Search Results (19)

This present study proposes different machine learning-based predictors for the assessment of the residual compressive strength of Self-Compacting Concrete (SCC) subjected to high temperatures. The investigation is based on several literature algorithmic approaches based on Artificial Neural Networks with distinct training algorithms (Bayesian Regularization, Levenberg–Marquardt, Scaled Conjugate Gradient, and Resilient Backpropagation), Support Vector Regression, and Random Forest methods. A training database of 150 experimental data points is derived from a careful literature review, incorporating temperature (20–800 °C), geometric ratio (height/diameter), and corresponding compressive strength values. A statistical analysis revealed complex non-linear relationships between variables, with strong negative correlation between temperature and strength and heteroscedastic data distribution, justifying the selection of advanced machine learning techniques. Feature engineering improved model performance through the incorporation of quadratic terms, interaction variables, and cyclic transformations. The Resilient Backpropagation algorithm demonstrated superior performance with the lowest prediction errors, followed by Bayesian Regularization. Support Vector Regression achieved competitive accuracy despite its simpler architecture. Experimental validation using specimens tested up to 800 °C showed a good reliability of the developed systems, with prediction errors ranging from 0.33% to 23.35% across different temperature ranges. Full article

Artificial neural network (ANN) models have been used in the past to model surface roughness in manufacturing processes. Specifically, different parameters influence surface roughness in fused filament fabrication (FFF) processes. In addition, the characteristics of the networks have a direct impact on the performance of the models. In this work, a study about the use of ANN to model surface roughness in FFF processes is presented. The main objective of the paper is discovering how key ANN parameters (specifically, the number of neurons, the training algorithm, and the percentage of training and validation datasets) affect the accuracy of surface roughness predictions. To address this question, 125 3D printing experiments were conducted changing orientation angle, layer height and printing temperature, and measuring average roughness Ra as response. A multilayer perceptron neural network model with backpropagation algorithm was used. The study evaluates the effect of three ANN parameters: (1) number of neurons in the hidden layer (4, 5, 6 or 7), (2) training algorithm (Levenberg–Marquardt, Resilient Backpropagation or Scaled Conjugate Gradient), and (3) data splitting ratios (70%–15%–15% vs. 55%–15%–30%). Mean Absolute Error (MAE) was used as the performance metric. The Resilient Backpropagation algorithm, 7 neurons, and using 55% of training data yielded the best predictive performance, minimizing the MAE. Additionally, the impact of the dataset size on prediction accuracy was analysed. It was observed that the performance of the ANN gets worse as the number of datasets is reduced, emphasizing the importance of having sufficient data. This study will help to select appropriate values for the printing parameters in FFF processes, as well as to define the characteristics of the ANN to be used to model surface roughness. Full article

In urban Vehicular Ad Hoc Network (VANET) environments, buildings play a crucial role as they can act as obstacles that attenuate the transmission signal between vehicles. Such obstacles lead to multipath effects, which could substantially impact data transmission due to fading. Therefore, quantifying the impact of buildings on transmission quality is a key parameter of the propagation model, especially in critical scenarios involving emergency vehicles where reliable communication is of utmost importance. In this research, we propose a supervised learning approach based on Artificial Neural Networks (ANNs) to develop a predictive model capable of estimating the level of signal degradation, represented by the Bit Error Rate (BER), based on the obstacles perceived by moving emergency vehicles. By establishing a relationship between the level of signal degradation and the encountered obstacles, our proposed mechanism enables efficient routing decisions being made prior to the transmission process. Consequently, data packets are routed through paths that exhibit the lowest BER. To collect the training data, we employed Network Simulator 3 (NS-3) in conjunction with the Simulation of Urban MObility (SUMO) simulator, leveraging real-world data sourced from the OpenStreetMap (OSM) geographic database. OSM enabled us to gather geospatial data related to the Two-Dimensional (2D) geometric structure of buildings, which served as input for our Artificial Neural Network (ANN). To determine the most suitable algorithm for our ANN, we assessed the accuracy of ten learning algorithms in MATLAB, utilizing five key metrics: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Correlation Coefficient (R), and Maximum Prediction Error (MaxPE). For each algorithm, we conducted fifteen iterations based on ten hidden neurons and gauged its accuracy against the aforementioned metrics. Our analysis highlighted that the ANN underpinned by the Conjugate Gradient With Powell/Beale Restarts (CGB) learning algorithm exhibited superior performance in terms of MSE, RMSE, MAE, R, and MaxPE compared to other algorithms such as Levenberg–Marquardt (LM), Bayesian Regularization (BR), BFGS Quasi-Newton (BFG), Resilient Backpropagation (RP), Scaled Conjugate Gradient (SCG), Fletcher–Powell Conjugate Gradient (CGF), Polak–Ribiére Conjugate Gradient (CGP), One-Step Secant (OSS), and Variable Learning Rate Backpropagation (GDX). The BER prediction by our ANN incorporates the TWO-RAY Ground (TRG) propagation model, an adjustable parameter within NS-3. When subjected to 300 new samples, the trained ANN’s simulation outcomes illustrated its capability to learn, generalize, and successfully predict the BER for a new data instance. Overall, our research contributes to enhancing the performance and reliability of communication in urban VANET environments, especially in critical scenarios involving emergency vehicles, by leveraging supervised learning and artificial neural networks to predict signal degradation levels and optimize routing decisions accordingly. Full article

