实验数据进行辨识和预测。实验验证预测的均方差不超过0.004,实验结果验证了该算法对碟形潜水器的姿态运动控制系统的辨识与预测有着良好的效果。     

基于支持向量机的航空电机表面温升预测

针对航空电机发热及散热受电机功率、结构,以及由于海拔高度改变带来的大气温度、粘度、压力变化等众多因素影响,温升模型难以准确建立的问题,通过已有试验数据,建立起遗传算法-支持向量机表面温升模型,解决了遗传算法局部搜索能力差的问题,有效地利用支持向量机学习速度快的特点,在小样本情况下具有良好的非线性建模和泛化能力。基于LS-SVM的预测控制算法具有很好的控制性能。试验表明,该模型实现了对航空电机表面温升的智能预测。由于支持向量机具有自学习功能,可在应用中不断提高预测精度,因而这种方法在电机设计中具有广阔的应用前景。     

基于支持向量机的纺纱质量预测模型研究

纱线的生产是一个多环节的复杂工业过程．其质量控制大多需要依赖领域专家的个人经验,为此提出一种基于支持向量机的纱线质量预测模型．探讨了模型选择与验证问题,并利用“网格搜索”法对模型参数进行了优化,试验结果表明,在小样本和“噪音”数据环境下．支持向量机模型仍能保持一定的预测精度,与人工神经网络模型相比,更适应于真实纺纱生产过程．     

Predicting business failure using multiple case-based reasoning combined with support vector machine

Financial distress prediction of business institutions is a long cherished topic concentrating on reducing loss of the society. Case-based reasoning (CBR) is an easily understandable methodology for problem solving. Support vector machine (SVM) is a new technology developed recently with high classification performance. Combining-classifiers system is capable of taking advantages of various single techniques to produce high performance. In this research, we develop a new combining-classifiers system for financial distress prediction, where four independent CBR systems with k-nearest neighbor (KNN) algorithms are employed as classifiers to be combined, and SVM is utilized as the algorithm fulfilling combining-classifiers. The new combining-classifiers system is named as Multiple CBR systems by SVM (Multi-CBR–SVM). The four CBR systems, respectively, are found on similarity measure on the basis of Euclidean distance metric, Manhattan distance metric, Grey coefficient metric, and Outranking relation metric. Outputs of independent CBRs are transferred as inputs of SVM to carry out combination. How to implement the combining-classifiers system with collected data is illustrated in detail. In the experiment, 83 pairs of sample companies in health and distress from Shanghai and Shenzhen Stock Exchange were collected, the technique of grid-search was utilized to get optimal parameters, leave-one-out cross-validation (LOO-CV) was used as assessment in parameter optimization, and predictive performances on 30-times hold-out data were used to make comparisons among Multi-CBR–SVM, its components and statistical models. Empirical results have indicated that Multi-CBR–SVM is feasible and validated for listed companies’ business failure prediction in China.     

Support vector machine-based multi-model predictive control

In this paper, a support vector machine-based multi-model predictive control is proposed, in which SVM classification combines well with SVM regression. At first, each working environment is modeled by SVM regression and the support vector machine network-based model predictive control （SVMN-MPC） algorithm corresponding to each environment is developed, and then a multi-class SVM model is established to recognize multiple operating conditions. As for control, the current environment is identified by the multi-class SVM model and then the corresponding SVMN-MPC controller is activated at each sampling instant. The proposed modeling, switching and controller design is demonstrated in simulation results.     

支持向量机在商场客流预测中的应用

支持向量机是数据挖掘中的一个新兴技术,它对非线性的决策边界间的建模能力是高度准确的.本文通过分析支持向量机的原理和算法,并给出基于支持向量机的客流量预测模型,最后通过试验结果说明了支持向量机在预测中的有效性.