Search Results (137)

The accurate estimation and prediction of internal states in lithium-ion (Li-Ion) batteries, such as State of Charge (SoC) and Remaining Useful Life (RUL), are vital for optimizing battery performance, safety, and longevity in electric vehicles and other applications. This paper presents a unified, modular, and extensible machine learning (ML) framework designed to address the heterogeneity and complexity of battery state prediction tasks. The proposed framework supports flexible configurations across multiple dimensions, including feature engineering, model selection, and training/testing strategies. It integrates standardized data processing pipelines with a diverse set of ML models, such as a long short-term memory neural network (LSTM), a convolutional neural network (CNN), a feedforward neural network (FFNN), automated machine learning (AutoML), and classical regressors, while accommodating heterogeneous datasets. The framework’s applicability is demonstrated through five distinct use cases involving SoC estimation and RUL prediction using real-world and benchmark datasets. Experimental results highlight the framework’s adaptability, methodological transparency, and robust predictive performance across various battery chemistries, usage profiles, and degradation conditions. This work contributes to a standardized approach that facilitates the reproducibility, comparability, and practical deployment of ML-based battery analytics. Full article

Gait, the fundamental form of human locomotion, occurs across diverse environments. The technology for recognizing environmental changes during walking is crucial for preventing falls and controlling wearable robots. This study collected gait data on level ground (LG), ramps, and stairs using a feed-forward neural network (FFNN) to classify the corresponding gait environments. Gait experiments were performed on five non-disabled participants using an inertial measurement unit, a galvanic skin response sensor, and a smart insole. The collected data were preprocessed through time synchronization and filtering, then labeled according to the gait environment, yielding 47,033 data samples. Gait data were used to train an FFNN model with a single hidden layer, achieving a high accuracy of 98%, with the highest accuracy observed on LG. This study confirms the effectiveness of classifying gait environments based on signals acquired from various wearable sensors during walking. In the future, these research findings may serve as basic data for exoskeleton robot control and gait analysis. Full article

This study investigated the application of neural network techniques to predict the mechanical properties of low-carbon hot-rolled steel strips using industrial data. A feedforward neural network (FFNN) model was developed to predict the yield strength (YS), ultimate tensile strength (UTS), and elongation (%EL) based on the chemical composition and processing parameters. For the low-carbon hot-rolled steel strip (C: 0.02–0.06%, Mn: 0.17–0.38%), 435 datasets were utilized with 17 input parameters, including 15 composition elements, finish rolling temperature (FRT), and coil target temperature (CTT). The model was trained using 335 datasets and tested using 100 randomly selected datasets. The optimum network architecture consisted of two hidden layers with 34 neurons each, achieving a mean squared error of 0.014 after 200,000 iterations. The model predictions showed excellent agreement with the actual values, with mean percentage errors of 4.44%, 3.54%, and 4.84% for the YS, UTS, and %EL, respectively. The study further examined the influence of FRT and CTT on mechanical properties, demonstrating that FRT has more complex effects on mechanical properties than CTT. The model successfully predicted property variations with different processing parameters, thereby providing a valuable tool for alloy design and process optimization in steel manufacturing. Full article

Vertical ground reaction force (vGRF) plays an important role in the study of running-related injuries (RRIs). This study explores the synchronization method between inertial measurement unit (IMU) and vGRF data of running and develops ANN models to accurately predict vGRF. Fifteen runners participated in this study. Acceleration data and vGRF values of eight rearfoot strikers and seven forefoot strikers running at 12, 14, and 16 km/h were collected by a single IMU and an instrumented treadmill. The sliding time window synchronization (STWS) algorithm was developed to sync IMU data with vGRF data. The wavelet neural network model (WNN) and feed-forward neural network model (FFNN) were adapted to predict vGRF using three-axis or sagittal-axis acceleration data in the stance phase, respectively. One rearfoot striker and one forefoot striker were randomly selected as a test set, while the other participants formed training sets. After synchronization, mean absolute errors for stride time of the IMU and vGRF data were less than 11.2 ms. The coefficient of multiple correlations for vGRF measured curves and predicted curves was more than 0.97. The normalized root mean square errors (NRMSEs) between two curves were 4.6~9.2%, and R² was 0.93~0.99. For peak vGRF, the NRMSEs were 1.6~8.2%, except for rearfoot strike runners at 16 km/h using the FFNN model (10.7% and 11.1%). The Bland–Altman plots indicate that the errors for both the WNN and FFNN models are within acceptable limits. The STWS algorithm can effectively achieve the data synchronization between the IMU and the force plate during running. Both WNN and FFNN models demonstrated good accuracy and agreement in predicting vGRF. Using sagittal-axis acceleration data may be an ideal model with good prediction accuracy and less input data. This work provides direction for developing ANN models of personalized monitoring of lower limb load. Full article

