Committee Machine Research Articles

Jointly stochastic fully symmetric interpolatory rules and local approximation for scalable Gaussian process regression

When exploring the broad application prospects of large-scale Gaussian process regression (GPR), three core challenges significantly constrain its full effectiveness: firstly, the O(n3) time complexity of computing the inverse covariance matrix of n training points becomes an insurmountable performance bottleneck when processing large-scale datasets; Secondly, although traditional local approximation methods are widely used, they are often limited by the inconsistency of prediction results; The third issue is that many aggregation strategies lack discrimination when evaluating the importance of experts (i.e. local models), resulting in a loss of overall prediction accuracy. In response to the above challenges, this article innovatively proposes a comprehensive method that integrates third-degree stochastic fully symmetric interpolatory rules (TDSFSI), local approximation, and Tsallis mutual information (TDSFSIRLA), aiming to fundamentally break through existing limitations. Specifically, TDSFSIRLA first introduces an efficient third-degree stochastic fully symmetric interpolatory rules, which achieves accurate approximation of Gaussian kernel functions by generating adaptive dimensional feature maps. This innovation not only significantly reduces the number of required orthogonal nodes and effectively lowers computational costs, but also maintains extremely high approximation accuracy, providing a solid theoretical foundation for processing large-scale datasets. Furthermore, in order to overcome the inconsistency of local approximation methods, this paper adopts the Generalized Robust Bayesian Committee Machine (GRBCM) as the aggregation framework for local experts. GRBCM ensures the harmonious unity of the prediction results of each local model through its inherent consistency and robustness, significantly improving the stability and reliability of the overall prediction. More importantly, in response to the issue of uneven distribution of expert weights, this article creatively introduces Tsallis mutual information as a metric for weight allocation. Tsallis mutual information, with its sensitive ability to capture information complexity, assigns weights to different local experts that match their contribution, effectively solving the problem of prediction bias caused by uneven weight distribution and further improving prediction accuracy. In the experimental verification phase, this article conducted comprehensive testing on multiple synthetic datasets and seven representative real datasets. The results show that the TDSFSIRLA method not only achieves significant reduction in time complexity, but also demonstrates excellent performance in prediction accuracy, fully verifying its significant advantages and broad application prospects in the field of large-scale Gaussian process regression.

Integrated metaheuristic approaches for estimation of fracture porosity derived from fullbore formation micro-imager logs: Reaping the benefits of stand-alone and ensemble machine learning models

Fracture porosity is one of the most effective parameters for reservoir productivity and recovery efficiency. This study aims to predict and improve the accuracy of fracture porosity estimation through the application of advanced machine learning (ML) algorithms. A novel approach was introduced for the first time to estimate fracture porosity by reaping the benefits of petrophysical and fullbore formation micro-imager (FMI) data based on employing various stand-alone, ensemble, optimisation and multi-variable linear regression (MVLR) algorithms. This study proposes a ground-breaking two-step committee machine (CM) model. Petrophysical data containing compressional sonic-log travel time, deep resistivity, neutron porosity and bulk density (inputs), along with FMI-derived fracture porosity values (outputs), were employed. Nine stand-alone ML algorithms, including back-propagation neural network, Takagi and Sugeno fuzzy system, adaptive neuro-fuzzy inference system, decision tree, radial basis function, extreme gradient boosting, least-squares boosting, least squares support vector regression and k-nearest neighbours, were trained for initial estimation. To improve the efficacy of stand-alone algorithms, their outputs were combined in CM structures using optimisation algorithms. This integration was applied through five optimisation algorithms, including genetic algorithm, ant colony, particle swarm, covariance matrix adaptation evolution strategy (CMA-ES) and Coyote optimisation algorithm. Considering the lowest error, the CM with CMA-ES showed superior performance. Subsequently, MVLR was employed to improve the CMs further. Employing MVLR to combine the CMs yielded a 57.85% decline in mean squared error and a 4.502% improvement in the correlation coefficient compared to the stand-alone algorithms. The results of the benchmark analysis validated the efficacy of this approach.

