基于IFOA-KELM-MEA模型的游梁式抽油机采油系统井下工况的短期预测

实现对井下工况的预测是及时掌握抽油井生产状态的有效方法,对提高油井生产效率和降低维护成本具有十分重要的意义。采用混沌理论实现抽油井井下工况的短期预测,首先将所提取的示功图的不变曲线矩特征向量作为预测变量,在证明其数据序列具有混沌特性后,由核极限学习机(kernel extreme learning machine,ELM)建立混沌时间序列预测模型,对其中的几个不确定参数采用改进的果蝇优化算法(improved fruit fly optimization algorithm,IFOA)进行优化选取,IFOA采用全局群体多样进化和局部个体随机变异的策略,最后,对模型所预测的结果进行物元分析(matter-element analysis,MEA),诊断其属于哪种故障类型。由某油田作业区的两口生产井进行实例验证,结果表明所提出的IFOA-KELM-MEA预测模型是合理有效的。     

基于自动谱聚类与多极端学习机模型的油井油液含水率软测量

油井油液的含水率是石油生产中的一个重要参数,及时、准确的测量对提高采油生产效率具有重要的意义。针对传统人工测量所存在的不足,引入软测量技术,建立基于自动谱聚类与多极端学习机（automatic spectral clustering-multiple extreme learning machines, ASC-MELM）的软测量模型。提出一种自动谱聚类（spectral clustering, SC）算法,由改进的萤火虫算法（firefly algorithm, FA）对聚类数目和尺度参数进行优化选取,所提出的改进萤火虫算法（improved firefly algorithm, IFA）采用以一定概率跳出当前解的机制,避免传统FA过早陷入局部最优解的不足;对聚类后的不同训练子集,分别由极端学习机（extreme learning machine, ELM）建立子模型,由IFA对其中的隐含层输入权值、隐含层神经元的偏置和隐含层节点个数进行优化选取;最后,将多个子模型的结果取加权平均值输出。由国内某油田作业区一口生产井进行实例验证,结果表明所提出方法具有较高的预测精度,对于实现油井油液含水率的软测量是合理有效的。     

基于电流图变化判断抽油井井下异常工况可行性探讨

采油井的日常维护保养在花土沟油田的生产中占有重要地位,而电流图是抽油井每天录取的资料之一,具有可延续性和及时性的特点,可比性强,因此电流图是观察油井动态变化的重要参数之一。通过对抽油机工作原理的描述,进而简化抽油机四连杆运动,在此基础上通过数学模型的演练,建立电流变化和抽油机运动的简化数学模型,探讨基于电流图变化判断抽油井井下异常工况的可行性,及时推断出油井的工作状况,提高油井机采效率。     

提高机采系统效率在文留油田的研究与应用

油田生产过程中,机采系统效率是对整个抽油井系统工况的标明,其实质是采油各环节的技术水平、管理水平和能源利用效率的具体体现。该指标的高低,直接表明了抽油井采油管理水平的高低,极大程度上影响着抽油井采油的生产成本。通过对抽油机拖动系统、抽油机及参数优化、配套技术不断研究和综合管理,提高抽油机井系统效率,对油田增产降耗、提高经济效益具有重要意义。     

油井示功图监控诊断计量系统应用分析

《油井示功图监控诊断计量系鳓的应用是采油队生产管理一个重要的信息化手段,可实现对抽油井工况的及时掌握,优化措施,细化采油队各节点管理;通过时时准确的单井产液量计量,实现生产管理压力的层层传递。     

应用“抽油机工况管理”改善油田开发效果探讨

胜利油田桩西采油厂有杆泵抽油井占生产井90％以上,抽油井管理水平的高低,直接影响原油生产任务的完成.技术人员采用“抽油井工况管理”这一方法,对油井进行综合分析,提出针对性措施,改善油田整体开发效果.     

