微粒群神经网络在常压塔汽油干点软测量建模中的应用

首先将微粒群优化算法用于神经网络连接权值和阈值的训练,构造微粒群神经网络,然后将微粒群神经网络用于常压塔汽油干点软测量建模。通过与实际值的对比,结果表明基于微粒群神经网络的软测量模型具有良好的性能和极好的应用前景。     

微粒群算法在干燥器多变量优化设计中的应用

建立了以具有废气循环的回转干燥系统年总费用为目标函数的优化设计数学模型,在此基础上探讨了惯性权因子对微粒群算法性能的影响,并应用微粒群算法求解干燥器优化设计数学模型,对干燥器出口废气温度与循环比进行优化设计。结果表明,带动态非线性惯性因子的微粒群算法对求解多变量的干燥优化设计问题具有方法简单、所需微粒群规模小、收敛速度快等特点;采用部分废气循环并进行优化设计对干燥系统的节能具有十分重要的意义,对湿空气出口温度和废气循环比进行优化设计,其年总费用比无废气循环的常规设计节省18．2%,比循环比为0．2时的常规设计节省12．6%。     

噪声环境下参数估计和模型降阶的一种有效方法

针对噪声环境下的非线性系统参数估计和模型降阶问题,提出了一种带假设检验的微粒群优化算法(PSOHT),以最小化平均平方误差为目标,结合统计意义下的评价和比较,通过微粒群操作进行参数估计.基于典型非线性时滞系统的仿真实验,验证了所提算法的有效性和抗噪声能力.     

微粒群算法在模拟移动床色谱分离过程优化中的应用

运用微粒群算法开发出一种非线性模拟移动床(SMB)色谱分离过程的优化策略.该优化策略将模拟移动床的最大吸附剂生产率作为优化问题的目标函数,采用模拟移动床的TMB模型来计算微粒群优化算法的适应值.采用该优化算法对手性化合物萘酚对映体(bi-naphthol)的模拟移动床色谱分离操作条件进行了优化,仿真结果表明了该优化策略的有效性.     

微粒群算法在地震波阻抗反演中的应用

地震波阻抗反演是油藏描述和储层预测中的关键技术,其本质属于多参数的非线性组合优化问题,诸如人工神经网络、模拟退火、遗传算法等非线性反演方法在地震波阻抗反演中已经得到了运用。起源于生物社会学研究和生物行为学模拟的微粒群算法,在多参数、非线性、多极值函数优化问题中具有较强的优越性。通过分析微粒群算法的原理,本文用该非线性算法实现了地震波阻抗反演,并且在理论模型的实验中,证明了算法的可行性。     

变邻域宽度的爬山微粒群优化算法及其应用

微粒群优化算法（particle swarm optimization algorithm,PSO）是由Kennedy和Eberhart 1995年提出的进化计算算法．PSO简单且具有许多良好的优化性能,但对一些复杂优化问题存在容易陷入局部极值的缺陷．本文提出一种变邻域宽度的爬山微粒群优化算法（hill-climbing PSO with variable width neighborhood,vwnHCPSO）,并用5种测试函数进行测试和比较,然后将vwnHCPSO用于催化裂化装置（FCCU）主分馏塔轻柴油闪点软测量．     

微粒群算法参数的理论分析

微粒群算法是近年来兴起的一种智能优化算法,而算法参数是影响算法性能和效率的关键。用基于常系数非齐次差分方程求解的分析、基于动态系统理论的分析和基于离散系统稳定判据的分析三种不同的方式对微粒的位置和速度两个变量进行了深入理论分析,最终得出了一个共同的结论,即保证微粒收敛的参数取值区域约束在一个直角梯形的内部,这将对算法的实际应用起到很重要的作用。     

群体智能优化算法

讨论四种群体智能优化算法--蚁群算法、微粒群算法、人工鱼群算法和混合蛙跳算法,对其算法的原理、发展及应用进行了综述.提出了群体智能优化算法统一框架模式,并对群体智能优化算法进一步发展进行了讨论.     

基于自适应粒子群优化算法的节点定位研究

针对粒子群优化定位算法易陷入局部极值的缺点,提出了一种基于自适应粒子群优化算法的无线传感器节点定位方法。该方法在迭代前期ω取较大值实现快速收敛到最优解附近,后期取较小值求高精度解。在适应度值越大时全局搜索能力越强,加快向全局最优位置的聚集速度;适应度值越小局部搜索能力越强,可得到高精度的解,并通过对全局最优位置进行自适应变异操作,保证算法能跳出当前的搜索区域。仿真结果表明:与常用的极大似然估计对比,该算法具有收敛快、能耗小、精度高和稳定性好的优点,适合应用在无线传感器网络的定位中。     

