Extreme Learning Machine Research Articles

An improved golden jackal optimization-hybrid kernel extreme learning machine approach for fault diagnosis of chillers

Abstract In the context of the dual carbon target, reducing energy consumption and carbon emissions in the building sector is crucial. Chillers, being the primary source of energy consumption in building equipment, can significantly increase energy use if they fail. To enhance the detection rate of early chiller failures, this paper proposes an improved chiller fault diagnosis model based on the golden jackal optimization algorithm (IGJO) and the hybrid kernel extremum learning machine (HKELM). To address the slow convergence speed and local optimization issues of the original golden jackal algorithm (GJO), we introduce four strategies: improved sine chaotic mapping, the simplex method, a fusion of the golden sinusoidal formula, and Cauchy’s variant. These strategies form the IGJO algorithm, which improves convergence speed and global optimization ability. Next, the IGJO algorithm is used to optimize the hyperparameters of the HKELM for fast search and efficient chiller fault detection. Finally, we simulate the performance of the proposed model using ASHRAE RP-1043. The results show that the IGJO-HKELM model achieves a chiller fault diagnosis accuracy of 99.83%, demonstrating a clear advantage over other algorithms.

Nanosatellite Autonomous Navigation via Extreme Learning Machine Using Magnetometer Measurements

This work presents an algorithm to perform autonomous navigation in spacecraft using onboard magnetometer data during GPS outages. An Extended Kalman Filter (EKF) exploiting magnetic field measurements is combined with a Single-Hidden-Layer Feedforward Neural Network (SLFN) trained via the Extreme Learning Machine to improve the accuracy of the state estimate. The SLFN is trained using GPS data when available and predicts the state correction to be applied to the EKF estimates. The CHAOS-7 magnetic field model is used to generate the magnetometer measurements, while a 13th-order IGRF model is exploited by the EKF. Tests on simulated data showed that the algorithm improved the state estimate provided by the EKF by a factor of 2.4 for a total of 51 days when trained on 5 days of GPS data.

Financial Time Series Prediction Using Pelican Optimized Extreme Learning Machine with Reduced Weights

Financial Time Series Prediction Using Pelican Optimized Extreme Learning Machine with Reduced Weights

State estimation with quantum extreme learning machines beyond the scrambling time

Quantum extreme learning machines (QELMs) leverage untrained quantum dynamics to efficiently process information encoded in input quantum states, avoiding the high computational cost of training more complicated nonlinear models. On the other hand, quantum information scrambling (QIS) quantifies how the spread of quantum information into correlations makes it irretrievable from local measurements. Here, we explore the tight relation between QIS and the predictive power of QELMs. In particular, we show efficient state estimation is possible even beyond the scrambling time, for many different types of dynamics — in fact, we show that in all the cases we studied, the reconstruction efficiency at long interaction times matches the optimal one offered by random global unitary dynamics. These results offer promising venues for robust experimental QELM-based state estimation protocols, as well as providing novel insights into the nature of QIS from a state estimation perspective.

An efficient energy management method for plug-in hybrid electric vehicles based on multi-source and multi-feature velocity prediction and improved extreme learning machine

The dynamic changes in vehicle status and road conditions pose a challenge to plug-in hybrid electric vehicles in achieving real-time optimal energy management, leading to increased energy consumption and pollutant emissions. In this paper, an efficient energy management control architecture is proposed based on the principle of model predictive control (MPC) to balance optimization effectiveness and operational efficiency. First, a multi-source and multi-feature velocity predictor based on long short-term memory networks is established using information from the driver, vehicle, and road. Subsequently, the latest bio-inspired algorithm, namely the white shark optimizer, is utilized to optimize the input weights and hidden biases of the extreme learning machine. Additionally, the concept of environmental treatment cost is introduced, and a comprehensive driving cost estimation model based on the enhanced extreme learning machine is established to improve computational efficiency by replacing complex nonlinear operations in optimization. The results demonstrate that the proposed velocity predictor and cost estimation model outperform existing methods. The proposed control framework reduces driving costs and emissions by 73.4 % and 78.4 %, respectively, compared to the rule-based method. Meanwhile, the optimization performance of the proposed method within the 5 s prediction horizon is 19.6 % better than that of the conventional MPC method. Although the inclusion of the cost estimation model leads to a slight 4.5 % reduction in optimization performance compared to the method using only the proposed velocity predictor, this is balanced by a 64.6 % improvement in computational efficiency, allowing the proposed method to maintain a significant overall advantage over conventional MPC.

