Solution of the isochoric–isoenergetic flash problem by direct entropy maximization

A new algorithm to find the phase equilibrium conditions in systems with specified values of internal energy, volume, and number of moles of each component present (isochoric–isoenergetic flash problem) is proposed. The core of the procedure consists of maximizing the system entropy by iterating on the values, in each phase, of internal energy, volume, and number of moles of each component. Analytical expressions for the physical properties and derivatives required by the calculations were generated by computer algebra. The algorithm tests for the possible need to add or remove phases during the course of iterations. The paper discusses possible numerical difficulties during application of the procedure and how to overcome them. The algorithm has shown to be robust and capable of solving multiphase equilibrium problems, avoiding trivial solutions.     

Artificial neural network for quantitative determination of total protein in yogurt by infrared spectrometry

A method has been introduced for quantitative determination of protein content in yogurt samples based on the characteristic absorbance of protein in 1800-1500 cm^− 1 spectral region by mid-FTIR spectroscopy and chemometrics. Successive Projection Algorithm (SPA) wavelength selection procedure, coupled with feed forward Back-Propagation Artificial Neural Network (BP-ANN) model was the benefited chemometric technique. Relative Error of Prediction (REP) in BP-ANN and SPA-BP-ANN methods for training set was 7.25 and 3.70 respectively. Considering the complexity of the sample, the ANN model was found to be reliable, while the proposed method is rapid and simple, without any sample preparation step.     

A new algorithm for the intersection of a line with the independent set polytope of a matroid

We present a new algorithm for the problem of determining the intersection of a half-line with the independent set polytope of a matroid. We show it can also be used to compute the strength of a graph and the corresponding partition using successive contractions. The algorithm is based on the maximization of successive linear forms on the boundary of the polytope. We prove it is a polynomial algorithm in probability with average number of iterations in O(n⁵). Finally, numerical tests reveal that it should only require O(n²) iterations in practice.     

On the complexity of paths avoiding forbidden pairs

Given a graph G=(V,E), two fixed vertices s,t∈V and a set F of pairs of vertices (called forbidden pairs), the problem of a path avoiding forbidden pairs is to find a path from s to t that contains at most one vertex from each pair in F. The problem is known to be NP-complete in general and a few restricted versions of the problem are known to be in P. We study the complexity of the problem for directed acyclic graphs with respect to the structure of the forbidden pairs.We write x?y if and only if there exists a path from x to y and we assume, without loss of generality, that for every forbidden pair (x,y)∈F we have x?y. The forbidden pairs have a halving structure if no two pairs (u,v),(x,y)∈F satisfy v?x or v=x and they have a hierarchical structure if no two pairs (u,v),(x,y)∈F satisfy u?x?v?y. We show that the PAFP problem is NP-hard even if the forbidden pairs have the halving structure and we provide a surprisingly simple and efficient algorithm for the PAFP problem with the hierarchical structure.     

模糊支付矩阵混合策略模型及其求解

在各局中人没有最优纯策略时,考虑选取不同策略的概率分布和支付矩阵的模糊性,建立模糊支付混合策略的对策模型.基于线性规划和三角模糊数排序准则,给出模型求解方法.模型在实例中得到验证,其计算简单,结果合理,可以为决策提供科学依据.     

基于MIV-BP型网络实验的房地产项目风险识别研究

为了更准确更客观地识别房地产项目中的风险,为房地产项目投资决策提供科学依据和参考,有效地规避风险,本研究在BP神经网络 (Back-Propagation Neural Network)建模的基础上,采取MIV(Mean Impact Value)算法对BP神经网络模型进行变量筛选的网络优化和改良,从而形成新的优化后的MIV-BP(Mean Impact Value Back-Propagation Neural Network)神经网络,并以此用于评价房地产项目中的风险度以及各因素在风险度中的影响作用大小;同时选取目前相关的房地产项目数据进行仿真实证分析和验证。验证实验结果表明,MIV-BP型神经网络对于房地产项目风险度识别具有良好的适应性和准确性,实验结果客观,达到专家评价的要求,并在风险因素作用度分析上具有良好的应用价值。     

基于遗传算法的变电站巡检机器人任务路径规划方法研究

近年来,随着变电站巡检机器人在变电站中的广泛使用,巡检机器人路径规划问题越来越成为亟待解决的问题。巡检机器人在已知的拓扑地图中标记了待执行巡检任务的停靠点,不同任务需要从初始点出发经过不同的一系列停靠点再返回初始点,如何规划路径是机器人面临的问题。首先分析了路径规划面临的问题,然后通过分析拓扑地图的特征,对地图进行等价简化,再对问题进行建模使用遗传算法求解巡检任务路径规划的近似最优解。通过仿真实验证明,提出的基于遗传算法的路径规划方法是可行有效的,为变电站巡检机器人任务路径规划提供了一种有效方法。     

基于遗传算法的安溪铁观音品质快速评价研究

为探究一种快速无损的安溪铁观音品质评价方法,利用遗传算法(GA)对茶样的近红外光谱特征波长进行筛选,结合偏最小二乘(PLS),建立全谱段的PLS定量模型与GA-PLS模型。结果表明,傅里叶变换近红外(FT-NIR)全谱段光谱在经过平滑+二阶导数+归一化处理后,PLS模型预测性能最高,建模结果为：校正集相关系数R_C=0.921,校正集均方根误差RMSEC=0.543,验证集相关系数R_P=0.913,验证集均方根误差RMSEP=0.665。选用近红外光谱6 670～4 000 cm-1谱区,采用遗传算法进行特征波长筛选,参与建模数据点数从1 557缩减到408个。优选波段后,GA-PLS建模结果为：校正集相关系数R_C=0.959,校正集均方根误差RMSEC=0.413,验证集相关系数R_P=0.940,验证集均方根误差RMSEP=0.587。可见,GA-PLS模型的校正集和验证集的预测结果均优于全谱段PLS模型。结果说明,在传统的近红外光谱技术结合化学计量学方法的建模基础上,加入遗传算法进行波长筛选,能有效提高模型预测能力,实现方法学的创新研究,且GA-PLS品质评价模型具有较强的参考和推广价值,为提高我国茶叶品质的检测技术水平提供新的方法借鉴。     

基于特征融合的手背静脉识别

图像平移和旋转会降低手背静脉识别的准确性,针对该问题,本文提出了一种特征融合的手背静脉识别法.该方法充分考虑图像的细节特征和全局特征,首先选取图像中的交叉点和端点作为特征点,再从特征点中提取出图像匹配的基准点,计算基准点至特征点间的相对距离及基准点与特征点连线间相邻连线产生的夹角作为细节特征;然后利用不变矩方法提取图像特征作为全局特征;最后将两种图像特征融合,进行手背静脉识别.实验模拟结果表明,该方法可快速有效地实现手背静脉识别.