两群微粒群算法及其在油品调和优化中的应用

针对复杂的非线性约束优化问题,提出了一种含变异算子的两群微粒群算法。算法构造了两个粒子群,分别设置了不同的搜索速度上限,并设计了粒子群间的协调机制和变异算子,使算法的寻优能力得到增强。针对油品调和配方优化进行了实例仿真,研究结果表明所提出的算法能获得较理想的调和配方,在满足调和利润最大的同时能保证对调和指标的卡边,使调和成品油的指标富余量大大降低。     

Diesel Oil Removal by Froth Flotation Under Low Interfacial Tension Conditions II: Continuous Mode of Operation

Abstract

The objective of this study was to investigate the relationship between interfacial tension (IFT) and foam characteristics and the efficiency of diesel oil removal from water in a continuous froth flotation column. The effects of operational parameters, including surfactant concentration, salinity, oil-to-water ratio, foam height, air flow rate, and hydraulic retention time (HRT) on the oil removal were investigated in the continuous mode of a froth flotation operation and compared to batch operation results. Unlike the batch system, for the continuous system used in the present study, having only branched alcohol propoxylate sulfate sodium salt surfactant (C_14–15(PO)₅SO₄Na) and NaCl present in the solution yielded such poor foam characteristics that a stable froth which overflowed the flotation column could not be produced, so the addition of sodium dodecyl sulfate (SDS) as a froth promoter was used to improve the foam stability. Unlike the batch froth flotation system with only C_14–15(PO)₅SO₄Na, the continuous froth flotation with the mixture of C_14–15(PO)₅SO₄Na and SDS, it was not possible to find a SDS and a NaCl concentration at which both ultralow IFT and good foaming were both achieved. Foam formation, stability, and production rate were found to be crucial parameters to the froth flotation efficiency. The continuous froth flotation system offers a high diesel oil removal of 96% in the single stage unit. Demonstration of efficient operation in the continuous mode in this work is important to the practical application of froth flotation in large scale processing.     

Application of Image Processing and Adaptive Neuro-fuzzy System for Estimation of the Metallurgical Parameters of a Flotation Process

It is a well-known fact in the literature and practice that flotation froth features are closely related to process conditions and performance. The authors have already developed some reliable algorithms for measurement of the froth surface visual parameters such as bubble size distribution, froth color, velocity and stability. Furthermore, the metallurgical parameters of a laboratory flotation cell were successfully predicted from the extracted froth features.

In this research study, the fuzzy c-mean clustering technique is utilized to classify the froth images (collected under different process conditions) based on the extracted visual characteristics. The classification of the images is actually necessary to determine the ideal froth structure and the target set-points for a machine vision control system. The results show that the captured froth images are well-classified into five categorizes on the basis of the extracted features. The correlation between the visual properties of froth (in different classes) and the metallurgical parameters is discussed and modeled by the adaptive neuro-fuzzy inference system (ANFIS). The promising results illustrate that the performance of the existing batch flotation system can be satisfactorily estimated from the measured froth characteristics. Therefore, the outputs from the current machine vision system can be inputted to a process control system.     

DTCWT-based zinc fast roughing working condition identification

The surface texture of mineral flotation froth is well acknowledged as an important index of the flotation process. The surface texture feature closely relates to the flotation working conditions and hence can be used as a visual indicator for the zinc fast roughing working condition. A novel working condition identification method based on the dual-tree complex wavelet transform (DTCWT) is proposed for process monitoring of zinc fast roughing. Three-level DTCWT is implemented to decompose the froth image into different directions and resolutions in advance, and then the energy parameter of each sub-image is extracted as the froth texture feature. Then, an improved random forest integrated classification (iRFIC) with 10-fold cross-validation model is introduced as the classifier to identify the roughing working condition, which effectively improves the shortcomings of the single model and overcomes the characteristic redundancy but achieves higher generalization performance. Extensive experiments have verified the effectiveness of the proposed method.     

基于图像空间结构统计分布的浮选泡沫状态识别

通过泡沫图像统计建模,实现了基于图像空间结构感知的浮选泡沫状态自动识别与客观评价。首先,采用Weibull分布建立了泡沫图像各方向边缘响应结构的统计分布模型,有效获取了泡沫图像空间结构的统计分布细节;然后,通过统计学习获得各典型工况状态下的泡沫图像边缘响应统计分布的混合高斯（MoG）模型;最后,通过简单的贝叶斯推理推断出测试泡沫图像对应的工况状态。结果表明：所提出的方法因有效获取了与浮选生产性能直接相关的泡沫空间结构的统计分布特征,可以实时监视泡沫空间结构的变化情况,泡沫生产状态识别准确率高。     

CHARACTERIZATION OF FLOTATION PROCESSES WITH SELF-ORGANIZING NEURAL NETS

Flotation processes are difficult to describe fundamentally, owing to the stochastic nature of the froth structures and the ill-defined chemorheology of the froth. Considerable information on the process is reflected by the structure of the froth. In previous work it has been shown that structural features extracted from flotation froths can be related to the behavior of flotation processes in a qualitative way through the identification of certain behavioral regimes or classes by using a supervised neural net as classifier. Although useful as an aid to control decisions, this method is less suitable for quantitative or dynamic analysis of the behavior of flotation plants. In this paper a new method for the analysis of flotation plants is consequently proposed, based on the use of order preserving maps of features extracted from digitized images of the froth phase. The construction of these maps by means of a self-organizing neural net is demonstrated by way of examples concerning the analysis of industrial copper and platinum flotation plants.     

