控制点优化磨光算法在长江径流量预测中的应用

盈亏修正磨光法所得到的逼近效果仍然很差,通过控制点的参数优化和目标函数的最小,提出一种控制点优化磨光算法,利用这个算法得到参数后代入模型,使预测的精度得到提高.通过实例,该算法简单易行,并通过相对误差进行了分析,控制点优化磨光算法所得到的预测值好于神经网络模型、PPAR和小波网络模型的预测值,这为研究磨光法提供了较好的分析方法.     

基于综合效益的城市地下物流网络设计与优化方法

针对城市地下物流系统(Underground Logistics System,ULS)的特征,对一系列ULS网络节点选址与优化问题进行了建模分析.1)从解决城市交通拥堵的角度出发,探讨了物流地上地下分配的三种方案,并建立了地下货运OD评价模型.2)综合权衡货运量与货源距离,基于改进的模糊C均值聚类确定ULS—级节点的选址和辐射范围.3)对每个一级区域构建了ULS二级节点选址优化模型,通过人工免疫算法搜索最少覆盖节点群及节点的最优归属.4)建立多目标ULS网络规划模型,结合Prim算法与Dijkstra算法实现货物地下运输路径的最优选择,并采用栅格覆盖的思路在节点服务范围内对ULS网络进行费用优化.5)提出ULS网络效能评估指标,设置中心节点以提高系统运输效率和抗风险能力.     

生理药动学模型参数估计问题

根据生理药动学模型的特点,把非线性药动学模型转化为线性模型,并验证线性模型的精确性.在此基础上,构造估计药动学模型参数的目标函数,并利用非线性优化算法求解模型参数.仿真结果表明,我们的算法具有快速、精确和稳定的特点.给出了一种快速估计复杂生理药动学模型参数的方法,这为解决复杂生理药动学模型的参数估计问题提供了一种有效工具.     

多式联运分运人选择的模型和算法

在对多式联运分运人选择问题分析的基础上,运用图论技术构建了基于联运分运人选择的多式联运网络。综合考虑了运输的规模经济性、时效性、风险性和联运联盟的稳定性,建立了基于综合运输成本最小、运输风险最小和合作强度最大的多目标优化选择模型。通过主要目标法,将模型转化为单目标模型。借鉴生物免疫原理,设计了基于克隆增扩的人工免疫算法对问题进行求解。最后通过算例对方法进行了验证。     

ABC-PSO算法优化混合核SVM参数及应用

针对混合核支持向量机(SVM)中的可调参数一般是根据经验或人工随机调试得到,不能确保参数最优的局限性,提出用粒子群和人工蜂群的并行混合优化(ABC-PSO)算法来优化混合核SVM参数,找出满足条件的最优参数组合.将该SVM模型应用到语音识别中,通过对三个不同语种的语音数据库的实验仿真,验证了混合算法优化SVM参数所得的优化SVM模型比PSO算法优化SVM所得的模型,具有良好的泛化能力和语音识别能力.     

深穿透低伤害酸液酸岩反应动力学模型的建立

"深穿透低伤害酸液"体系具有酸蚀速率慢、酸蚀深度大、无二次伤害、储层改造效果明显等优点,此类酸液体系的酸岩反应动力学参数测试及酸蚀机理研究是开展深度酸化和酸压工作的基础.采用深穿透低伤害酸液体系配方,并通过新型实验装置对储层岩心开展酸蚀实验,在Hekim分布参数模型的基础上,提出深穿透低伤害酸液在储层中反应的动力学模型.通过对深穿透低伤害酸酸岩反应动力学方程的确定,可以研究酸岩反应动力学机理;研究浓度、温度、压力等外界因素对酸岩反应速度的影响;确定氢离子的有效传质速度与酸岩表面反应速率的关系等.     

灰色GM[1,1,t~(1/2)]模型及在药动学中的应用

将灰色理论用于人体药物代谢动力学的研究 ,建立了一种新的灰色模型 GM 1 ,1 ,t ,适合于那些单峰形态的原始数据列 .并根据抗高血压药复方卡托普利片的血药浓度给出了灰色药物动力模型 .结果表明 ,精度优于传统的一室模型 ,而且所用参数少 ,便于计算 .灰色理论可用于药物动力学中单峰序列的研究 .     

灰色GM(1,1,(√t))模型及在药动学中的应用

将灰色理论用于人体药物代谢动力学的研究,建立了一种新的灰色模型GM(1,1,(√t)),适合于那些单峰形态的原始数据列.并根据抗高血压药复方卡托普利片的血药浓度给出了灰色药物动力模型.结果表明,精度优于传统的一室模型,而且所用参数少,便于计算.灰色理论可用于药物动力学中单峰序列的研究.     

