Multilayer Extreme Learning Machine Research Articles

Hybrid Machine-Learning Model for Accurate Prediction of Filtration Volume in Water-Based Drilling Fluids

Accurately predicting the filtration volume (FV) in drilling fluid (DF) is crucial for avoiding drilling problems such as a stuck pipe and minimizing DF impacts on formations during drilling. Traditional FV measurement relies on human-centric experimental evaluation, which is time-consuming. Recently, machine learning (ML) proved itself as a promising approach for FV prediction. However, existing ML methods require time-consuming input variables, hindering the semi-real-time monitoring of the FV. Therefore, employing radial basis function neural network (RBFNN) and multilayer extreme learning machine (MELM) algorithms integrated with the growth optimizer (GO), predictive hybrid ML (HML) models are developed to reliably predict the FV using only two easy-to-measure input variables: drilling fluid density (FD) and Marsh funnel viscosity (MFV). A 1260-record dataset from seventeen wells drilled in two oil and gas fields (Iran) was used to evaluate the models. Results showed the superior performance of the RBFNN-GO model, achieving a root-mean-square error (RMSE) of 0.6396 mL. Overfitting index (OFI), score, dependency, and Shapley additive explanations (SHAP) analysis confirmed the superior FV prediction performance of the RBFNN-GO model. In addition, the low RMSE (0.3227 mL) of the RBFNN-NGO model on unseen data from a different well within the studied fields confirmed the strong generalizability of this rapid and novel FV prediction method.

Sub-surface geospatial intelligence in carbon capture, utilization and storage: A machine learning approach for offshore storage site selection

This study introduces an innovative data-driven and machine-learning framework designed to accurately predict site scores in the site screening study for specific offshore CO2 storage sites. The framework seamlessly integrates diverse sub-surface geospatial data sources with human aided expert-weighted criteria, thereby providing a high-resolution screening tool. Tailored to accommodate varying data accessibility and the significance of criteria, this approach considers both technical and non-technical factors. Its purpose is to facilitate the identification of priority locations for projects associated with Carbon Capture, Utilization, and Storage (CCUS). Through aggregating and analyzing geospatial datasets, the study employs machine learning algorithms and an expert-weighted model to identify suitable geologic CCUS regions. This process adheres to stringent safety, risk control, and environmental guidelines, addressing situations where human analysis may fail to recognize patterns and provide detailed insights in suitable site screening techniques. The primary emphasis of this research is to bridge the gap between scientific inquiry and practical application, facilitating informed decision-making in the implementation of CCUS projects. Rigorous assessments encompassing geological, oceanographic, and eco-sensitivity metrics contribute valuable insights for policymakers and industry leaders. To ensure the accuracy, efficiency, and scalability of the established offshore CO2 storage facilities, the proposed machine learning approach undergoes benchmarking. This comprehensive evaluation includes the utilization of machine learning algorithms such as Extreme Gradient Boosting (XGBoost), Random Forest (RF), Multilayer Extreme Learning Machine (MLELM), and Deep Neural Network (DNN) for predicting more suitable site scores. Among these algorithms, the DNN algorithm emerges as the most effective in site score prediction. The strengths of the DNN algorithm encompass nonlinear modeling, feature learning, scale invariance, handling high-dimensional data, end-to-end learning, transfer learning, representation learning, and parallel processing. The evaluation results of the DNN algorithm demonstrate high accuracy in the testing subset, with values of AAPD (Average Absolute Percentage Difference) = 1.486 %, WAAPD (Weighted Average Absolute Percentage Difference) = 0.0149 %, VAF (Variance Accounted For) = 0.9937, RMSE (Root Mean Square Error) = 0.9279, RSR (Root Sum of Squares Residuals) = 0.0068, and R2 (Coefficient of Determination) = 0.9937.

