Search

Forecasting has extreme importance in industry due to the numerous competitive advantages that it provides, allowing to foresee what might happen and adjust management decisions accordingly. Industries increasingly use sensors, which allow for large-scale data collection. Big datasets enable training, testing and application of complex predictive algorithms based on machine learning models. The present paper focuses on predicting values from sensors installed on a pulp paper press, using data collected over three years. The variables analyzed are electric current, pressure, temperature, torque, oil level and velocity. The results of XGBoost and artificial neural networks, with different feature vectors, are compared. They show that it is possible to predict sensor data in the long term and thus predict the asset's behaviour several days in advance. [ABSTRACT FROM AUTHOR]

Demand response management (DRM) plays a crucial role in the prospective development of smart grids. The precise estimation of electricity demand for individual houses is vital for optimizing the operation and planning of the power system. Accurate forecasting of the required components holds significance as it can substantially impact the final cost, mitigate risks, and support informed decision-making. In this paper, a forecasting approach employing neural networks for smart grid demand-side management is proposed. The study explores various enhanced artificial neural network (ANN) architectures for forecasting smart grid consumption. The performance of the ANN approach in predicting energy demands is evaluated through a comparison with three statistical models: a time series model, an auto-regressive model, and a hybrid model. Experimental results demonstrate the ability of the proposed neural network framework to deliver accurate and reliable energy demand forecasts. [ABSTRACT FROM AUTHOR]

Nowadays, trajectory control is a significant issue for unmanned micro aerial vehicles (MAVs) due to large disturbances such as wind and storms. Trajectory control is typically implemented using a proportional–integral–derivative (PID) controller. In order to achieve high accuracy in trajectory tracking, it is essential to set the PID gain parameters to optimum values. For this reason, separate gain values are set for roll, pitch and yaw movements before autonomous flight in quadrotor systems. Traditionally, this adjustment is performed manually or automatically in autotune mode. Given the constraints of narrow orchard corridors, the use of manual or autotune mode is neither practical nor effective, as the quadrotor system has to fly in narrow apple orchard corridors covered with hail nets. These reasons require the development of an innovative solution specific to quadrotor vehicles designed for constrained areas such as apple orchards. This paper recognizes the need for effective trajectory control in quadrotors and proposes a novel neural network-based approach to tuning the optimal PID control parameters. This new approach not only improves trajectory control efficiency but also addresses the unique challenges posed by environments with constrained locational characteristics. Flight simulations using the proposed neural network models have demonstrated successful trajectory tracking performance and highlighted the superiority of the feed-forward back propagation network (FFBPN), especially in latitude tracking within 7.52745 × 10−5 RMSE trajectory error. Simulation results support the high performance of the proposed approach for the development of automatic flight capabilities in challenging environments. [ABSTRACT FROM AUTHOR]

During the development of the nervous system, neuronal cells extend axons and dendrites that form complex neuronal networks, which are essential for transmitting and processing information. Understanding the physical processes that underlie the formation of neuronal networks is essential for gaining a deeper insight into higher-order brain functions such as sensory processing, learning, and memory. In the process of creating networks, axons travel towards other recipient neurons, directed by a combination of internal and external cues that include genetic instructions, biochemical signals, as well as external mechanical and geometrical stimuli. Although there have been significant recent advances, the basic principles governing axonal growth, collective dynamics, and the development of neuronal networks remain poorly understood. In this paper, we present a detailed analysis of nonlinear dynamics for axonal growth on surfaces with periodic geometrical patterns. We show that axonal growth on these surfaces is described by nonlinear Langevin equations with speed-dependent deterministic terms and gaussian stochastic noise. This theoretical model yields a comprehensive description of axonal growth at both intermediate and long time scales (tens of hours after cell plating), and predicts key dynamical parameters, such as speed and angular correlation functions, axonal mean squared lengths, and diffusion (cell motility) coefficients. We use this model to perform simulations of axonal trajectories on the growth surfaces, in turn demonstrating very good agreement between simulated growth and the experimental results. These results provide important insights into the current understanding of the dynamical behavior of neurons, the self-wiring of the nervous system, as well as for designing innovative biomimetic neural network models. [ABSTRACT FROM AUTHOR]

Simple Summary: Due to various causes, such as the thinning of the ozone layer, climate change, or the fashion for artificial tanning, the incidence of skin cancer has increased recently. Early detection of cancerous skin lesions is the only chance to prevent dangerous developments. This paper proposes two models of support systems for the detection of some skin lesions including dangerous melanoma, based on two different methods of assembling neural networks and making global decisions by fusing individual decisions. The system based on the fusion of some intermediate binary classification subsystems is a new solution and offers an accuracy of 91.04%, better than that offered by the classic system based on the vote with multiple weights of the constituent networks. In addition, it is found that the individual performance indicators depend on the type of skin lesion. For example, the F1 score varied from 81.36% to 94.17%. Today, skin cancer, and especially melanoma, is an increasing and dangerous health disease. The high mortality rate of some types of skin cancers needs to be detected in the early stages and treated urgently. The use of neural network ensembles for the detection of objects of interest in images has gained more and more interest due to the increased performance of the results. In this sense, this paper proposes two ensembles of neural networks, based on the fusion of the decisions of the component neural networks for the detection of four skin lesions (basal cancer cell, melanoma, benign keratosis, and melanocytic nevi). The first system is based on separate learning of three neural networks (MobileNet V2, DenseNet 169, and EfficientNet B2), with multiple weights for the four classes of lesions and weighted overall prediction. The second system is made up of six binary models (one for each pair of classes) for each network; the fusion and prediction are conducted by weighted summation per class and per model. In total, 18 such binary models will be considered. The 91.04% global accuracy of this set of binary models is superior to the first system (89.62%). Separately, only for the binary classifications within the system was the individual accuracy better. The individual F1 score for each class and the global system varied from 81.36% to 94.17%. Finally, a critical comparison is made with similar works from the literature. [ABSTRACT FROM AUTHOR]