螺杆泵抽油优化设计软件的研制及应用

本文介绍了地面驱动螺杆泵抽油井生产系统优化设计软件的抽油杆柱动力学模型,应用有限元计算方法进行模型求解,实现了螺杆泵抽油井的产能分析、静力学工况分析、动力学工况分析、优化设计等功能,并在锦州采油厂进行了实际应用,分析设计结果与实际工况吻合较好。     

对产量、效率排序确定抽油井工作参数的方法

在抽油井生产管理中,以计算多种泵径、冲程、冲次组合下不同下泵深度的产液量、井下效率为基础,对不同泵径、冲程、冲次组合所对应的最大产液量、最大井下效率进行排序,以获得合理产液量和比较高的井下效率为目标,根据指标排序情况确定出合理的抽油井工作参数。这一方法在实际应用中获得了满意的结果,值得推广。     

抽油机井系统效率评价技术研究

石油工业的蓬勃发展以及各行各业能源的需求,对于石油企业抽油井系统的效率要求越来越高。抽油井系统效率预测与评价系统包括许多方面,如测试模块、文件模块、统计报表模块等,其不仅能够对于抽油井系统的效率进行全面的测试与检查,方便管理人员了解及准确掌握各个分系统的工作情况、效率及管理状态,还能够对其隐形的、尚未开发的效率进行定位及测评,同时管理人员可以根据该特性,设定一套适用地质各种条件、不同油层物质、不同层次供液能力等各种条件的效率管理评价指标,从而建立完整有效的效率管理系统[1]。能够以此为根据,有效的提高抽油井的生产效率、控制生产成本及保持抽油井的生产效率的稳定性。     

井下压力计在电潜泵井工况分析中的应用

电潜泵采油是海上油田最主要的人工举升方式,正确进行电潜泵井工况分析和故障诊断对保障油井生产时效、处理油井故障具有重要意义。依靠电流卡片、油井油套压、电机对地绝缘和相间电阻等资料对潜油电泵井进行定性分析的方法对人员的经验具有很强的依赖性,从而影响结论的准确性和全面性。利用直读式井下电子压力计测量的参数辅助进行生产井工况分析,可以直观定量的判断油井生产情况,查找故障原因,对指导电潜泵井的使用和管理具有重要的现实意义。     

Re-refining of used lubricant oil by solvent extraction using central composite design method

The primary aim of this study was to recover base oil from used oil using solvent extraction followed by the adsorption method. Many effective variables were examined within the solvent extraction method, including using different solvents, solvent/used oil, temperature and speed of blending. Central composite design (CCD) was applied as the statistical method. Response surface methodology was then used to find the optimum conditions in the process of extraction: ratio of solvent/used oil 2.4 and 3.12 vol/vol, temperature=54 and 18 °C, and speed of mixing=569 and 739 rpm for 1-butanol and methyl ethyl ketone (MEK), respectively. Various flocculation agents were used with the solvent, such as Sodium hydroxide (NaOH), Potassium hydroxide (KOH) and Monoethylamine (MEA); they provided an increase in the separation efficiency. The best result was obtained when using 2 grams of MEA/kg solvent; this amount of MEA increases sludge removal from 12.6% to 14.7%. In the process of clay adsorption, the variables that were tested included the ratio of clay/extract oil, temperature and time of contact. The best conditions in the process of adsorption by activated bentonite were a ratio of clay/extract oil=15 wt/vol%, temperature=120 °C, and time of contact=150 minutes. The recovered base oil was analyzed by Fourier transform infrared spectroscopy (FTIR) and compared to Iraqi specifications of base oils. The recovered base oil specifications were analyzed, including, viscosity @100 °C 8.32, 9.22 cSt, pour point ?17.35, ?22.23 °C, flash point 210.12, 223.04 °C, total acid number (TAN) 0.25, nill, total base number (TBN) nill, nill, ash 0.031, 0.0019 wt% and color 3.0, 2.5 for two types of base oil recovered using MEK, 1-butanol with activated bentonite, respectively.     