Application of power-law committee machine to combine five machine learning algorithms for enhanced oil recovery screening

One of the main challenges in screening of enhanced oil recovery (EOR) techniques is the class imbalance problem, where the number of different EOR techniques is not equal. This problem hinders the generalization of the data-driven methods used to predict suitable EOR techniques for candidate reservoirs. The main purpose of this paper is to propose a novel approach to overcome the above challenge by taking advantage of the Power-Law Committee Machine (PLCM) technique optimized by Particle Swam Optimization (PSO) to combine the output of five cutting-edge machine learning methods with different types of learning algorithms. The PLCM method has not been used in previous studies for EOR screening. The machine learning models include the Artificial Neural Network (ANN), CatBoost, Random Forest (RF), K-Nearest Neighbors (KNN), and Support Vector Machine (SVM). The CatBoost is used for the first time in this work for screening of EOR methods. The role of the PSO is to find the optimal values for the coefficients and exponents of the power-law model. In this study, a bigger dataset than those in previous studies, including 2563 successful worldwide EOR experiences, was gathered. A bigger dataset improves the generalization of the data-driven methods and prevents overfitting. The hyperparameters of the individual machine-learning models were tuned using the fivefold cross-validation technique. The results showed that all the individual methods could predict the suitable EOR method for unseen cases with an average score of 0.868. Among the machine learning models, the KNN and SVM had the highest scores with a value of 0.894 and 0.892, respectively. Nonetheless, after combining the output of the models using the PLCM method, the score of the predictions improved to 0.963, which was a substantial increase. Finally, a feature importance analysis was conducted to find out the most influential parameters on the output. The novelty of this work is having shown the ability of the PLCM technique to construct an accurate model to overcome the class-imbalance issue in EOR screening by utilizing different types of data-driven models. According to feature importance analysis, oil gravity and formation porosity were recognized as the most influential parameters on EOR screening.

Sub-seasonal soil moisture anomaly forecasting using combinations of deep learning, based on the reanalysis soil moisture records

Sub-seasonal drought forecasting is crucial for early warning in estimating agricultural production and optimizing irrigation management, as forecasting skills are relatively weak during this period. Soil moisture exhibits stronger persistence compared to other climate system quantities, which makes it especially influential in shaping land-atmosphere feedback, thus supplying a unique insight into drought forecasting. Relying on the soil moisture memory, this study investigates the combination of multiple deep-learning modules for sub-seasonal drought indices hindcast in the Huai River basin of China, using long-term ERA5-Land soil moisture records with a noise-assisted data analysis tool. The inter-compared deep-learning models include a hybrid model and a committee machine framework. The results show that the performance of the committee machine framework can be improved with the help of series decomposition and the forecasting skill is not impaired with the lead time increases. Overall, this study highlights the potential of combining deep-learning models with soil moisture memory analysis to improve sub-seasonal drought forecasting.

Active sparse Bayesian committee machine potential for isothermal-isobaric molecular dynamics simulations.

Recent advancements in machine learning potentials (MLPs) have significantly impacted the fields of chemistry, physics, and biology by enabling large-scale first-principles simulations. Among different machine learning approaches, kernel-based MLPs distinguish themselves through their ability to handle small datasets, quantify uncertainties, and minimize over-fitting. Nevertheless, their extensive computational requirements present considerable challenges. To alleviate these, sparsification methods have been developed, aiming to reduce computational scaling without compromising accuracy. In the context of isothermal and isobaric ML molecular dynamics (MD) simulations, achieving precise pressure estimation is crucial for reproducing reliable system behavior under constant pressure. Despite progress, sparse kernel MLPs struggle with precise pressure prediction. Here, we introduce a virial kernel function that significantly enhances the pressure estimation accuracy of MLPs. Additionally, we propose the active sparse Bayesian committee machine (BCM) potential, an on-the-fly MLP architecture that aggregates local sparse Gaussian process regression (SGPR) MLPs. The sparse BCM potential overcomes the steep computational scaling with the kernel size, and a predefined restriction on the size of kernel allows for fast and efficient on-the-fly training. Our advancements facilitate accurate and computationally efficient machine learning-enhanced MD (MLMD) simulations across diverse systems, including ice-liquid coexisting phases, Li10Ge(PS6)2 lithium solid electrolyte, and high-pressure liquid boron nitride.