Matsutake mushrooms, known for their high value, present challenges due to their seasonal availability, difficulties in harvesting, and short shelf life, making it crucial to extend their post-harvest preservation period. In this study, we developed three quality predictive models of Matsutake mushrooms using three different methods. The quality changes of Matsutake mushrooms were experimentally analyzed under two cases (case A: Temperature control and sealing measures; case B: Alteration of gas composition) with various parameters including the hardness, color, odor, pH, soluble solids content (SSC), and moisture content (MC) collected as indicators of quality changes throughout the storage period. Prediction models for Matsutake mushroom quality were developed using three different methods based on the collected data: multiple linear regression (MLR), support vector regression (SVR), and an artificial neural network (ANN). The comparative results reveal that the ANN outperforms MLR and SVR as the optimal model for predicting Matsutake mushroom quality indicators. To further enhance the ANN model’s performance, optimization techniques such as the Levenberg–Marquardt, Bayesian regularization, and scaled conjugate gradient backpropagation algorithm techniques were employed. The optimized ANN model achieved impressive results, with an R-Square value of 0.988 and an MSE of 0.099 under case A, and an R-Square of 0.981 and an MSE of 0.164 under case B. These findings provide valuable insights for the development of new preservation methods, contributing to the assurance of a high-quality supply of Matsutake mushrooms in the market. Full article

In this paper, force sensor signals are classified using a pattern recognition neural network (PRNN). The signals are classified to show if there is a collision or not. In our previous work, the joints positions of a 2-DOF robot were used to estimate the external force sensor signal, which was attached at the robot end-effector, and the external joint torques of this robot based on a multilayer feedforward NN (MLFFNN). In the current work, the estimated force sensor signal and the external joints’ torques from the previous work are used as the inputs to the proposed designed PRNN, and its output is whether a collision is found or not. The designed PRNN is trained using a scaled conjugate gradient backpropagation algorithm and tested and validated using different data from the training one. The results prove that the PRNN is effective in classifying the force signals. Its effectiveness for classifying the collision cases is 92.8%, and for the non-collisions cases is 99.4%. Therefore, the overall efficiency is 99.2%. The same methodology and work are repeated using a PRNN trained using another algorithm, which is the Levenberg–Marquardt (PRNN-LM). The results using this structure prove that the PRNN-LM is also effective in classifying the force signals, and its overall effectiveness is 99.3%, which is slightly higher than the first PRNN. Finally, a comparison of the effectiveness of the proposed PRNN and PRNN-LM with other previous different classifiers is included. This comparison shows the effectiveness of the proposed PRNN and PRNN-LM. Full article

