Study on Coal Consumption Curve Fitting of the Thermal Power Based on Genetic Algorithm

Coal consumption curve of the thermal power plant can reflect the function relationship between the coal consumption of unit and load, which plays a key role for research on unit economic operation and load optimal dispatch. Now get coal consumption curve is generally obtained by least square method, but which are static curve and these curves remain unchanged for a long time, and make them are incompatible with the actual operation situation of the unit. Furthermore, coal consumption has the characteristics of typical nonlinear and time varying, sometimes the least square method does not work for nonlinear complex problems. For these problems, a method of coal consumption curve fitting of the thermal power plant units based on genetic algorithm is proposed. The residual analysis method is used for data detection;quadratic function is employed to the objective function;appropriate parameters such as initial population size, crossover rate and mutation rate are set;the unit’s actual coal consumption curves are fitted, and comparing the proposed method with least squares method, the results indicate that fitting effect of the former is better than the latter, and further indicate that the proposed method to do curve fitting can best approximate known data in a certain significance, and they can real-timely reflect the interdependence between power output and coal consumption.

The rate of convergence for the least squares estimator in nonlinear regression model with dependent errors

We study the parameter estimation in a nonlinear regression model with a general error's structure,strong consistency and strong consistency rate of the least squares estimator are obtained.

Nonlinear unknown input observer using mean value theorem and genetic algorithm

In this note,we consider a new unknown input observer design for nonlinear systems.The main idea consists in determining the estimation error and mean value theorem parameters(β)to introduce them into proposed observer structure.This process is designed on the basis of mean value theorem and genetic algorithm.The stability study relies on the use of a classical quadratic Lyapunov function.The observer’s gains are determined systematically.For the validation of theoretical development proposed in this paper,we consider two practical realizations that deals with the secure communication problem.

A Study of EM Algorithm as an Imputation Method: A Model-Based Simulation Study with Application to a Synthetic Compositional Data

Compositional data, such as relative information, is a crucial aspect of machine learning and other related fields. It is typically recorded as closed data or sums to a constant, like 100%. The statistical linear model is the most used technique for identifying hidden relationships between underlying random variables of interest. However, data quality is a significant challenge in machine learning, especially when missing data is present. The linear regression model is a commonly used statistical modeling technique used in various applications to find relationships between variables of interest. When estimating linear regression parameters which are useful for things like future prediction and partial effects analysis of independent variables, maximum likelihood estimation (MLE) is the method of choice. However, many datasets contain missing observations, which can lead to costly and time-consuming data recovery. To address this issue, the expectation-maximization (EM) algorithm has been suggested as a solution for situations including missing data. The EM algorithm repeatedly finds the best estimates of parameters in statistical models that depend on variables or data that have not been observed. This is called maximum likelihood or maximum a posteriori (MAP). Using the present estimate as input, the expectation (E) step constructs a log-likelihood function. Finding the parameters that maximize the anticipated log-likelihood, as determined in the E step, is the job of the maximization (M) phase. This study looked at how well the EM algorithm worked on a made-up compositional dataset with missing observations. It used both the robust least square version and ordinary least square regression techniques. The efficacy of the EM algorithm was compared with two alternative imputation techniques, k-Nearest Neighbor (k-NN) and mean imputation (), in terms of Aitchison distances and covariance.

Orthogonal-Least-Squares Forward Selection for Parsimonious Modelling from Data

The objective of modelling from data is not that the model simply fits the training data well. Rather, the goodness of a model is characterized by its generalization capability, interpretability and ease for knowledge extraction. All these desired properties depend crucially on the ability to construct appropriate parsimonious models by the modelling process, and a basic principle in practical nonlinear data modelling is the parsimonious principle of ensuring the smallest possible model that explains the training data. There exists a vast amount of works in the area of sparse modelling, and a widely adopted approach is based on the linear-in-the-parameters data modelling that include the radial basis function network, the neurofuzzy network and all the sparse kernel modelling techniques. A well tested strategy for parsimonious modelling from data is the orthogonal least squares (OLS) algorithm for forward selection modelling, which is capable of constructing sparse models that generalise well. This contribution continues this theme and provides a unified framework for sparse modelling from data that includes regression and classification, which belong to supervised learning, and probability density function estimation, which is an unsupervised learning problem. The OLS forward selection method based on the leave-one-out test criteria is presented within this unified data-modelling framework. Examples from regression, classification and density estimation applications are used to illustrate the effectiveness of this generic parsimonious modelling approach from data.

