Evolving an accurate model based on machine learning approach for prediction of dew-point pressure in gas condensate reservoirs

Over the years, accurate prediction of dew-point pressure of gas condensate has been a vital importance in reservoir evaluation. Although various scientists and researchers have proposed correlations for this purpose since 1942, but most of these models fail to provide the desired accuracy in prediction of dew-point pressure. Therefore, further improvement is still needed. The objective of this study is to present an improved artificial neural network (ANN) method to predict dew-point pressures in gas condensate reservoirs. The model was developed and tested using a total set of 562 experimental data point from different gas condensate fluids covering a wide range of variables. After a series of optimization processes by monitoring the networks performance, the best network structure was selected. This study also presents a detailed comparison between the results predicted by this ANN model and those of other universal empirical correlations for estimation dew-point pressure. The results showed that the developed model outperforms all the existing methods and provides predictions in acceptable agreement with experimental data. Also it is shown that the improved ANN model is capable of simulating the actual physical trend of the dew-point pressure versus temperature between the cricondenbar and cricondenterm on the phase envelope. Finally, an outlier diagnosis was performed on the whole data set to detect the erroneous measurements from experimental data.     

集气管压力无级模糊控制系统

焦炉集气管压力是焦炉生产的重要工艺参数,以前存在集气管压力大幅波动、焦炉冒黄烟等现象。通过采用无级模糊控制技术有效解决了集气管的压力稳定问题,控制精度可达设定值±20Pa。     

高压含气原油凝点测量新方法

针对常规测量方法只能测量常压下脱气原油凝点的不足,提出了落球法测量高压含气原油凝点的新方法,分析了测量的原理,设计了一套高压含气原油凝点测量装置,比较了该装置和常规方法对常压脱气原油的测量结果,测量了不同溶解气油比、不同压力下含气原油的凝点,分析了压力和溶解气对原油凝点的影响。结果表明,对于常压脱气原油,该装置的测量结果与常规方法吻合。压力对原油的凝点有一定的影响,随压力的升高凝点上升,在本试验条件下,压力每上升2MPa,凝点上升0.7℃;溶解气油比对原油的凝点有影响,含气原油的凝点低于脱气原油的凝点,对所研究的油样来说,溶解气油比每增加10m³·m^-3,凝点大约降低1℃。本研究结果可为高含蜡油田的生产管理和油气集输设计提供依据。     

Effect of coke shrinkage and plastic layer deformation on gas pressure in a coke oven

A plausible mechanism is provided for the generation of internal gas pressure (IGP) in the plastic layer of a coke charge, by relating the formation of low permeability at the outer edges of the plastic layer to the deformability of the plastic material. The amount of lateral shrinkage post-resolidification is shown to be important in constraining the deformation of the plastic material, and hence the generation of low permeability to gas flow. The mechanism provides a natural explanation for permeability variation in the plastic layer, as well as for the linkage between IGP/coking pressure generation and pushing difficulties.     

Development of a pilot-scale acid gas removal system for coal syngas

The Korean pilot-scale gasification facility consists of a coal gasifier, hot gas filtering system, and acid gas removal (AGR) system. The syngas stream from the coal gasification at the rate of 100–120 Nm³/hr included pollutants such as fly ash, H₂S, COS, etc. The acid gas, such as H₂S and COS, is removed in the AGR system before generating electricity by gas engine and producing chemicals like Di-methyl Ether (DME) in the catalytic reactor. A hydrolysis system was installed to hydrolyze COS into H₂S. The designed operation temperature and pressure of the COS hydrolysis system are 150 °C and 8 kg/cm². After the hydrolysis system, COS was reduced below 1 ppm at the normal operating condition. The normal designed operation temperature and pressure of the AGR system are below 40 °C and 8 kg/cm². Fe-chelate was used as an absorbent. H₂S was removed below 0.5 ppm in the AGR system when the maximum concentration of H₂S was 900 ppm. A small scale COS adsorber was also installed and tested to remove COS below 0.5 ppm. COS was removed below 0.1 ppm after the COS adsorbents such as the activated carbon and ion exchange resin. This work was presented at the 6 ^th Korea-China Workshop on Clean Energy Technology held at Busan, Korea, July 4–7, 2006.     