基于高斯混合模型与主元分析的多模型切换方法

针对多模型预测控制的模型切换问题,提出了一种基于工况判断的多模型切换方法,利用工业过程中的可测变量综合反映系统的动态特性,根据动态特性的变化进行多模型切换。首先利用高斯混合模型(GMM)将历史数据划分为若干个工况,然后利用不同工况下的历史数据建立负荷向量矩阵和预测模型,最后根据主元模型的平方预报误差(SPE)选择预测模型。以乙烯裂解炉的反应管出口温度(COT)的控制为例进行仿真,仿真结果表明:提出的方法实现了多个反应管出口温度的稳定均衡控制,当系统的工况发生改变时,通过不同主元模型的SPE统计量的比较,可以很容易地找到匹配的工况,并切换为相应的预测模型,解决了当系统动态特性发生改变时,预测模型切换滞后的问题。     

A Predictive Classification Model for the Management of Virgin Olive Oil Filtration at Industrial Scale

Precoat-bodyfeed filtration of virgin olive oil was investigated on an industrial filter-press plant. Several chemical parameters of the unfiltered oil were measured and the relationship with filtration performances was investigated by Principal Component Analysis. Further, Linear Discriminant Analysis was applied to develop a predictive model for oil filterability. Principal Component Analysis allowed the construction of latent variables which were used to separate oil groups and to select variables for Linear Discriminant Analysis. The developed linear model gave an overall correct recognition of about 88%, good enough for a convenient filterability prediction of oil at industrial scale.     

响应曲面法优化Fenton氧化处理稠油污染土壤的条件研究

采用响应曲面法优化Fenton氧化处理稠油污染土壤的条件。以反应体系的初始pH值、H20z用量、Fe2＋用量为影响因素，以石油烃降解率为响应值，采用Box-Behnken（BB）设计法考察了3个因素对Fenton氧化处理稠油污染土壤效果的影响。以二次多项式回归方程预测模型为基础，得出在初始pH值为5．0、6．0及7．0的条件下，石油烃的降解率分别达到35．12％、35．57％和33．43％，实验所得优化条件的实测值与预测值十分接近，标准方差小于1．3％，说明预测模型准确可靠。     

Preparing a catalyst layer in magnetic field to improve the performance of proton exchange membrane fuel cells

Electro catalyst Pt–Co/multi-walled C nanotubes were synthesized by using the modified polyol method with glycol as reducer. The magnetic-field-assisted fabrication of membrane electrode assemblies (MEAs) for proton exchange membrane fuel cells (PEMFCs) was proposed, to orient catalyst layers and increase the efficiency of catalyst utilization. PEMFCs with the magnetic-field-treated MEA (M-MEA) exhibited significant performance improvement over common MEA (C-MEA) without magnetic-field treatment. Under the same operating conditions, the maximum power density of MEA increased from 149.6 to 223.8 mW cm^?2 when C-MEA was replaced by M-MEA. Scanning electron microscope images showed that catalysts exhibited a “cluster-like structure” in M-MEA opposed to a chaotic arrangement in C-MEA. Electrochemical impedance spectroscopy measurements revealed that M-MEA reaction resistance was lower than that of C-MEA. Cyclic voltammetry data showed an increment of almost 29.6 % in electrochemical surface area as a result of the magnetic-field treatment.     

Chaotic forecasting of time series of heat-transfer coefficient for an evaporator with a two-phase flow

The time series of heat-transfer film coefficient for an evaporator with a vapor-liquid two-phase boiling flow were forecasted by the chaotic prediction or the local nonlinear short-term prediction. The forecasting method was based on the phase space reconstruction theory and a mathematic model in the form of datum driving had been developed to carry out the prediction. The signals of heat-transfer film coefficients predicted by this method were compared with those obtained from the experimental measurements. Different from the previous work, the comparisons in this work were done both from the point of view of the time trajectory and from the point of view of macroscopic or general characters. Besides the time-averaged statistics characteristics, such as the average value, average deviation and standard deviation, the power spectrum, phase plane map and chaotic invariants including the correlation dimension and Kolmogorov entropy had been calculated and been compared for the time series obtained both from the experimental measurement and from the forecast. The comparison for each parameter between the value calculated from the time series measured and that estimated from the time series forecasted by the chaotic forecast method shows satisfactory agreement. The limited length in time with an accuracy prediction indicates that the system is chaotic. The agreement of the comparison of the general parameter indicates that the chaotic prediction is effective for the estimations of the heat transfer characteristics of the two-phase flow boiling system and that the chaotic prediction method may be a potential tool for the effective thermo-fluid control for such evaporators.     