LIMITING BEHAVIOR OF RECURSIVE M-ESTIMATORS IN MULTIVARIATE LINEAR REGRESSION MODELS AND THEIR ASYMPTOTIC EFFICIENCIES

Recursive algorithms are very useful for computing M-estimators of regression coefficients and scatter parameters. In this article, it is shown that for a nondecreasing ul （t）, under some mild conditions the recursive M-estimators of regression coefficients and scatter parameters are strongly consistent and the recursive M-estimator of the regression coefficients is also asymptotically normal distributed. Furthermore, optimal recursive M-estimators, asymptotic efficiencies of recursive M-estimators and asymptotic relative efficiencies between recursive M-estimators of regression coefficients are studied.

Modified Levenberg-Marquardt algorithm for source localization using AOAs in the presence of sensor location errors

In this paper,by utilizing the angle of arrivals(AOAs) and imprecise positions of the sensors,a novel modified Levenberg-Marquardt algorithm to solve the source localization problem is proposed.Conventional source localization algorithms,like Gauss-Newton algorithm and Conjugate gradient algorithm are subjected to the problems of local minima and good initial guess.This paper presents a new optimization technique to find the descent directions to avoid divergence,and a trust region method is introduced to accelerate the convergence rate.Compared with conventional methods,the new algorithm offers increased stability and is more robust,allowing for stronger non-linearity and wider convergence field to be identified.Simulation results demonstrate that the proposed algorithm improves the typical methods in both speed and robustness,and is able to avoid local minima.

LS-SVM model based nonlinear predictive control for MCFC system

This paper describes a nonlinear model predictive controller for regulating a molten carbonate fuel cell (MCFC). In order to improve MCFC’s generating performance, prolong its life and guarantee safety, it must be controlled efficiently. First, the output voltage of an MCFC stack is identified by a least squares support vector machine (LS-SVM) method with radial basis function (RBF) kernel so as to implement nonlinear predictive control. And then, the optimal control sequences are obtained by applying genetic algorithm (GA). The model and controller have been realized in the MATLAB environment. Simulation results indicated that the proposed controller exhibits satisfying control effect.

正弦余弦算法求解含有异常值的非线性数据拟合

针对非线性数据拟合问题,建立以残差的平方和与绝对值和为目标的最小二乘与最小一乘模型,采用正弦余弦算法计算模型参数.计算结果表明:如果数据的分布是对称且无异常值,则最小二乘得到的结果与最小一乘得到的结果基本一致;如果数据存在异常值,则异常值对最小二乘有着较大的影响,而对最小一乘的影响较小.

A novel procedure for identifying a hybrid QTL-allele system for hybrid-vigor improvement, with a case study in soybean(Glycine max)yield

“Breeding by design” for pure lines may be achieved by construction of an additive QTL-allele matrix in a germplasm panel or breeding population, but this option is not available for hybrids, where both additive and dominance QTL-allele matrices must be constructed. In this study, a hybrid-QTL identification approach, designated PLSRGA, using partial least squares regression(PLSR) for model fitting integrated with a genetic algorithm(GA) for variable selection based on a multi-locus, multi-allele model is described for additive and dominance QTL-allele detection in a diallel hybrid population(DHP). The PLSRGA was shown by simulation experiments to be superior to single-marker analysis and was then used for QTL-allele identification in a soybean DPH yield experiment with eight parents. Twenty-eight main-effect QTL with 138 alleles and nine QTL × environment QTL with 46 alleles were identified, with respective contributions of 61.8% and 23.5% of phenotypic variation. Main-effect additive and dominance QTL-allele matrices were established as a compact form of the DHP genetic structure. The mechanism of heterosis superior-to-parents(or superior-to-parents heterosis, SPH) was explored and might be explained by a complementary locus-set composed of OD+(showing positive over-dominance, most often), PD+(showing positive partial-to-complete dominance, less often) and HA+(showing positive homozygous additivity, occasionally) loci, depending on the parental materials. Any locus-type, whether OD+, PD + and HA+, could be the best genotype of a locus. All hybrids showed various numbers of better or best genotypes at many but not necessarily all loci, indicating further SPH improvement. Based on the additive/dominance QTL-allele matrices, the best hybrid genotype was predicted, and a hybrid improvement approach is suggested. PLSRGA is powerful for hybrid QTL-allele detection and cross-SPH improvement.