Estimation of phase velocities at flooding point in packed columns for any gas/liquid system

A method for the estimation of gas velocity at the flooding point of various packings, based on such fundamental parameters as the porosity and specific surface of the packing itself, has been developed and introduced herein. The proposed relationships can be applied to the computation of rectification and absorption towers alike, within a wide range of physical/chemical properties of the gas and liquid phases. The method presented in this paper should be particularly useful for estimating the flooding-point gas velocity of new packing elements, mostly, because it needs no empirical constants to be determined.     

一种生化反应器智能自适应学习和预测神经元网络系统

The adaptive learning and prediction of a highly nonlinear and time-varying bioreactor benchmark process is studied using Neur-On-Line, a graphical tool kit for developing and deploying neural networks in the G2 real time intelligent environment,and a new modified Broyden, Fletcher, Goldfarb, and Shanno （BFGS） quasi-Newton algorithm. The modified BFGS algorithm for the adaptive learning of back propagation （BP） neural networks is developed and embedded into NeurOn-Line by introducing a new search method of learning rate to the full memory BFGS algorithm. Simulation results show that the adaptive learning and prediction neural network system can quicklv track the time-varving and nonlinear behavior of the bioreactor.     

A prediction model for the critical gas velocity and pressure gradient of continuous liquid-carrying in inclined pipes

A generic model for the prediction of critical gas velocity and pressure gradient in slightly inclined pipes (β ≤ 6°) is presented in this article. The gas–liquid configuration was determined based on the minimum energy principle and consideration of wettability and surface tension. A visualization experiment was conducted to obtain the critical gas velocity and the critical pressure gradient of a gas–liquid flow through the 40 and 60-mm pipe diameter. The theoretical study shows that the configuration is close to a convex interface shape at the critical conditions, which is in accord with the experimental phenomenon. Experimental study shows interfacial waves are the main cause of increased interfacial friction factor and a linear functional relationship between the inclination angle and the flow correction factor f(β). The results demonstrate that the new model is capable of providing satisfactory prediction results for the critical gas velocity, pressure drop, and liquid holdup.     

Development of a data‐driven fuzzy screening model for enhanced oil recovery methods using an adaptive weighting system

Screening of enhanced oil recovery (EOR) methods is a main prerequisite for EOR planning and design. In this study, an integrated data‐driven screening model (DDSM) is developed to improve EOR screening using the combined capabilities of the fuzzy expert approach (FEA) and support vector regression (SVR) techniques. In this study, EOR field data from the past 40 years were reviewed to generate an updated and reliable EOR criteria table as a basis to construct a fuzzy screening model. The DDSM was evaluated to determine the quantitative screening and ranking of EOR methods using seven field datasets, including the fast forecasting of the nominated EOR methods. In order to improve screening performance, a fuzzy model was integrated using 4 SVR models to predict the adaptive weights of the screening parameter for decision making. The SVR models can predict the recovery factor (RF) of EOR methods including gas, chemical, steam, and combustion to calculate the adaptive effective weight of the screening parameters. The SVR models were trained with datasets generated from simulations of the EOR process. The absolute average error (AAE) of the SVR models from the simulation varied within the range of 0.078–0.095 for the prediction of the RF. The DDSM results were compatible to the data published in other literature. In addition, the developed model can provide comparable results to common screening software. The results showed improvements due to the adaptive weighting system on the EOR methods’ screening for the studied reservoirs relative to the fuzzy engine with constant weights. The presented integrated model can guide the screening process to select the efficient EOR method in practical applications.     