A fuel cell is an energy conversion device that utilizes hydrogen energy through an electrochemical reaction. Despite their many advantages, such as high efficiency, zero emissions, and fast startup, fuel cells have not yet been fully commercialized due to deficiencies in service life, cost, and performance. Efficient evaluation methods for performance and service life are critical for the design and optimization of fuel cells. The purpose of this paper was to review the application of common machine learning algorithms in fuel cells. The significance and status of machine learning applications in fuel cells are briefly described. Common machine learning algorithms, such as artificial neural networks, support vector machines, and random forests are introduced, and their applications in fuel cell performance prediction and optimization are comprehensively elaborated. The review revealed that machine learning algorithms can be successfully used for performance prediction, service life prediction, and fault diagnosis in fuel cells, with good accuracy in solving nonlinear problems. Combined with optimization algorithms, machine learning models can further carry out the optimization of design and operating parameters to achieve multiple optimization goals with good accuracy and efficiency. It is expected that this review paper could help the reader comprehend the state of the art of machine learning applications in fuel fuels and shed light on further development directions in fuel cell research. [ABSTRACT FROM AUTHOR]

Short-term load forecasting (STLF) is critical for the energy industry. Accurate predictions of future electricity demand are necessary to ensure power systems' reliable and efficient operation. Various STLF models have been proposed in recent years, each with strengths and weaknesses. This paper comprehensively reviews some STLF models, including time series, artificial neural networks (ANNs), regression-based, and hybrid models. It first introduces the fundamental concepts and challenges of STLF, then discusses each model class's main features and assumptions. The paper compares the models in terms of their accuracy, robustness, computational efficiency, scalability, and adaptability and identifies each approach's advantages and limitations. Although this study suggests that ANNs and hybrid models may be the most promising ways to achieve accurate and reliable STLF, additional research is required to handle multiple input features, manage massive data sets, and adjust to shifting energy conditions. [ABSTRACT FROM AUTHOR]

Long series time forecasting has become a popular research direction in recent years, due to the ability to predict weather changes, traffic conditions and so on. This paper provides a comprehensive discussion of long series time forecasting techniques and their applications, using the Informer algorithm model as a framework. Specifically, we examine sequential time prediction models published in the last two years, including the tightly coupled convolutional transformer (TCCT) algorithm, Autoformer algorithm, FEDformer algorithm, Pyraformer algorithm, and Triformer algorithm. Researchers have made significant improvements to the attention mechanism and Informer algorithm model architecture in these different neural network models, resulting in recent approaches such as wavelet enhancement structure, auto-correlation mechanism, and depth decomposition architecture. In addition to the above, attention algorithms and many models show potential and possibility in mechanical vibration prediction. In recent state-of-the-art studies, researchers have used the Informer algorithm model as an experimental control, and it can be seen that the algorithm model itself has research value. The informer algorithm model performs relatively well on various data sets and has become a more typical algorithm model for time series forecasting, and its model value is worthy of in-depth exploration and research. This paper discusses the structures and innovations of five representative models, including Informer, and reviews the performance of different neural network structures. The advantages and disadvantages of each model are discussed and compared, and finally, the future research direction of long series time forecasting is discussed. [ABSTRACT FROM AUTHOR]

This paper presents an approach to estimate demand in the Polish Power System (PPS) using the historical electricity usage of 27 thousand commercial customers, observed between 2016 and 2020. The customer data were clustered and samples as well as features were created to build neural network models. The goal of this research is to analyze if the clustering of customers can help to explain demand in the PPS. Additionally, considering that the datasets available for commercial customers are typically much smaller, it was analyzed what a minimal sample size drawn from the clusters would have to be in order to accurately estimate demand in the PPS. The evaluation and experiments were conducted for each year separately; the results proved that, considering adjusted R2 and mean absolute percentage error, our clustering-based method can deliver a high accuracy in the load estimation. [ABSTRACT FROM AUTHOR]

Multiconnectivity allows user equipment/devices to connect to multiple radio access technologies simultaneously, including 5G, 4G (LTE), and WiFi. It is a necessity in meeting the increasing demand for mobile network services for the 5G and beyond wireless networks, while ensuring that mobile operators can still reap the benefits of their present investments. Multipath TCP (MPTCP) has been introduced to allow uninterrupted reliable data transmission over multiconnectivity links. However, energy consumption is a significant issue for multihomed wireless devices since most of them are battery-powered. This paper employs software-defined networking (SDN) and deep neural networks (DNNs) to manage the energy consumption of devices with multiconnectivity running MPTCP. The proposed method involves two lightweight algorithms implemented on an SDN controller, using a real hardware testbed of dual-homed wireless nodes connected to WiFi and cellular networks. The first algorithm determines whether a node should connect to a specific network or both networks. The second algorithm improves the selection made by the first by using a DNN trained on different scenarios, such as various network sizes and MPTCP congestion control algorithms. The results of our extensive experimentation show that this approach effectively reduces energy consumption while providing better network throughput performance compared to using single-path TCP or MPTCP Cubic or BALIA for all nodes. [ABSTRACT FROM AUTHOR]

Blade icing seriously affects wind turbines' aerodynamic performance and output power. Timely and accurately predicting blade icing status is crucial to improving the economy and safety of wind farms. However, existing blade icing prediction methods cannot effectively solve the problems of unbalanced icing/non-icing data and low prediction accuracy. In order to solve the above problems, this paper proposes a wind turbine blade icing prediction method based on the focal loss function and CNN-Attention-GRU. First, the recursive feature elimination method combined with the physical mechanism of icing is used to extract features highly correlated with blade icing, and a new feature subset is formed through a sliding window algorithm. Then, the focal loss function is utilized to assign more weight to the ice samples with a lower proportion, addressing the significant class imbalance between the ice and non-ice categories. Finally, based on the CNN-Attention-GRU algorithm, a blade icing prediction model is established using continuous 24-h historical data as the input and the icing status of the next 24 h as the output. The model is compared with advanced neural network models. The results show that the proposed method improves the prediction accuracy and F1 score by an average of 6.41% and 4.27%, respectively, demonstrating the accuracy and effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The rapidly evolving field of Virtual Reality (VR)-based Human–Computer Interaction (HCI) presents a significant demand for robust and accurate hand tracking solutions. Current technologies, predominantly based on single-sensing modalities, fall short in providing comprehensive information capture due to susceptibility to occlusions and environmental factors. In this paper, we introduce a novel sensor fusion approach combined with a Long Short-Term Memory (LSTM)-based algorithm for enhanced hand tracking in VR-based HCI. Our system employs six Leap Motion controllers, two RealSense depth cameras, and two Myo armbands to yield a multi-modal data capture. This rich data set is then processed using LSTM, ensuring the accurate real-time tracking of complex hand movements. The proposed system provides a powerful tool for intuitive and immersive interactions in VR environments. [ABSTRACT FROM AUTHOR]

