Feed-forward Artificial Neural Network Research Articles

A hybrid physics-based and data-driven model for intra-day and day-ahead wind power forecasting considering a drastically expanded predictor search space

This work presents a novel hybrid (physics- and data-driven) model for short-term (intra-day and day-ahead, 3h-24h) wind power forecasting (STWPF). Traditionally, STWPF predictors admitted very few meteorological variables only from the grid points closest to the turbines. Here, with the aim to further capture the underlying atmospheric processes ruling the wind variability in the wind farm, the approach relies on drastically expanding the predictor space, composed of numerous meteorological variables throughout a large geographical domain, retrieved from a weather forecasting model (COSMO-1). An ad-hoc genetic algorithm that optimizes the selection of predictors is designed and combined with feed-forward artificial neural networks for its cost function evaluation. The introduced model is compared to multiple benchmark models in a 16-turbine wind farm in the Swiss Jura mountains. For +12h and +24h lead times, the new approach shows a root-mean squared error normalized to the installed wind farm capacity of 11% and 11.6%, respectively. These values entail ∼16% higher forecasting skill compared to state-of-the-art predictor frameworks. Results highlight the ability of the presented approach to systematically select as predictors different variables with a well-known impact on the wind farm performance, such as the turbulent kinetic energy or the vertical wind shear. Clustering the data according to the wind direction provides substantial benefit. In addition, it provides a better understanding of the attained improvement: largest performances occur in those wind directions affected by highly complex terrain. This indicates that the proposed model can be especially suitable for wind farms in complex terrain.

Novel Advanced Artificial Neural Network-Based Online Stator and Rotor Resistance Estimator for Vector-Controlled Speed Sensorless Induction Motor Drives

This paper presents a new approach for the online estimation of stator and rotor resistance of induction motors for speed sensorless vector-controlled drives, using feed-forward artificial neural networks with advanced adaptive learning rates. For the rotor resistance estimation, a neural network model based on rotor speed and stator currents is developed. The rotor flux linkages acquired from the voltage model are compared with the neural network model. The feed-forward neural network employs an adaptive learning rate as the function of the obtained error during training for quick convergence with minimal estimation error. A two-layered neural network model based on the stator voltage and current equations is developed for the stator resistance estimation. The d-q axes stator currents obtained from the developed model are compared with the acquired d-q axes stator currents. For the fast convergence with minimal estimation error, an adaptive learning rate as the function of error is adopted during training. Furthermore, the neural network estimates the induction motor’s speed. The simulation and experimental results justify that the developed algorithms track variation in the resistances quickly and precisely along with the speed as compared with the conventional constant learning rate algorithm, leading to reliable operation of the drive.

Efficient calibration of cost-efficient particulate matter sensors using machine learning and time-series alignment

Atmospheric particulate matter (PM) poses a significant threat to human health, infiltrating the lungs and brain and leading to severe issues such as heart and lung diseases, cancer, and premature death. The main sources of PM pollution are vehicular and industrial emissions, construction and agricultural activities, and natural phenomena such as wildfires. Research underscores the absence of a safe threshold for particulate exposure, highlighting the crucial need for monitoring PM levels to develop and implement effective risk mitigation measures. Notwithstanding, accurate measurement of PM concentration relies on expensive and cumbersome equipment. Despite the rising popularity of low-cost alternatives, their reliability remains questionable, given their sensitivity to environmental conditions, inherent instability, and manufacturing imperfections. This article proposes a novel approach to efficient correction of low-cost PM sensors. The primary calibration model is a feedforward artificial neural network (ANN), which directly renders predicted output of the corrected sensor based on environmental variables such as temperature, humidity, and atmospheric pressure. The ANN hyper-parameters are identified by aligning time series of prior reference and low-cost sensor readings, which enables the network to learn typical temporal changes of the sensor outcome as a function of the aforementioned parameters as well as operational relationships between the sensor and the reference device. The architecture of the ANN is optimized in terms of the number of neurons in each layer to enhance its generalization capability. Our methodology has been demonstrated using a custom-designed portable monitoring platform and reference data acquired from public stations in Gdansk. The results are indicative of excellent calibration reliability. The achieved correlation coefficients w.r.t. the reference readings are 0.86, 0.88, and 0.72 for PM1, PM10, and PM2.5, respectively, whereas RMSE values are only 3.0, 3.9, and 5.4 µg/m³.

