Search

Solar energy is one of the most promising renewable energy resources. Over the last few decades, photovoltaic (PV) systems have grown in popularity. Since the maximum power point (MPP) of a solar system changes with environmental circumstances, the maximum power point tracking (MPPT) technique is required to get the most power out of the solar system. Various MPPT techniques based on classical and artificial intelligence (AI) methodologies have been proposed in the literature so far. In this paper, we aim to provide a thorough comparative analysis of the most widely used MPPT algorithms based on AI. The MPPT techniques discussed are based on fuzzy logic (FL), artificial neural networks (ANN), and the suggested hybrid approach ANN-fuzzy. The designed MPPT controllers are evaluated in the same PV system, which consists of a PV module, a DC-DC boost converter, and a DC load, under the same weather profile. Using the MATLAB/Simulink simulation tool, the tracking accuracy, response time, overshoot, and steady-state ripple of each method are tested in different weather conditions. The simulation results show that the ANN-fuzzy proposed tactic outperforms both the FL and the ANN MPPT controllers in correctly and successfully tracking the maximum power under diverse atmospheric conditions.

Despite all of the technical advancements in writing and text editing with keyboards on numerous devices, writing with a pen remains a fundamental ability in modern human existence. Handwriting disabilities are referred to as dysgraphia. Nonetheless, how well they are taught to write in school, 10-30% of children never attain a respectable level of handwriting. Early identification is critical because it can help children avoid difficulties in their behavioral and academic development. On blank papers attached to digital tablets, 280 individuals were asked to complete the concise evaluation scale for children’s handwriting (BHK), with 218 having typical handwriting and 62 having dysgraphia. In addition to their age and BHK quality and speed scores, 12 variables identifying digital handwriting across several domains (static, kinematic, pressure, and tilt) were collected. In this paper, we provided a rapid and automated dysgraphia classification approach using an artificial neural network (ANN) model. Using digital handwriting traits as an input to the ANN approach, the prediction findings were encouraging and very accurate, reaching 96% accuracy, and they could lead to the development of a new self-administered dysgraphia screening tool.

In this work, the Copper composites Cu-x wt. % B4C (x = 0, 5, 10, 15, 20) were fabricated for the metallurgical and mechanical property evaluation as per ASTM standards. The metallurgical characterization tests on the samples include x-ray diffraction, optical microscopy, and scanning electron microscopy with EDX. Further, pin-on-disc apparatus was used to investigate the tribological behavior of composite specimens. An SEM micrograph of the worn surface and wear debris, along with the Gwyddion software, has been used to discuss the wear mechanisms in detail. The Artificial Neural Networks (ANN) classifier model is also constructed to describe the wear behavior in more detail. The experimental results inferred that the addition of Boron carbide particles has enhanced the Copper's corrosion resistance in a 1 M HCl electrolyte solution from 30.34% to 74.2%, 75.08%, and 83.29% with B and C ions. Also, it significantly enhance the mechanical and tribological characteristics considerably. KEY WORDS: Powder metallurgy, Cu-B4C, Gwyddion, Wear, Artificial Neural Network Bull. Chem. Soc. Ethiop. 2023, 37(4), 959-972. DOI: https://dx.doi.org/10.4314/bcse.v37i4.12 &nbsp

AI edge devices require local intelligence for the concerns of latency and privacy. Given the accuracy and energy constraints, low-power convolutional neural networks (CNNs) are gaining popularity. To alleviate the high memory access energy and computational cost of large CNN models, prior works have proposed promising approaches including in-memory-computing (IMC) [1], mixed-signal multiply-and-accumulate (MAC) calculation [2], and reduced resolution network –[4]. With weights and activations restricted to ±1, binary neural network (BNN) combining with IMC greatly improves the storage and computation efficiency, making it well-suited for edge-based applications, and has demonstrated state-of-the-art energy efficiency in image classification problems [5]. However, compared to full resolution network, BNN requires larger model thus more operations (OPs) per inference for a certain accuracy. To address such challenge, we propose a mixed-signal ternary CNN based processor featuring higher energy efficiency than BNN. It confers several key improvements: 1) the proposed ternary network provides 1.5-b resolution (0/+1/-1), leading to 3.9x OPs/inference reduction than BNN for the same MNIST accuracy; 2) a 1.5b MAC is implemented by V CM -based capacitor switching scheme, which inherently benefits from the reduced signal swing on the capacitive DAC (CDAC); 3) the V CM -based MAC introduces sparsity during training, resulting in lower switching rate. With a complete neural network on chip, the proposed design realizes 97.1% MNIST accuracy with only 0.18uJ per classification, presenting the highest power efficiency for comparable MNIST accuracy.

In this article, we present an extended data-fitting model which involves different and conflicting criteria, and we propose an algorithm based on a scalarization technique to solve it. Our model integrates in a unique framework three different criteria, namely, a data-fitting term, and the entropy and the sparsity of the set of unknown parameters. This model can be analyzed by means of multiple criteria decision-making techniques. We then validate the proposed modified algorithm using two computational experiments: We analyze the problem of handwritten digit recognition using a logistic regression model and a deep neural network model, respectively. In the final part of the article, we employ this methodology to forecasting instead. Given the importance of forecasting techniques to predict the future, which in turn can lead to positive impacts on firm performance, we propose two numerical experiments focusing on the forecast of the US GDP. In the first one, we proceed by means of a modified iterated function system with grayscale maps-type fractal operator, and, in the second one, we implement a modified neural network-based model.

With the dramatic progress of wearable devices, continuous collection of single lead ECG wave is able to be implemented in a comfortable fashion. Data mining on single lead ECG wave is therefore attracting increasing attention, where atrial fibrillation (AF) detection is a hot topic. In this paper, we propose a dual-channel neural network for AF detection from a single lead ECG wave. Two primary phases are included, the data preprocessing part followed by a dual-channel neural network. A two-stage denoising procedure is developed for data preprocessing, so as to tackle the high noise and disturbance which generally resides in the ECG wave collected by wearable devices. Then the time-frequency spectrum and Poincare plot of the denoised ECG signal are imported into the developed dual-channel neural network for feature extraction and AF detection. On the 2017 PhysioNet/CinC Challenge database, the F1 values were 0.83, 0.90, and 0.75 for AF rhythm and normal rhythm, and other rhythm, respectively. The results well validate the effectiveness of the proposed method for AF detection from a single lead ECG wave, and also indicate its performance advantages over some state-of-the-art counterparts.