Deterministic Weight Modification-based Extreme Learning Machine for Stock Price Prediction

Background: The prediction of the stock price is considered to be one of the most fascinating and important research and patent topics in the financial sector. Aims: Making more accurate predictions is a difficult and significant task because the financial industry supports investors and the national economy. Objectives: The DWM is used to adjust the connection weights and biases to enhance prediction precision and convergence rate. DWM was proposed as a method to reduce system error by changing the weights of various levels. The methods for predictable changes in weight were provided together with the computational difficulty. Methods: An extreme learning machine (ELM) is a fast-learning method for training a singlehidden layer neural network (SLFN). However, the model's learning process is ineffective or incomplete due to the randomly chosen weights and biases of the input's hidden layers. Hence, this article presents a deterministic weight modification (DWM) based ELM called DWM-ELM for predicting the stock price. Results: The calculated results showed that DWM-ELM had the best predictive performance, with RMSE (root mean square error) of 0.0096, MAE (mean absolute error) of 0.0563, 0.0428, MAPE (mean absolute percentage error) of 1.7045, and DS (Directional Symmetry) of 89.34. Conclusion: The experimental results showed that, in comparison to other well-known prediction algorithms, the suggested DWM+ELM prediction model offers better prediction performance.

Deep photonic reservoir computing based on a distributed feedback laser array

Photonic reservoir computing (RC) is emerging as a competitive candidate for ultra-fast and energy-efficient neuromorphic computing, noted for its physical compatibility and straightforward training process. However, this widely appreciated form of machine learning typically employs only a single hidden layer with a feedback loop (FL), which essentially constrains the capability for complex task processing and poses challenges to integration. Here, we present an integrated deep photonic RC setup based on a distributed feedback laser array. This setup eliminates the dependence of the FL through the pre-processing termed quasi-convolution coding (QC), where the QC-based RC (QRC) significantly reduces the complexity of the network under the ensured performance. In particular, we extend this QRC into an on-chip deep structure, where this deep QRC (DQRC) exhibits remarkable superiority to its shallow counterparts. We demonstrate the augmented capabilities of DQRC through both simulations and experimental comparisons with QRC, extreme learning machine, and time-delay RC in time-dependent tasks. Moreover, the proposed deep configuration also excels in static image processing. We confirm the potential for simplifying the hardware implementations of deep neural networks, revealing a promising solution to satisfy the urgent demand for high-integration brain-inspired systems.

Extended physics-informed extreme learning machine for linear elastic fracture mechanics

Extended physics-informed extreme learning machine for linear elastic fracture mechanics

Secretary bird optimization algorithm based on quantum computing and multiple strategies improvement for KELM diabetes classification.

The classification of chronic diseases has long been a prominent research focus in the field of public health, with widespread application of machine learning algorithms. Diabetes is one of the chronic diseases with a high prevalence worldwide and is considered a disease in its own right. Given the widespread nature of this chronic condition, numerous researchers are striving to develop robust machine learning algorithms for accurate classification. This study introduces a revolutionary approach for accurately classifying diabetes, aiming to provide new methodologies. An improved Secretary Bird Optimization Algorithm (QHSBOA) is proposed in combination with Kernel Extreme Learning Machine (KELM) for a diabetes classification prediction model. First, the Secretary Bird Optimization Algorithm (SBOA) is enhanced by integrating a particle swarm optimization search mechanism, dynamic boundary adjustments based on optimal individuals, and quantum computing-based t-distribution variations. The performance of QHSBOA is validated using the CEC2017 benchmark suite. Subsequently, QHSBOA is used to optimize the kernel penalty parameter [Formula: see text] and bandwidth [Formula: see text] of the KELM. Comparative experiments with other classification models are conducted on diabetes datasets. The experimental results indicate that the QHSBOA-KELM classification model outperforms other comparative models in four evaluation metrics: accuracy (ACC), Matthews correlation coefficient (MCC), sensitivity, and specificity. This approach offers an effective method for the early diagnosis and prediction of diabetes.