煤泥浮选泡沫图像纹理特征的提取及泡沫状态的识别

用煤泥浮选泡沫数字图像获取系统获取了51幅煤泥精矿泡沫图像；引入了空间灰度相关矩阵和邻域灰度相关矩阵来提取泡沫的纹理特性，并提取基于这两种算法的一系列特征参数来描述泡沫的结构；分析了各泡沫特征参数随浮选时间(泡沫纹理)的变化关系，定性地指出了各泡沫特征参数与泡沫纹理的相关性；并利用自组织神经网络对煤泥浮选泡沫的状态进行了识别，分类识别的平均正确率达76.5%.     

Froth flotation as primary treatment of flowback water: Removal of total organic carbon and prediction of kinetics

Petroleum and exploration industries employ a hydrofracking process where a large volume of water (fracturing fluid) is injected and a fraction (known as flowback water) is returned to the surface. Froth flotation is a typical process employed for the primary treatment of water. In the present work, froth flotation has been used as a pretreatment method for real flowback water sourced from the petroleum and shale gas exploration industry. In the present work, a first-principle based convective mass transfer model has been developed to describe the froth flotation performance. The resultant equation was solved analytically and compared with the numerical solution, and a parametric sensitivity analysis of the process performance was also undertaken. In addition, a correlation to estimate the flotation rate constant was proposed, thereby circumventing the need to obtain a large number of cumbersome parameters experimentally. Overall, this study proposes froth flotation as an efficient primary treatment method towards the separation of dispersed oil droplets from the flowback water and the corresponding prediction of kinetics using a first-principle based transport model.     

Processing of coal fines in a water-only cyclone

It is reported in the literature that a water-only cyclone (WOC), a centrifugal gravity concentrator, is an alternative to froth flotation to treat coal fines (below 0.5 mm). This unit overcomes the inherent limitations of froth flotation and the dense-medium cyclone techniques as it requires no chemicals or artificial medium. The literature dealing with WOC performance to treat coal fines is also limited and as a result it is not well established how the design variables affect the performance of a WOC while treating coal fines. Therefore, an attempt has been made to develop regression models based on factorial design of experiments to quantify the effects of major design variables of a WOC on the beneficiation characteristics of a typical coal fine sample. Further attempts have been made to provide possible explanations on the observed trends of the data based on simple hydrodynamic analyses.     

Diesel Oil Removal by Froth Flotation Under Low Interfacial Tension Conditions I: Foam Characteristics,and Equilibration Time

Abstract

Froth flotation is a surfactant‐based separation process which is suitable for treating dilute wastewaters. To achieve high performance for the froth flotation operation, the combination of an ultra‐low interfacial tension (IFT) between excess oil and excess water phases, high foam production rates, and high stability of the foam produced, must be attained. To obtain the ultra‐low interfacial tensions, a Winsor Type III or middle phase microemulsion has to be formed. In this study, branched alcohol propoxylate sulfate sodium salt with 14–15 carbon number and 4 PO groups (Alfoterra 145–4PO) was used to form microemulsions with diesel oil. From the results of this work, an increase in surfactant concentration decreased the IFT, and increased foam stability. To obtain the minimum IFT in the region of a Winsor Type III microemulsion, the addition of 5 wt.% NaCl was needed. However, this optimum salinity does not result in effective froth flotation due to poor foam characteristics. The results indicate that both the IFT and the foam characteristics should be optimized to achieve high efficiency of oil removal in a froth flotation operation. Unlike the previously‐studied ethylbenzene system, agitation of the solution before introduction into the flotation column yielded the lowest diesel oil removal efficiency because of the poor foam characteristics compared to either unagitated systems or systems allowed to equilibrate for one month.     

Identification of abnormal conditions for gold flotation process based on multivariate information fusion and double-channel convolutional neural network

Accurate and in-time working condition identification plays a great role in industrial processes. However, most of the current flotation process identification models only use the characteristics of flotation froth as an identification basis, which often causes identification errors due to the instability of the froth, and a large amount of process data is not fully utilized. In this paper, an abnormal condition identification method based on multivariate information fusion and double-channel convolutional neural network (double-channel CNN) is proposed to achieve higher accuracy. First, a double-channel CNN is used to extract depth features from different distributions of froth images and process data in parallel. Then, double normalized attention mechanism (double normalized AM) and multivariate information fusion methods are used to attach weights to the features and fuse them so as to ensure a higher response of key features and increase the reliability of the identification model. The method shows better performance than existing methods in offline simulations and has been validated online at a mineral processing plant in Shandong.