This article is dedicated to investigating the impulsive-based almost surely synchronization issue of neural network systems (NSSs) with quality-of-service constraints. First, the communication network considered suffers from random double deception attacks, which are modeled as a nonlinear function and a desynchronizing impulse sequence, respectively. Meanwhile, the impulsive instants and impulsive gains are randomly and only their expectations are available. Second, by taking two different types of random deception attacks into consideration, a novel mathematical model for vulnerable NSSs is constructed. Then, almost surely synchronization criteria are established by using Borel-Cantelli lemma. Furthermore, based on the derived strong and weak sufficient conditions, the almost surely synchronization of NSSs is achieved. Finally, the section of numerical example is shown to illustrate the effectiveness of the proposed method.

Real neural networks are characterized by large-scale and complex topology. However, the current dynamical analysis is limited to low-dimensional models with simplified topology. Therefore, there is still a huge gap between neural network theory and its application. This article proposes a class of large-scale neural networks with a ring-hub structure, where a hub node is connected to n peripheral nodes and these peripheral nodes are linked by a ring. In particular, there exists a hybrid coupling mode in the network topology. The mathematical model of such systems is described by fractional-order delayed differential equations. The aim of this article is to investigate the local stability and Hopf bifurcation of this high-dimensional neural network. First, the Coates flow graph is employed to obtain the characteristic equation of the linearized high-dimensional neural network model, which is a transcendental equation including multiple exponential items. Then, the sufficient conditions ensuring the stability of equilibrium and the existence of Hopf bifurcation are achieved by taking time delay as a bifurcation parameter. Finally, some numerical examples are given to support the theoretical results. It is revealed that the increasing time delay can effectively induce the occurrence of periodic oscillation. Moreover, the fractional order, the self-feedback coefficient, and the number of neurons also have effects on the onset of Hopf bifurcation.

In this article, two novel neural adaptive attitude control strategies are presented for the attitude tracking of rigid spacecraft affected by inertia uncertainty, external disturbance, and torque saturation with guaranteed prescribed performance. Under the backstepping framework, a neural adaptive state feedback controller is designed first by utilizing the neural network (NN) approximation and barrier Lyapunov function (BLF). Then, on the basis of the state feedback controller, a neural adaptive output feedback controller is exploited by integrating with a high-gain observer. Lyapunov stability analysis shows that all the closed-loop error signals are uniformly ultimately bounded under the both proposed neural adaptive attitude controllers. In comparisons with the most existing researches, the distinctive advantages of the proposed neural adaptive attitude controllers are threefold. (1) With the help of NN approximation, the proposed attitude controllers are model-independent and still applicable even when the spacecraft attitude dynamics are completely unknown. (2) Benefiting from the BLF, the proposed attitude controllers can guarantee the attitude tracking error always satisfies the preassigned performance. (3) Both the state feedback control and output feedback control are considered. The angular velocity information of the spacecraft is not required for the output feedback control design. At last, numerical simulations and comparisons are conducted to verify and highlight the proposed neural adaptive attitude control strategies.

Forecast skill from dynamical forecast models decreases quickly with projection time due to various errors. Therefore, postprocessing methods, from simple bias correction methods to more complicated multiple linear regression–based model output statistics, are used to improve raw model forecasts. Usually, these methods show clear forecast improvement over the raw model forecasts, especially for short-range weather forecasts. However, linear approaches have limitations because the relationship between predictands and predictors may be nonlinear. This is even truer for extended range forecasts, such as week-3–4 forecasts. In this study, neural network techniques are used to seek or model the relationships between a set of predictors and predictands, and eventually to improve week-3–4 precipitation and 2-m temperature forecasts made by the NOAA/NCEP Climate Forecast System. Benefitting from advances in machine learning techniques in recent years, more flexible and capable machine learning algorithms and availability of big datasets enable us not only to explore nonlinear features or relationships within a given large dataset, but also to extract more sophisticated pattern relationships and covariabilities hidden within the multidimensional predictors and predictands. Then these more sophisticated relationships and high-level statistical information are used to correct the model week-3–4 precipitation and 2-m temperature forecasts. The results show that to some extent neural network techniques can significantly improve the week-3–4 forecast accuracy and greatly increase the efficiency over the traditional multiple linear regression methods.

Mid-term load forecasting (MTLF) is of great significance for power system planning, operation, and power trading. However, the mid-term electrical load is affected by the coupling of multiple factors and demonstrates complex characteristics, which leads to low prediction accuracy in MTLF. Furthermore, MTLF is faced with the ``curse of dimensionality'' problem due to a large number of variables. This article proposes an MTLF method based on manifold learning, which can extract the underlying factors of load variations to help improve the accuracy of MTLF and significantly reduce the calculation. Unlike linear dimensionality reduction methods, manifold learning has better nonlinear feature extraction ability and is more suitable for load data with nonlinear characteristics. Furthermore, long short-term memory (LSTM) neural networks are used to establish forecasting models in the low-dimensional space obtained by manifold learning. The proposed MTLF method is tested on independent system operator (ISO) New England datasets, and load forecasting in 24, 168, and 720 h ahead is carried out. The numerical results validate that the proposed method has higher prediction accuracy than many mature methods in the mid-term time scale.

In this study, an artificial neural network (ANN) model was applied to predict the thermo-hydraulic performance of the offset-strip fin heat exchanger. The ANN model was used for predicting the thermo-hydraulic performance to improve the prediction accuracy and extend the applicable range compared to the results of using the empirical equations reported in previous studies. The main parameters of these empirical equations were fin height (0.134 < α < 0.997), fin length (0.012 < δ < 0.048), fin width (0.041 < γ < 0.121), and Reynolds number (120 < Re 4). In addition, the Fanning friction factor f and the Colburn factor j were considered as the outputs of the ANN model. The impact of the parameters on the thermo-hydraulic performance of the heat exchanger was quantitatively evaluated, and the prediction accuracy was improved over a wide range for the thermo-hydraulic performance generated by the ANN model. Thus, the results obtained using the ANN model agreed well with the experimental data over a wider range than possible for the previous empirical correlations, showing extremely high accuracy and validity of the ANN model in comparison to the empirical equations.