The accurate real-time estimation of the remaining useful life (RUL) of lithium-polymer (LiPo) batteries is a critical enabler for ensuring the safety, reliability, and operational efficiency of unmanned aerial vehicles (UAVs). Nevertheless, achieving such prognostics on resource-constrained embedded platforms remains a considerable technical challenge. This study proposes an end-to-end TinyML-based framework that integrates embedded sensor data fusion with an optimized feedforward neural network (FFNN) model for efficient RUL estimation under strict hardware limitations. The system collects voltage, discharge time, and capacity measurements through a lightweight data fusion pipeline and leverages the Edge Impulse platform with the EON™Compiler for model optimization. The trained model is deployed on a dual-core ARM Cortex-M0+ Raspberry Pi RP2040 microcontroller, communicating wirelessly with a LabVIEW-based visualization system for real-time monitoring. Experimental validation on an 80-gram UAV equipped with a 1100 mAh LiPo battery demonstrates a mean absolute error (MAE) of 3.46 cycles and a root mean squared error (RMSE) of 3.75 cycles. Model testing results show an overall accuracy of $98.82\%$, with a mean squared error (MSE) of 55.68, a mean absolute error (MAE) of 5.38, and a variance score of 0.99, indicating strong regression precision and robustness. Furthermore, the quantized (int8) version of the model achieves an inference latency of 2 ms, with memory utilization of only 1.2 KB RAM and 11 KB flash, confirming its suitability for real-time deployment on resource-constrained embedded devices. Overall, the proposed framework effectively demonstrates the feasibility of combining embedded sensor data fusion and TinyML to enable accurate, low-latency, and resource-efficient real-time RUL estimation for UAV battery health management. Full article

Maintaining optimal microclimatic conditions within greenhouses represents a significant challenge in modern agricultural contexts, where prediction systems play a crucial role in regulating temperature and humidity, thereby enabling timely interventions to prevent plant diseases or adverse growth conditions. In this work, we propose a novel approach which integrates a cascaded Feed-Forward Neural Network (FFNN) with the Granular Computing paradigm to achieve accurate microclimate forecasting and reduced computational complexity. The experimental results demonstrate that the accuracy of our approach is the same as that of the FFNN-based approach but the complexity is reduced, making this solution particularly well suited for deployment on edge devices with limited computational capabilities. Our innovative approach has been validated using a real-world dataset collected from four greenhouses and integrated into a distributed network architecture. This setup supports the execution of predictive models both on sensors deployed within the greenhouse and at the network edge, where more computationally intensive models can be utilized to enhance decision-making accuracy. Full article

As the solar share in energy generation is expanding globally, solar nowcasting is becoming increasingly important for the efficient and economical management of the power grid. This study leveraged the spatial context provided by all-sky images (ASI) in addition to the meteorological records for improved nowcasting of global, direct, and diffuse irradiance components. The proposed methodology consists of two branches for processing the multimodal data of ASIs and meteorological data. Due to its capability of understanding the overall characteristics of the image through self-attention, a vision transformer is utilized for the image branch while normal dense layers process the tabular meteorological data. The proposed architecture is compared against the baselines of the Ineichen clear sky model, a feedforward neural network (FFNN) where cloud coverage is computed from the ASIs by a simple color-channel threshold algorithm, and a hybrid of FFNN and U-Net model, which replaces the color threshold algorithm with fully convolutional layers for cloud segmentation. The models are trained, validated, and tested using the quality-assured ground-truth data collected in Ulaanbaatar, Mongolia, from May to August 2024, under one-minute intervals with a random split of 70%, 15%, and 15%. Our approach exhibits superior performance to baselines with a significantly lower mean absolute error (MAE) of 15–33 W/m² and root mean square error (RMSE) of 26–72 W/m², thus potentially aiding grid operators’ decision-making in real-time. Full article