Predicting the solubility of hydrogen in hydrocarbon fractions: Advanced data-driven machine learning approach and equation of state

BackgroundHydrogen is a free carbon source for energy that attracts massive interest to be utilized during energy transition period in the near future. Molecular hydrogen soon will be a key component in fuels productions. It can be produced through catalytic steam and natural gas. For such processes the accurate estimation of solubility of hydrogen in many alkanes’ solution can have a major impact on the design and optimization of such petrochemical processes. MethodsIn this study, advanced data-driven scheme consisting in a committee machine intelligent system (CMIS) model was proposed for predicting the solubility of hydrogen in many hydrocarbon liquids raging from light n-alkanes C1 to heavy n-alkanes of C46. The CMIS was gained by combining multilayer perceptron (MLP) and cascaded forward neural network (CFNN) beneath a sole model using gene expression programming (GEP). A wide range database of hydrogen solubility in a binary system of hydrogen and various n-alkanes types was considered for developing the intelligent model. Besides, the reliability of the physics-based models, namely the Peng–Robinson equations of state (PR EOS) was checked for predicting the solubility of H2 in n-alkanes. Significant findingsThe main results demonstrated that the developed CMIS scheme can estimate the solubility of hydrogen in n-alkanes with a quite low computational cost and excellent performance, where the yielded overall RMSE and R2 values were 0.0033 and 0.9972, respectively. Besides, the established machine learning based model is capable of providing a good match against a wide range of experiment data. Furthermore, it is found that the implemented CMIS paradigm is superior to the existing intelligent and physics-based models in terms of accuracy. Lastly, the result of this work is vital for increasing the accuracy of hydrogen solubility in various hydrocarbon solvents required for process optimization of the hydrogen production units.

A committee machine-based estimation of shear velocity log by combining intelligent systems and rock-physics model using metaheuristic algorithms

Estimation of reservoir parameters such as porosity, lithology and determination of elastic moduli is based on well-logging data and seismic velocity information. However, the lack of shear velocity log in many wells, and the high costs of its acquisition have led to developing different approaches for shear velocity estimation. Currently, shear velocity estimation is accomplished by common industrial methods such as empirical equations, regression analysis, intelligent methods and rock-physics models (RPMs). The current study proposes that robust, accurate and cost-effective intelligent methods coupled with suitable RPM provide reasonable estimations of the shear velocity in an optimal processing time. Hybridizing intelligent models with RPMs in the framework of a committee machine is an optimal solution for shear velocity estimation of carbonate reservoirs. The intelligent methods used in this study include neural network (NN), fuzzy logic (FL) and an adaptive-neuro fuzzy inference system (ANFIS). In this research, for each individual intelligent model and a developed RPM, the optimal weights were allocated by Metaheuristic optimization algorithms including genetic algorithm (GA), ant colony optimization routing (ACOR) and simulated annealing (SA)). To obtain the best results with the lowest mean squared error (MSE), a total of 9408, 10801 and 12288 models were created and tested in MATLAB to optimally parametrize the GA, ACOR and SA, respectively. The results showed that the combination of the applied intelligent models with the developed RPMs leads to a higher correlation coefficient with reference to the measured Shear velocity in test dataset. The use of a combination of intelligent methods in the form of committee-machine instead of individual methods has led to a reduction in error and improved the results by up to 2% Considering the high sensitivity of shear wave velocity in geomechanical and reservoir characterization studies, even small improvements in its estimation can significantly reduce error in the characterization of hydrocarbon reservoirs.