As a category of the recurrent neural network (RNN), zeroing neural network (ZNN) can effectively handle time-variant optimization issues. Compared with the fixed-parameter ZNN that needs to be adjusted frequently to achieve good performance, the conventional variable-parameter ZNN (VPZNN) does not require frequent adjustment, but its variable parameter will tend to infinity as time grows. Besides, the existing noise-tolerant ZNN model is not good enough to deal with time-varying noise. Therefore, a new-type segmented VPZNN (SVPZNN) for handling the dynamic quadratic minimization issue (DQMI) is presented in this work. Unlike the previous ZNNs, the SVPZNN includes an integral term and a nonlinear activation function, in addition to two specially constructed time-varying piecewise parameters. This structure keeps the time-varying parameters stable and makes the model have strong noise tolerance capability. Besides, theoretical analysis on SVPZNN is proposed to determine the upper bound of convergence time in the absence or presence of noise interference. Numerical simulations verify that SVPZNN has shorter convergence time and better robustness than existing ZNN models when handling DQMI.

Scene parsing, or semantic segmentation, aims at labeling all pixels in an image with the predefined categories of things and stuff. Learning a robust representation for each pixel is crucial for this task. Existing state-of-the-art (SOTA) algorithms employ deep neural networks to learn (discover) the representations needed for parsing from raw data. Nevertheless, these networks discover desired features or representations only from the given image (content), ignoring more generic knowledge contained in the dataset. To overcome this deficiency, we make the first attempt to explore the meaningful supportive knowledge, including general visual concepts (i.e., the generic representations for objects and stuff) and their relations from the whole dataset to enhance the underlying representations of a specific scene for better scene parsing. Specifically, we propose a novel supportive knowledge mining module (SKMM) and a knowledge augmentation operator (KAO), which can be easily plugged into modern scene parsing networks. By taking image-specific content and dataset-level supportive knowledge into full consideration, the resulting model, called knowledge augmented neural network (KANN), can better understand the given scene and provide greater representational power. Experiments are conducted on three challenging scene parsing and semantic segmentation datasets: Cityscapes, Pascal-Context, and ADE20K. The results show that our KANN is effective and achieves better results than all existing SOTA methods.

Artificial intelligence is used for various applications and is promising as an indispensable infrastructure in future societies. Neural networks are representative technologies that imitate human brains and exhibit various advantages. However, the size is bulky, the power is huge, and some advantages are not demonstrated because they are executed on Neumann-type computers. Neuromorphic systems are biomimetic systems from the hardware level to implement neuron and synapse elements, and the size is compact, the power is low, and the operation is robust. However, because the conventional ones are not composed of fully optimized hardware, the power is not yet minimal, and extra control circuits must be used. In this article, we developed a neuromorphic system using memcapacitors and autonomous local learning. By using memcapacitors, the power can be minimal, and by using autonomous local learning, the control circuits to handle the synapse elements can be deleted. First, the memcapacitors are completed in a cross-bar array, where the ferroelectric layers are sandwiched between the horizontal and perpendicular electrodes. The polarization and capacitance exhibit hysteresis due to the dielectric polarization. Next, autonomous local learning is introduced as follows. During the training phase, associative patterns to be memorized are directly sent, relatively high voltages are applied, and dielectric polarizations are induced. During the operation phase, relatively low voltages are applied, and input signals are weighted with the capacitances of the memcapacitors, summed, and transferred as the output signals. Finally, the experimental system is set up, and the experimental results are acquired. The memorized patterns during the training phase, distorted patterns as the input signals during the operation phase, and retrieved patterns as the output signals in the operation phase are shown. Researchers found that the retrieved patterns are completely the same as the memorized patterns. This means that the neuromorphic system works as an associative memory.

This article addresses the robust coordination problem for nonlinear uncertain second-order multiagent networks with motion constraints, including velocity saturation and collision avoidance. A single-critic neural network-based approximate dynamic programming approach and exact estimation of unknown dynamics are employed to learn online the optimal value function and controller. By incorporating avoidance penalties into tracking variable, constructing a novel value function, and designing of suitable learning algorithms, multiagent coordination and collision avoidance are achieved simultaneously. We show that the developed feedback-based coordination strategy guarantees uniformly ultimately bounded convergence of the closed-loop dynamical stability and all underlying motion constraints are always strictly obeyed. The effectiveness of the proposed collision-free coordination law is finally illustrated using numerical simulations.

The presented study focuses on a numerical simulation of the transport phenomena inside a solid oxide fuel cell anode. The classical mathematical model leads to a notable discrepancy between measured and predicted overpotentials. One of the possible reasons is the assumption of the constant values of electrochemical reaction charge transfer coefficients. A modified formulation of the problem includes data-driven correction of reaction charge transfer coefficients in the electrochemical reaction model. Here we show a dedicated computational scheme in which an artificial neural network updates the charge transfer coefficients depending on the operational conditions and the available datasets. The neural network was trained on twelve experimental data points of an anode's polarization curve obtained from the literature. The training set contained data for the anode operating in two different temperatures - 800 °C and 900 °C. The test set contained additional six data points for an anode operating at 1000 °C. Charge transfer coefficients were proposed by the Artificial Neural Network as a functional relation of the temperature and withdrawn current. The results of the predictions are juxtaposed with the experimental data from the literature. It was shown that an Artificial Neural Network could improve an electrochemical reaction model in Solid Oxide Fuel Cell modeling.

In this article, the local stabilization problem is investigated for a class of memristive neural networks (MNNs) with communication bandwidth constraints and actuator saturation. To overcome these challenges, a discontinuous event-trigger (DET) scheme, consisting of the rest interval and work interval, is proposed to cut down the triggering times and save the limited communication resources. Then, a novel relaxed piecewise functional is constructed for closed-loop MNNs. The main advantage of the designed functional consists in that it is positive definite only in the work intervals and the sampling instants but not necessarily inside the rest intervals. With the aid of extended reciprocally convex combination lemma, generalized sector condition, and some inequality techniques, two local stabilization criteria are established on the basis of both the discrete- and continuous-time Lyapunov methods. The proposed analysis technique fully takes advantage of the looped-functional and the event-trigger mechanism. Moreover, two optimization schemes are, respectively, established to design the control gain and enlarge the estimates of the admissible initial conditions (AICs) and the upper bound of rest intervals. Finally, some comparison results are given to validate the superiority of the proposed method.