In this paper, the Levenberg–Marquardt (LM), Bayesian regularization (BR), resilient backpropagation (RP), gradient descent momentum (GDM), Broyden–Fletcher–Goldfarb–Shanno (BFGS), and scaled conjugate gradient (SCG) algorithms constructed using artificial neural networks (ANN) are applied to the problem of MPPT energy harvesting in solar photovoltaic (PV) systems for the purpose of creating a comparative evaluation of the performance of the six distinct algorithms. The goal of this analysis is to determine which of the six algorithms has the best overall performance. In the study, the performance of managing the training dataset is compared across the algorithms. The maximum power point tracking energy harvesting system is created using the environment of MATLAB or Simulink, and the produced model is examined using the artificial neural network toolkit. A total of 1000 datasets of solar irradiance, temperature, and voltage were used to train the suggested model. The data are split into three categories: training, validation, and testing. Eighty percent of the total data is used for training the model, and the remaining twenty percent is divided equally for testing and validation. According to the results, the regression values of LM, RP, BR, and BFGS are 1, whereas the regression values for SCG and GDM are less than 1. The gradient values for LM, RP, BFGS, SCG, BR, and GDM are 7.983 × 10⁻⁶, 0.033415, 1.0211 × 10⁻⁷, 0.14161, 0.00010493, and 11.485, respectively. Similarly, the performance values for these algorithms are 2.0816 × 10⁻¹⁰, 2.8668 × 10⁻⁶, 9.98 × 10⁻¹⁷, 0.052985, 1.583 × 10⁻⁷, and 0.15378. Overall, the results demonstrate that the LM and BFGS algorithms exhibit superior performance in terms of gradient and overall performance. The RP and BR algorithms also perform well across various metrics, while the SCG and GDM algorithms show comparatively less effectiveness in addressing the proposed problem. These findings provide valuable insights into the relative performance of the six evaluated algorithms for MPPT energy harvesting in solar PV systems. Full article

Novel or unconventional technologies are critical to providing cost-competitive natural gas supplies to meet rising demands and provide more opportunities to develop low-quality gas fields with high contaminants, including high carbon dioxide (CO₂) fields. High nitrogen concentrations that reduce the heating value of gaseous products are typically associated with high CO₂ fields. Consequently, removing nitrogen is essential for meeting customers’ requirements. The intensification approach with a rotating packed bed (RPB) demonstrated considerable potential to remove nitrogen from natural gas under cryogenic conditions. Moreover, the process significantly reduces the equipment size compared to the conventional distillation column, thus making it more economical. The prediction model developed in this study employed artificial neural networks (ANN) based on data from in-house experiments due to a lack of available data. The ANN model is preferred as it offers easy processing of large amounts of data, even for more complex processes, compared to developing the first principal mathematical model, which requires numerous assumptions and might be associated with lumped components in the kinetic model. Backpropagation algorithms for ANN Lavenberg–Marquardt (LM), scaled conjugate gradient (SCG), and Bayesian regularisation (BR) were also utilised. Resultantly, the LM produced the best model for predicting nitrogen removal from natural gas compared to other ANN models with a layer size of nine, with a 99.56% regression (R²) and 0.0128 mean standard error (MSE). Full article

Accurately determining rock elastic modulus (EM) and uniaxial compressive strength (UCS) using laboratory methods requires considerable time and cost. Hence, the development of models for estimating the mechanical properties of rock is a very attractive alternative. The current research was conducted to predict the UCS and EM of sandstone rocks using quartz%, feldspar%, fragments%, compressional wave velocity (PW), the Schmidt hardness number (SN), porosity, density, and water absorption via simple regression, multivariate regression (MVR), K-nearest neighbor (KNN), support vector regression (SVR) with a radial basis function, the adaptive neuro-fuzzy inference system (ANFIS) using the Gaussian membership (GM) function, and the back-propagation neural network (BPNN) based on various training algorithms. The samples were categorized as litharenite and feldspathic litharenite. By increasing the feldspar% and quartz% and decreasing the fragments%, the static properties increased. The results of the statistical analysis showed that the SN and porosity have the greatest effect on the UCS and EM, respectively. Among the Levenberg–Marquardt (LM), Bayesian regularization, and Scaled Conjugate Gradient training algorithms using the BPNN method, the LM achieved the best results in forecasting the UCS and EM. The ideal obtained BPNN, using a trial-and-error process, contains four neurons in a hidden layer with eight inputs. All five models attained acceptable accuracy (correlation coefficient greater than 70%) for estimating the static properties. By comparing the methods, the ANFIS showed higher precision than the other methods. The UCS and EM of the samples can be determined with very high accuracy (R² > 99%). Full article