Revolutionizing Anemia Classification with Multilayer Extremely Randomized Tree Learning Machine for Unprecedented Accuracy

Anemia is a prevalent global health issue that is characterized by a deficit in red blood cells or low levels of hemoglobin. This condition is influenced by various causes, including nutritional inadequacies, chronic diseases, and genetic predisposition. The incidence of the phenomenon exhibits variation across different geographical regions and demographic groups. This pioneering research investigates the identification and classification of anemia, potentially leading to transformative advancements in the discipline. The classification of anemia encompasses four distinct groups, namely Beta Thalassemia Trait, Iron Deficiency Anemia, Hemoglobin E, and Combination. This comprehensive categorization offers clinicians a more refined and detailed comprehension of the condition. The integration of deep learning and machine learning in the Multilayer Extremely Randomized Tree Learning Machine (MERTLM) model represents a departure from traditional approaches and a significant advancement in the field of medical categorization accuracy. The MERTLM approach integrates randomized tree with multilayer extreme learning machine (M-ELM) representation learning, hence emphasizing the possibility of interdisciplinary collaboration in the field of diagnostics. In addition to its impact on anemia, artificial intelligence (AI) is playing a significant role in revolutionizing medical diagnosis by emphasizing the integration of innovative methods. This study utilizes the combined capabilities of machine learning and deep learning to improve accuracy. Notably, recent developments have resulted in an exceptional accuracy rate of 99.67%, precision of 99.60%, sensitivity of 99.47%, and an amazing F1-Score of 99.53%. This study represents a significant advancement in the field of anemia research, providing valuable insights that may be applied to intricate medical issues and enhancing the quality of patient care.

A Physics-Informed Low-Shot Adversarial Learning for sEMG-Based Estimation of Muscle Force and Joint Kinematics.

Muscle force and joint kinematics estimation from surface electromyography (sEMG) are essential for real-time biomechanical analysis of the dynamic interplay among neural muscle stimulation, muscle dynamics, and kinetics. Recent advances in deep neural networks (DNNs) have shown the potential to improve biomechanical analysis in a fully automated and reproducible manner. However, the small sample nature and physical interpretability of biomechanical analysis limit the applications of DNNs. This paper presents a novel physics-informed low-shot adversarial learning method for sEMG-based estimation of muscle force and joint kinematics. This method seamlessly integrates Lagrange's equation of motion and inverse dynamic muscle model into the generative adversarial network (GAN) framework for structured feature decoding and extrapolated estimation from the small sample data. Specifically, Lagrange's equation of motion is introduced into the generative model to restrain the structured decoding of the high-level features following the laws of physics. A physics-informed policy gradient is designed to improve the adversarial learning efficiency by rewarding the consistent physical representation of the extrapolated estimations and the physical references. Experimental validations are conducted on two scenarios (i.e. the walking trials and wrist motion trials). Results indicate that the estimations of the muscle forces and joint kinematics are unbiased compared to the physics-based inverse dynamics, which outperforms the selected benchmark methods, including physics-informed convolution neural network (PI-CNN), vallina generative adversarial network (GAN), and multi-layer extreme learning machine (ML-ELM).

A hybrid clustering-based type-2 adaptive neuro-fuzzy forecasting model for smart control systems

The practical applicability of smart control systems suffers from the existence of uncertainty in the physical parameters, sensor measurements and external excitations. Effective methods and its ease of implementation are required in the real-time active control strategies. To this end, in this research, we introduce an overlapping clusters and Multilayer Extreme Learning Machine (ML-ELM)-aided interval type-2 Takagi-Sugeno fuzzy inference system for vibration mitigation of smart structures. The combined Fuzzy C-Means (FCM) and imperialist competitive algorithm (ICA) with respect to degree of fuzzy overlap between clusters was utilized for characterizing the upper and lower membership functions forming the antecedent part. Furthermore, indeterminacy was modeled as footprint of uncertainty (FOU) and automatically formed based on statistical characteristics of soft clusters. The consequent parameters were adjusted by hybridization of ML-ELM and overlapping clusters, then the adaptivity was guaranteed. The predictive accuracy and generalization capability of the proposed method were demonstrated over the estimation of active control effort. Finally, to prove the practicability of the implemented framework, it was applied to four high-rise, mid-rise and low-rise benchmark building structures equipped with active mass damper (AMD), active tendon and active bracing system (ABS). The obtained results demonstrated the higher predictive accuracy and generalization ability of the presented data-driven control scheme compared to the other standard technical machine learning strategies. This practical guideline is easy to implement and automates control engineering tasks in real-time applications, solving one of the most challenging aspects of smart control systems.

An Efficient Recognition Method for Orbital Angular Momentum via Adaptive Deep ELM.