Deep neural network (DNN)-task enabled mobile edge computing (MEC) is gaining ubiquity due to outstanding performance of artificial intelligence. By virtue of characteristics of DNN, this paper develops a joint design of task partitioning and offloading for a DNN task enabled MEC network that consists of a single server and multiple mobile devices (MDs), where the server and each MD employ the welltrained DNNs for task computation. The main contributions of this paper are as follows: First, we propose a layer-level computation partitioning strategy for DNN to partition each MDs task into the subtasks that are either locally computed at the MD or offloaded to the server. Second, we develop a delay prediction model for DNN to characterize the computation delay of each subtask at the MD and the server. Third, we design a slot model and a dynamic pricing strategy for the server to efficiently schedule the offloaded subtasks. Fourth, we jointly optimize the design of task partitioning and offloading to minimize each MDs cost that includes the computation delay, the energy consumption, and the price paid to the server. In particular, we propose two distributed algorithms based on the aggregative game theory to solve the optimization problem.

This paper introduces an algorithm for Facial Expression Recognition (FER) using deep Siamese Neural Networks (SNNs) that preserve the local structure of images in the embedding similarity space. We designed the network to reveal the input pairs similarity by comparing features through a designed metric. Furthermore, we developed a novel image pairing (i.e., positive and negative pairs) strategy technique to train our Siamese model. Our Siamese model comprises of a verification framework and an identification framework to learn a joint embedding space. The verification path reduces the intra-class variations by minimizing the distance between the extracted features from the same class, while the identification path increases the inter-class variations by maximizing the distance between the features extracted from different classes. We apply transfer learning to only use the identification model for facial expression classification. We evaluated our algorithm using AffectNet, FER2013, and Compound Facial Expressions of Emotion (CFEE) datasets, where better results are achieved compared to other deep learning-based approaches.

The adaptive neurofuzzy inference system (ANFIS) is a structured multioutput learning machine that has been successfully adopted in learning problems without noise or outliers. However, it does not work well for learning problems with noise or outliers. High-accuracy real-time forecasting of traffic flow is extremely difficult due to the effect of noise or outliers from complex traffic conditions. In this study, a novel probabilistic learning system, probabilistic regularized extreme learning machine combined with ANFIS (probabilistic R-ELANFIS), is proposed to capture the correlations among traffic flow data and, thereby, improve the accuracy of traffic flow forecasting. The new learning system adopts a fantastic objective function that minimizes both the mean and the variance of the model bias. The results from an experiment based on real-world traffic flow data showed that, compared with some kernel-based approaches, neural network approaches, and conventional ANFIS learning systems, the proposed probabilistic R-ELANFIS achieves competitive performance in terms of forecasting ability and generalizability.

Being able to recognize emotional intensity is a desirable feature for a facial emotional recognition (FER) system. However, the development of such a feature is hindered by the paucity of intensity-labeled data for model training. To ameliorate the situation, the present study proposes using face morphing as a way of data augmentation to synthesize faces that express different degrees of a designated emotion. Such an approach has been successfully validated on humans and machines. Specifically, humans indeed perceived different levels of intensified emotions in these parametrically synthesized faces, and FER systems based on neural networks indeed showed improved sensitivities to intensities of different emotions when additionally trained on the synthesized faces. Overall, the proposed data augmentation method is not only simple and effective but also useful for building FER systems that recognize facial expressions of mixed emotions.

Unmanned Aerial Vehicles (UAVs) have been utilized to serve on-ground users with various services, e.g., computing, communication and caching, due to their mobility and flexibility. The main focus of many recent studies on UAVs is to deploy a set of homogeneous UAVs with identical capabilities controlled by one UAV owner/company to provide services. However, little attention has been paid to the issue of how to enable different UAV owners to provide services with differentiated service capabilities in a shared area. To address this issue, we propose a multi-agent imitation learning enabled UAV deployment approach to maximize both profits of UAV owners and utilities of on-ground users. Specially, a Markov game is formulated among UAV owners and we prove that a Nash equilibrium exists based on the full knowledge of the system. For online scheduling with incomplete information, we design agent policies by imitating the behaviors of corresponding experts. A novel neural network model, integrating convolutional neural networks, generative adversarial networks and a gradient-based policy, can be trained and executed in a fully decentralized manner with a guaranteed -Nash equilibrium. Performance results show that our algorithm has significant superiority on average profits, utilities and execution time compared with other representative algorithms.

Heterogeneous information network (HIN) embedding, aiming to map the structure and semantic information in a HIN to distributed representations, has drawn considerable research attention. Graph neural networks for HIN embeddings typically adopt a hierarchical attention (including node-level and meta-path-level attentions) to capture the information from meta-path-based neighbors. However, this complicated attention structure often cannot achieve the function of selecting meta-paths due to severe overfitting. Moreover, when propagating information, these methods do not distinguish direct (one-hop) meta-paths from indirect (multi-hop) ones. But from the perspective of network science, direct relationships are often believed to be more essential, which can only be used to model direct information propagation. To address these limitations, we propose a novel neural network method via implicitly utilizing attention and meta-paths, which can relieve the severe overfitting brought by the current over-parameterized attention mechanisms on HIN. We first use the multi-layer graph convolutional network (GCN) framework, which performs a discriminative aggregation at each layer, along with stacking the information propagation of direct linked meta-paths layer-by-layer, realizing the function of attentions for selecting meta-paths in an indirect way. We then give an effective relaxation and improvement via introducing a new propagation operation which can be separated from aggregation. That is, we first model the whole propagation process with well-defined probabilistic diffusion dynamics, and then introduce a random graph-based constraint which allows it to reduce noise with the increase of layers. Extensive experiments demonstrate the superiority of the new approach over state-of-the-art methods.

This work introduces an efficient load scheduling method for handling the day-to-day power supply needs. At peak load times, due to its instabilitythe power generation system fails and as a measure, the load shedding process is followed. The presented method overcomes this problem by scheduling the load based on necessity. For this load scheduling is handled with an artificial neural network (ANN). For the training purpose the backpropagation (BP) algorithm is used. The whole load essential is the input of the neural network (NN). The power generation of all resources and power losses at the instant of transmission is the NN output. The optimum scheduling of different power sources is important when considering all the available sources. Load scheduling shares the feasibility of entire load and losses. It is well-known as optimal scheduling if the constraints such as availability of power, load requirement, cost and power losses are considered. Training the system using a large number of parameters would be a difficult task. So, finest number of communally independent inputs is selected. The presented method aims to lower the power generation expenditure and formulate the power available on demand without alteration. The network is designed using MATLAB.

Wind energy conversion systems (WECS) now play a significant role in meeting the world's energy needs. Several different approaches are used to try to increase the reliability of these renewable energy systems. Smart systems are designed to be more proactive to improve the performance of renewable energy equipment. Artificial neural networks (ANNs) have a variety of applications, including controlling renewable energy systems. Using optimal torque control (OTC) system based prediction techniques, a controller to monitor a maximum point of wind turbine output power (MPPT) is designed and modeled in this paper. MATLAB/Simulink package tools for neural networks are also used to design the controller and simulate it to achieve the necessary results and to obtain an appropriate analysis for the controller. The results show that the ANN is more active and delivers better output than the traditional controller.