Real-time and accurate state-of-charge estimation performs an important role in the smooth operation of various electric vehicle battery management systems. Neural network theory represents one of the most effective and commonly used methods of SOC prediction. However, traditional neural network methods are disadvantaged by such issues as the limited range of application, limited generalization ability, and low accuracy, which makes it difficult to meet the increasing safety requirements on electric vehicles. In view of these problems, an ensemble learning algorithm based on the AdaBoost.Rt is proposed in this paper. AdaBoost.Rt recurrent neural network model is purposed to ensure the accurate prediction of lithium battery SOC. Relying on a chain-connected recurrent neural network model, this method enables the correlation adaptability of sample data in the spatio-temporal dimension. The ensemble learning method was adopted to devise a method of multi-RNN model integration, with the RNN model as the base learner, thus constructing the AdaBoost.Rt-RNN strong learner model. According to the results of simulation and experimental comparisons, the integrated algorithm proposed in this paper is applicable to improve the accuracy of SOC prediction and the generalization performance of the model. [ABSTRACT FROM AUTHOR]

Deep unrolling networks (DUNs) have emerged as a promising approach for solving compressed sensing (CS) problems due to their superior explainability, speed, and performance compared to classical deep network models. However, the CS performance in terms of efficiency and accuracy remains a principal challenge for approaching further improvements. In this paper, we propose a novel deep unrolling model, SALSA-Net, to solve the image CS problem. The network architecture of SALSA-Net is inspired by unrolling and truncating the split augmented Lagrangian shrinkage algorithm (SALSA) which is used to solve sparsity-induced CS reconstruction problems. SALSA-Net inherits the interpretability of the SALSA algorithm while incorporating the learning ability and fast reconstruction speed of deep neural networks. By converting the SALSA algorithm into a deep network structure, SALSA-Net consists of a gradient update module, a threshold denoising module, and an auxiliary update module. All parameters, including the shrinkage thresholds and gradient steps, are optimized through end-to-end learning and are subject to forward constraints to ensure faster convergence. Furthermore, we introduce learned sampling to replace traditional sampling methods so that the sampling matrix can better preserve the feature information of the original signal and improve sampling efficiency. Experimental results demonstrate that SALSA-Net achieves significant reconstruction performance compared to state-of-the-art methods while inheriting the advantages of explainable recovery and high speed from the DUNs paradigm. [ABSTRACT FROM AUTHOR]

Geometallurgical models are commonly built by combining explanatory variables to obtain the response that requires prediction. This study presented a phosphate plant with three concentration steps: magnetic separation, desliming and flotation, where the yields and recoveries corresponding to each process unit were predicted. These output variables depended on the ore composition and the collector concentration utilized. This paper proposed a solution based on feature engineering to select the best set of explanatory variables and a subset of them able to keep the model as simple as possible but with enough precision and accuracy. After choosing the input variables, two neural network models were developed to simultaneously forecast the seven geometallurgical variables under study: the first, using the best set of variables; and the second, using the reduced set of inputs. The forecasts obtained in both scenarios were compared, and the results showed that the mean squared error and the root mean squared error increase in all output variables evaluated in the test set was smaller than 2.6% when the reduced set was used. The trade-off between simplicity and the quality of the model needs to be addressed when choosing the final neural network to be used in a 3D-block model. [ABSTRACT FROM AUTHOR]

Huntington Disease (HD) is a degenerative neurological disease that causes a significant impact on the quality of life of the patient and eventually death. In this paper we present an approach to create a biomarker using as an input DNA CpG methylation data to identify HD patients. DNA CpG methylation is a well-known epigenetic marker for disease state. Technological advances have made it possible to quickly analyze hundreds of thousands of CpGs. This large amount of information might introduce noise as potentially not all DNA CpG methylation levels will be related to the presence of the illness. In this paper, we were able to reduce the number of CpGs considered from hundreds of thousands to 237 using a non-linear approach. It will be shown that using only these 237 CpGs and non-linear techniques such as artificial neural networks makes it possible to accurately differentiate between control and HD patients. An underlying assumption in this paper is that there are no indications suggesting that the process is linear and therefore non-linear techniques, such as artificial neural networks, are a valid tool to analyze this complex disease. The proposed approach is able to accurately distinguish between control and HD patients using DNA CpG methylation data as an input and non-linear forecasting techniques. It should be noted that the dataset analyzed is relatively small. However, the results seem relatively consistent and the analysis can be repeated with larger data-sets as they become available. [ABSTRACT FROM AUTHOR]

This paper introduces a comprehensive examination of vector-controlled- (VC-) based techniques intended for induction motor (IM) drives. In addition, the evaluation and critique of modern control techniques that improve the performance of IM drives are discussed by considering a systematic literature survey. Detailed research on variable-speed drive control, for instance, VC and scalar control (SCC), was conducted. The SCC-based systems' speed and V/f control purposes are clarified in closed and open loops of IM drives. The operations, benefits, and drawbacks of the direct and indirect field-oriented control systems are illustrated. Furthermore, the direct torque control (DTC) method for IMs is reviewed. Numerous VC methods established along with microprocessor/digital control, model reference adaptive control (MRAC), sliding mode control (SMC), and intelligent control (in terms of fuzzy logic (FL) and artificial neural networks (ANNs)) are described and examined. Uncertainties in the IM parameter are a considerable problem in VC drives. Therefore, this problem is addressed, and some studies that attempted to provide solutions are listed. Magnetic saturation and core loss impact are mentioned, as they are important issues in IM drives. Toward demonstrating the strengths and limitations of various VC configurations, a few experiments were simulated via MATLAB® and Simulink® that show the influence of machine parameter variation. Efforts are made to supply powerful guidelines for practicing engineers and researchers in AC drives. [ABSTRACT FROM AUTHOR]