For orbital angular momentum (OAM) recognition in atmosphere turbulence, how to design a self-adapted model is a challenging problem. To address this issue, an efficient deep learning framework that uses a derived extreme learning machine (ELM) has been put forward. Different from typical neural network methods, the provided analytical machine learning model can match the different OAM modes automatically. In the model selection phase, a multilayer ELM is adopted to quantify the laser spot characteristics. In the parameter optimization phase, a fast iterative shrinkage-thresholding algorithm makes the model present the analytic expression. After the feature extraction of the received intensity distributions, the proposed method develops a relationship between laser spot and OAM mode, thus building the steady neural network architecture for the new received vortex beam. The whole recognition process avoids the trial and error caused by user intervention, which makes the model suitable for a time-varying atmospheric environment. Numerical simulations are conducted on different experimental datasets. The results demonstrate that the proposed method has a better capacity for OAM recognition.

A comparative evaluation of Stacked Auto-Encoder neural network and Multi-Layer Extreme Learning Machine for detection and classification of faults in transmission lines using WAMS data

Smart grid is envisaged as a power grid that is extremely reliable and flexible. The electrical grid has wide-area measuring devices like Phasor measurement units (PMUs) deployed to provide real-time grid information and resolve issues effectively and speedily without compromising system availability. The development and application of machine learning approaches for power system protection and state estimation have been facilitated by the availability of measurement data. This research proposes a transmission line fault detection and classification (FD&C) system based on an auto-encoder neural network. A comparison between a Multi-Layer Extreme Learning Machine (ML-ELM) network model and a Stacked Auto-Encoder neural network (SAE) is made. Additionally, the performance of the models developed is compared to that of state-of-the-art classifier models employing feature datasets acquired by wavelet transform based feature extraction as well as other deep learning models. With substantially shorter testing time, the suggested auto-encoder models detect faults with 100% accuracy and classify faults with 99.92% and 99.79% accuracy. The computational efficiency of the ML-ELM model is demonstrated with high accuracy of classification with training time and testing time less than 50 ms. To emulate real system scenarios the models are developed with datasets with noise with signal-to-noise-ratio (SNR) ranging from 10 dB to 40 dB. The efficacy of the models is demonstrated with data from the IEEE 39 bus test system.

A new robust predictive model for lost circulation rate using convolutional neural network: A case study from Marun Oilfield

A major cause of some of serious issues encountered in a drilling project, including wellbore instability, formation damage, and drilling string stuck – which are known to increase non-productive time (NPT) and hence the drilling cost – is what we know as mud loss. The mud loss can be prevented or at least significantly reduced by taking proper measures beforehand provided the position and intensity of such loss can be properly predicted using an accurate predictor model. Accordingly, in this study, we used the convolutional neural network (CNN) and hybridized forms of multilayer extreme learning machine (MELM) and least square support vector machine (LSSVM) with the Cuckoo optimization algorithm (COA), particle swarm optimization (PSO), and genetic algorithm (GA) for modeling the mud loss rate based on drilling data, mud properties, and geological information of 305 drilling wells penetrating the Marun Oilfield. For this purpose, we began by a pre-processing step to attenuate the effect of noise using the Savitzky-Golay method. The whole set of available data was divided into the modeling (including 2300 data points) and the validation (including 483 data points) subsets. Next, the second generation of the non-dominated sorting genetic algorithm (NSGA-II) was applied to the modeling data to identify the most significant features for estimating the mud loss. The results showed that the prediction accuracy increased with the number of selected features, but the increase became negligible when the number of selected features exceeded 9. Accordingly, the following 9 features were selected as input to the intelligent algorithms (IAs): pump pressure, mud weight, fracture pressure, pore pressure, depth, gel 10 min/gel 10 s, fan600/fan300, flowrate, and formation type. Application of the hybrid algorithms and simple forms of LSSVM and CNN to the training data (80% of the modeling data, i.e. 1840 data points) showed that all of the models tend to underestimate the mud loss at higher mud loss rates, although the CNN exhibited lower underestimation levels. Error analysis on different models showed that the CNN provided for a significantly higher degree of accuracy, as compared to other models. The more accurate outputs of the hybrid LSSVM model than those of the simple LSSVM indicated the large potentials of metaheuristic algorithms for achieving optimal solutions. The lower error levels obtained with the CNN model in the testing phase highlighted the excellent generalizability of this model for unseen data. The more accurate predictions obtained with this model, rather than the other models, in the validation phase further proved this latter finding. Therefore, application of this method to other wells in the same field is highly recommended.