The mathematical study known as queueing theory has recently become a major point of interest for many government agencies and private companies for increasing efficiency. One such application is vehicle queueing at an international port-of-entry (POE). When queueing, fumes from idling vehicles negatively affect the overall health and well-being of the community, especially the U.S. Customs and Border Protection (CBP) agents that work at the POEs. As such, there is a need to analyze and optimize the border crossing queuing operations to minimize wait times and number of vehicles in the queue and, thus, reduce the vehicle emissions. For this research, the U.S.–Mexico POE located at The Gateway International Bridge in Brownsville, Texas, is used as a case study. Due to data privacy concerns, the hourly wait times for vehicles arriving at the border had to be extracted manually each day using a live wait time tracker online. The data extraction was performed for the month of March 2022. Using these wait times, a queueing simulation software, SIMIO, was used to develop an interactive simulation model and calibrate the service rates. The output from the SIMIO model was then used to develop an artificial neural network (ANN) to predict hourly particulate matter content with an R2 of 0.402. From the ANN, a predictive equation has been developed, which may be used by CBP to make operational decisions and improve the overall efficiency of this POE. Thus, lowering the average wait times and the emissions from idling vehicles in the queue.

Freezing of gait (FoG) is identified as a sudden and brief episode of movement cessation despite the intention to continue walking. It is one of the most disabling symptoms of Parkinson's disease (PD) and often leads to falls and injuries. Many computer-aided FoG detection methods have been proposed to use data collected from unimodal sources, such as motion sensors, pressure sensors, and video cameras. However, there are limited efforts of multimodal-based methods to maximize the value of all the information collected from different modalities in clinical assessments and improve the FoG detection performance. Therefore, in this study, a novel end-to-end deep architecture, namely graph fusion neural network (GFN), is proposed for multimodal learning-based FoG detection by combining footstep pressure maps and video recordings. GFN constructs multimodal graphs by treating the encoded features of each modality as vertex-level inputs and measures their adjacency patterns to construct complementary FoG representations, thus reducing the representation redundancy among different modalities. In addition, since GFN is devised to process multimodal graphs of arbitrary structures, it is expected to achieve superior performance with inputs containing missing modalities, compared to the alternative unimodal methods. A multimodal FoG dataset was collected, which included clinical assessment videos and footstep pressure sequences of 340 trials from 20 PD patients. Our proposed GFN demonstrates a great promise of multimodal FoG detection with an area under the curve (AUC) of 0.882. To the best of our knowledge, this is one of the first studies to utilize multimodal learning for automated FoG detection, which offers significant opportunities for better patient assessments and clinical trials in the future.

The generation of a virtual, personal, auditory space to obtain a high-quality sound experience when using headphones is of great significance. Normally this experience is improved using personalized head-related transfer functions (HRTFs) that depend on a large degree of personal anthropometric information on pinnae. Most of the studies focus their personal auditory optimization analysis on the study of amplitude versus frequency on HRTFs, mainly in the search for significant elevation cues of frequency maps. Therefore, knowing the HRTFs of each individual is of considerable help to improve sound quality. The following work proposes a methodology to model HRTFs according to the individual structure of pinnae using multilayer perceptron and linear regression techniques. It is proposed to generate several models that allow knowing HRTFs amplitude for each frequency based on the personal anthropometric data on pinnae, the azimuth angle, and the elevation of the sound source, thus predicting frequency magnitudes. Experiments show that the prediction of new personal HRTF generates low errors, thus this model can be applied to new heads with different pinnae characteristics with high confidence. Improving the results obtained with the standard KEMAR pinna, usually used in cases where there is a lack of information. The authors wish to thank to the Basque Government for its support through the KK-2019-00033 METALCR2, and the University of the Basque Country UPV/EHU for its support through the MOV21/03.

Problem definition: We present a data-driven study of the secondary ticket market. In particular, we are primarily concerned with accurately estimating price sensitivity for listed tickets. In this setting, there are many issues including endogeneity, heterogeneity in price sensitivity for different tickets, binary outcomes, and nonlinear interactions between ticket features. Our secondary goal is to highlight how this estimation can be integrated into a prescriptive trading strategy for buying and selling tickets in an active marketplace. Academic/practical relevance: We present a novel method for demand estimation with heterogeneous treatment effect in the presence of confounding. In practice, we embed this method within an optimization framework for ticket reselling, providing the ticket reselling platform with a new framework for pricing tickets on its platform. Methodology: We introduce a general double/orthogonalized machine learning method for classification problems. This method allows us to isolate the causal effects of price on the outcome by removing the conditional effects of the ticket and market features. Furthermore, we introduce a novel loss function that can be easily incorporated into powerful, off-the-shelf machine learning algorithms, including gradient boosted trees. We show how, in the presence of hidden confounding variables, instrumental variables can be incorporated. Results: Using a wide range of synthetic data sets, we show this approach beats state-of-the-art machine learning and causal inference approaches for estimating treatment effects in the classification setting. Furthermore, using National Basketball Association ticket listings from the 2014–2015 season, we show that probit models with instrumental variables, previously used for price estimation of tickets in the resale market, are significantly less accurate and potentially misspecified relative to our proposed approach. Through pricing simulations, we show our proposed method can achieve an 11% return on investment by buying and selling tickets, whereas existing techniques are not profitable. Managerial implications: The knowledge of how to price tickets on its platform offers a range of potential opportunities for our collaborator, both in terms of understanding sellers on their platform and in developing new products to offer them. History: This paper has been accepted as part of the 2019 Manufacturing & Service Operations Management Practice-Based Research Competition. Funding: This work was supported by the National Science Foundation [Grant CMMI-1563343]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/msom.2021.1065 .