Electrical impedance tomography (EIT) is a non-invasive imaging modality used for estimating the conductivity of an object Ω from boundary electrode measurements. In recent years, researchers achieved substantial progress in analytical and numerical methods for the EIT inverse problem. Despite the success, numerical instability is still a major hurdle due to many factors, including the discretization error of the problem. Furthermore, most algorithms with good performance are relatively time consuming and do not allow real-time applications. In our approach, the goal is to separate the unknown conductivity into two regions, namely the region of homogeneous background conductivity and the region of non-homogeneous conductivity. Therefore, we pose and solve the problem of shape reconstruction using machine learning. We propose a novel and simple jet intriguing neural network architecture capable of solving the EIT inverse problem. It addresses previous difficulties, including instability, and is easily adaptable to other ill-posed coefficient inverse problems. That is, the proposed model estimates the probability for a point of whether the conductivity belongs to the background region or to the non-homogeneous region on the continuous space R d ∩ Ω with d ∈ { 2 , 3 } . The proposed model does not make assumptions about the forward model and allows for solving the inverse problem in real time. The proposed machine learning approach for shape reconstruction is also used to improve gradient-based methods for estimating the unknown conductivity. In this paper, we propose a piece-wise constant reconstruction method that is novel in the inverse problem setting but inspired by recent approaches from the 3D vision community. We also extend this method into a novel constrained reconstruction method. We present extensive numerical experiments to show the performance of the architecture and compare the proposed method with previous analytic algorithms, mainly the monotonicity-based shape reconstruction algorithm and iteratively regularized Gauss–Newton method. [ABSTRACT FROM AUTHOR]

Piezoelectric actuators (PEA) are high-precision devices used in applications requiring micrometric displacements. However, PEAs present non-linearity phenomena that introduce drawbacks at high precision applications. One of these phenomena is hysteresis, which considerably reduces their performance. The introduction of appropriate control strategies may improve the accuracy of the PEAs. This paper presents a high precision control scheme to be used at PEAs based on the model-based predictive control (MPC) scheme. In this work, the model used to feed the MPC controller has been achieved by means of artificial neural networks (ANN). This approach simplifies the obtaining of the model, since the achievement of a precise mathematical model that reproduces the dynamics of the PEA is a complex task. The presented approach has been embedded over the dSPACE control platform and has been tested over a commercial PEA, supplied by Thorlabs, conducting experiments to demonstrate improvements of the MPC. In addition, the results of the MPC controller have been compared with a proportional-integral-derivative (PID) controller. The experimental results show that the MPC control strategy achieves higher accuracy at high precision PEA applications such as tracking periodic reference signals and sudden reference change. [ABSTRACT FROM AUTHOR]

In this paper, the useability of feedforward and recurrent neural networks for fusion of data from impulse-radar sensors and depth sensors, in the context of healthcare-oriented monitoring of elderly persons, is investigated. Two methods of data fusion are considered, viz., one based on a multilayer perceptron and one based on a nonlinear autoregressive network with exogenous inputs. These two methods are compared with a reference method with respect to their capacity for decreasing the uncertainty of estimation of a monitored person's position and uncertainty of estimation of several parameters enabling medical personnel to make useful inferences on the health condition of that person, viz., the number of turns made during walking, the travelled distance, and the mean walking speed. Both artificial neural networks were trained on the synthetic data. The numerical experiments show the superiority of the method based on a nonlinear autoregressive network with exogenous inputs. This may be explained by the fact that for this type of network, the prediction of the person's position at each time instant is based on the position of that person at the previous time instants. [ABSTRACT FROM AUTHOR]

This paper presents a method for the transparent, robust, and highly capacitive watermarking of video signals using an information mapper. The proposed architecture is based on the use of deep neural networks to embed the watermark in the luminance channel in the YUV color space. An information mapper was used to enable the transformation of a multi-bit binary signature of varying capacitance reflecting the entropy measure of the system into a watermark embedded in the signal frame. To confirm the effectiveness of the method, tests were carried out for video frames with a resolution of 256 × 256 pixels, with a watermark capacity of 4 to 16,384 bits. Transparency metrics (SSIM and PSNR) and a robustness metric—the bit error rate (BER)—were used to assess the performance of the algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we explore a process called neural teleportation, a mathematical consequence of applying quiver representation theory to neural networks. Neural teleportation teleports a network to a new position in the weight space and preserves its function. This phenomenon comes directly from the definitions of representation theory applied to neural networks and it turns out to be a very simple operation that has remarkable properties. We shed light on the surprising and counter-intuitive consequences neural teleportation has on the loss landscape. In particular, we show that teleportation can be used to explore loss level curves, that it changes the local loss landscape, sharpens global minima and boosts back-propagated gradients at any moment during the learning process. [ABSTRACT FROM AUTHOR]

In the field of natural language processing, machine translation is a colossally developing research area that helps humans communicate more effectively by bridging the linguistic gap. In machine translation, normalization and morphological analyses are the first and perhaps the most important modules for information retrieval (IR). To build a morphological analyzer, or to complete the normalization process, it is important to extract the correct root out of different words. Stemming and lemmatization are techniques commonly used to find the correct root words in a language. However, a few studies on IR systems for the Urdu language have shown that lemmatization is more effective than stemming due to infixes found in Urdu words. This paper presents a lemmatization algorithm based on recurrent neural network models for the Urdu language. However, lemmatization techniques for resource-scarce languages such as Urdu are not very common. The proposed model is trained and tested on two datasets, namely, the Urdu Monolingual Corpus (UMC) and the Universal Dependencies Corpus of Urdu (UDU). The datasets are lemmatized with the help of recurrent neural network models. The Word2Vec model and edit trees are used to generate semantic and syntactic embedding. Bidirectional long short-term memory (BiLSTM), bidirectional gated recurrent unit (BiGRU), bidirectional gated recurrent neural network (BiGRNN), and attention-free encoder–decoder (AFED) models are trained under defined hyperparameters. Experimental results show that the attention-free encoder-decoder model achieves an accuracy, precision, recall, and F-score of 0.96, 0.95, 0.95, and 0.95, respectively, and outperforms existing models. [ABSTRACT FROM AUTHOR]