基于GRA-CS-BP算法的期货价格预测

针对在采用BP神经网络进行期货价格预测时,存在的模型结构复杂,易陷入局部极小值,模型无法收敛问题.考虑从网络结构和网络参数两个方面对BP网络模型进行优化,由此提出基于GRA-CS-BP算法的期货价格预测方法.首先用灰色关联度分析法进行输入变量筛选,找出和预测价格关联度大的重要因素作为网络输入,简化网络模型整体结构.然后采用布谷鸟算法对网络权阈值参数进行优化,将经过选择优化后建立的BP神经网络模型用于期货价格预测.仿真结果表明,新模型不仅具有更高的预测精度,同时其运行的稳定性也要好于单纯BP神经网络模型,为期货价格预测提出了一种新的方法.     

基于粒子群算法的大型网络计划资源优化

现有求解网络计划资源优化的方法中,解析法不能解决大型复杂网络优化问题,启发式方法过多依赖具体问题、求解效率低,遗传算法生成新一代优化解种群依据的三个算子的实现参数选择,大部分依靠经验并严重影响解的品质,粒子群算法存在大型网络计划资源优化计算量过大和缺少大型网络计划资源优化算例问题.借助设计网络计划时间参数的计算机算法、建立评价函数、设计进化方程等基础工作,选择与工作开始时间相关的变量作为粒子空间位置,用蒙特卡洛方法和限制条件优化初始粒子群,设置可行解范围,用二维动态数组解决大型网络计划资源优化运行image超限问题,通过粒子群算法进化,寻求大型网络计划资源优化解,算例表明基于粒子群算法的大型网络计划资源优化效果明显,粒子群算法参数分析表明:粒子群算法的参数会影响网络计划资源优化结果,而且初始粒子群限制条件和优化目标设置的影响程度较大.     

基于组合神经网络的时间分数阶扩散方程计算方法

该文首次采用一种组合神经网络的方法，求解了一维时间分数阶扩散方程.组合神经网络是由径向基函数（RBF）神经网络与幂激励前向神经网络相结合所构造出的一种新型网络结构.首先，利用该网络结构构造出符合时间分数阶扩散方程条件的数值求解格式，同时设置误差函数，使原问题转化为求解误差函数极小值问题；然后，结合神经网络模型中的梯度下降学习算法进行循环迭代，从而获得神经网络的最优权值以及各项最优参数，最终得到问题的数值解.数值算例验证了该方法的可行性、有效性和数值精度.该文工作为时间分数阶扩散方程的求解开辟了一条新的途径.     

Artificial immune algorithm-based credit evaluation for mobile telephone customers

The arrearage problem is a critical concern for China’s mobile communication services industry. Analysis of customer credit evaluation provides this study with a potential viable solution to the arrearage problem in China. By employing an artificial immune algorithm (AIA), a measure of customer credit based on customer attributes is proposed. This method was applied to one China mobile communication services company with approximately 400?000 customers yielding satisfying results. Utilizing traditional predictive accuracy and alternative metrics, performance comparisons of the proposed AIA were made using the feed-forward back propagation artificial neural network and the logistic regression model. A decision tree analysis of anticipated benefits was performed and indicates workability of the proposed method based on customer credit evaluation.     

Maximum likelihood estimation for a road traffic network model

The use of the maximum likelihood method for parameter estimation in a probabilistic path choice model used for road traffic networks is described. The method requires observation of mean travel costs in a given network rather than the traffic flows throughout the network, and thus data collection efforts may be reduced. A test network, for which many sets of artificial data were generated, is examined so that the usefulness of the parameter estimation procedure may be assessed. It is shown that the procedure is both convenient and quick, and that accurate estimates of model parameters may be obtained.     

Study on network traffic forecast model of SVR optimized by GAFSA

There are some problems, such as low precision, on existing network traffic forecast model. In accordance with these problems, this paper proposed the network traffic forecast model of support vector regression (SVR) algorithm optimized by global artificial fish swarm algorithm (GAFSA). GAFSA constitutes an improvement of artificial fish swarm algorithm, which is a swarm intelligence optimization algorithm with a significant effect of optimization. The optimum training parameters used for SVR could be calculated by optimizing chosen parameters, which would make the forecast more accurate. With the optimum training parameters searched by GAFSA algorithm, a model of network traffic forecast, which greatly solved problems of great errors in SVR improved by others intelligent algorithms, could be built with the forecast result approaching stability and the increased forecast precision. The simulation shows that, compared with other models (e.g. GA-SVR, CPSO-SVR), the forecast results of GAFSA-SVR network traffic forecast model is more stable with the precision improved to more than 89%, which plays an important role on instructing network control behavior and analyzing security situation.     