Decoding auditory-evoked response in affective states using wearable around-ear EEG system

Objective. In this paper, an around-ear EEG system is investigated as an alternative methodology to conventional scalp-EEG-based systems in classifying human affective states in the arousal-valence domain evoked in response to auditory stimuli. Approach. EEG recorded from around the ears is compared to EEG collected according to the international 10–20 system in terms of efficacy in an affective state classification task. A wearable device with eight dry EEG channels is designed for ear-EEG acquisition in this study. Twenty-one subjects participated in an experiment consisting of six sessions over three days using both ear and scalp-EEG acquisition methods. Experimental tasks consisted of listening to an auditory stimulus and self-reporting the elicited emotion in response to the said stimulus. Various features were used in tandem with asymmetry methods to evaluate binary classification performances of arousal and valence states using ear-EEG signals in comparison to scalp-EEG. Main results. We achieve an average accuracy of 67.09% ± 6.14 for arousal and 66.61% ± 6.14 for valence after training a multi-layer extreme learning machine with ear-EEG signals in a subject-dependent context in comparison to scalp-EEG approach which achieves an average accuracy of 68.59% ± 6.26 for arousal and 67.10% ± 4.99 for valence. In a subject-independent context, the ear-EEG approach achieves 63.74% ± 3.84 for arousal and 64.32% ± 6.38 for valence while the scalp-EEG approach achieves 64.67% ± 6.91 for arousal and 64.86% ± 5.95 for valence. The best results show no significant differences between ear-EEG and scalp-EEG signals for classifications of affective states. Significance. To the best of our knowledge, this paper is the first work to explore the use of around-ear EEG signals in emotion monitoring. Our results demonstrate the potential use of around-ear EEG systems for the development of emotional monitoring setups that are more suitable for use in daily affective life log systems compared to conventional scalp-EEG setups.

Predicting uniaxial compressive strength from drilling variables aided by hybrid machine learning

Awareness of uniaxial compressive strength (UCS) as a key rock formation parameter for the design and development of gas and oil field plays. It plays an essential role in the selection of the drill bits and stability of the wellbore’s wall. Precise prediction of UCS before or during the drilling, especially in exploration wellbores, is necessary to improve the drilling speed and reduce the instability of the wellbore walls. UCS predictor machine-learning (ML) models are developed in this study using drilling parameters recorded during drilling using least-squares support-vector machine (LSSVM) and multi-layer extreme learning machine (MELM) algorithms hybridized with cuckoo optimization algorithm (COA), particle swarm optimization (PSO) and genetic algorithm (GA) optimizers. In addition, stand-alone LSSVM and convolutional neural network (CNN) models without optimizer enhancements are evaluated. Drilling and petrophysical data recorded for two wells (A and B) from the Rag-e-Safid oil field in southwest Iran were compiled to form the studied dataset. UCS was initially calculated numerically based on data from laboratory tests from petrophysical logs. The Well A dataset was pre-processed to remove outlying data records by applying the quantile regression algorithm. That analysis indicated that 9 data records should be removed from the Well A dataset. A decision tree model was employed for feature selection purposes to identify the more influential variables with respect to UCS. Depth, weight on the drill bit (WOB), drill-string rotation speed (RPM), rate of penetration (ROP), and torque (Trq) were the variables identified as being highly influential on UCS values. Application of the ML models on the training data subset (75% of Well A data records) revealed that the MELM-COA algorithm achieved the lowest root mean squared error (4.6945 MPa) and a higher coefficient of determination (0.9873) value than the other models when predicting UCS in the Well A training and validation data subsets. The Well-A-trained MELM-COA model confirmed its generalizability within the studied field by generating low UCS prediction errors when applied to the independent Well B testing dataset.