Enhancing municipal solid waste leachate treatment efficiency: AI-based prediction of electrocoagulation/flocculation recovery using iron electrodes

ABSTRACT This study addresses a gap in municipal leachate (MUPL) treatment by introducing a pioneering application of artificial intelligence (AI) in the electrocoagulation/electroflocculation (EC/EF) process utilizing iron electrodes. The overarching aim is to demonstrate the efficacy of AI, particularly a multi-layer perceptron (MLP)-based feed-forward artificial neural network (ANN) incorporating the Levenberg-Marquardt (LMb) algorithm, in predicting and optimizing EC/EF outcomes for turbidity (TDY) removal. The research methodology involved experimentation and robust ANN data modeling. The significance of this work emerges from the successful integration of AI, showcasing its potential in advancing wastewater, demonstrated through a strong positive correlation (0.994) between the ANN model predictions and experimental outcomes. The study achieves a remarkable 99.4% TDY removal at an electrolysis time of 10 min and contributes valuable insights into the critical parameters influencing the EC/EF process. Results from the ANN modeling exhibit high predictive accuracy, supported by elevated R-squared values and minimal mean square error. Statistical analyses underscore the significance of key process parameters, highlighting the influential roles of current intensity and settling time. The study emphasized the favourable impact of maintaining an acidic pH range, as it reduced electrostatic repulsion between particles, facilitating pollutant agglomeration, and identified electrolysis time as a key factor in enhancing treatment efficiency, supported by a strong positive correlation between electrolysis time and TDY reduction. Energy cost savings were realized by not requiring temperature elevation. Achieving a 99.4% TDY removal translates to substantial reductions in other pollutants present in the MUPL, thereby elevating water quality and ensuring compliance.

Approximation of One-Dimensional Darcy–Brinkman–Forchheimer Model by Physics Informed Deep Learning Feedforward Artificial Neural Network and Finite Element Methods: A Comparative Study

Approximation of One-Dimensional Darcy–Brinkman–Forchheimer Model by Physics Informed Deep Learning Feedforward Artificial Neural Network and Finite Element Methods: A Comparative Study

Predictive modeling of copper (II) adsorption from aqueous solutions by sawdust: a comparative analysis of adaptive neuro-fuzzy interference system (ANFIS) and artificial neural network (ANN) approaches

Heavy metal ions are considered to be the most prevalent and toxic water contaminants. The objective of thois work was to investigate the effectiveness of employing the adsorption technique in a laboratory-size reactor to remove copper (II) ions from an aqueous medium. An adaptive neuro-fuzzy inference system (ANFIS) and a feed-forward artificial neural network (ANN) were used in this study. Four operational factors were chosen to examine their influence on the adsorption study: pH, contact duration, initial Cu (II) ions concentration, and adsorbent dosage. Using sawdust from wood, prediction models of copper (II) ions adsorption were optimized, created, and developed using the ANN and ANFIS models for tests. The result indicates that the determination coefficient for copper (II) metal ions in the training dataset was 0.987. Additionally, the ANFIS model’s R 2 value for both pollutants was 0.992. The findings demonstrate that the models presented a promising predictive approach that can be applied to successfully and accurately anticipate the simultaneous elimination of copper (II) and dye from the aqueous solution.

A New Global Gas Dynamics Performance Optimization Method for Refrigeration Centrifugal Compressor Stage Based on the Immune Algorithm

Abstract This study proposes a new global gas dynamics optimization method, which was applied to a multi-objective optimization task of centrifugal compressor performance with the aim of determining the improvement probability for achieving high efficiency across a wide operating range. Initially, the original nondominated neighbor immune algorithm (NNIA) was extended to solve constrained multi-objective optimization problems for the first time, which mainly incorporated a procedure for handling inequality and equality constraints without additional parameters. Subsequently, an adaptive topological back-propagation multilayer feed-forward artificial neural network (BP-MLFANN) was trained using the modified NNIA to quickly evaluate the fitness value of the centrifugal compressor stage performance during the optimization. The feasibility of the method was validated using the first stage of a refrigeration centrifugal compressor. The results indicated a substantial enhancement in the stage efficiency of the optimized impeller at the Near-stall, Design, and Near-choke operating points, with increasement of 1.8%, 1.9%, and 4%, respectively, as compared to the baseline stage. The flow field analysis shows that the impact loss at impeller leading edge and flow separation in the passage reduced greatly, the mixing process between the leakage flow and mainstream in the channel is significantly weakened, thus the flow field becomes more uniform after optimization. The new global gas dynamics optimization method provides a reference for the development of efficient and rapid optimization techniques for centrifugal compressor.