Comparison of ML–EM algorithm and ART for reconstruction of gas hold-up profile in a bubble column

Algebraic reconstruction technique (ART) and maximum likelihood–expectation maximization (ML–EM) algorithm have been applied for image reconstruction using gamma-ray tomography. This methodology can be of immense help in establishing the hydrodynamics of several multiphase systems such as two-phase and three-phase bubble column reactors. The effect of various image processing parameters such as initial guess, grid size, stopping criteria and gamma-ray measurement parameters like beam configuration, number of projection, number of views on the quality of reconstructed image has been studied in present work. It has been observed that ML–EM algorithm shows more precise and faster results as compared to ART and it serve as a preferential tool in image reconstruction. These techniques were then used in the estimation of gas hold-up profile in a two-phase aqueous system. Average gas hold-up values in bubble column based on reconstructed local hold-up values based on the above two techniques were found to be in good agreement with the experimental data within ±10% accuracy, however, ML–EM algorithm may be preferred due to better capability of incorporating the modalities of data collection.     

Development of an artificial neural network model for prediction of cell voltage and current efficiency in a chlor-alkali membrane cell

This paper presents the development of an artificial neural network (ANN) model for the prediction of cell voltage and caustic current efficiency (CCE) versus various operating parameters in a lab scale chlor-alkali membrane cell. In order to validate the model predictions, the effects of various operating parameters on the cell voltage and current efficiency of the membrane cell were experimentally studied. The membrane cell incorporated a standard DSA/Cl₂ electrode as the anode, a nickel electrode as the cathode and a Flemion 892 polymer film as the membrane. Each of the six process parameters including anolyte pH (2–5), operating temperature (25–90 °C), electrolyte velocity (2.2–5.9 cm/s), brine concentration (200–300 g/L), current density (1–4 kA/m²), and run time were thoroughly studied at four levels and low caustic concentrations (5–22 g/L). The predictions of ANN model as well as those from other statistical methods were evaluated versus the measured values of cell voltages.

The developed ANN model is not only capable to predict the cell voltage and caustic current efficiency but also to reflect the impacts of process parameters on the same functions. The predicted cell voltages and current efficiencies using ANN modeling were found to be close to the measured values, particularly at higher current densities.     

Development of a poloxamer analogs/bioadhesive polymers‐based in situ gelling ophthalmic delivery system for tiopronin

The purpose of this study was to develop a poloxamer analogs/bioadhesive polymers‐based in situ gelling ophthalmic delivery system aiming at enhancing bioavailability and anticataract effect. The effect of poloxamer 407 (P407), poloxamer 188 (P188), carbopol 934P (C934), and sodium hyaluronate (NaHA) concentration on the gelation temperature (GT) was examined. The GT of P407 based in situ gel increased with an increase in the P188 concentration. NaHA and C934 lowered the GT of poloxamer analogs based in situ gel. Correlation analysis demonstrated that in vitro drug release from in situ gel was controlled by gel dissolution and followed zero‐order kinetics. Tiopronin in vitro transcorneal transit accorded with zero‐order kinetics. Twenty‐two percent P407 and 6% P188 containing 0.2% NaHA based formulation can be chosen as in situ gel matrix of tiopronin because of proper GT and sustained releasing ability. In vivo study showed that the area under the aqueous humor–concentration time curve of tiopronin increased by 1.6 folds for in situ gel, compared with tiopronin aqueous solution. High‐dose tiopronin in situ gel and solution delayed the development of selenite cataract 6 d and 4 d, respectively. The results showed that tiopronin in situ gel exhibits higher bioavailability and therapeutical effect. © 2009 Wiley Periodicals, Inc. J Appl Polym Sci, 2009     

Taguchi's parametric design approach for the selection of optimization variables in a refrigerated gas plant

This paper presents a systematic procedure for selecting optimization variables in a refrigerated gas plant (RGP) using the Taguchi method. A dynamic RGP model developed under HYSYS environment is utilized as a test bed. Nine variables with three levels each are employed for optimizing RGP profit. These optimization variables are selected due to their roles as manipulated variables in controlling the process. Results are validated by comparing optimum values of RGP profit obtained through a set of HYSYS experiments and those from analyses of means.     