RETRACTED: Continual Learning Approach for Continuous Data Stream Analysis in Dynamic Environments

Continuous data stream analysis primarily focuses on the unanticipated changes in the transmission of data distribution over time. Conceptual change is defined as the signal distribution changes over the transmission of continuous data streams. A drift detection scenario is set forth to develop methods and strategies for detecting, interpreting, and adapting to conceptual changes over data streams. Machine learning approaches can produce poor learning outcomes in the conceptual change environment if the sudden change is not addressed. Furthermore, due to developments in concept drift, learning methodologies have been significantly systematic in recent years. The research introduces a novel approach using the fully connected committee machine (FCM) and different activation functions to address conceptual changes in continuous data streams. It explores scenarios of continual learning and investigates the effects of over-learning and weight decay on concept drift. The findings demonstrate the effectiveness of the FCM framework and provide insights into improving machine learning approaches for continuous data stream analysis. We used a layered neural network framework to experiment with different scenarios of continual learning on continuous data streams in the presence of change in the data distribution using a fully connected committee machine (FCM). In this research, we conduct experiments in various scenarios using a layered neural network framework, specifically the fully connected committee machine (FCM), to address conceptual changes in continuous data streams for continual learning under a conceptual change in the data distribution. Sigmoidal and ReLU (Rectified Linear Unit) activation functions are considered for learning regression in layered neural networks. When the layered framework is trained from the input data stream, the regression scheme changes consciously in all scenarios. A fully connected committee machine (FCM) is trained to perform the tasks described in continual learning with M hidden units on dynamically generated inputs. In this method, we run Monte Carlo simulations with the same number of units on both sides, K and M, to define the advancement of intersections between several hidden units and the calculation of generalization error. This is applied to over-learnability as a method of over-forgetting, integrating weight decay, and examining its effects when a concept drift is presented.

Integrating three smart predictive models using a power-law committee machine for the prediction of compressive strength in masonry made of clay bricks and cement mortar

Due to the importance of the compressive strength of masonry (CSM) prisms made of clay bricks and cement mortar in the design of masonry structures, it is crucial to develop a reliable method for accurate prediction. This paper presents an innovative approach for estimating CSM by integrating an optimized convolutional neural network (OCNN), extreme learning machine (ELM), and decision tree (DT) using a power-law committee machine (PLCM). The strategy consists of two steps: firstly, three regression approaches (OCNN, ELM, and DT) are developed to model CSM, and secondly, the results of these methods are combined through a PLCM. The convolutional neural network is enhanced using the cuckoo search (CS) algorithm to determine its weights and bias, thereby improving its ability to model CSM accurately. The CS algorithm is also employed in the PLCM structure to extract the optimal contribution of each individual model in the overall prediction. To statistically evaluate the effectiveness and efficiency of the developed models and compare them with previously established empirical correlations for estimating CSM values, various visualization methods and statistical parameters are utilized. The evaluation demonstrates that OCNN, ELM, and DT can estimate CSM with remarkable accuracy and further confirms that PLCM outperforms its individual elements as well as empirical methods. This study provides evidence of the significant improvement in the performance of PLCM in modeling CSM.

Machine learning assisted predictions for hydrogen storage in metal-organic frameworks

Metal organic frameworks (MOFs) have been studied vastly for hydrogen storage purposes due to their unique properties. In the present work hydrogen storage capacity is modeled using machine learning approaches including Multi-layer Perceptron (MLP), Support Vector Machines (SVM), Random Forest (RF), CatBoost, LightGBM, XGBoost, and Committee Machine Intelligence System (CMIS) trained on experimental data gathered from various experimental studies. CMIS model shows the best results with R2=0.982 and Root Mean Square Error = 0.088. Additionally, this study uses the Leverage method to detect and eliminate suspected data points in order to improve predictability. Only 2.04% of data points are detected as suspected data points. Sensitivity analysis demonstrates that surface area, pore volume, pressure, and temperature have almost the same contribution on hydrogen storage. The findings of this research offer a valuable model for MOF selection and optimization of operational conditions in hydrogen storage processes that utilizes MOFs.