人工神经网络BP算法的改进和结构的自调整

本文解决了BP神经网络结构参数和学习速率的选取问题,并对传统的BP算法进行了改进,提出了BP神经网络动态全参数自调整学习算法,又将其编制成计算机程序,使得隐层节点和学习速率的选取全部动态实现,减少了人为因素的干预,改善了学习速率和网络的适应能力。计算结果表明：BP神经网络动态全参数自调整算法较传统的方法优越。训练后的神经网络模型不仅能准确地拟合训练值,而且能较精确地预测未来趋势。     

A comparative assessment of classification methods for resonance frequency prediction of Langevin piezoelectric transducers

A Langevin piezoelectric transducer is used as a physical element for transmitting and receiving sound waves. The operating frequency of a transducer determines the distance that the sound wave can travel, so it is important to measure it. Due to the fact the structure of a transducer is quite complicated, it is quite difficult to estimate the precise physical parameters for the simulation model. Therefore, it takes a long time to measure the resonance frequency in the laboratory and fix the parameters by trial and error methods. This study applies a learning method to estimate a transducer frequency instead by trial and error experiments. The learning methods applied and compared including artificial neural network, support vector machine, C4.5, neuro-fuzzy, and ega-fuzzification. Compared with the theoretical one-dimensional model (simple lump element model), the results indicate that a learning method is an efficient way to estimate the piezoelectric transducer resonance frequency. The mega-fuzzification method is the best compared with other methods in this study.     

Optimizing fracture design based on a novel RBF neural network

The factors affecting performance of fractured wells are analyzed in this work. The static and dynamic geologic data of fractured well and fracturing treatment parameters obtained from 51 fractured wells in sand reservoirs of Zhongyuan oilfield are analyzed by applying the grey correlation method. Ten parameters are screened, including penetrability, porosity, net thickness, oil saturation, water cut, average daily production, and injection rate, amount cementing front spacer, amount sand-carrying agent and amount sand. With the novel Radial Basis Function neural network model based on immune principles, 13 parameters of 42 wells out of 51 are used as the input samples and the stimulation ratios as the output samples. The nonlinear interrelationship between the input samples and output samples are investigated, and a productivity prediction model of optimizing fracture design is established. The data of the rest 7 wells are used to test the model. The results show that the relative errors are all less than 7%, which proves that the novel Radial Basis Function neural network model based on immune principles has less calculation, high precision and good generalization ability.     

Chaotic artificial immune approach applied to economic dispatch of electric energy using thermal units

The economic dispatch problem (EDP) is an optimization problem useful in power systems operation. The objective of the EDP of electric power generation, whose characteristics are complex and highly non-linear, is to schedule the committed generating unit outputs so as to meet the required load demand at minimum operating cost while satisfying system constraints. Recently, as an alternative to the conventional mathematical approaches, modern heuristic optimization techniques have been given much attention by many researchers due to their ability to find an almost global optimal solution in EDPs. As special mechanism to avoid being trapped in local minimum, the ergodicity property of chaotic sequences has been used as optimization technique in EDPs. Based on the chaos theory, this paper discusses the design and validation of an optimization procedure based on a chaotic artificial immune network approach based on Zaslavsky’s map. The optimization approach based on chaotic artificial immune network is validated for a test system consisting of 13 thermal units whose incremental fuel cost function takes into account the valve-point loading effects. Simulation results and comparisons show that the chaotic artificial immune network approach is competitive in performance with other optimization approaches presented in literature and is also an attractive tool to be used on applications in the power systems field.     

Forecasting with cue information: A comparison of multiple regression with alternative forecasting approaches

Multiple linear regression (MLR) is a popular method for producing forecasts when data on relevant independent variables (or cues) is available. The accuracy of the technique in forecasting the impact on Greek TV audience shares of programmes showing sport events is compared with forecasts produced by: (1) a simple bivariate regression model, (2) three different types of artificial neural network, (3) three forms of nearest neighbour analysis and (4) human judgment. MLR was found to perform relatively poorly. The application of Theil’s bias decomposition and a Brunswik lens decomposition suggested that this was because of its inability to handle complex non-linearities in the relationship between the dependent variable and the cues and its tendency to overfit the in-sample data. Much higher accuracy was obtained from forecasts based on a simple bivariate regression model, a simple nearest neighbour procedure and from two of the types of artificial neural network.