Relation classification (RC) task is one of fundamental tasks of information extraction, aiming to detect the relation information between entity pairs in unstructured natural language text and generate structured data in the form of entity-relation triple. Although distant supervision methods can effectively alleviate the problem of lack of training data in supervised learning, they also introduce noise into the data, and still cannot fundamentally solve the long-tail distribution problem of the training instances. In order to enable the neural network to learn new knowledge through few instances like humans, this work focuses on few-shot relation classification (FSRC), where a classifier should generalize to new classes that have not been seen in the training set, given only a number of samples for each class. To make full use of the existing information and get a better feature representation for each instance, we propose to encode each class prototype in an adaptive way from two aspects. First, based on the prototypical networks, we propose an adaptive mixture mechanism to add label words to the representation of the class prototype, which, to the best of our knowledge, is the first attempt to integrate the label information into features of the support samples of each class so as to get more interactive class prototypes. Second, to more reasonably measure the distances between samples of each category, we introduce a loss function for joint representation learning to encode each support instance in an adaptive manner. Extensive experiments have been conducted on FewRel under different few-shot (FS) settings, and the results show that the proposed adaptive prototypical networks with label words and joint representation learning has not only achieved significant improvements in accuracy, but also increased the generalization ability of few-shot RC models.

This article is concerned with the neural network (NN)-based event-triggered control problem for discrete-time networked Markov jump systems with hybrid cyberattacks and unmeasured states. The event-triggered mechanism (ETM) is used to reduce the communication load, and a Luenberger observer is introduced to estimate the unmeasured states. Two kinds of cyberattacks, denial-of-service (DoS) attacks and deception attacks, are investigated due to the vulnerability of cyberlayer. For the sake of mitigating the impact of these two types of cyberattacks on system performance, the ETM under DoS jamming attacks is discussed first, and a new estimation of such mechanism is given. Then, the NN technique is applied to approximate the injected false information. Some sufficient conditions are derived to guarantee the boundedness of the closed-loop system, and the observer and controller gains are presented by solving a set of matrix inequalities. The effectiveness of the presented control method is demonstrated by a numerical example.

In this article, the exponential synchronization control issue of reaction-diffusion neural networks (RDNNs) is mainly resolved by the sampling-based event-triggered scheme under Dirichlet boundary conditions. Based on the sampled state information, the event-triggered control protocol is updated only when the triggering condition is met, which effectively reduces the communication burden and saves energy. In addition, the proposed control algorithm is combined with sampled-data control, which can effectively avoid the Zeno phenomenon. By thinking of the proper Lyapunov-Krasovskii functional and using some momentous inequalities, a sufficient condition is obtained for RDNNs to achieve exponential synchronization. Finally, some simulation results are shown to demonstrate the validity of the algorithm.

Amidst a world of never-ending waste production and waste disposal crises, scientists have been working their ways to come up with solutions to serve the earth better. Two such commonly found trashes deteriorating the environment are glass and tin can waste. This study aims to investigate the comparative suitability of response surface methodology (RSM) and artificial neural network (ANN) in predicting the mechanical strength of concrete prepared with fine glass aggregate (GFA) and condensed milk can (tin) fibers (CMCF). An experimental scheme has been designed in this study with two input variables as GFA and CMCF, and two output variables as compressive and splitting tensile strength. The results show that both variables influenced the compressive and splitting tensile strength of concrete at 7, 28, and 56 days (p

The goal of hyperspectral image fusion (HIF) is to reconstruct high spatial resolution hyperspectral images (HR-HSI) via fusing low spatial resolution hyperspectral images (LR-HSI) and high spatial resolution multispectral images (HR-MSI) without loss of spatial and spectral information. Most existing HIF methods are designed based on the assumption that the observation models are known, which is unrealistic in many scenarios. To address this blind HIF problem, we propose a deep learning-based method that optimizes the observation model and fusion processes iteratively and alternatively during the reconstruction to enforce bidirectional data consistency, which leads to better spatial and spectral accuracy. However, general deep neural network inherently suffers from information loss, preventing us to achieve this bidirectional data consistency. To settle this problem, we enhance the blind HIF algorithm by making part of the deep neural network invertible via applying a slightly modified spectral normalization to the weights of the network. Furthermore, in order to reduce spatial distortion and feature redundancy, we introduce a Content-Aware ReAssembly of FEatures module and an SE-ResBlock model to our network. The former module helps to boost the fusion performance, while the latter make our model more compact. Experiments demonstrate that our model performs favorably against compared methods in terms of both nonblind HIF fusion and semiblind HIF fusion.

Forest production and growth are obtained from statistical models that allow the generation of information at the tree or forest stand level. Although the use of regression models is common in forest measurement, there is a constant search for estimation procedures that provide greater accuracy. Recently, machine learning techniques have been used with satisfactory performance in measuring forests. However, methods such as Adaptive Neuro-Fuzzy Inference System (ANFIS) and Random Forest are relatively poorly studied for predicting the volume of wood in eucalyptus plantations in Brazil. Therefore, it is essential to check whether these techniques can provide gains in terms of accuracy. Thus, this study aimed to evaluate the use of Random Forest and ANFIS techniques in the prognosis of forest production. The data used come from continuous forest inventories carried out in stands of eucalyptus clones. The data were divided into 70% for training and 30% for validation. The algorithms used to generate rules in ANFIS were Subtractive Clustering and Fuzzy-C-Means. Besides, training was done with the hybrid algorithm (descending gradient and least squares) with the number of seasons ranging from 1 to 20. Several RFs were trained, varying the number of trees from 50 to 850 and the number of observations by five leaves to 35. Artificial neural networks and decision trees were also trained to compare the feasibility of the techniques. The evaluation of the estimates generated by the techniques for training and validation was calculated based on the following statistics: correlation coefficient (r), relative Bias (RB), and the relative root mean square error (RRMSE) in percentage. In general, the techniques studied in this work showed excellent performance for the training and validation data set with RRMSE values 0.98. The RF presented inferior statistics about the ANFIS for the prognosis of forest production. The Subtractive Clustering (SC) and Fuzzy-C-Means (FCM) algorithms provide accurate baseline and volume projection estimates; both techniques are good alternatives for selecting variables used in modeling forest production.