Estimation of geomechanical units using petrophysical logs, core and supervised intelligent committee machine method to optimize exploration drilling operations

Estimation of geomechanical units using petrophysical logs, core and supervised intelligent committee machine method to optimize exploration drilling operations

Prediction of asphaltene adsorption capacity of clay minerals using machine learning

AbstractA thorough understanding of asphaltene adsorption on clay minerals is particularly important in oil production and contaminated soil remediation using clay‐based adsorbents. In this paper, we introduced a machine learning approach as a reliable alternative for commonly used adsorption isotherms that suffer from inherent limitations in the prediction of asphaltene adsorption onto clay minerals. Machine learning (ML) models, namely multilayer perceptron (MLP), support vector machine (SVM), decision tree (DT), random forest (RF), and committee machine intelligent system (CMIS) combined with two optimizers were used. Experimental data (142 data points for six different clay minerals) was used for the modelling. To improve the accuracy of the smart models, a comprehensive data preparation such as outlier removal and feature selection was carried out. The results showed that relatively all the proposed models predict asphaltene adsorption on clay minerals with acceptable precision. Nevertheless, the MLP model showed superior performance compared with other models in which the overall root mean square error (RMSE) and coefficient of determination (R2) values of 6.72 and 0.93 were obtained, respectively. Finally, the developed MLP model was compared with the well‐known adsorption isotherms of Langmuir and Freundlich and exhibited superior performance.

Handling dataset dependence with model ensembles for skin lesion classification from dermoscopic and clinical images

AbstractDataset dependence affects many real‐life applications of machine learning: the performance of a model trained on a dataset is significantly worse on samples from another dataset than on new, unseen samples from the original one. This issue is particularly acute for small and somewhat specific databases in medical applications; the automated recognition of melanoma from skin lesion images is a prime example. We document dataset dependence in dermoscopic skin lesion image classification using three publicly available medium size datasets. Standard machine learning techniques aimed at improving the predictive power of a model might enhance performance slightly, but the gain is small, the dataset dependence is not reduced, and the best combination depends on model details. We demonstrate that simple differences in image statistics account for only 5% of the dataset dependence. We suggest a solution with two essential ingredients: using an ensemble of heterogeneous models, and training on a heterogeneous dataset. Our ensemble consists of 29 convolutional networks, some of which are trained on features considered important by dermatologists; the networks' output is fused by a trained committee machine. The combined International Skin Imaging Collaboration dataset is suitable for training, as it is multi‐source, produced by a collaboration of a number of clinics over the world. Building on the strengths of the ensemble, it is applied to a related problem as well: recognizing melanoma based on clinical (non‐dermoscopic) images. This is a harder problem as both the image quality is lower than those of the dermoscopic ones and the available public datasets are smaller and scarcer. We explored various training strategies and showed that 79% balanced accuracy can be achieved for binary classification averaged over three clinical datasets.

Optimization of Logging Interpretation Parameters Based on Committee Machine

Aiming at the problems of low precision and low applicability of the selection method of macroscopic capture cross-section parameters of rock skeleton, formation water, oil and gas, and mud in the volume model of remaining oil saturation logging, this study proposes a method for optimizing the capture section parameters based on the committee machine regression model. To begin, we select well sections from well logging data in a reasonable manner, allocate well sections with different component parameters in the same proportion, and construct sample datasets. The basic experts are then selected as multiple regression models, particle swarm optimization, and robust regression methods to train and learn the input parameters. By combining multiple experts, the regression committee machine improves the overall performance of the intelligent model. Finally, the genetic algorithm is used as a combiner to determine the contribution of each basic expert network in the final output, and the optimized parameters are obtained, which are then fed into the volume model to calculate the remaining oil saturation of the newly developed production wells, guiding the perforation and development work. The model is used to evaluate the remaining oil in the X oilfield, and the calculated water saturation matches the oil test results, proving the model’s accuracy and availability. The use of real-world data demonstrates that this method can effectively characterize the four parameter values in the volume model and provide reliable geophysical technical support for the evaluation of remaining oil.