基于LS-SVM的宽带接收前端非线性补偿算法

针对目前常用的基于参数化非线性模型(Parameterized Nonlinear Model,PNM)的补偿算法存在易陷入局部最小值,导致补偿性能不稳的问题,该文提出了基于最小二乘支持向量机(Least Squares Support Vector Machine,LS-SVM)的宽带接收前端非线性补偿算法.该算法基于减谱-时频变换法(Spectrum Reduction Algorithm based on Time-Frequency Conversion,SRA-TFC)盲分离接收前端输出信号中的大功率基波信号和其他小功率信号,并以此作为LS-SVM逆模型的训练输入-输出样本对.引入最小二乘支持向量回归(Least Squares Support Vector Regression,LS-SVR)算法高精度拟合接收前端非线性逆模型.通过以宽带接收前端的输出信号为测试样本消除其非线性失真分量.仿真与实测结果表明:该算法可使宽带接收前端的无杂散失真动态范围(Spurs-Free-Dynamic-Range,SFDR)提高约20 dB,较基于PNM的补偿算法提高了约5 dB.

工厂化水产养殖溶解氧预测模型优化

为准确预测溶解氧变化趋势,降低水产养殖风险,提出混沌变异的分布估计(CMEDA)算法优化最小二乘支持向量机模型(LSSVR),提高了溶解氧预测精度。并对粒子群算法和遗传算法分别优化的LSSVR模型(PSOLSSVR、GA-LSSVR)以及传统的LSSVR模型与CMEDA优化的LSSVR模型(CMEDA-LSSVR)进行了比较研究。利用该模型对江苏省扬中市红鲷鱼工厂化养殖鱼塘溶解氧含量进行了预测。实验结果表明,CMEDA-LSSVR的预测精度高于其他3种算法,CMEDA-LSSVR、PSO-LSSVR、GA-LSSVR、LSSVR 4种模型预测精度评价指标平均绝对百分比误差分别为0.32%、1.27%、1.98%和2.56%。实际应用结果表明该模型可以为鱼塘水质决策管理提供依据,具有一定的应用价值。

LS-SVM与多层前向网络的非线性回归性能比较

在阐述支持向量机(SVM)和最小二乘支持向量机(LS-SVM)的原理并比较了两者的优缺点后,将LS-SVM与多层前向网络中的两种典型网络BP网络和RBF网络,分别应用于装载机载重动态测量的非线性函数回归估计中,对这三种网络在函数逼近和泛化能力两方面的性能进行比较研究。仿真结果表明,LS-SVM在精度和泛化性能两方面做到了最好的折衷,是用于非线性函数回归分析的一种很有效的方法。

混合威布尔分布参数估计的L-M算法

混合威布尔分布模型常用来分析具有多种失效模式的机械系统或零部件的可靠性寿命数据,为提高混合威布尔分布未知参数估计的精度,利用非线性最小二乘理论,建立了小子样条件下两重混合威布尔分布参数优化估计模型,将L-M算法用于优化求解。以概率图参数估计法的结果作为迭代初始值,提高了迭代求解的速度。计算实例表明利用该方法估计混合威布尔分布参数是可行的,而且能够获得较精确的结果。