Complex network science is an interdisciplinary field of study based on graph theory, statistical mechanics, and data science. With the powerful tools now available in complex network theory for the study of network topology, it is obvious that complex network topology models can be applied to enhance artificial neural network models. In this paper, we provide an overview of the most important works published within the past 10 years on the topic of complex network theory-based optimization methods. This review of the most up-to-date optimized neural network systems reveals that the fusion of complex and neural networks improves both accuracy and robustness. By setting out our review findings here, we seek to promote a better understanding of basic concepts and offer a deeper insight into the various research efforts that have led to the use of complex network theory in the optimized neural networks of today. [ABSTRACT FROM AUTHOR]

Monitoring the condition of industrial equipment is fundamental to avoid failures and maximize uptime. The present work used supervised and unsupervised learning methods to create models for predicting the condition of an industrial machine. The main objective was to determine when the asset was either in its nominal operation or working outside this zone, thus being at risk of failure or sub-optimal operation. The results showed that it is possible to classify the machine state using artificial neural networks. K-means clustering and PCA methods showed that three states, chosen through the Elbow Method, cover almost all the variance of the data under study. Knowing the importance that the quality of the lubricants has in the functioning and classification of the state of machines, a lubricant classification algorithm was developed using Neural Networks. The lubricant classifier results were 98% accurate compared to human expert classifications. The main gap identified in the research is that the found classification works only carried out classifications of present, short-term, or mid-term failures. To close this gap, the work presented in this paper conducts a long-term classification. [ABSTRACT FROM AUTHOR]

The seasonal autoregressive integrated moving average with exogenous factors (SARIMAX) has shown promising results in modeling small and sparse observed time-series data by capturing linear features using independent and dependent variables. Long short-term memory (LSTM) is a promising neural network for learning nonlinear dependence features from data. With the increase in wildlife roadkill patterns, the SARIMAX-only and LSTM-only models would likely fail to learn the precise endogenous and/or exogenous variables driven by this wildlife roadkill data. In this paper, we design and implement an error correction mathematical framework based on LSTM-only. The framework extracts features from the residual error generated by a SARIMAX-only model. The learned residual features correct the output time-series prediction of the SARIMAX-only model. The process combines SARIMAX-only predictions and LSTM-only residual predictions to obtain a hybrid SARIMAX-LSTM. The models are evaluated using South African wildlife–vehicle collision datasets, and the experiments show that compared to single models, SARIMAX-LSTM increases the accuracy of a taxon whose linear components outweigh the nonlinear ones. In addition, the hybrid model fails to outperform LSTM-only when a taxon contains more nonlinear components rather than linear components. Our assumption of the results is that the collected exogenous and endogenous data are insufficient, which limits the hybrid model's performance since it cannot accurately detect seasonality on residuals from SARIMAX-only and minimize the SARIMAX-LSTM error. We conclude that the error correction framework should be preferred over single models in wildlife time-series modeling and predictions when a dataset contains more linear components. Adding more related data may improve the prediction performance of SARIMAX-LSTM. [ABSTRACT FROM AUTHOR]

Adversarial examples easily mislead vision systems based on deep neural networks (DNNs) trained with softmax cross entropy (SCE) loss. The vulnerability of DNN comes from the fact that SCE drives DNNs to fit on the training examples, whereas the resultant feature distributions between the training and adversarial examples are unfortunately misaligned. Several state-of-the-art methods start from improving the inter-class separability of training examples by modifying loss functions, where we argue that the adversarial examples are ignored, thus resulting in a limited robustness to adversarial attacks. In this paper, we exploited the inference region, which inspired us to apply margin-like inference information to SCE, resulting in a novel inference-softmax cross entropy (I-SCE) loss, which is intuitively appealing and interpretable. The inference information guarantees that it is difficult for neural networks to cross the decision boundary under an adversarial attack, and guarantees both the inter-class separability and the improved generalization to adversarial examples, which was further demonstrated and proved under the min-max framework. Extensive experiments show that the DNN models trained with the proposed I-SCE loss achieve a superior performance and robustness over the state-of-the-arts under different prevalent adversarial attacks; for example, the accuracy of I-SCE is 63% higher than SCE under the P G D 50 u n attack on the MNIST dataset. These experiments also show that the inference region can effectively solve the misaligned distribution. [ABSTRACT FROM AUTHOR]

Currently, Android is the most popular operating system among mobile devices. However, as the number of devices with the Android operating system increases, so does the danger of using them. This is especially important as smartphones increasingly authenticate critical activities(e-banking, e-identity). BotSense Mobile is a tool already integrated with some critical applications (e-banking, e-identity) to increase user safety. In this paper, we focus on the novel functionality of BotSense Mobile: the detection of malware applications on a user device. In addition to the standard blacklist approach, we propose a machine learning-based model for unknown malicious application detection. The lightweight neural network model is deployed on an edge device to avoid sending sensitive user data outside the device. For the same reason, manifest-related features can be used by the detector only. We present a comprehensive empirical analysis of malware detection conducted on recent data (May–June, 2022) from the Koodous platform, which is a collaborative platform where over 70 million Android applications were collected. The research highlighted the problem of machine learning model aging. We evaluated the lightweight model on recent Koodous data and obtained f 1 = 0.77 and high precision ( 0.9 ). [ABSTRACT FROM AUTHOR]

In the past two years, medical researchers and data scientists worldwide have focused their efforts on containing the pandemic of coronavirus disease 2019 (COVID-19). Deep learning models have been proven to be capable of efficient medical diagnosis and prognosis in cancer, common lung diseases, and COVID-19. On the other hand, artificial neural networks have demonstrated their potential in pattern recognition and classification in various domains, including healthcare. This literature review aims to report the state of research on developing neural network models to diagnose COVID-19 from cough sounds to create a cost-efficient and accessible testing tool in the fight against the pandemic. A total of 35 papers were included in this review following a screening of the 161 outputs of the literature search. We extracted information from articles on data resources, model structures, and evaluation metrics and then explored the scope of experimental studies and methodologies and analyzed their outcomes and limitations. We found that cough is a biomarker, and its associated information can determine an individual's health status. Convolutional neural networks were predominantly used, suggesting they are particularly suitable for feature extraction and classification. The reported accuracy values ranged from 73.1% to 98.5%. Moreover, the dataset sizes ranged from 16 to over 30,000 cough audio samples. Although deep learning is a promising prospect in identifying COVID-19, we identified a gap in the literature on research conducted over large and diversified data sets. [ABSTRACT FROM AUTHOR]

