Adaptive neuro-fuzzy algorithm applied to predict and control multi-phase flow rates through wellhead chokes

A Takagi-Sugeno adaptive neuro-fuzzy inference system (TSFIS) model is developed and applied to a dataset of wellhead flow-test data for the Resalat oil field located offshore southern Iran, the objective is to assist in the prediction and control of multi-phase flow rates of oil and gas through the wellhead chokes. For this purpose, 182 test data points (Appendix 1) related to the Resalat field are evaluated. In order to predict production flow rate (Q_L) expressed as stock-tank barrels per day (STB/D), this dataset includes four selected input variables: upstream pressure (Pwh); wellhead choke sizes (D64); gas to liquid ratio (GLR); and, base solids and water including some water-soluble oil emulsion (BS&W). The test data points evaluated include a wide range of oil flow rate conditions and values for the four input variables recorded. The TSFIS algorithm applied involves five data processing steps: a) pre-processing, b) fuzzification, c) rules base and adaptive neuro-fuzzy inference engine, d) defuzzification, and e) post-processing of the fuzzy model. The developed TSFIS model for the Resalat oil field database predicted oil flow rate to a high degree of accuracy (root mean square error = 247 STB/D, correlation coefficient = 0.9987), which improves substantially on the commonly used empirical algorithms used for such predictions. TSFIS can potentially be applied in wellhead choke fuzzy controllers to stabilize flow in specific wells based on real-time input data records.     

Multiple Empirical Kernel Learning with Majority Projection for imbalanced problems

Traditional Multiple Empirical Kernel Learning (MEKL) expands the expressions of the sample and brings better classification ability by using different empirical kernels to map the original data space into multiple kernel spaces. To make MEKL suit for the imbalanced problems, this paper introduces a weight matrix and a regularization term into MEKL. The weight matrix assigns high misclassification cost to the minority samples to balanced misclassification cost between minority and majority class. The regularization term named Majority Projection (MP) is used to make the classification hyperplane fit the distribution shape of majority samples and enlarge the between-class distance of minority and majority class. The contributions of this work are: (i) assigning high cost to minority samples to deal with imbalanced problems, (ii) introducing a new regularization term to concern the property of data distribution, (iii) and modifying the original PAC-Bayes bound to test the error upper bound of MEKL-MP. Through analyzing the experimental results, the proposed MEKL-MP is well suited to the imbalanced problems and has lower generalization risk in accordance with the value of PAC-Bayes bound.     

Predicting hydrogen adsorption and desorption rates in cylindrical metal hydride beds: Empirical correlations and machine learning

A major limitation in chemisorptive hydrogen storage in metal hydrides is the long time required for the adsorption reaction during charging. This study investigates how the shape and material of the reaction chamber influences the adsorption and desorption rates. Numerical simulations of hydrogen storage in a cylindrical reaction chamber filled with LaNi₅ hydride are conducted for a range of chamber thermal conductivities and aspect ratios. The results show that adsorption and desorption processes are limited by thermal diffusion in the hydride bed and storage chamber. A storage efficiency is proposed based on an ideal isothermal process and used to evaluate the impact of chamber thermal conductivity and aspect ratio on the adsorption and desorption rates. Empirical correlations are proposed for predicting the adsorption and desorption efficiency of cylindrical LaNi₅ hydride beds. Finally, a machine-learning based data model for predicting storage efficiency in metal hydride chambers is presented. Comparison against the empirical correlations highlights that the machine learning-based data model can predict the storage efficiency more accurately.     

Prediction of sediment,particulate nutrient and dissolved nutrient concentrations in a dry tropical river to provide input to a mechanistic coastal water quality model

A Generalised Additive Modelling (GAM) approach is applied to prediction of both particulate and dissolved nutrient concentrations in a wet-tropical river (the Fitzroy River, Queensland, Australia). In addition to covariant terms considered in previous work (i.e. flow, discounted flow and a rising-falling limb term), we considered several new potential covariates: meteorological and hydrological variables that are routinely monitored, available in near-real time, and were considered to have potential predictive power. Of the additional terms considered, only flows from three tributaries of the Fitzroy River (namely, the Nogoa, Comet and Isaac Rivers) were found to significantly improve the model. Inclusion of one or more of these additional flow terms greatly improved results for dissolved nitrogen and dissolved phosphorus concentrations, which were not otherwise amenable to prediction. In particular, the Nogoa sub-catchment, dominated by pasture for cattle, was found to be important in determining dissolved inorganic nitrogen and phosphorus concentrations reaching the river mouth. This insight may direct further research, including future refinement of processed-based catchment models. The GAMs described here are used to provide near real-time river boundary conditions for a complex coupled hydrodynamic and biogeochemical model of the Great Barrier Reef Lagoon, and can be coupled with a forecasting hydrological model to allow integrated forecasting simulations of the catchment to coast system.     