Development of a galvanic sensor system for detecting the corrosion damage of the steel embedded in concrete structures: Part 1. Laboratory tests to correlate galvanic current with actual damage

The correlation between sensor output and corrosion rate of reinforcing steel was evaluated by laboratory electrochemical tests in saturated Ca(OH)₂ with 3.5 wt.% NaCl. In this paper, two types of electrochemical probes were developed: galvanic cells containing of steel/copper and steel/stainless steel couples. The corrosion behavior in saturated Ca(OH)₂ solution with and without 3.5 wt.% NaCl addition for the different electrodes was investigated by potentiodynamic test. Weight loss measurement and galvanic corrosion test were conducted to obtain the corrosion rate of reinforcing steel and the charge of sensor in saturated Ca(OH)₂ solution with 3.5 wt.% NaCl addition, respectively.

The results of the potentiodynamic test indicated the possibility of detecting an ingress point of chlorides by measuring the galvanic current. In galvanic corrosion tests, the galvanic current of steel/copper couple was higher than that of steel/stainless steel couple, i.e., the steel/copper sensor is more suitable for high resistance environment. The steel/stainless sensor showed a better linear correlation than the steel/copper sensor. Through the relationship between the sensor system output and the weight loss (mg/cm²) of steel, real corrosion damage of the steel embedded in concrete can be detected.     

Development of a galvanic sensor system for detecting the corrosion damage of the steel embedded in concrete structure: Part 2. Laboratory electrochemical testing of sensors in concrete

The correlation between sensor output and the corrosion rate of steel bar was confirmed in concrete environment. Open-circuit potential, linear polarization resistance (LPR) measurement and electrochemical impedance spectroscopy (EIS) were used to evaluate the corrosion behavior of steel bar embedded in concrete. Also, galvanic current measurements of designed sensors were conducted to obtain the charge of sensor embedded in concrete.In this study, the results of corrosion behavior of reinforcing steel showed a consistence among the data obtained by open-circuit potential monitoring, LPR and EIS measurements. Steel/copper sensor showed a good correlation in concrete environment between sensor output and corrosion rate of steel bar. However, there was no relationship between steel/stainless steel sensor output and corrosion rate of steel bar due to the low galvanic current output. Through the relationship between the steel/copper sensor output and the corrosion rate of reinforcing steel, the real corrosion damage of the reinforcing steel can be detected. Consequently, this confirms that the galvanic sensor system is a good method for detection of corrosion in reinforced concrete.     

Experimental analysis of alternative fuel impact on a new “torque-controlled” light-duty diesel engine for passenger cars

The present paper describes the results of an experimental study performed burning alternative fuels, different per quality and feedstock, in a modern diesel engine compliance the Euro 5 emission standards. Three alternative fuels were tested on the engine and compared with a reference fossil fuel in terms of combustion characteristics, fuel consumption, noise and emissions. The alternative fuels were two biodiesels (RME and SME) and a Fischer-Tropsh (GTL), while the reference fuel was an EU certification diesel fuel. The engine employed in the study was a light-duty diesel engine developed for passenger car and light truck application, and equipped with the new generation ECU able to drive the engine under “torque-controlled” mode by means of instrumented glow-plugs with pressure sensor. The experiments were carried out in a fully instrumented test bench fuelling the engine with the various fuels. The tests were done in a wide range of engine operation points for the complete characterization of the biodiesels performance in the NEDC cycle. Moreover, the trade-off NO_x-PM by EGR sweep in the three most critical test points for the engine emission performance was carried out for all fuels. The test methodology was selected carefully in order to evaluate the interaction between the fuel quality and the engine management strategy. The results put in evidence a strong interaction between the alternative fuel quality and the engine control mode highlighting the great benefits reachable by exploiting simultaneously the alternative fuel quality and the flexibility of the new engine management strategies.