While thunderstorms can pose severe risks to property and life, forecasting remains challenging, even at short lead times, as these often arise in meta-stable atmospheric conditions. In this paper, we examine the question of how well we could perform short-term (up to 180 min) forecasts using exclusively multi-spectral satellite images and past lighting events as data. We employ representation learning based on deep convolutional neural networks in an "end-to-end" fashion. Here, a crucial problem is handling the imbalance of the positive and negative classes appropriately in order to be able to obtain predictive results (which is not addressed by many previous machine-learning-based approaches). The resulting network outperforms previous methods based on physically based features and optical flow methods (similar to operational prediction models) and generalizes across different years. A closer examination of the classifier performance over time and under masking of input data indicates that the learned model actually draws most information from structures in the visible spectrum, with infrared imaging sustaining some classification performance during the night. [ABSTRACT FROM AUTHOR]

This paper presents and analyzes a programming tool that implements a method for classification and function regression problems. This method builds new features from existing ones with the assistance of a hybrid algorithm that makes use of artificial neural networks and grammatical evolution. The implemented software exploits modern multi-core computing units for faster execution. The method has been applied to a variety of classification and function regression problems, and an extensive comparison with other methods of computational intelligence is made. [ABSTRACT FROM AUTHOR]

In this paper, we study the learnability of the Boolean inner product by a systematic simulation study. The family of the Boolean inner product function is known to be representable by neural networks of threshold neurons of depth 3 with only 2 n + 1 units (n the input dimension)—whereas an exact representation by a depth 2 network cannot possibly be of polynomial size. This result can be seen as a strong argument for deep neural network architectures. In our study, we found that this depth 3 architecture of the Boolean inner product is difficult to train, much harder than the depth 2 network, at least for the small input size scenarios n ≤ 16 . Nonetheless, the accuracy of the deep architecture increased with the dimension of the input space to 94% on average, which means that multiple restarts are needed to find the compact depth 3 architecture. Replacing the fully connected first layer by a partially connected layer (a kind of convolutional layer sparsely connected with weight sharing) can significantly improve the learning performance up to 99% accuracy in simulations. Another way to improve the learnability of the compact depth 3 representation of the inner product could be achieved by adding just a few additional units into the first hidden layer. [ABSTRACT FROM AUTHOR]

When computers started to become a dominant part of technology around the 1950s, fundamental questions about reliable designs and robustness were of great relevance. Their development gave rise to the exploration of new questions, such as what made brains reliable (since neurons can die) and how computers could get inspiration from neural systems. In parallel, the first artificial neural networks came to life. Since then, the comparative view between brains and computers has been developed in new, sometimes unexpected directions. With the rise of deep learning and the development of connectomics, an evolutionary look at how both hardware and neural complexity have evolved or designed is required. In this paper, we argue that important similarities have resulted both from convergent evolution (the inevitable outcome of architectural constraints) and inspiration of hardware and software principles guided by toy pictures of neurobiology. Moreover, dissimilarities and gaps originate from the lack of major innovations that have paved the way to biological computing (including brains) that are completely absent within the artificial domain. As it occurs within synthetic biocomputation, we can also ask whether alternative minds can emerge from A.I. designs. Here, we take an evolutionary view of the problem and discuss the remarkable convergences between living and artificial designs and what are the pre-conditions to achieve artificial intelligence. [ABSTRACT FROM AUTHOR]

A lightweight on-device liquid consumption estimation system involving an energy-aware machine learning algorithm is developed in this work. This system consists of two separate on-device neural network models that carry out liquid consumption estimation with the result of two tasks: the detection of sip from gestures with which the bottle is handled by its user and the detection of first sips after a bottle refill. This predictive volume estimation framework incorporates a self-correction mechanism that can minimize the error after each bottle fill-up cycle, which makes the system robust to errors from the sip classification module. In this paper, a detailed characterization of sip detection is performed to understand the accuracy-complexity tradeoffs by developing and implementing a variety of different ML models with varying complexities. The maximum energy consumed by the entire framework is around 119 mJ during a maximum computation time of 300 μ s . The energy consumption and computation times of the proposed framework is suitable for implementation in low-power embedded hardware that can be incorporated in consumer grade water bottles. [ABSTRACT FROM AUTHOR]

This paper deals with the design of sliding mode control and neural network compensation for a sensorless permanent magnet synchronous motor (PMSM) controlled system that is able to improve both power consumption and speed response performance. The position sensor of PMSM is unreliable in harsh environments. Therefore, the sensorless control technique is widely proposed in industry. A sliding mode observer can estimate the rotor angle and has the robustness to load disturbance and parameter variations. However, the sliding mode observer is not conducive to standstill and low speed conditions because the amplitude of the back EMF is almost zero. As a result, this paper combines an iterative sliding mode observer (ISMO) and neural networks (NNs) as an angle compensator to improve the above problems. A dsPIC30F6010A-based PMSM sensorless drive system is implemented to validate the proposed algorithm. The simulation and experimental results prove its effectiveness. [ABSTRACT FROM AUTHOR]

Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calibration process of this kind of sensor could be a time-consuming task because it is usually done by identification in a manual and repetitive way. The resulting obstacles detection models are usually nonlinear functions that can be different for each proximity sensor attached to the robot. In addition, the model is highly dependent on the type of sensor (e.g., ultrasonic or infrared), on changes in light intensity, and on the properties of the obstacle such as shape, colour, and surface texture, among others. That is why in some situations it could be useful to gather all the measurements provided by different kinds of sensor in order to build a unique model that estimates the distances to the obstacles around the robot. This paper presents a novel approach to get an obstacles detection model based on the fusion of sensors data and automatic calibration by using artificial neural networks. [ABSTRACT FROM AUTHOR]