Effective prediction of turning movement counts at intersections through efficient and accurate methods is essential and needed for various applications. Commonly predictive methods require extensive data collection, calibration, and modeling efforts to estimate turning movements. In this study, three models were proposed to estimate turning movements at signalized intersections using approach volumes. Two sets of data from the United States and Canada were obtained to develop and test the proposed models. Machine learning-based regression models, including random forest regressor (RFR) and multioutput regressor (MOR) in addition to an artificial neural network (ANN) model, were developed and trained to analyze the relationship between approach volumes and corresponding turning movements. Multiple evaluation measurements were utilized to compare the models. All models produced satisfactory results. The RFR regression model outperformed the MOR model. However, the ANN model had the best performance when compared to the other models. The proposed models provide traffic engineers and planners with reliable and fast methods to estimate turning movements. 2022 Tongji University, Tongji University Press The authors would like to thank the reviewers for their dedicated work and insightful comments and recommendations. Scopus

Drug-drug interactions are one of the main concerns in drug discovery. Accurate prediction of drug-drug interactions plays a key role in increasing the efficiency of drug research and safety when multiple drugs are c o-prescribed. With various data sources that describe the relationships and properties between drugs, the comprehensive approach that integrates multiple data sources would be considerably effective in making high-accuracy prediction. In this paper, we propose a Deep Attention Neural Network based Drug-Drug Interaction prediction framework, abbreviated as DANN-DDI, to predict unobserved drug-drug interactions. First, we construct multiple drug feature networks and learn drug representations from these networks using the graph embedding method; then, we concatenate the learned drug embeddings and design an attention neural network to learn representations of drug-drug pairs; finally, we adopt a deep neural network to accurately predict drug-drug interactions. The experimental results demonstrate that our model DANN-DDI has improved prediction performance compared with state-of-the-art methods. Moreover, the proposed model can predict novel drug-drug interactions and drug-drug interaction-associated events.

Morphing is a promising bio-inspired technology, with the potential to make aircraft more economical and sustainable through adaptation of the wing shape for best efficiency at any flight condition. This paper proposes an online black-box performance optimization strategy for a seamless wing with distributed morphing control. Pursuing global performance, the presented method integrates a global radial basis function neural network (RBFNN) surrogate model with a derivative-free evolutionary optimization algorithm. The effectiveness of the optimization strategy was validated on a vortex lattice method (VLM) aerodynamic model of an over-actuated morphing wing augmented by wind tunnel experiment data. Simulations show that the proposed method is able to control the morphing shape and angle of attack to achieve various target lift coefficients with better aerodynamic efficiency than the unmorphed wing shape. The global nature of the on-board model allows the presented method to find shape solutions for a wide range of target lift coefficients without the need for additional model excitation maneuvers. Compared to the unmorphed shape, up to 14.6% of lift-to-drag ratio increase is achieved.

In this study, an adaptive neural network (NN) control using nonlinear information (NI) gain for permanent magnet synchronous motors (PMSMs) is proposed to improve control and estimation performance. The proposed method consists of a nonlinear controller, a three-layer NN approximator, and NI gain. The nonlinear controller is designed via a backstepping procedure for position tracking. The commutation scheme is designed to implement the PMSM control without the direct-quadrature (DQ) transform. The three-layer NN approximator is designed to estimate the unknown complex function generated by the recursive backstepping process. The NI gains are designed to enhance the control and estimation performance according to the increased tracking errors owing to the load torque and the desired position variations. All of signals in the closed-loop system guarantee the semiglobal uniformly ultimately boundness (UUB) using the Lyapunov stability theorem and the input-to-state stability (ISS) property. The performance of the proposed method was validated by experiments performed using a PMSM testbed.

Missing data widely exist in the raw or processed data, implying information loss. In many cases, missing values have to be accurately imputed for further use. In this paper, an extreme case, consecutively missing data in large-length and mainly remaining data in small-length, is discussed for time series varying with operating conditions, very universal in industrial processes. Firstly, to fully utilize the information of remaining data, a similar conditions screening scheme is provided, efficient to improve imputation accuracy. Then, multiple receding imputation via Gaussian process regression (GPR) and long short term memory (LSTM) neural network are proposed, deducing generic multiple combination imputation and bidirectional imputation structures. At last, applied for data imputation of extremely missing wind power data, condition-dependent on wind speed, imputation effects of the proposed methods are carefully compared. Simulation results reveal effectiveness of these methods to impute missing data under the extreme case, laying very important foundation for data-driven applications.

For bonded Fibre Reinforced Polymer (FRP) strengthening systems in civil engineering projects, the adhesive joint performance is a key factor in the effectiveness of the strengthening; however, it is known that the material properties of structural epoxy adhesives change with temperature. This present paper examines the implied relationship between the curing regimes and the storage modulus response of the adhesive using a Machine Learning (ML) approach. A dataset containing 157 experimental data collected from the scientific papers and academic theses was used for training and testing an Artificial Neural Network (ANN) model. The sensitivity analysis reveals that the curing conditions have a significant effect on the glass transition temperatures (T g) of the adhesive, and consequently on the storage modulus response at elevated temperatures. Curing at an extremely high temperature for a long time does not, however, guarantee a better thermal performance. For the studied adhesive, curing in a warm (≥ 45°C) and dry (near 0 % RH) environment for 21 days is recommended for practical applications. A software with a Graphical User Interface (GUI) was established, which can predict the storage modulus response of the adhesive, plot the corresponding response curve, and estimate the optimum curing condition.

У роботі пропонується метод динамічного управління буфером запасів на основі м’яких обчислень. Новизна дослідження полягає в тому, що для динамічного управління буфером запасів було створено метод на основі нечіткої логіки та штучної нейронної мережі, дві моделі штучної нейро-нечіткої мережі динамічного управління буфером запасів, вибрано три критерії оцінки ефективності запропонованих моделей, ідентифіковано параметри запропонованих моделей на основі методу зворотного поширення в пакетному режимі, орієнтовані на технологію паралельної обробки інформації. Запропоновані моделі та процедури їх параметричної ідентифікації дозволяють підвищити швидкодію, точність і надійність прийняття рішення. Запропонований метод динамічного управління буфером запасів на основі м’яких обчислень може використовуватись у різних інтелектуальних системах. Currently, more and more companies seek to improve and optimize their business processes based on the implementation of the technology of the theory of constraints, which provides dynamic management of the inventory buffer and is used to manage supply chains. As a result, the relevance of the development of methods of intellectualization of the technology of the theory of constraints is increasing significantly. To date, there are no computer systems for dynamic management by the inventory buffer, which are based on soft calculations. The aim of the work is to improve the efficiency of dynamic management of the inventory buffer by means of an artificial neuro-fuzzy network, which is trained on the basis of the back-propagation method. In order to solve the problem of increasing the efficiency of the dynamic management of the inventory buffer, appropriate methods of artificial intelligence were investigated. Research data has shown that the most effective method today is the use of artificial neural networks in combination with a fuzzy inference system. The paper proposes a method of dynamic management of the inventory buffer based on soft calculations. The novelty of the research is that for dynamic management of the inventory buffer a method based on fuzzy logic and an artificial neural network, and also two models of artificial neuro-fuzzy network of the dynamic management of the inventory buffer have been created, three criteria for evaluating the effectiveness of the proposed models have been selected, the parameters of the proposed models based on the method of back propagation in batch mode, oriented on the technology of information parallel processing, have been identified. As a result of numerical study, it is established that the proposed method of neuro-fuzzy dynamic management of the inventory buffer provides a probability of incorrectly made decisions regarding the dynamic management of the inventory buffer of 0.07, and a root mean square error of 0.10. The proposed models and procedures for their parametric identification allow to increase the speed, accuracy and reliability of decision-making. The proposed method of dynamic management of the inventory buffer based on soft calculations can be used in various intelligent systems.