Hybridized machine-learning for prompt prediction of rheology and filtration properties of water-based drilling fluids

Careful design and preparation of drilling fluids with appropriate rheology and filtration properties, combined with operational monitoring, is essential for successful drilling operations. Field results reveal that most drilling-fluid problems encountered are avoidable based on prompt detection of unexpected changes in fluid rheology and filtration behavior. Drilling-fluid rheology and filtration properties are typically only checked once or twice a day, whereas other drilling-fluid properties, such as fluid density (FD), solid percentage (S%), and March funnel viscosity (MFV), tend to be monitored several times per hour. Machine learning is therefore applied to estimate rheology and filtration properties with FD, S%, and MFV as input variables. A 1160-record field dataset collected from 14 wells drilled in two oil and gas fields in southwest Iran with water-based drilling fluids is used to predict the drilling fluid’s rheological and filtration characteristics. Plastic viscosity (PV), yield point (YP), and filtrate volume (FV) are the targeted prediction objectives. Of six models tested, Multilayer extreme learning machine (MELM) hybridized with the cuckoo optimization algorithm (COA) provides the best PV, YP, and FV predictions. It achieves root mean squared error (RMSE) values of 0.6357 mL (FV), 0.6086 cP (PV), and 0.6796 lb/100 ft 2 (YP). MELM-COA generates rapid and accurate estimations of rheology and filtration properties with potential for real-time monitoring during drilling operations, without recourse to time-consuming laboratory filtration and rheological tests. This work delivers, in a novel way, accurate and reliable predictions of drilling fluid filtration properties using only the more readily available FD, MFV, and S% variables as input features.

Deep incremental random vector functional-link network: A non-iterative constructive sketch via greedy feature learning

The incremental version of randomized neural networks provides a greedy constructive algorithm for the shallow network, which adds new nodes through different stochastic methods rather than gradient optimization. However, the potential of the random incremental mechanism is still underutilized in deep structures. To address this research gap, we propose an unsupervised algorithm termed the incremental randomization-based autoencoder (IR-AE) for greedy feature learning, which applies an integrated optimized constructive algorithm to train the feature extractor. Using IR-AE as a hierarchical stacked block, we synthesize the deep incremental random vector functional-link (DI-RVFL) network that builds a deep structure with overall feature-output links through a feedforward approach. Furthermore, it is a novel data-driven initialization to implement the feedforward constructive sketch (CoSketch) as a pre-trained model for multi-layer perceptron. The simulation results empirically demonstrate that the proposed IR-AE can realize a higher reconstruction efficiency than AE and randomization-based AE. Moreover, DI-RVFL shows the advantages of deep structures in higher-level feature extraction compared to other stacked random structures. The overall performance of deep RVFLs outperforms those of multi-layer extreme learning machines. As data-driven initialization, CoSketch significantly improves the convergence performance of gradient descent.

A Review of multilayer extreme learning machine neural networks

A Review of multilayer extreme learning machine neural networks

Research on 3-DOF stabilized control system of ship propulsion-assisting sail

In recent years, wind-assisted ship propulsion (WASP) has become an important technology that can reduce the energy consumption of shipping. With different wind directions, a propulsion-assisting sail is rotated in azimuth to achieve maximum thrust along the ship's course. Most existing servo systems enable azimuth control of sails. However, the ship's attitude motion will affect the energy reduction of the sail. In this paper, this negative influence is described and demonstrated. The idea of three-degree-of-freedom (3-DOF) sail control is presented, and the energy reduction improved by the method is verified. Then, a 3-DOF stabilized control system of the ship propulsion-assisting sail is proposed to achieve azimuth tracking of the sail and compensate for attitude deviation caused by the ship. Considering the attitude target and the asymmetric structure of the sail, a kinematics solution and a dynamic model are established. Regarding the coupling characteristics of multivariable systems, inverse dynamics with a kernel-based multilayer extreme learning machine (ML-KELM) are designed to decouple the origin system. Finally, a numerical simulation was performed, and a principle prototype was developed. Results show that the designed control strategy eliminates the coupling of the original system, and the proposed system ensures the attitude stability of the sail.