This study introduces an innovative Emotional Artificial Neural Network (EANN) model to predict ship fuel consumption with high accuracy, addressing the challenges posed by complex environmental conditions and operational variability. This research examines the impact of climate change on maritime operations and fuel efficiency by analyzing climatic variables such as wave period, wind speed, and sea-level rise. The model’s performance is assessed using two ship types (bulk carrier and container ship with max 60,000 dead weight tonnage (DWT)) under various climate scenarios. A comparative analysis demonstrates that the EANN model significantly outperforms the conventional Feedforward Neural Network (FFNN) in predictive accuracy. For bulk carriers, the EANN achieved a Root Mean Squared Error (RMSE) of 5.71 tons/day during testing, compared to 9.91 tons/day for the FFNN model. Similarly, for container ships, the EANN model achieved an RMSE of 5.97 tons/day, significantly better than the FFNN model’s 10.18 tons/day. A sensitivity analysis identified vessel speed as the most critical factor, contributing 33% to the variance in fuel consumption, followed by engine power and current speed. Climate-change simulations showed that fuel consumption increases by an average of 22% for bulk carriers and 19% for container ships, highlighting the importance of operational optimizations. This study emphasizes the efficacy of the EANN model in predicting fuel consumption and optimizing ship performance. The proposed model provides a framework for improving energy efficiency and supporting compliance with International Maritime Organization Standards (IMO) environmental standards. Meanwhile, the Carbon Intensity Indicator (CII) evaluation results emphasize the urgent need for measures to reduce carbon emissions to meet the IMO’s 2030 standards. Full article

The primary objective of this study is to evaluate and compare the classification performance of feed-forward neural networks (FFNNs) and one-dimensional convolutional neural networks (1D-CNNs) on magnetoencephalography (MEG) signals from epileptic patients. MEG signals were recorded using the NEUROMAG-122 whole-brain superconducting quantum interference device (SQUID), installed, and operated in our laboratory. The dataset comprised over 5000 MEG segments, each one with a duration of 1 s and sampled at a frequency of 256 Hz. Each segment was classified by expert neurologists as either epileptic or non-epileptic. The FFNN with five hidden layers demonstrated promising results, achieving a classification accuracy of approximately 92%. The 1D-CNN, utilizing four layers, achieved an accuracy of 90.4%, with a significantly reduced training time. Building on these findings, the study’s secondary objective was to enhance the artificial neural network (ANN) model by incorporating transfer learning–stacked generalization for FFNN in various configurations. These enhancements successfully improved the performance of the pretrained network by approximately 1%. Full article

Recently, fault detection has played a crucial role in ensuring the safety and reliability of inverter operation. Switch failures are primarily classified into Open-Circuit (OC) and short-circuit faults. While OC failures have limited negative impacts, prolonged system operation under such conditions may lead to further malfunctions. This paper demonstrates the effectiveness of employing Artificial Intelligence (AI) approaches for detecting single OC faults in a Packed E-Cell (PEC) inverter. Two promising strategies are considered: Random Forest Decision Tree (RFDT) and Feed-Forward Neural Network (FFNN). A comprehensive literature review of various fault detection approaches is first conducted. The PEC inverter’s modulation scheme and the significance of OC fault detection are highlighted. Next, the proposed methodology is introduced, followed by an evaluation based on five performance metrics, including an in-depth comparative analysis. This paper focuses on improving the robustness of fault detection strategies in PEC inverters using MATLAB/Simulink software. Simulation results show that the RFDT classifier achieved the highest accuracy of 93%, the lowest log loss value of 0.56, the highest number of correctly predicted estimations among the total samples, and nearly perfect ROC and PR curves, demonstrating exceptionally high discriminative ability across all fault categories. Full article

Heavy metals, extensively used in various industrial applications, are among the most significant environmental pollutants due to their hazardous effects on human health and other living organisms. Removing these pollutants from the environment is essential. In this study, activated carbon (AC) (Carbon C) was employed to eliminate Pb(II) from water. The optimal removal conditions were determined as follows: a 50 mg dose of activated carbon, an initial Pb(II) concentration of 100 mg/L, pH 4, a temperature of 30 °C, and a contact time of 60 min Under these conditions, activated carbon achieved a Pb(II) removal efficiency of approximately 97.86%. The adsorption data for Pb(II) closely aligned with the 2nd-order kinetic model, and the equilibrium data were effectively described by the Langmuir isotherm equation. The maximum adsorption capacity of Pb(II), as determined by the Langmuir model, was 48.75 mg/g. These methods were successfully applied to remove Pb(II) from various environmental and industrial wastewater samples. To accurately predict the percentage of Pb(II) removal based on parameters such as pollutant type, carbon dosage, pH, initial concentration, temperature, and treatment duration, feed-forward neural networks (FFNNs) were utilized. The FFNN model demonstrated outstanding predictive accuracy, achieving a root mean square error (RMSE) of 0.03 and an R² value of 0.996. Full article

The daily use of devices generating electric and magnetic fields has led to potential human overexposure in home and work environments. This paper assesses the possible effects of electric fields on human health at low and high frequencies. It presents an electronic monitoring device that captures the incidence of specific absorption rate (SAR) and temperature variation (∆T) on the human body. The system transmits data to a cloud platform, where a feedforward neural network (FFNN) processes the received information. SAR and surface temperature values are detected in an indoor environment, monitoring stationary and moving subjects. The results effectively assess temperature distribution due to electromagnetic fields. The prototype detected temperature peaks and high SAR values when the subjects remained motionless. Predictive analysis confirms the need for workplaces with materials shielding external electromagnetic signals and attenuating internal sources. Moderate mobile phone use could lower SAR and temperature values. Full article