Spatial Air Quality Assessment in Peninsular Malaysia: Insights from Environmetric Analyses and Artificial Neural Networks

The research investigated the spatial distribution of air quality in Peninsular Malaysia to offer crucial insights for mitigating air quality degradation. Utilizing historical data obtained from the Malaysian Department of Environment, eleven years' worth (2011-2021) of daily readings of PM10, O3 , SO2 , NO2 , and CO from thirty-four monitoring stations were subjected to environmetric techniques and Artificial Neural Network analysis. HACA successfully classified seven stations into the High Pollution Cluster (HPC), seven stations into the Moderate Pollution Cluster (MPC), and twenty stations into the Low Pollution Cluster (LPC). The discriminant analysis demonstrated a correct assignment rate of 91.87%, suggesting that all five parameters were able to discriminate with a significance level of p<0.0001. Principal Component Analysis (PCA) revealed two Variance Factors (VFs) across all clusters, with cumulative variances of 69.43% (HPC), 82.32% (MPC), and 62.16% (LPC) respectively. Furthermore, employing a Multi-Layer Perceptron Feed-Forward Artificial Neural Network (MLP-FF-ANN) model to predict API readings yielded a significant and strong correlation, as indicated by R2 =0.7774 and RMSE=9.9048. These findings hold significant potential in informing the adoption of effective preventative and management strategies.

Optimization of α-L-arabinofuranosidase CcABF on clarification and beneficial active substances in fermented ginkgo kernel juice by artificial neural network and genetic algorithm

This study aimed at using α-L-arabinofuranosidase CcABF to improve the clarity and active substances in fermented ginkgo kernel juice by artificial neural network (ANN) modeling and genetic algorithm (GA) optimization. A credible three-layer feedforward ANN model was established to predict the optimal parameters for CcABF clarification. The experiments proved the highest transmittance of 89.40% for fermented ginkgo kernel juice with this understanding, which exhibited a 25.56% increase over the unclarified group. With the clarification of CcABF, the antioxidant capacity in juice was enhanced with the increase of total phenolic and flavone contents, and the maximum DPPH and hydroxyl radical scavenging rates were increased by 89.71% and 26.65%, respectively. The contents of toxic ginkgolic acids declined markedly, while the active ingredients of ginkgetin and ginkgolide B showed a modest increase. Moreover, changes in free amino acids and volatile compounds improved the nutritive value and flavor of clarified fermented ginkgo kernel juice.

Estimating cost of pothole repair from digital images using Stereo Vision and Artificial Neural Network

A significant amount of road maintenance cost goes into pothole repairs. The primary cost factors related to potholes are their size and depth, as larger and thicker potholes incur higher repair costs. However, existing methods for estimating pothole repair in developing countries rely on manual size measurements, which is time consuming, labor intensive, subjective and can lead to poor estimation of repair cost. This paper presents a system that can automatically determine the size of potholes from digital images and estimate the cost of repair. In this study, the stereo vision method was used to automatically estimate the depths of potholes from digital camera images. A feed-forward backward propagation Artificial Neural Network (ANN) was trained using pothole images acquired using mobile phones. The predicted depths and sizes of the potholes were then used to estimate the quantity of materials required to fill the potholes and subsequently, the cumulative cost of repair. Marking out and manual size measurements were performed for twenty randomly selected potholes in the Ugbowo Campus of the University of Benin, Nigeria. These measurements were compared against the estimated sizes of potholes predicted by the ANN model. A system was developed to automatically compute these material costs and considering other cost components such as transportation, labor, and equipment. Results obtained showed that the mean errors for depth, width and height estimates were 3.403%, 3.789% and 5.2617% respectively. Consequently, the developed system correctly estimated the cost of repair of the potholes considered in this study. A significant contribution of the paper is the speed and convenience of acquiring pothole data using a mobile phones without the need for on spot assessment of potholes or use of relatively more expensive stereoscopic camera setup.