Prediction of Color Change in Heat-Treated Wood Based on Improved Zebra Algorithm Optimized Deep Hybrid Kernel Extreme Learning Machine Model (IZOA-DHKELM)

In this study, an Improved Zebra Optimization Algorithm (ZOA) is proposed based on the search mechanism of the Sparrow Optimization Algorithm (SSA), the perturbation mechanism of the Particle Swarm Algorithm (PSO), and the adaptive function. Then, Improved Zebra Optimization Algorithm (IZOA) was used to optimize the Deep Hybrid Kernel Extreme Learning Machine Model (DHKELM), and the IZOA-DHKELM was obtained. The model has been used to predict the color of heat-treated wood for different species, temperatures, times, media, and profile types. In this article, the original DHKELM and the ZOA-DHKELM were compared to verify the validity and accuracy of the model. The results indicated that the IZOA-DHKELM decreased the mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) by 56.2%, 67.4%, and 34.2%, respectively, while enhancing the coefficient of determination, R2, to 0.9952 compared to the ZOA-DHKELM. This demonstrated that the model was significantly optimized, with improved generalization ability and prediction accuracy. It can better meet the actual engineering needs.

Classification of peanut variety based on hyperspectral imaging and improved extreme learning machine

Classification of peanut variety based on hyperspectral imaging and improved extreme learning machine

Research on Evaluation of Hotel Digitization Effect Assisted by Extreme Learning Machine in Artificial Intelligence Environment

Hotels streamline operations, provide more personalized service, streamline communication, automate procedures, and gather feedback for ongoing development using modern digital technologies like customer care chatbots, mobile guest applications, and property management systems. Batch, sequential, and incremental learning all make heavy use of the Extreme Learning Machine (ELM) because of its efficient and rapid learning rate, rapid convergence, strong generalizability, and simplicity of implementation. By automating repetitive tasks, Artificial Intelligence (AI) improves the hospitality sector’s efficiency and quality on service it provides to customers and employees. Using AI-driven automated systems, hotels may improve inventory management, track and manage energy use, and simplify cleaning schedules. The challenging characteristics of smart flood detection and alert systems are that they lack expertise, streamline operations and provide personalized guest services. Hence, in the proposed method, the Artificial Intelligence Environment enabled Wireless Sensor Network (AIE-WSN), which integrates a hotel digitization effect in an ELM. In addition, case studies and expert interviews provide practical insights into how hotels have used and benefited from financial management systems driven by AI. The research shows that hotels may greatly benefit from AI tools, including personalized pricing strategies, spending monitoring systems, fraud detection algorithms, and predictive analytics when making financial decisions through WSN. The hotel industry is undergoing digitalization, adopting AI-WSN solutions to track products’ provenance and monitor their history, among the many benefits of the proposed approach. It can help with supply chain management and adopting more efficient practices.

Random resistive memory-based deep extreme point learning machine for unified visual processing

Visual sensors, including 3D light detection and ranging, neuromorphic dynamic vision sensor, and conventional frame cameras, are increasingly integrated into edge-side intelligent machines. However, their data are heterogeneous, causing complexity in system development. Moreover, conventional digital hardware is constrained by von Neumann bottleneck and the physical limit of transistor scaling. The computational demands of training ever-growing models further exacerbate these challenges. We propose a hardware-software co-designed random resistive memory-based deep extreme point learning machine. Data-wise, the multi-sensory data are unified as point set and processed universally. Software-wise, most weights are exempted from training. Hardware-wise, nanoscale resistive memory enables collocation of memory and processing, and leverages the inherent programming stochasticity for generating random weights. The co-design system is validated on 3D segmentation (ShapeNet), event recognition (DVS128 Gesture), and image classification (Fashion-MNIST) tasks, achieving accuracy comparable to conventional systems while delivering 6.78 × /21.04 × /15.79 × energy efficiency improvements and 70.12%/89.46%/85.61% training cost reductions.