Prediction and Online Optimization of Strip Shape in Hot Strip Rolling Process Using Sparrow Search Algorithm‐Online Sequential‐Deep Multilayer Extreme Learning Machine Algorithm

In the hot strip process, the prediction of strip shape is a key factor to improve the quality of products. However, strip‐shape prediction models based on traditional machine learning algorithms with relatively simple network structures often rely on data preprocessing and feature selection. Herein, five strip‐shape prediction models are proposed by combining sparrow search algorithm (SSA), particle swarm optimization algorithm (PSO), extreme learning machine (ELM), and deep multilayer extreme learning machine (DELM). The prediction experiments show that the SSA‐DELM model has highest accuracy, and the determination coefficient (R2) of crown prediction and flatness prediction of SSA‐DELM model is 0.971 and 0.979. 100% of the samples have a prediction error of less than ±7 μm and ±7 IU. Furthermore, to solve the problem of concept drift affecting prediction accuracy in industrial processes, online optimization method is proposed. Especially, online sequential deep ELM model (OS‐DELM) optimized by SSA and guided by double window drift detection is proposed to fill the gap. Compared with other online optimization methods, the update time with this method can be reduced by up to 83.05% averagely, which is more suitable for online strip‐shape prediction in hot rolling process.

Deep Extreme Learning Machines Based Two-Phase Spatiotemporal Modeling for Distributed Parameter Systems

Accurate and robust modeling of complex distributed parameter systems (DPSs) is a challenge for three reasons: 1) they have infinite-dimensional characteristics; 2) they are time/space coupled; and 3) there are model uncertainties. In this article, a two-phase spatiotemporal (S/T) modeling framework based on deep extreme learning machine (DELM) is proposed for DPSs. The modeling process consists of two S/T models in two phases: Phase I: a DELM model and Phase II: a Karhunen–Loève (KL) based ELM (KL-ELM) model. In phase I, the DELM model is constructed by combing the multilayer ELM (ML-ELM), ELM, and kernel-based ELM (K-ELM) to approximate the dominant S/T dynamics of DPSs. Since DPSs have an infinite-dimensional characteristic that can hardly be handled directly, ML-ELM is first employed to transform the infinite-dimensional systems into finite-dimensional systems. Then, the ELM model is adopted to further approximate the finite-dimensional systems to ensure the model can predict future dynamic behavior. Finally, the K-ELM is used to reconstruct the infinite-dimensional systems, which can be considered as the inverse process of ML-ELM. Thus, the final DELM model can be used for prediction in both space and time directions. In phase II, a KL-ELM model is constructed to compensate for modeling errors caused by reconstruction error or unknown nonlinear dynamics. By integrating the obtained DELM and KL-ELM models, the proposed two-phase S/T model can be constructed. Experiments on a typical industrial thermal process verified that the proposed method may work better in complex DPSs.

A Hybrid Method to Measure Distribution Consistency of Mixed-Attribute Datasets

Random sample partition (RSP) is a newly developed data management and processing model for Big Data processing and analysis. To apply the RSP model for Big Data computation tasks, it is very important to measure the distribution consistency of different datasets. Existing measurement methods for continuous-attribute and discrete-attribute datasets cannot directly deal with mixed-attribute datasets. In this article, we design a hybrid method to measure the distribution consistency among different mixed-attribute datasets by using a multilayer extreme learning machine (MLELM) and the generalized maximum mean discrepancy (GMMD) criterion, abbreviated as MLELM-GMMD. MLELM is first used to transform original mixed-attribute datasets into corresponding deep encoding datasets. Then, the GMMD criterion is applied to check the distribution consistency of the deep encoding datasets. Four experiments have been done to validate the feasibility and effectiveness of MLELM-GMMD, i.e., the impact of MLELM on the amount of information during mixed-attribute data transformation, the impact of MLELM on distributions of mixed-attribute data, the distribution consistencies of RSP and non-RSP data blocks, and the comparison with other measurement methods. Experimental results show that the proposed MLELM-GMMD method can measure the distribution consistency of mixed-attribute datasets more accurately than one-hot encoding-based methods.