Using Deep Neural Networks and Big Data to Predict Microscopic Travel Time in Work Zones

Traffic management in work zones is a challenging task that requires a balance between creating a safe environment for workers and minimizing traffic delays. Accurately predicting travel times through work zones is essential for dynamic traffic management and reducing congestion. However, complex interrupted traffic patterns in work zones (also known as construction zones) make this challenging compared with regular traffic congestion and free-flow conditions. To address this complexity, the paper develops a data-driven deep feed-forward artificial neural network to forecast under-construction travel times in work zones using an integrated data set of almost half a million observations. The variables considered in the neural network include the work zone characteristics, road design features, weather information, and traffic flow information. The raw data set comprised approximately 15 million travel time observations collected at 674 work zones. After cleaning and preprocessing the data and applying feature engineering techniques, the raw data set was reduced to 81 work zones spread along a 700 km corridor between the western borders of Alberta and Vancouver, British Columbia. The data were split into training and test data using a 90:10 ratio. Training data were used to develop a 27-input neural network with four hidden layers. After validating the test data, the neural network achieved a root mean squared error of 0.150 min and an R-squared score of 0.945, indicating a high accuracy level in estimating the under-construction travel time.

Energy and economic analysis of building integrated photovoltaic thermal system: Seasonal dynamic modeling assisted with machine learning-aided method and multi-objective genetic optimization

Building integrated photovoltaic thermal (BIPV/T) systems offer a highly effective means of generating clean energy for both electricity and heating purposes in residential buildings. Hence, this article introduces a new BIPV/T system to optimally minimize the energy consumption of a household residential building. The meticulous design of the proposed BIPV/T system is accomplished through MATLAB/Simulink® dynamic modeling. Performance analysis for the BIPV/T system is performed under different seasonal conditions with in-depth techno-economic analyses to estimate the expected enhancement in the thermal, electrical, and economic performance of the system. Moreover, a sensitivity analysis is conducted to explore the impact of various factors on the energetic and economic performances of the proposed BIPV/T system. More so, the two-layer feed-forward back-propagation artificial neural network modeling is developed to accurately predict the hourly solar radiation and ambient temperature for the BIPV/T. Additionally, a multi-objective optimization using the NSGA-II method is also conducted for the minimization of the total BIPV/T plant area and maximization of the total efficiency and net thermal power of the system as well as to estimate the optimized operating conditions for input variables across different seasons within the provided ranges. The sensitivity analysis revealed that higher solar flux levels lead to increased electric output power of the BIPV/T plant, but total efficiency decreases due to higher thermal losses. Moreover, the proposed NSGA-II shows a feasible method to attain a maximum net thermal power and optimal total efficiency of 5320 W and 63% with a minimal total plant area of 32.89 m2 that attained a very low deviation index from the ideal solution. The levelised cost of electricity is obtained as 0.10 $/kWh under the optimal conditions. Thus, these findings offer valuable insights into the potential of BIPV/T systems as a sustainable and efficient energy solution for residential applications.

Impact of Arrhenius activation energy on magnetic nanofluid flow over a slendering stretchable sheet with nonlinear radiative heat transfer: A machine learning algorithm

Thermal systems in the contemporary world need effective cooling and heating methods. Heat transfer technology plays a vital role in various industries, including aerospace, electronic equipment, automobiles, and transportation. Considering the importance of effective cooling and heating processes, this framework explores the hydromagnetic stagnation point flow with binary chemical reactions, ohmic heating, Arrhenius activation energy, nonlinear thermal radiation, thermophoresis, and Brownian motion. At first, the flow equations are determined in a formal way as dimensional partial differential equations. Then, they are transformed into simpler, dimensionless ordinary differential equations using self-similar variables. The present study describes a novel approach to intelligent numerical computing, utilizing a Levenberg-Marquard algorithm-based MLP (Multi-Layer Perceptron) feed-forward back-propagation ANN (Artificial Neural Networks) implementation. The collection of data was conducted to facilitate the evaluation, validation, and instruction of the artificial neural network model. The utilization of appropriate self-similarity variables has facilitated the conversion of fluid transport equations into ordinary differential equations. The Runge–Kutta–Fehlberg (RKF) technique was used to solve these equations. The numerical results are presented visually for numerous constraints of the governing parameters. The increased values of the velocity ratio enhanced the fluid velocity. The nonlinear thermal radiation and temperature differential characteristics significantly raise the temperature of the fluid. Increasing the nonlinear radiation and magnetic field parameters improves the rate of heat transfer. The temperature of the fluid decreases as the size of stagnation increases. The skin friction coefficients decrease as the wall thickness and magnetic parameters increase.