A Spatiotemporal Fuzzy Modeling Approach Combining Automatic Clustering and Hierarchical Extreme Learning Machines for Distributed Parameter Systems

Modeling distributed parameter systems (DPSs) is challenging due to their strong nonlinearity and spatiotemporal coupling. In this study, a three-dimensional fuzzy modeling method combining genetic algorithm (GA)-based automatic clustering and hierarchical extreme learning machine (HELM) is proposed for DPS modeling. The method utilizes GA-based automatic clustering to learn the premise part of 3D fuzzy rules, while HELM is employed to learn spatial basis functions and construct a complete fuzzy rule base. This approach effectively captures the spatiotemporal coupling characteristics of the system and mitigates the information loss commonly observed in dimensionality reduction in traditional fuzzy modeling methods. Through experimental verification, the proposed method is successfully applied to a rapid thermal chemical vapor deposition system. The experimental results demonstrate that the method can accurately predict temperature distribution and maintain good robustness under noise and disturbances.

Scalable Multi-Clustering Aggregation Scheme in WSN Using Machine Learning

Due to Limited resource constraints in WSN data packets collide while being routed to sink, redundant data can be eliminated by data aggregation, which minimizes overall amount of data transferred and increases network's lifespan. Minimizing energy consumption and boost data aggregation rate is most crucial factor in WSN. A Scalable Multi-Clustering Aggregation utilizing Machine Learning (SMCA-ML) focuses on data aggregation approach of heterogeneous wireless sensor networks, using neurons as wireless sensor network nodes in a machine learning method. Machine Learning method accumulates the captured data collected by senor nodes and integrates the accumulated data with multi-clustering route. Threshold value of hidden layer and weight of input layer are randomly generated by the proposed method prior to training. This results in an unstable output that affects the efficiency of data aggregation and causes a long delay. More crucially, distinct threshold settings were made in accordance with the features of uneven energy consumption in wireless sensor networks (WSNs) allow data packets more swiftly by setting smaller threshold in far sink with enough energy. To maximize data aggregation, close sink area with tight energy employs a greater threshold. This way, the combination can result in high data fusion, efficient energy consumption, and little delay. The results of simulation suggested that SMCA-ML based data aggregation algorithm can significantly extend lifespan of network, substantially decrease energy consumption, enhance network energy, expand network performance and improve data aggregation efficiency when compared to conventional Stable Election Protocol (SEP), Back Propagation algorithm, and Extreme Learning Machine.

An Ensemble Approach with Evolutionary Algorithm for Hand Posture Classification

Grasping is a fundamental action in daily life and particularly evident during mealtime situations where various grasping actions occur with tableware such as chopsticks, spoons, forks, bowls, and cups, each serving specific purposes. While tableware usage varies across regions and cultures, recognizing grasping actions is crucial for assessing performance in daily activities. In this study, we focus on assessing grasping functionality in terms of tableware usage during meals and propose a method for identifying hand movements. In recent years, there has been a surge in developing approaches for hand pose estimation and gesture recognition using deep learning. However, these approaches encounter common challenges, including the need for large-scale datasets, hyperparameter tuning, significant time and computational costs, and limited applicability to incremental learning. To address these challenges, we propose an ensemble approach employing extreme learning machines to recognize grasp postures. In addition, we apply spatiotemporal modeling to extract the relationship between grasp postures and the surrounding tools during mealtimes.