In this paper, the synchronization problem of fractional-order complex-valued neural networks with discrete and distributed delays is investigated. Based on the adaptive control and Lyapunov function theory, some sufficient conditions are derived to ensure the states of two fractional-order complex-valued neural networks with discrete and distributed delays achieve complete synchronization rapidly. Finally, numerical simulations are given to illustrate the effectiveness and feasibility of the theoretical results. [ABSTRACT FROM AUTHOR]

This paper presents a novel methodology of generic nature for controlling nonlinear systems, using inverse radial basis function neural network models, which may combine diverse data originating from various sources. The algorithm starts by applying the particle swarm optimization-based non-symmetric variant of the fuzzy means (PSO-NSFM) algorithm so that an approximation of the inverse system dynamics is obtained. PSO-NSFM offers models of high accuracy combined with small network structures. Next, the applicability domain concept is suitably tailored and embedded into the proposed control structure in order to ensure that extrapolation is avoided in the controller predictions. Finally, an error correction term, estimating the error produced by the unmodeled dynamics and/or unmeasured external disturbances, is included to the control scheme to increase robustness. The resulting controller guarantees bounded input-bounded state (BIBS) stability for the closed loop system when the open loop system is BIBS stable. The proposed methodology is evaluated on two different control problems, namely, the control of an experimental armature-controlled direct current (DC) motor and the stabilization of a highly nonlinear simulated inverted pendulum. For each one of these problems, appropriate case studies are tested, in which a conventional neural controller employing inverse models and a PID controller are also applied. The results reveal the ability of the proposed control scheme to handle and manipulate diverse data through a data fusion approach and illustrate the superiority of the method in terms of faster and less oscillatory responses. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a study of the odometric system for the autonomous cart Verdino, which is an electric vehicle based on a golf cart, is presented. A mathematical model of the odometric system is derived from cart movement equations, and is used to compute the vehicle position and orientation. The inputs of the system are the odometry encoders, and the model uses the wheels diameter and distance between wheels as parameters. With this model, a least square minimization is made in order to get the nominal best parameters. This model is updated, including a real time wheel diameter measurement improving the accuracy of the results. A neural network model is used in order to learn the odometric model from data. Tests are made using this neural network in several configurations and the results are compared to the mathematical model, showing that the neural network can outperform the first proposed model. [ABSTRACT FROM AUTHOR]

Despite recent progress in the study of complex systems, reconstruction of damaged networks due to random and targeted attack has not been addressed before. In this paper, we formulate the network reconstruction problem as an identification of network structure based on much reduced link information. Furthermore, a novel method based on multilayer perceptron neural network is proposed as a solution to the problem of network reconstruction. Based on simulation results, it was demonstrated that the proposed scheme achieves very high reconstruction accuracy in small-world network model and a robust performance in scale-free network model. [ABSTRACT FROM AUTHOR]

Regularization of neural networks can alleviate overfitting in the training phase. Current regularizationmethods, such as Dropout and DropConnect, randomly drop neural nodes or connections based on a uniform prior. Such a data-independent strategy does not take into consideration of the quality of individual unit or connection. In this paper, we aim to develop a data-dependent approach to regularizing neural network in the framework of Information Geometry. A measurement for the quality of connections is proposed, namely confidence. Specifically, the confidence of a connection is derived from its contribution to the Fisher information distance. The network is adjusted by retaining the confident connections and discarding the less confident ones. The adjusted network, named as ConfNet, would carry the majority of variations in the sample data. The relationships among confidence estimation, Maximum Likelihood Estimation and classical model selection criteria (like Akaike information criterion) is investigated and discussed theoretically. Furthermore, a Stochastic ConfNet is designed by adding a self-adaptive probabilistic sampling strategy. The proposed data-dependent regularization methods achieve promising experimental results on three data collections including MNIST, CIFAR-10 and CIFAR-100. [ABSTRACT FROM AUTHOR]

This paper presents a non-intrusive approach for monitoring driver drowsiness using the fusion of several optimized indicators based on driver physical and driving performance measures, obtained from ADAS (Advanced Driver Assistant Systems) in simulated conditions. The paper is focused on real-time drowsiness detection technology rather than on long-term sleep/awake regulation prediction technology. We have developed our own vision system in order to obtain robust and optimized driver indicators able to be used in simulators and future real environments. These indicators are principally based on driver physical and driving performance skills. The fusion of several indicators, proposed in the literature, is evaluated using a neural network and a stochastic optimization method to obtain the best combination. We propose a new method for ground-truth generation based on a supervised Karolinska Sleepiness Scale (KSS). An extensive evaluation of indicators, derived from trials over a third generation simulator with several test subjects during different driving sessions, was performed. The main conclusions about the performance of single indicators and the best combinations of them are included, as well as the future works derived from this study. [ABSTRACT FROM AUTHOR]

With the automotive industry moving towards automated driving, sensing is increasingly important in enabling technology. The virtual sensors allow data fusion from various vehicle sensors and provide a prediction for measurement that is hard or too expensive to measure in another way or in the case of demand on continuous detection. In this paper, virtual sensing is discussed for the case of vehicle suspension control, where information about the relative velocity of the unsprung mass for each vehicle corner is required. The corresponding goal can be identified as a regression task with multi-input sequence input. The hypothesis is that the state-of-art method of Bidirectional Long–Short Term Memory (BiLSTM) can solve it. In this paper, a virtual sensor has been proposed and developed by training a neural network model. The simulations have been performed using an experimentally validated full vehicle model in IPG Carmaker. Simulations provided the reference data which were used for Neural Network (NN) training. The extensive dataset covering 26 scenarios has been used to obtain training, validation and testing data. The Bayesian Search was used to select the best neural network structure using root mean square error as a metric. The best network is made of 167 BiLSTM, 256 fully connected hidden units and 4 output units. Error histograms and spectral analysis of the predicted signal compared to the reference signal are presented. The results demonstrate the good applicability of neural network-based virtual sensors to estimate vehicle unsprung mass relative velocity. [ABSTRACT FROM AUTHOR]