Prediction of reservoir brittleness from geophysical logs using machine learning algorithms

Brittleness is an important geomechanical property of reservoirs, which is usually estimated from cores or sonic logs that are expensive to acquire. In this study, we report data-driven, machine learning workflows to predict brittleness from less expensive, readily available conventional logs. We propose three strategies to predict brittleness using gamma ray, neutron porosity, density, caliper, sonic, and photoelectric factor logs by utilizing gradient boosting (GB), support vector regression (SVR), and neural networks (NN). The first strategy involves predicting brittleness directly from the logs while the second strategy predicts shear sonic logs used for the estimation of brittleness. The performance of the models given as R2 on deployment on the testing set for the first strategy is: GB (0.87), SVR (0.73), and NN (0.82), while for the second strategy: GB (0.94), SVR (0.87), and NN (0.94). In the third strategy, we convert the prediction into a classification task by grouping the brittleness estimate into ductile, transition, and brittle. The accuracy of model deployment on the testing set for the third strategy is: GB (89.37%), SVM (89.06%), and NN (89.16%). We demonstrate that depending on the strategy adopted, the gradient boosting algorithm outperforms the other peer algorithms in terms of training and validation scores. Furthermore, we combine the three algorithms using a committee machine to improve the performance of the model. The workflow in this study can be adopted to predict other reservoir properties from available logs. The workflow can also be adopted to characterize reservoir heterogeneity from seismic traces trained by well logs.

Flood River Water Level Forecasting using Ensemble Machine Learning for Early Warning Systems

Flood forecasting is crucial for early warning system and disaster risk reduction. Yet the flood river water levels are difficult and challenging task that it cannot be easily captured with classical time-series approaches. This study proposed a novel intelligence system utilised various machine learning techniques as individual models, including radial basis function neural network (RBF-NN), adaptive neuro-fuzzy inference system (ANFIS), support vector machine (SVM), and long short-term memory network (LSTM) to establish intelligent committee machine learning flood forecasting (ICML-FF) framework. The combination of these individual models achieved through simple averaging method, and further optimised using weighted averaging by K-nearest neighbour (K-NN) and genetic algorithm (GA). The effectiveness of the proposed model was evaluated using real case study for Malaysia’s Kelantan River. The results show that ANFIS outperforms as individual model, while ICML-FF-based model produced better accuracy and lowest error than any one of the individuals. In general, it is found that the proposed ICML-FF is capable of robust forecasting model for flood early warning systems.

Hybrid artificial intelligence paradigms for modeling of water-gas (pure/mixture) interfacial tension

There are many applications with the two-phase flow of gas (hydrocarbon, non-hydrocarbon, and their mixture) and water in different courses of gas recovery from natural gas resources and gas storage/sequestration programs. As the interface of gas-water is crucial in such systems, precise prediction of gas-water interfacial tension (IFT) can aid in the simulation and development of such processes. Artificial intelligence techniques (AIT) are being used to estimate IFT. In this paper, the IFT of the gas and water system was estimated based on models built using a comprehensive data set comprised of 2658 experimental data points. These cover a wide range of input parameters, i.e., specific gravity (0.5539–1.5225), temperature (278.1–477.5944 K), pressure (0.01–280 MPa), and water salinity (0–200,000 ppm). The intelligent models include Least-Squares Boosting (LS-Boost), Multilayer perceptron (MLP), Least Square Support Vector Machine (LSSVM), and Committee machine intelligent system (CMIS). The models reproduce the IFT data in 7.4–81.69 mN/m. The modeling approaches contain new hybrid forms in which Imperialist Competitive Algorithm (ICA), Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), Levenberg-Marquardt algorithm (LM), Bayesian regularization algorithm (BR), Scaled conjugate gradient algorithm (SCG), and Coupled Simulated Annealing (CSA) were used for optimization and learning purposes. Statistical and graphical analyses were implemented to check the agreement between the prediction and evaluation data. The results show a reasonable coherence for most models, among which the CMIS approach exhibited a promising performance. CMIS was accurate even in conditions of varying specific gravity, pressure, temperature, and salinity. The findings were also compared with available models in the literature and demonstrated superior predictions of the CMIS model. Also, outlier detection by the Leverage approach demonstrates the validity of the gathered dataset and, subsequently, the CMIS model.