The Senise red pepper, known as peperone crusco, is a protected geographical indication (PGI) product from Basilicata, Italy, traditionally consumed dried. Producers use semi-open greenhouses to meet PGI standards, but significant losses are caused by rot from microorganisms thriving in high moisture, temperature, and humidity, which also encourage pest infestations. To minimize losses, a low-cost alert system was developed. The study, conducted in summer 2022 and 2023, used external parameters from the ALSIA Senise weather station and internal sensors monitoring the air temperature and humidity inside the greenhouse. Since rot is complex and difficult to model, an artificial intelligence (AI)-based approach was adopted. A feed forward neural network (FFNN) estimated greenhouse climate conditions as if it were empty, comparing them with actual values when peppers were present. This revealed the most critical period was the first 3–4 days after introduction and identified a critical air relative humidity threshold. The system could also predict microclimatic parameters inside the greenhouse with red peppers, issuing warnings one hour before risk conditions arose. In 2023, it was tested by comparing predicted values with previously identified thresholds. When critical levels were exceeded, greenhouse operators were alerted to adjust conditions. In 2023, pepper rot decreased. Full article

Background: Non-small cell lung cancer (NSCLC) has been the most common cancer globally in the recent decade. CT is the most common imaging modality for the initial diagnosis of NSCLC. The gold standard for definitive diagnosis is the histological evaluation of a biopsy or surgical sample, which usually requires a long processing time for the confirmation of diagnosis. This study aims to develop artificial intelligence models to predict overall staging based on patient demographics and radiomics retrieved from the initial CT images, so as to prioritize later-stage patients for histology evaluation to facilitate cancer diagnosis. Method: Two cohorts of NSCLC patient datasets were utilized for this study. The NSCLC-radiomics dataset from The Cancer Imaging Archive (TCIA) was divided into 70% for the training group and 30% for the internal testing group. Another cohort from a local hospital was collected for the an external testing group. Patient demographics and 107 radiomic features were retrieved from the gross tumor volume delineated by clinical oncologists on CT images. Artificial neural networks were used to build models for NSCLC overall staging (stage I, II, or III) prediction. Four traditional classifiers were also adopted to build models for comparison. Result: The proposed feed-forward neural network (FFNN) model showed good performance in predicting overall staging with an accuracy of 88.84%, 76.67%, and 74.52% in overall accuracies in validation, internal cohort testing, and external cohort testing, respectively. The sensitivity and specificity are balanced in all the stages, with average precision and F1 score in each of the stages. Conclusion: The FFNN demonstrated good performance in overall staging prediction for NSCLC patients. It has the benefit of predicting multiple overall stages in a single model. The software required and the proposed model are simple. It can be operated on a general-purpose computer in the radiology department. The application will eventually be used as a prediction tool to prioritize the biopsy or surgery sample for histological analysis and molecular investigation, thus shortening the time for diagnosis by pathologists, which supports the triage of patients for further testing. Full article

Atmospheric correction is essential in remote sensing, as it reduces the effects of light absorption and scattering by suspended particles and gases, enabling accurate surface reflectance computation from the observed Top-of-Atmosphere (TOA) reflectance. Each satellite sensor requires a customized atmospheric correction processor due to its unique system characteristics. Currently, PeruSAT-1, the first Peruvian remote sensing satellite, does not include this capability in its image processing pipeline, which poses challenges for its effectiveness in defense, security, and natural disaster management applications. This research investigated pixel-based machine learning methods for atmospheric correction of PeruSAT-1, using Sentinel-2 harmonized Bottom-of-Atmosphere (BOA) surface reflectance images as a benchmark, alongside additional atmospheric, terrain, and acquisition parameters. A robust dataset was developed to align data across temporal, spatial, geometric, and contextual conditions. Experimental results showed R² values between 0.886 and 0.938, and RMSE values ranging from 0.009 to 0.025 compared to the benchmarks. Among the models tested, the Feedforward Neural Network (FFNN) using the Leaky ReLU activation function achieved the best overall performance. These findings confirm the robustness of this approach, offering a scalable methodology for satellites with similar characteristics and establishing a foundation for a customized atmospheric correction pipeline for PeruSAT-1. Future work will focus on diversifying the dataset across spectral and seasonal conditions and optimizing the modeling to address challenges in shorter wavelengths and high-reflectance surfaces. Full article