With the rapid progress of wireless communication technologies along with their digital revolutions, the quantity of Internet of Things (IoT) has been increased by manifolds resulting in a huge increase in data volume and network traffic. It became easier for an intruder to pretend as a valid service provider, and generate different types of network attacks. This becomes even more severe when the service involves digital financial transactions. This paper proposes an intrusion detection system (IDS) based on a stacked autoencoder (AE) and a deep neural network (DNN). The stacked AE learns the features of the input network record in an unsupervised manner to decrease the feature width. Then, the DNN is trained in a supervised manner to extract deep-learned features for the classifier. In the proposed system, the stacked AE has two latent layers and the DNN has two or three layers, where each layer has a fully-connected layer, a batch normalization, and a dropout. The system was evaluated on three publicly available datasets: KDDCup99, NSL-KDD, and AWID. Experimental results exhibited that the proposed IDS achieved 94.2%, 99.7%, 99.9% accuracy, respectively, for multiclass classification.

The state of health (SOH) estimation is of great importance in the battery management system. However, the accurate real time state of health estimation is still a challenge because the capacity of the battery cannot be precisely monitored and measured in real-time. In response to this problem, this paper proposes an online SOH monitoring method for lithium-ion battery (LIB) based on the charge-discharge feature of battery in real time. Firstly, the relationship between the charge-discharge feature and SOH through principal component analysis (PCA) is analyzed, and the appropriate feature which can precisely describe the battery aging process is then selected. Secondly, as to decompose and denoise the collected charge-discharge feature data, the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) is applied. The major trend obtained is predicted by logistic regression based on the sliding time window (LR-STW), and the minor fluctuation is predicted by Kalman Filter (KF). Finally, an online SOH estimation method based on RBF neural network is proposed to establish the mapping relationship between feature and SOH. The proposed method was verified by NASA experimental data. The experimental results show that the proposed method can accurately predict the LIB SOH in real time, and the robustness of the proposed method is strengthened and promised.

The article investigates full information control problem for switched neural networks subject to fault and disturbance. First, the main objective is realizing interval stability and zero tracking error under condition that neither of the neuron states' vectors including the plant and reference models is available. Second, the desired full information controller and neural networks' observer are designed to ensure observer-based dynamic error system mean-square exponentially stable with sufficient condition of strict weight ${H}_{∞} /{H}_-$ performance levels. Finally, we concentrate on stability analyses and fault tolerance for switched neural networks with fault accompanied by disturbance through linear matrix inequalities (LMIs), Lyapunov function, and average dwell time, discussing it according to different values of fault. Finally, simulation examples are listed to account for the availability and effectiveness of the research methodology.

This paper aims to predict a set of apps a user will open on her mobile device in the next time slot. Such an information is essential for many smartphone operations, e.g., app pre-loading and content pre-caching, to improve user experience. However, it is hard to build an explicit model that accurately captures the complex environment context and predicts a set of apps at one time. This paper presents a deep reinforcement learning framework, named as DeepAPP, which learns a model-free predictive neural network from historical app usage data. Meanwhile, an online updating strategy is designed to adapt the predictive network to the time-varying app usage behavior. To transform DeepAPP into a practical deep reinforcement learning system, several challenges are addressed by developing a context representation method for complex contextual environment, a general agent for overcoming data sparsity and a lightweight personalized agent for minimizing the prediction time. Extensive experiments on a large-scale anonymized app usage dataset reveal that DeepAPP provides high accuracy (precision 70.6% and recall of 62.4%) and reduces the prediction time of the state-of-the-art by 6.58 times. A field experiment of 29 participants demonstrates DeepAPP can effectively reduce launch time of apps.

International audience; New investigations were carried out to demonstrate the ability of Trichoderma viride dead biomass for the bioremoval of Cr(VI) from aqueous solutions. The gathered experimental data were initially modelled considering both adsorption and reduction kinetic models, followed by equilibrium models. In the next step, Design of Experiments method (DoE) and a Differential Evolution (DE) algorithm were applied for determining the optimal Response Surface Methodology and Artificial Neural Network (ANN) models which are able to correlate the process efficiency with the experimental conditions. In addition, DE (in combination with the previously determined ANN) was used to obtain the optimal working conditions for maximum process efficiency. These optimization procedures allow setting various process parameters and working conditions which ensure the maximum bioremoval efficiency. DoE highlighted that all the study parameters have statistically significant individual influence on Cr(VI) bioremoval efficiency and, for example, at pH 2 and a temperature of 50 °C, 100% of Cr(VI) is bioremoved if the biosorbent dosage is 10 g/L, considering a minimum contact time of 2 h and given a maximum initial concentration of 122.5 mg/L. According to the ANN model, a contact time set at 24 h yields optimal values for Cr(VI) bioremoval by dead biomass considering pH 1.00, 3 g/L biosorbent dosage, 23 mg/L metallic ion concentration and a temperature of 50 °C. pHpzc, FT-IR and SEM-EDX analyses were included for biomass characterization and to underline the bioremoval mechanism of Cr(VI). Our results are providing a solid basis for the development of technologies for detoxification of contaminated effluents. © 2022 Institution of Chemical Engineers

This article investigates the problem of global neural network (NN) tracking control for uncertain nonlinear systems in output feedback form under disturbances with unknown bounds. Compared with the existing NN control method, the differences of the proposed scheme are as follows. The designed actual controller consists of an NN controller working in the approximate domain and a robust controller working outside the approximate domain, in addition, a new smooth switching function is designed to achieve the smooth switching between the two controllers, in order to ensure the globally uniformly ultimately bounded of all closed-loop signals. The Lyapunov analysis method is used to strictly prove the global stability under the combined action of unmeasured states and system uncertainties, and the output tracking error is guaranteed to converge to an arbitrarily small neighborhood through a reasonable selection of design parameters. A numerical example and a practical example were put forward to verify the effectiveness of the control strategy.