用于催化裂化装置产率预测的Filter-Wrapper特征变量选择方法

催化裂化过程是重质油轻质化的重要手段,为了研究操作条件、原料性质等因素对产品分布的影响,通常需要对催化裂化过程建立准确可靠的数学模型。选择合适的输入变量对模型预测效果有着较大的影响,而在现有的催化裂化装置模型中,输入变量的选取主要依赖于对催化裂化机理的理解。本文从数据驱动建模的角度出发,提出一种Filter法与Wrapper法联合使用的特征子集选择方法。该方法在输入变量选取的过程中不依赖于催化裂化的先验知识,是一种数据驱动的自发的特征变量选择过程。以某炼油厂催化裂化装置为研究对象,利用该装置的生产数据分别选择用于干气和焦炭产率预测模型的输入变量,建立了预测精度高、输入变量数目适中的模型。此外,该方法为催化裂化装置建模的变量选取提供了新角度。     

加热炉炉温的模糊广义预测控制仿真研究

针对加热炉炉温的大惯性、大滞后及非线性等特点,提出一种基于T-S模糊模型的模糊广义预测控制策略。T-S模糊模型的前件和后件参数分别采用粒子群优化的模糊C-均值算法和递推最小二乘法辨识,根据输入变量更新模型隶属度并将T-S模糊模型等价转换为线性模型,以此作为预测模型应用于广义预测控制。仿真结果表明:该方法在不同工况下均具有较短的调节时间,在扰动作用下有很强的鲁棒性。     

SOFT COMPUTING METHOD FOR PREDICTION OF CO2 CORROSION IN FLOW LINES BASED ON NEURAL NETWORK APPROACH

An important aspect of corrosion prediction for oil/gas wells and pipelines is to obtain a realistic estimate of the corrosion rate. Corrosion rate prediction involves developing a predictive model that utilizes commonly available operational parameters, existing lab/field data, and theoretical models to obtain realistic assessments of corrosion rates. This study presents a new model to predict corrosion rates by using artificial neural network (ANN) systems. The values of pH, velocity, temperature, and partial pressure of the CO₂ are input variables of the network and the rate of corrosion has been set as the network output. Among the 718 data sets, 503 of the data were implemented to find the best ANN structure, and 108 of the data that were not used in the development of the model were used to examine the reliability of this method. Statistical error analysis was used to evaluate the performance and the accuracy of the ANN system for predicting the rate of corrosion. It is shown that the predictions of this method are in acceptable agreement with experimental data, indicating the capability of the ANN for prediction of CO₂ corrosion rate in production flow lines.     

Applicability domains and accuracy of prediction of soft sensor models

Soft sensors are used to estimate process variables that are difficult to measure online. However, the predictive accuracy gradually decreases with changes in the state of chemical plants. Regression models can be updated, but if the model is updated with abnormal data, the predictive ability deteriorates. In practice, when the prediction error of an objective variable exceeds a threshold, an abnormal situation is detected. However, no effective method exists to decide this threshold. We have proposed a method to estimate the relationships between applicability domains and the accuracy of prediction of soft sensor models quantitatively. The larger the distances to models (DMs), the lower the estimated accuracy of prediction. Hence, the model between DMs and accuracy can separate variations in process variables and y‐analyzer fault. This method was applied to real industrial data. The fault detection ability of the proposed method was better than that of the traditional one. © 2010 American Institute of Chemical Engineers AIChE J, 2011