In the study, rupture energy values of Deveci and Abate Fetel pear fruits were predicted using artificial neural network (ANN). This research aimed to develop a simple, accurate, rapid, and economic model for harvest/post-harvest loss of efficiently predicting rupture energy values of Deveci and Abate Fetel pear fruits. The breaking energy of the pears was examined in terms of storage time and loading position. The experiments were carried out in two stages, with samples kept in cold storage immediately after harvest and 30 days later. Rupture energy values were estimated using four different single and multi-layer ANN models. Four different model results obtained using Levenberg–Marquardt, Scaled Conjugate Gradient, and resilient backpropagation training algorithms were compared with the calculated values. Statistical parameters such as R², RMSE, MAE, and MSE were used to evaluate the performance of the methods. The best-performing model was obtained in network structure 5-1 that used three inputs: the highest R² value (0.90) and the lowest square of the root error (0.018), and the MAE (0.093). Full article

Some of the effects of climate change may be related to a change in patterns of rainfall intensity or scarcity. Therefore, humanity is facing environmental challenges due to an increase in the occurrence and intensity of droughts. The forecast of droughts can be of great help when trying to reduce the adverse effects that the scarcity of water brings, particularly in agriculture. When evaluating the conditions of water scarcity, as well as in the identification and characterization of droughts, the use of predictive models of drought indices could be a very useful tool. In this research, the utility of Artificial Neural Networks with exogenous inputs was tested, with the aim of predicting the monthly Standardized Precipitation Index in 4 regions (Semi-desert, Highlands, Canyons and Mountains) of north-central México using predictor data from 1979 to 2014. The best model was found using the scaled conjugate gradient backpropagation algorithm as the optimization method and was set to the following architecture: 6-25-1 network. The correlation coefficient of predicted and observed Standardized Precipitation Index values for the test dataset was between 0.84 and 0.95. As a result, the Artificial Neural Network models performed successfully in predicting Standardized Precipitation Index at the four analyzed regions. The developed and tested Artificial Neural Network models in this research suggest remarkable prediction abilities of the monthly Standardized Precipitation Index in the study region. Full article

A considerable amount of discarded building materials are produced each year worldwide, resulting in ecosystem degradation. Self-compacting concrete (SCC) has 60–70% coarse and fine particles in its composition, so replacing this material with another waste material, such as recycled aggregate (RA), reduces the cost of SCC. This study compares novel Artificial Neural Network algorithm techniques—Levenberg–Marquardt (LM), Bayesian regularization (BR), and Scaled Conjugate Gradient Backpropagation (SCGB)—to estimate the 28-day compressive strength (f’c) of SCC with RA. A total of 515 samples were collected from various published papers, randomly splitting into training, validation, and testing with percentages of 70, 10 and 20. Two statistical indicators, correlation coefficient (R) and mean squared error (MSE), were used to assess the models; the greater the R and lower the MSE, the more accurate the algorithm. The findings demonstrate the higher accuracy of the three models. The best result is achieved by BR (R = 0.91 and MSE = 43.755), while the accuracy of LM is nearly the same (R = 0.90 and MSE = 48.14). LM processes the network in a much shorter time than BR. As a result, LM and BR are the best models in forecasting the 28 days f’c of SCC having RA. The sensitivity analysis showed that cement (28.39%) and water (23.47%) are the most critical variables for predicting the 28-day compressive strength of SCC with RA, while coarse aggregate contributes the least (9.23%). Full article

Wind energy is a valuable source of electric power as its motion can be converted into mechanical energy, and ultimately electricity. The significant variability of wind speed calls for highly robust estimation methods. In this study, the mechanical power of wind turbines (WTs) is successfully estimated using input variables such as wind speed, angular speed of WT rotor, blade pitch, and power coefficient (Cp). The feed-forward backpropagation neural networks (FFBPNNs) and recurrent neural networks (RNNs) are incorporated to perform the estimations of wind turbine output power. The estimations are performed based on diverse parameters including the number of hidden layers, learning rates, and activation functions. The networks are trained using a scaled conjugate gradient (SCG) algorithm and evaluated in terms of the root mean square error (RMSE) and mean absolute percentage error (MAPE) indices. FFBPNN shows better results in terms of RMSE (0.49%) and MAPE (1.33%) using two and three hidden layers, respectively. The study indicates the significance of optimal selection of input parameters and effects of changing several hidden layers, activation functions, and learning rates to achieve the best performance of FFBPNN and RNN. Full article

