遗传算法的改进及其在电力系统无功优化中的应用

针对无功优化的动态、多目标、多约束以及非线性特点,提出了基于简单遗传算法的改进算法,在简单遗传算法的基础上,添加了自适应控制和适应度排序思想,并对算法终止准则进行了改进。以IEEE14节点标准测试系统为例对该算法进行了有效性验证。仿真结果表明,该方法对电力系统的无功优化效果良好。     

Ranking Functions Induced by Weighted Average of Fuzzy Numbers

In this paper we present two definitions of possibilistic weighted average of fuzzy numbers, and by them we introduce two different rankings on the set of real fuzzy numbers. The two methods are dependent on several parameters. In the first case, the parameter is constant and the results generalize what Carlsson and Fuller have obtained in (2001). In the second case, the parameter is a function, not fixed a priori by the decision maker, but it depends on the position of the interval on the real axe. In all the two cases we call the parameter degree of risk, which takes into account of a risk-tendency or aversion of the decision maker.     

The de novo design of median molecules within a property range of interest

Summary In this paper an application is presented of the median molecule workflow to the de novo design of novel molecular entities with a property profile of interest. Median molecules are structures that are optimised to be similar to a set of existing molecules of interest as an approach for lead exploration and hopping. An overview of this workflow is provided together with an example of an instance using the similarity to camphor and menthol as objectives. The methodology of the experiments is defined and the workflow is applied to designing novel molecules for two physical property datasets: mean molecular polarisability and aqueous solubility. This paper concludes with a discussion of the characteristics of this method.     

基于背景先验与中心先验的显著性目标检测北大核心CSCD

针对传统的图流行排序显著性目标检测算法存在先验信息单一,显著目标检测不完整的问题,提出一种新的基于背景先验与中心先验的显著性目标检测算法。首先将图像边界节点作为背景种子进行流行排序获得粗略的前景区域,将其再次流行排序得到初步显著图;然后利用Harris角点检测、聚类实现中心先验显著性检测,捕获中心显著信息;最后在初步显著图上融合图像中心显著性,得到最终显著图。本文对综合指标、精确率-召回率曲线、F-measure值以及平均绝对误差(mean absolute error,MAE)值进行实验评估,在公开数据集MSRA-10K和ECSSD上进行的实验结果表明:对比10种主流算法,本文算法在不同的评估指标上都具有较好的表现,且能准确地突出显著目标,提升背景抑制效果。     

Simultaneous kinetic determination of thiocyanate and sulfide using eigenvalue ranking and correlation ranking in principal component-wavelet neural network

In this work, two toxic compound, sulfide and thiocyanate were determined simultaneously using kinetic spectrophotometry. These anions have shown the catalytic effects on the reaction between iodine and azide. Since the system was nonlinear, a nonlinear model, principal component-wavelet neural network (PC-WNN) was used as the multivariate calibration method. The principal component analysis was used to decrease the dimension of the original matrix. In other words, the scores of the PCs, 5, instead of the original variables, 301, were used as the input for the model. Two methods were used to select the most relevant principal components: eigenvalue ranking and correlation ranking. In this work, eigenvalue and correlation ranking methods have shown better results for thiocyanate and sulfide, respectively, and it can be concluded that these methods are complementary. The WNN has several advantages relative to other types of neural network such as better convergence ability. The data set was divided to calibration, prediction and validation sets. Each set was selected so that the concentrations of the analytes were approximately covered the entire ranges of the analytes. Mean relative error for thiocyanate and sulfide in validation set were 8.5 and 10.6, respectively. Thiocyanate and sulfide can be determined in the range of 60–700 ng ml⁻¹ and 20–400 ng ml⁻¹, respectively. The proposed method was applied for the determination of sulfide and thiocyanate in real samples such as tap, waste and river waters with satisfactory results.     

New indices for analysing partial ranking diagrams

Interest is growing in decision making strategies and several techniques are now available. The assessment of priorities is a typical premise before final decisions are taken. Total and partial order ranking (POR) strategies, which from a mathematical point of view are based on elementary methods of discrete mathematics, appear as an attractive and simple tool to asses priorities. Despite the well-known total ranking strategies, which are scalar methods combining the different criteria values into a global index which always ranks elements in an ordered sequence, the partial order ranking is a vectorial approach which recognises that not all the elements can be directly compared with all the others. In fact when many criteria are considered, contradictions in the ranking are bound to exist and the higher the number of criteria, the higher the probability that contradictions in the ranking occur. The Hasse diagram technique (HDT) is a very useful tool to perform partial order ranking. The results of the partial order ranking are visualised in a diagram, called Hasse diagram. Incomparable elements are located at the same geometrical height and as high as possible in the diagram, thus incomparable elements are arranged in levels. The quality of a ranking procedure has to be evaluated by a deep analysis and by several indices, i.e. scalar functions that describe features of an ordered set and allow comparison among different rankings. For this purpose, new indices for ranking analysis are proposed here, compared with the ones found in literature and tested on theoretical examples and on real data.     

Development of a multicriteria assessment model for ranking biomass feedstock collection and transportation systems

This study details multicriteria assessment methodology that integrates economic, social, environmental, and technical factors in order to rank alternatives for biomass collection and transportation systems. Ranking of biomass collection systems is based on cost of delivered biomass, quality of biomass supplied, emissions during collection, energy input to the chain operations, and maturity of supply system technologies. The assessment methodology is used to evaluate alternatives for collecting 1.8×10⁶ dry t/yr based on assumptions made on performance of various assemblies of biomass collection systems is based on cost of delivered biomass, quality of biomass supplied, emissions during collection, energy input to the chain operations, and maturity of supply system technologies. The assessment methodology is used to evaluate alternatives for collecting 1.8×10⁶ dry t/yr based on assumptions made on performance of various assemblies of biomass collection systems. A proposed collection option using loafer/stacker was shown to be the best option followed by ensiling and baling. Ranking of biomass transport systems is based on cost of biomass transport, emissions during transport, traffic congestion, and maturity of different technologies. At a capacity of 4×10⁶ dry t/yr, rail transport was shown to be the best option, followed by truck transport and pipeline transport, respectively. These rankings depend highly on assumed maturity of technologies and scale of utilization. These may change if technologies such as loafing or ensiling (wet storage) methods are proved to be infeasible for large-scale collection systems.     

四川省市州经济发展差异的统计分析

采用因子分析研究四川省21市州之间的相关性,利用主因子计算出各个市州得分,以市州得分作为新变量进行集团排序,再将基于相对位置排序法引入到集团排序中,进而得到最终的市州分类结果,在统计分析的基础上对市州经济发展给出一些建议.     

排序支持的交互数据分类算法及其应用

交互分类是解决数据复杂分类问题的主要手段之一。在现有的大多交互分类系统中,用户能准确识别数据类别,但在有些分类场景中,类别之间的顺序关系更容易被识别,为此,提出一种排序支持的交互数据分类算法。为提升交互分类精度,引入数据的顺序信息,为降低标记难度,提出候选样本推荐策略。另外,提出一种评估分类算法性能的可视化方法,用包含基本车况、交通违法记录、交通事故记录等信息的车辆数据集进行实验验证,将相关车辆分为高危车辆、中危车辆、低危车辆3类,算法的分类结果模型一致度达近98%,验证了方法的有效性。     

区间数的排序方法研究

指出献^[3-5]所定义的用于区间数排序的可能度存在的不足,分别给出了刻画区间数大小比较的相对优势度的定义和模糊互补矩阵排序的一种新方法。在此基础上,给出了区间数的一种排序方法。