Anomalous hall and skyrmion topological hall resistivity in magnetic heterostructures for the neuromorphic computing applications

Topologically protected spin textures, such as magnetic skyrmions, have shown the potential for high-density data storage and energy-efficient computing applications owing to their particle-like behavior, small size, and low driving current requirements. Evaluating the writing and reading of the skyrmion’s magnetic and electrical characteristics is crucial to implementing these devices. In this paper, we present the magnetic heterostructure Hall bar device and study the anomalous Hall and topological Hall signals in these devices. Using different measurement techniques, we investigate the magnetic and electrical characteristics of the magnetic structure. We measure the skyrmion topological resistivity and the magnetic field at different temperatures. MFM imaging and micromagnetic simulations further explain the anomalous Hall and topological Hall resistivity characteristics at various magnetic fields and temperatures. The study is extended to propose a skyrmion-based synaptic device showing spin-orbit torque-controlled plasticity. The resistance states are read using the anomalous Hall measurement technique. The device integration in a neuromorphic circuit is simulated in a 3-layer feedforward artificial neural network ANN. Based on the proposed synapses, the neural network is trained and tested on the MNIST data set, where a recognition accuracy performance of about 90% is achieved. Considering the nanosecond reading/writing time scale and a good system level performance, these devices exhibit a substantial prospect for energy-efficient neuromorphic computing.

Optimizing Short-Term Photovoltaic Power Forecasting: A Novel Approach with Gaussian Process Regression and Bayesian Hyperparameter Tuning

The inherent volatility of PV power introduces unpredictability to the power system, necessitating accurate forecasting of power generation. In this study, a machine learning (ML) model based on Gaussian process regression (GPR) for short-term PV power output forecasting is proposed. With its benefits in handling nonlinear relationships, estimating uncertainty, and generating probabilistic forecasts, GPR is an appropriate approach for addressing the problems caused by PV power generation’s irregularity. Additionally, Bayesian optimization to identify optimal hyper-parameter combinations for the ML model is utilized. The research leverages solar radiation intensity data collected at 60-min and 30-min intervals over periods of 1 year and 6 months, respectively. Comparative analysis reveals that the data set with 60-min intervals performs slightly better than the 30-min intervals data set. The proposed GPR model, coupled with Bayesian optimization, demonstrates superior performance compared to contemporary ML models and traditional neural network models. This superiority is evident in 98% and 90% improvements in root mean square errors compared to feed-forward neural network and artificial neural network models, respectively. This research contributes to advancing accurate and efficient forecasting methods for PV power output, thereby enhancing the reliability and stability of power systems.

Artificial Neural Network-Based Mechanism to Detect Security Threats in Wireless Sensor Networks.

Wireless sensor networks (WSNs) are essential in many areas, from healthcare to environmental monitoring. However, WSNs are vulnerable to routing attacks that might jeopardize network performance and data integrity due to their inherent vulnerabilities. This work suggests a unique method for enhancing WSN security through the detection of routing threats using feed-forward artificial neural networks (ANNs). The proposed solution makes use of ANNs' learning capabilities to model the network's dynamic behavior and recognize routing attacks like black-hole, gray-hole, and wormhole attacks. CICIDS2017 is a heterogeneous dataset that was used to train and test the proposed system in order to guarantee its robustness and adaptability. The system's ability to recognize both known and novel attack patterns enhances its efficacy in real-world deployment. Experimental assessments using an NS2 simulator show how well the proposed method works to improve routing protocol security. The proposed system's performance was assessed using a confusion matrix. The simulation and analysis demonstrated how much better the proposed system performs compared to the existing methods for routing attack detection. With an average detection rate of 99.21% and a high accuracy of 99.49%, the proposed system minimizes the rate of false positives. The study advances secure communication in WSNs and provides a reliable means of protecting sensitive data in resource-constrained settings.

Comprehensive investigation of isotherm, RSM, and ANN modeling of CO2 capture by multi-walled carbon nanotube