Artificial neural network, support vector machine, decision tree, random forest, and committee machine intelligent system help to improve performance prediction of low salinity water injection in carbonate oil reservoirs

A large body of experimental research supports the effectiveness of Low Salinity Water Injection (LSWI) for enhanced oil recovery from carbonate reservoirs in laboratory scale. Development of robust predictive smart models connecting effective parameters controlling this complex process to Final Recovery Factor (RFf), as the target parameter, is of a paramount importance. The main objective of this research work is to develop intelligent models using Artificial Neural Network (ANN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Committee Machine Intelligent System (CMIS) to forecast performance of LSWI in carbonates using experimental data reported in the literature. Random Search (RS) and Anneal (AL) algorithms were used for optimization of hyperparameters. After data collection from 47 reliable coreflooding studies (582 data points), a rigorous data preprocessing was conducted to ensure quality of the database. Features selection process was used to determine the main parameters controlling LSWI performance in carbonates: brine permeability (Kb), core diameter (d), porosity (Φ), and residual water saturation (Swi) of the core, HCO3− concentration, and salinity (S) of the connate brine, the salinity (S) of the injected brine, and initial recovery factor (RFi) which were used for development of the models. We considered initial oil recovery (RFi) in this research work, which was not considered in previous works reported in the literature. The applicability domain analysis showed that training and testing response outliers were zero and 9, respectively, indicating acceptable quality of the database. Performance of the developed smart models was analyzed and compared using statistical and graphical error analysis methods. The best performance was obtained for the RF model with Root Mean Square Error (RMSE) of 2.497 and 5.757 for training and testing datasets, respectively, which exhibits a very good agreement with the experimental data.

Asphaltene Precipitation Prediction during Bitumen Recovery: Experimental Approach versus Population Balance and Connectionist Models.

Deasphalting bitumen using paraffinic solvent injection is a commonly used technique to reduce both its viscosity and density and ease its flow through pipelines. Common modeling approaches for asphaltene precipitation prediction such as population balance model (PBM) contains complex mathematical relation and require conducting precise experiments to define initial and boundary conditions. Machine learning (ML) approach is considered as a robust, fast, and reliable alternative modeling approach. The main objective of this research work was to model the effect of paraffinic solvent injection on the amount of asphaltene precipitation using ML and PBM approaches. Five hundred and ninety (590) experimental data were collected from the literature for model development. The gathered data was processed using box plot, data scaling, and data splitting. Data pre-processing led to the use of 517 data points for modeling. Then, multilayer perceptron, random forest, decision tree, support vector machine, committee machine intelligent system optimized by annealing, and random search techniques were used for modeling. Precipitant molecular weight, injection rate, API gravity, pressure, C5 asphaltene content, and temperature were determined as the most relevant features for the process. Although the results of the PBM model are precise, the AI/ML model (CMIS) is the preferred model due to its robustness, reliability, and relative accuracy. The committee machine intelligent system is the superior model among the developed smart models with an RMSE of 1.7% for the testing dataset and prediction of asphaltene precipitation during bitumen recovery.

Prediction of hydraulic conductivity based on the soil grain size using supervised committee machine artificial intelligence

Prediction of hydraulic conductivity based on the soil grain size using supervised committee machine artificial intelligence