SEMG Signal-Based Human Lower-Limb Intention Recognition Algorithm Using Improved Extreme Learning Machine

To address the difficulty in identifying human lower-limb movement intentions, low accuracy of classification models, and weak generalization ability, this study proposes a motion intention recognition method that combines an improved pelican optimization algorithm (IPOA) and a hybrid kernel extreme learning machine (HKELM). First, we collect the surface electromyography (sEMG) signals of subjects in six motion modes and perform feature parameter extraction under non-ideal conditions. On this basis, we establish a dataset of the relationships between the feature parameters and gait movements. Second, we build a motion intention classification model based on relational data using the HKELM to solve the problems of low modeling accuracy and weak generalizability. Third, the IPOA is used to optimize the parameters related to the HKELM, and a differential evolution algorithm is introduced to improve the population quality and prevent the algorithm from falling into a local optimal solution. The experimental results show that the IPOA exhibits better optimization accuracy and convergence speed for four classical benchmark functions. Its average classification accuracy, average classification recall, and average F-value are 94.45%, 94.47%, and 94.46%, respectively, which are significantly higher than those of other intention recognition algorithms. Therefore, the proposed method has high classification accuracy and generalization performance.

Advanced Efficient Feature Selection Integrating Augmented Extreme Learning Machine and Particle Swarm Optimization for Predicting Nitrogen Use Efficiency and Yield in Corn

Advanced Efficient Feature Selection Integrating Augmented Extreme Learning Machine and Particle Swarm Optimization for Predicting Nitrogen Use Efficiency and Yield in Corn

Optimizing papermaking wastewater treatment by predicting effluent quality with node-level capsule graph neural networks.

Papermaking wastewater consists of a sizable amount of industrial wastewater; hence, real-time access to precise and trustworthy effluent indices is crucial. Because wastewater treatment processes are complicated, nonlinear, and time-varying, it is essential to adequately monitor critical quality indices, especially chemical oxygen demand (COD). Traditional models for predicting COD often struggle with sensitivity to parameter tuning and lack interpretability, underscoring the need for improvement in industrial wastewater treatment. In this manuscript, an optimized papermaking wastewater treatment method is proposed that predicts effluent quality using node-level capsule graph neural networks (PWWT-PEQ-NLCGNN). To improve the accuracy of predicting important effluent COD quality indices, the NLCGNN weight parameters are optimized using the hermit crab optimization (HCO) algorithm. The performance of the proposed PWWT-PEQ-NLCGNN technique demonstrated improvements over existing techniques. Specifically, the proposed strategy achieved 30.53%, 23.34%, and 32.64% higher accuracy; 20.53%, 25.34%, and 29.64% higher precision; and 20.53%, 25.34%, and 29.64% higher sensitivity compared to the water quality prediction model using Gaussian process regression based on deep learning for carbon neutrality in papermaking wastewater treatment system (WQP-GPR-DL-CLPWWTS), the prediction of effluent quality in papermaking wastewater treatment processes using dynamic kernel-based extreme learning machine (POEQ-PWWTP-DKBELM), and the quality-related monitoring of papermaking wastewater treatment processes using dynamic multi-block partial least squares (QRM-PWWTP-DMPLS). These results highlight the potential of the PWWT-PEQ-NLCGNN method for enabling timely and accurate monitoring of wastewater treatment processes.

Short-Term Photovoltaic Power Forecasting Based on the VMD-IDBO-DHKELM Model

A short-term photovoltaic power forecasting method is proposed, integrating variational mode decomposition (VMD), an improved dung beetle algorithm (IDBO), and a deep hybrid kernel extreme learning machine (DHKELM). First, the weather factors less relevant to photovoltaic (PV) power generation are filtered using the Spearman correlation coefficient. Historical data are then clustered into three categories—sunny, cloudy, and rainy days—using the K-means algorithm. Next, the original PV power data are decomposed through VMD. A DHKELM-based combined prediction model is developed for each component of the decomposition, tailored to different weather types. The model’s hyperparameters are optimized using the IDBO. The final power forecast is determined by combining the outcomes of each individual component. Validation is performed using actual data from a PV power plant in Australia and a PV power station in Kashgar, China demonstrates. Numerical evaluation results show that the proposed method improves the Mean Absolute Error (MAE) by 3.84% and the Root-Mean-Squared Error (RMSE) by 3.38%, confirming its accuracy.