On the probability distribution of faults in complex software systems

ContextThere are several empirical principles related to the distribution of faults in a software system (e.g. the Pareto principle) widely applied in practice and thoroughly studied in the software engineering research providing evidence in their favor. However, the knowledge of the underlying probability distribution of faults, that would enable a systematic approach and refinement of these principles, is still quite limited.ObjectiveIn this paper we study the probability distribution of faults detected during verification in four consecutive releases of a large-scale complex software system for the telecommunication exchanges. This is the first such study analyzing closed software system, replicating two previous studies for open source software.MethodWe take into consideration the Weibull, lognormal, double Pareto, Pareto, and Yule–Simon probability distributions, and investigate how well these distributions fit our empirical fault data using the non-linear regression.ResultsThe results indicate that the double Pareto distribution is the most likely choice for the underlying probability distribution. This is not consistent with the previous studies on open source software.ConclusionThe study shows that understanding the probability distribution of faults in complex software systems is more complicated than previously thought. Comparison with previous studies shows that the fault distribution strongly depends on the environment, and only further replications would make it possible to build up a general theory for a given context.     

A new method of predicting the saturation pressure of oil reservoir and its application

Saturation pressure is a vital parameter of oil reservoir which can reflect the oilfield characteristics and determine the oilfield development process, and it is determined by experiments in the laboratory in general. However, there was only one well with saturation pressure test in this target reservoir, and it is necessary to determine whether this parameter is right or not.In this work, we present a new method for quickly determining saturation pressure using machine learning algorithms, including random forest regressor (RF), support vector machine (SVM), decision trees (DT), and artificial neural network (ANN or NN). Using these approaches, saturation pressure was obtained by using the initial solution gas-oil ratio (GOR), temperature, API gravity and other reservoir-fluid data available in the oilfields. Compared with the empirical formula for saturation pressure calculation, the calculated result shows that the accuracy given from machine learning is higher than that from other formulas at home and abroad, and has a good match with the lab test. On the basis of the calculated saturation pressure, it can determine whether the reservoir enters into the stage of dissolved gas drive or not, which also provides the basis for maintaining the reservoir pressure by water injection in advance, rational development decision-making and work over measures.This approach above can provide technical guidance for predicting the saturation pressure in the development of different kinds of reservoirs, including the sandstone reservoirs and carbonate reservoirs.     

Model of reactive precipitation process of preparation nano-ZnO in rotating packed bed: Contribution ratio

With excellent micromixing characteristic of rotating packed bed (RPB), many nanoparticles with small average size, narrower distribution and good morphology had been successfully and continuously prepared. To reveal complex crystal process, an empirical model were developed to simulate nano-ZnO by considering mass changed, population balance equation, growth rate G, nucleation rate B, drop sizes D_i, and resident time t. The predicted particle sizes were shown good agreement with experimental data with error of ±10%. Therefore, it was further adopted to predict the effects of rotating speed, liquid flow rate and reactant concentration on the mean particle size. To look more deeply insight in this process, their contribution ratios were further analyzed. The proposed empirical models were of great helpful to obtain suitable operation conditions for preparing much better properties of nanoparticles with fewer experiments. It was also beneficial to produce other nanoparticles in RPB.     

Role of mass transfer in overall substrate removal rate in a sequential aerobic sludge blanket reactor treating a non-inhibitory substrate

A laboratory study was undertaken to explore the role of mass transfer in overall substrate removal rate and the subsequent kinetic behavior in a glucose-fed sequential aerobic sludge blanket (SASB) reactor. At the organic loading rates (OLRs) of 2-8 kg chemical oxygen demand (COD)/m³-d, the SASB reactor removed over 98% of COD from wastewater. With an increase in OLR, the average granule diameter (d_p = 1.1-1.9 mm) and the specific oxygen utilization rate increased; whereas biomass density of granules and solids retention time decreased (13-32 d). The intrinsic and apparent kinetic parameters were evaluated using break-up and intact granules, respectively. The calculated COD removal efficiencies using the kinetic model (incorporating intrinsic kinetics) and empirical model (incorporating apparent kinetics) agreed well with the experimental results, implying that both models can properly describe the overall substrate removal rate in the SASB reactor. By applying the validated kinetic model, the calculated mass transfer parameter values and the simulated substrate concentration profiles in the granule showed that the overall substrate removal rate is intra-granular diffusion controlled. By varying different d_p within a range of 0.1-3.5 mm, the simulated COD removal efficiencies disclosed that the optimal granular size could be no greater than 2.5 mm.     

EMD-based random decrement technique for modal parameter identification of an existing railway bridge

Vibrational measurement data are often nonstationary and modal parameter identification based on these data is of practical value for structural health monitoring and condition assessment. The empirical mode decomposition (EMD) is a most recent tool for analysis of nonstationary signals. An EMD-based random decrement (RD) technique is presented to identify modal parameters from monitoring vibrational data. The nonstationary measurement data are first decomposed into a series of quasi-stationary intrinsic mode functions (IMFs) by EMD. The RD technique is then applied to the selected IMFs to obtain the free-decay response. The modal frequencies and damping ratios are finally identified from the free-decay response by minimizing the error between the measured free-decay responses and the predicted responses from a parametric model. The present method is applied to extract the modal parameters of the Nanjing Yangtze River Bridge from the measured responses. The identification result is compared to those from finite element analysis as well as from the experimental result identified with the peak-picking (PP) method. In addition, the modal frequencies of the bridge loaded with heavy trains are also identified and compared to the ‘empty’ bridge. The EMD-based random decrement (RD) technique provides an effective and promising tool for modal parameter identification for large bridges and other structures.