Rapid Recognition of Different Sources of Heroin Drugs by Using a Hand-Held Near-Infrared Spectrometer Based on a Multi-Layer Extreme Learning Machine Algorithm

Rapid recognition of the sources of drugs can provide some valuable clues and the basis for determining the nature of the case. A novel recognition method was put forward to identify the sources of heroin drugs rapidly and non-destructively by using a hand-held near infrared (NIR) spectrometer and a multi-layer-extreme learning machine (ML-ELM) algorithm. In contrast to traditional linear discriminant analysis (LDA), support vector machine (SVM) and extreme learning machine (ELM) algorithms, the accuracy, sensitivity and specificity were the highest for the proposed ML-ELM algorithm. The prediction accuracy of the ML-ELM algorithm was 25.33, 20.00, 17.33% higher than that of LDA, SVM and ELM algorithm, respectively, for 4 cases. The ML-ELM models for recognizing the different sources of heroin drugs had the best generalization ability and prediction results. The experimental results indicated that the combination of the hand‑held NIR technology and ML-ELM algorithm can recognize the different sources of heroin drugs rapidly, accurately, and non-destructively on the spot.

Application of Multilayer Extreme Learning Machine for Efficient Building Energy Prediction

Building energy efficiency is vital, due to the substantial amount of energy consumed in buildings and the associated adverse effects. A high-accuracy energy prediction model is considered as one of the most effective ways to understand building energy efficiency. In several studies, various machine learning models have been proposed for the prediction of building energy efficiency. However, the existing models are based on classical machine learning approaches and small datasets. Using a small dataset and inefficient models may lead to poor generalization. In addition, it is not common to see studies examining the suitability of machine learning methods for forecasting the energy consumption of buildings during the early design phase so that more energy-efficient buildings can be constructed. Hence, for these purposes, we propose a multilayer extreme learning machine (MLELM) for the prediction of annual building energy consumption. Our MLELM fuses stacks of autoencoders (AEs) with an extreme learning machine (ELM). We designed the autoencoder based on the ELM concept, and it is used for feature extraction. Moreover, the autoencoders were trained in a layer-wise manner, employed to extract efficient features from the input data, and the extreme learning machine model was trained using the least squares technique for a fast learning speed. In addition, the ELM was used for decision making. In this research, we used a large dataset of residential buildings to capture various building sizes. We compared the proposed MLELM with other machine learning models commonly used for predicting building energy consumption. From the results, we validated that the proposed MLELM outperformed other comparison methods commonly used in building energy consumption prediction. From several experiments in this study, the proposed MLELM was identified as the most efficient predictive model for energy use before construction, which can be used to make informed decisions about, manage, and optimize building design before construction.

Prediction of permeability of highly heterogeneous hydrocarbon reservoir from conventional petrophysical logs using optimized data-driven algorithms

Permeability is an important parameter in the petrophysical study of a reservoir and serves as a key tool in the development of an oilfield. This is while its prediction, especially in carbonate reservoirs with their relatively lower levels of permeability compared to sandstone reservoirs, is a complicated task as it has larger contributions from heterogeneously distributed vugs and fractures. In this respect, the present research uses the data from two wells (well A for modeling and well B for assessing the generalizability of the developed models) drilled into a carbonate reservoir to estimate the permeability using composite formulations based on least square support vector machine (LSSVM) and multilayer extreme learning machine (MELM) coupled with the so-called cuckoo optimization algorithm (COA), particle swarm optimization (PSO), and genetic algorithm (GA). We further used simple forms of convolutional neural network (CNN) and LSSVM for the sake of comparison. To this end, firstly, the Tukey method was applied to identify and remove the outliers from modeling data. In the next step, the second version of the nondominated sorting genetic algorithm (NSGA-II) was applied to the training data (70% of the entire dataset, selected randomly) to select an optimal group of features that most affect the permeability. The results indicated that although including more input parameters in the modeling added to the resultant coefficient of determination (R2) while reducing the error successively, yet the slope of the latter reduction got much slow as the number of input parameters exceeded 4. In this respect, petrophysical logs of P-wave travel time, bulk density, neutron porosity, and formation resistivity were identified as the most effective parameters for estimating the permeability. Evaluation of the results of permeability modeling based on root-mean-square error (RMSE) and R2 shed light on the MELM-COA as the best-performing model in the training and testing stages, as indicated by (RMSE = 0.5600 mD, R2 = 0.9931) and (RMSE = 0.6019 mD, R2 = 0.9919), respectively. The generalizability assessment conducted on the prediction of permeability in well B confirmed the MELM-COA can provide reliable permeability predictions by achieving an RMSE of 0.9219 mD. Consequently, the mentioned methodology is strongly recommended for predicting the permeability with high accuracy in similar depth intervals at other wells in the same field should the required dataset be available.