岩溶地下河日流量预测的小样本非线性时间序列模型

针对岩溶含水系统高度的非线性特征,在小样本时间序列条件下,引入了能较好解决小样本、非线性问题的支持向量回归方法,利用偏最小二乘回归对影响地下河流量的诸多因素进行综合分析,并提取主成分作为支持向量机的输入变量,采用遗传算法优化模型参数,建立了地下河日流量预测的偏最小二乘-遗传-支持向量回归模型;将该模型用于后寨典型岩溶地下河流域日流量模拟和预测,并与BP人工神经网络、多元线性回归模型预测结果进行对比。偏最小二乘-遗传-支持向量回归模型模拟期的均方误差(MSE)、平均绝对百分比误差(MAPE)分别为0.25%、6.89%,预测期为0.65%、6.03%;BP神经网络模拟期的MSE、MAPE分别为0.24%、7.30%,预测期为0.84%、7.39%;多元线性回归模型模拟期的MSE、MAPE分别为0.28%、9.30%,预测期为1.10%、10.54%。结果表明,偏最小二乘-遗传-支持向量回归模型预测精度明显优于BP人工神经网络和多元线性回归模型。

非线性最小二乘估计的遗传算法

探讨了用遗传算法进行非线性模型参数估计的可能性,设计了非线性最小二乘估计的遗传算法,并用实例验证了该算法的有效性。通过比较该算法与其它算法的结果,得出了一些具有参考价值的结论。

遗传门限自回归模型在气象时间序列预测中的应用

提出了建立门限自回归模型（TAR）的一套简便通用的方法。用基于实码的改进遗传算法,可同时优化门限值和自回归系数,解决了TAR建模过程所涉及的大量复杂寻优工作这一难题,为TAR模型在气象预测中的广泛应用提供了有力工具。实例计算的结果说明:通过门限值的控制作用,TAR模型可有效地利用气象时序资料所隐含的时序分段相依性这一重要信息,限制了模型误差,保证了TAR模型预测性能的稳健性,提高了预测精度。该方法具有通用性,在各种气象非线性时序预测中具有广泛的实用价值。

基于LS-SVM的混合动力镍氢电池组SOC预测

在电池管理系统中为了使荷电状态量SOC(state of charge)估计精确,提出以遗传算法优化最小二乘支持向量机(LS-SVM)的方法对电池的SOC进行预测的模型。在电池变流情况下对SOC进行研究,以标准工况下的实验数据作为样本,以电池的电流、电压及温度作为训练模型的输入,SOC作为输出建立模型,使之能很好地适用于混合动力汽车用电池在变电流状态下的实时SOC估计。研究结果表明:该预测模型预测精度高,其最大相对误差小于3%,平均相对误差小于2%,且与神经网络预测结果相比具有更强的实用性。

一种新的非线性回归模型参数估计算法

提出一种新的基于混合基因算法 (HGA)的非线性回归模型参数估计算法 .新算法通过对问题的解空间交替进行全局和局部搜索 ,达到快速收敛至全局最优解 ,较好地解决了传统算法通用性差、易陷入局部极小的问题 .实验验证了算法的通用性和有效性 .

导数同步荧光光谱-小波-SGA-LSSVR联用快速测定鸭蛋蛋清中新霉素残留含量

新霉素在巯基乙醇存在的条件下与邻苯二甲醛生成的衍生物具有强荧光特性,可实现鸭蛋蛋清中新霉素残留含量的荧光测定。在模型建立过程中,分析了波长为280～390 nm光谱范围内的三维同步荧光光谱,确定检测鸭蛋蛋清中的新霉素含量的最佳波长差Δλ为110 nm;然后利用db10小波的2层分解对一阶导数同步荧光光谱进行去噪处理,并利用分段遗传算法(SGA)优选出了14个特征波长;最后应用最小二乘支持向量回归(LSSVR)建立了鸭蛋蛋清中的新霉素含量的预测模型,其模型预测集的决定系数(R2)和预测均方根误差(RMSEP)分别为0.9671和1.713。结果表明,SGA能有效提取出鸭蛋蛋清中新霉素对应的特征波长,有利于提高LSSVR模型预测精度,可实现鸭蛋蛋清中新霉素残留含量的快速测定。