Location prediction has attracted much attention due to its important role in many location-based services, such as food delivery, taxi-service, real-time bus system, and advertisement posting. Traditional prediction methods often cluster track points into regions and mine movement patterns within the regions. Such methods lose information of points along the road and cannot meet the demand of specific services. Moreover, traditional methods utilizing classic models may not perform well with long location sequences. In this paper, a spatial-temporal-semantic neural network algorithm (STS-LSTM) has been proposed, which includes two steps. First, the spatial-temporal-semantic feature extraction algorithm (STS) is used to convert the trajectory to location sequences with fixed and discrete points in the road networks. The method can take advantage of points along the road and can transform trajectory into model-friendly sequences. Then, a long short-term memory (LSTM)-based model is constructed to make further predictions, which can better deal with long location sequences. Experimental results on two real-world datasets show that STS-LSTM has stable and higher prediction accuracy over traditional feature extraction and model building methods, and the application scenarios of the algorithm are illustrated. [ABSTRACT FROM AUTHOR]

Many of the next generation of adaptive optics systems on large and extremely large telescopes require tomographic techniques in order to correct for atmospheric turbulence over a large field of view. Multi-object adaptive optics is one such technique. In this paper, different implementations of a tomographic reconstructor based on a machine learning architecture named "CARMEN" are presented. Basic concepts of adaptive optics are introduced first, with a short explanation of three different control systems used on real telescopes and the sensors utilised. The operation of the reconstructor, along with the three neural network frameworks used, and the developed CUDA code are detailed. Changes to the size of the reconstructor influence the training and execution time of the neural network. The native CUDA code turns out to be the best choice for all the systems, although some of the other frameworks offer good performance under certain circumstances. [ABSTRACT FROM AUTHOR]

Improving the effectiveness of spatial shape features classification from 3D lidar data is very relevant because it is largely used as a fundamental step towards higher level scene understanding challenges of autonomous vehicles and terrestrial robots. In this sense, computing neighborhood for points in dense scans becomes a costly process for both training and classification. This paper proposes a new general framework for implementing and comparing different supervised learning classifiers with a simple voxel-based neighborhood computation where points in each non-overlapping voxel in a regular grid are assigned to the same class by considering features within a support region defined by the voxel itself. The contribution provides offline training and online classification procedures as well as five alternative feature vector definitions based on principal component analysis for scatter, tubular and planar shapes. Moreover, the feasibility of this approach is evaluated by implementing a neural network (NN) method previously proposed by the authors as well as three other supervised learning classifiers found in scene processing methods: support vector machines (SVM), Gaussian processes (GP), and Gaussian mixture models (GMM). A comparative performance analysis is presented using real point clouds from both natural and urban environments and two different 3D rangefinders (a tilting Hokuyo UTM-30LX and a Riegl). Classification performance metrics and processing time measurements confirm the benefits of the NN classifier and the feasibility of voxel-based neighborhood. [ABSTRACT FROM AUTHOR]

Electroencephalography is a non-invasive measure of the brain electrical activity generated by millions of neurons. Feature extraction in electroencephalography analysis is a core issue that may lead to accurate brain mental state classification. This paper presents a new feature selection method that improves left/right hand movement identification of a motor imagery brain-computer interface, based on genetic algorithms and artificial neural networks used as classifiers. Raw electroencephalography signals are first preprocessed using appropriate filtering. Feature extraction is carried out afterwards, based on spectral and temporal signal components, and thus a feature vector is constructed. As various features might be inaccurate and mislead the classifier, thus degrading the overall system performance, the proposed approach identifies a subset of features from a large feature space, such that the classifier error rate is reduced. Experimental results show that the proposed method is able to reduce the number of features to as low as 0.5% (i.e., the number of ignored features can reach 99.5%) while improving the accuracy, sensitivity, specificity, and precision of the classifier. [ABSTRACT FROM AUTHOR]

This paper presents a multiple artificial neural networks (MANN) method with interaction noise for estimating the occurrence probabilities of different classes at any site in space. The MANN consists of several independent artificial neural networks, the number of which is determined by the neighbors around the target location. In the proposed algorithm, the conditional or pre-posterior (multi-point) probabilities are viewed as output nodes, which can be estimated by weighted combinations of input nodes: two-point transition probabilities. The occurrence probability of a certain class at a certain location can be easily computed by the product of output probabilities using Bayes' theorem. Spatial interaction or redundancy information can be measured in the form of interaction noises. Prediction results show that the method of MANN with interaction noise has a higher classification accuracy than the traditional Markov chain random fields (MCRF) model and can successfully preserve small-scale features. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of energy consumption prediction using neural networks over a set of public buildings. Since energy consumption in the public sector comprises a substantial share of overall consumption, the prediction of such consumption represents a decisive issue in the achievement of energy savings. In our experiments, we use the data provided by an energy consumption monitoring system in a compound of faculties and research centers at the University of Granada, and provide a methodology to predict future energy consumption using nonlinear autoregressive (NAR) and the nonlinear autoregressive neural network with exogenous inputs (NARX), respectively. Results reveal that NAR and NARX neural networks are both suitable for performing energy consumption prediction, but also that exogenous data may help to improve the accuracy of predictions. [ABSTRACT FROM AUTHOR]

For decades, PID (Proportional + Integral + Derivative)-like controllers have been successfully used in academia and industry for many kinds of plants. This is thanks to its simplicity and suitable performance in linear or linearized plants, and under certain conditions, in nonlinear ones. A number of PID controller gains tuning approaches have been proposed in the literature in the last decades; most of them off-line techniques. However, in those cases wherein plants are subject to continuous parametric changes or external disturbances, online gains tuning is a desirable choice. This is the case of modular underwater ROVs (Remotely Operated Vehicles) where parameters (weight, buoyancy, added mass, among others) change according to the tool it is fitted with. In practice, some amount of time is dedicated to tune the PID gains of a ROV. Once the best set of gains has been achieved the ROV is ready to work. However, when the vehicle changes its tool or it is subject to ocean currents, its performance deteriorates since the fixed set of gains is no longer valid for the new conditions. Thus, an online PID gains tuning algorithm should be implemented to overcome this problem. In this paper, an auto-tune PID-like controller based on Neural Networks (NN) is proposed. The NN plays the role of automatically estimating the suitable set of PID gains that achieves stability of the system. The NN adjusts online the controller gains that attain the smaller position tracking error. Simulation results are given considering an underactuated 6 DOF (degrees of freedom) underwater ROV. Real time experiments on an underactuated mini ROV are conducted to show the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]