Chemical vapor deposition was used to produce multi-walled carbon nanotubes (MWCNTs), which were modified by Fe–Ni/AC catalysts to enhance CO2 adsorption. In this study, a new realm of possibilities and potential advancements in CO2 capture technology is unveiled through the unique combination of cutting-edge modeling techniques and utilization of the recently synthesized Fe–Ni/AC catalyst adsorbent. SEM, BET, and FTIR were used to analyze their structure and morphology. The surface area of MWCNT was found to be 240 m2/g, but after modification, it was reduced to 11 m2/g. The modified MWCNT showed increased adsorption capacity with higher pressure and lower temperature, due to the introduction of new adsorption sites and favorable interactions at lower temperatures. At 25 °C and 10 bar, it reached a maximum adsorption capacity of 424.08 mg/g. The optimal values of the pressure, time, and temperature parameters were achieved at 7 bar, 2646 S and 313 K. The Freundlich and Hill models had the highest correlation with the experimental data. The Second-Order and Fractional Order kinetic models fit the adsorption results well. The adsorption process was found to be exothermic and spontaneous. The modified MWCNT has the potential for efficient gas adsorption in fields like gas storage or separation. The regenerated M-MWCNT adsorbent demonstrated the ability to be reused multiple times for the CO2 adsorption process, as evidenced by the study. In this study, a feed-forward MLP artificial neural network model was created using a back-propagation training approach to predict CO2 adsorption. The most suitable and efficient MLP network structure, selected for optimization, consisted of two hidden layers with 25 and 10 neurons, respectively. This network was trained using the Levenberg–Marquardt backpropagation algorithm. An MLP artificial neural network model was created, with a minimum MSE performance of 0.0004247 and an R2 value of 0.99904, indicating its accuracy. The experiment also utilized the blank spreadsheet design within the framework of response surface methodology to predict CO2 adsorption. The proximity between the Predicted R2 value of 0.8899 and the Adjusted R2 value of 0.9016, with a difference of less than 0.2, indicates a high level of similarity. This suggests that the model is exceptionally reliable in its ability to predict future observations, highlighting its robustness.

Modeling of compressive stress in AlSi10Mg alloys using feed-forward neural networks

This study addresses the challenge of modeling compressive stress in AlSi10Mg composites by introducing a method that employs feedforward artificial neural networks (ANNs) and their interpretability, which helps to simulate and analyze material behavior under various conditions. The main objective is to develop a predictive ANN model that can effectively simulate material responses under several factors, incorporating diverse testing parameters and material specifications related with its synthesis. An optimized ANN model, featuring eleven neurons in its hidden layer, was used and demonstrated high predictive accuracy, achieving R 2 values exceeding 0.94. Additionally, a SHAP interpretability analysis was conducted to assess the influence of key factors such as strain and material conditions on the stress response. The results highlight the significant role of material synthesis processes, compared to the strain rate, in the stress response. In conclusion, this method presents a comprehensive tool for studying complex stress behaviors in AlSi10Mg-based composites , offering insights that could guide future material development and research.

Neural networks modeling of strain energy density and Tsai-Wu index in laminated composites

In laminated composite materials design, optimization mainly targets the stacking sequence configuration, which is defined by the lamina thickness and fiber orientations within each layer. Recent studies emphasize the increasing role of Machine Learning in promoting innovative composite designs by facilitating the accurate modeling of essential properties such as strength and stiffness. This study introduces two metamodels that utilize feed-forward artificial neural networks, taking laminate thickness and fiber steering angles as input parameters. The output variables, including strain energy density and the Tsai-Wu failure index, enable the prediction of stacking sequence configurations for laminated materials, a capability confirmed in a case study. The results showcase neural network models with the ability to predict these variables, achieving coefficients of determination above 0.90 for testing data. Consequently, this modeling approach has the potential to be a tool for designers, aiding in decision-making processes for the subsequent optimization of stiffness and strength in structural components made of laminated composite materials.

Development and Validation of a Dry Electrode Array for sEMG Recording and Hand Movement Recognition

Surface electromyography (sEMG) signals are an indirect measurement of muscle activity, and their applications range from biomechanics to control and rehabilitation. Hand movement recognition is a very difficult endeavor due to forearm anatomy. Hence, a multichannel approach for signal acquisition and processing is required. Conventional electrodes can limit the ease-of-use and repeatability of multi-channel sEMG recordings. New techniques have been proposed in this regard, with dry electrodes being one of them. Dry electrode technology has enabled the design of better donning and doffing procedures for multichannel sEMG recording, particularly for rehabilitation and prosthetic applications. However, there is a debate about the quality of the signals recorded with them and their usefulness for the recognition of multiple hand movements. To mitigate these quality issues, this work proposes an array of reusable stainless steel dry electrodes for multichannel sEMG recording with a design that facilitates its positioning on the forearm. The dry electrodes were characterized through electrical impedance measures and a Bland-Altman test. They were found to have similar characteristics to standard, disposable sEMG pre-gelled electrodes. For placement repeatability and application feasibility, an anatomy-based electrode positioning protocol was implemented with 17 healthy subjects and six hand movements. To evaluate the application feasibility of the electrode array, a feed-forward artificial neural network was trained to classify signals from the six movements, with a 97,86±0,58% accuracy. The amplitude of the sEMG signals for two antagonist movements was compared, finding a 24,81% variation. The dry electrode array showed feasibility in acquiring and classifying sEMG signals of hand movements with high accuracy.