The composition of self-compacting concrete (SCC) contains 60–70% coarse and fine aggregates, which are replaced by construction waste, such as recycled aggregates (RA). However, the complexity of its structure requires a time-consuming mixed design. Currently, many researchers are studying the prediction of concrete properties using soft computing techniques, which will eventually reduce environmental degradation and other material waste. There have been very limited and contradicting studies regarding prediction using different ANN algorithms. This paper aimed to predict the 28-day splitting tensile strength of SCC with RA using the artificial neural network technique by comparing the following algorithms: Levenberg–Marquardt (LM), Bayesian regularization (BR), and Scaled Conjugate Gradient Backpropagation (SCGB). There have been very limited and contradicting studies regarding prediction by using and comparing different ANN algorithms, so a total of 381 samples were collected from various published journals. The input variables were cement, admixture, water, fine and coarse aggregates, and superplasticizer; the data were randomly divided into three sets—training (60%), validation (10%), and testing (30%)—with 10 neurons in the hidden layer. The models were evaluated by the mean squared error (MSE) and correlation coefficient (R). The results indicated that all three models have optimal accuracy; still, BR gave the best performance (R = 0.91 and MSE = 0.2087) compared with LM and SCG. BR was the best model for predicting TS at 28 days for SCC with RA. The sensitivity analysis indicated that cement (30.07%) was the variable that contributed the most to the prediction of TS at 28 days for SCC with RA, and water (2.39%) contributed the least. Full article

With the development of autonomous driving technology, the requirements for machine perception have increased significantly. In particular, camera-based lane detection plays an essential role in autonomous vehicle trajectory planning. However, lane detection is subject to high complexity, and it is sensitive to illumination variation, appearance, and age of lane marking. In addition, the sheer infinite number of test cases for highly automated vehicles requires an increasing portion of test and validation to be performed in simulation and X-in-the-loop testing. To model the complexity of camera-based lane detection, physical models are often used, which consider the optical properties of the imager as well as image processing itself. This complexity results in high efforts for the simulation in terms of modelling as well as computational costs. This paper presents a Phenomenological Lane Detection Model (PLDM) to simulate camera performance. The innovation of the approach is the modelling technique using Multi-Layer Perceptron (MLP), which is a class of Neural Network (NN). In order to prepare input data for our neural network model, massive driving tests have been performed on the M86 highway road in Hungary. The model’s inputs include vehicle dynamics signals (such as speed and acceleration, etc.). In addition, the difference between the reference output from the digital-twin map of the highway and camera lane detection results is considered as the target of the NN. The network consists of four hidden layers, and scaled conjugate gradient backpropagation is used for training the network. The results demonstrate that PLDM can sufficiently replicate camera detection performance in the simulation. The modelling approach improves the realism of camera sensor simulation as well as computational effort for X-in-the-loop applications and thereby supports safety validation of camera-based functionality in automated driving, which decreases the energy consumption of vehicles. Full article

Neuro-muscular disorders and diseases such as cerebral palsy and Duchenne Muscular Dystrophy can severely limit a person’s ability to perform activities of daily living (ADL). Exoskeletons can provide an active or passive support solution to assist these groups of people to perform ADL. This study presents an artificial neural network-trained adaptive controller mechanism that uses surface electromyography (sEMG) signals from the human forearm to detect hand gestures and navigate an in-house-built wheelchair-mounted upper limb robotic exoskeleton based on the user’s intent while ensuring safety. To achieve the desired position of the exoskeleton based on human intent, 10 hand gestures were recorded from 8 participants without upper limb movement disabilities. Participants were tasked to perform water bottle pick and place activities while using the exoskeleton, and sEMG signals were collected from the forearm and processed through root mean square, median filter, and mean feature extractors prior to training a scaled conjugate gradient backpropagation artificial neural network. The trained network achieved an average of more than 93% accuracy, while all 8 participants who did not have any prior experience of using an exoskeleton were successfully able to perform the task in less than 20 s using the proposed artificial neural network-trained adaptive controller mechanism. These results are significant and promising thus could be tested on people with muscular dystrophy and neuro-degenerative diseases. Full article