Multilayer Perceptron Artificial Neural Network Research Articles

Dairy factory milk product processing and sustainable of the shelf-life extension with artificial intelligence: a model study

This study models milk product processing and sustainable of the shelf-life extension in a dairy factory using artificial intelligence. The Cappadocia dairy factory was used to study chemical processes and computational system modeling and simulation. Levenberg–Marquardt algorithm was used to create an artificial neural network model from real-time data. An AI-based method utilizing a Multilayer Perceptron (MLP) Artificial Neural Network (ANN) model was employed to precisely analyze productivity data in dairy factories. There are 9 product types and production quantities used as input parameters, and 90 datasets of actual dairy products used as output values. The model was trained using the Levenberg–Marquardt algorithm on 62 datasets for training, 14 for validation, and 14 for testing. The accuracy of the model is affected by the optimal data segmentation. The model showed how AI algorithms can improve processes and industrial production by increasing dairy production efficiency from 20 to 40%. Model efficiency values were compared to observed values to determine prediction accuracy. Model mean squared error was 4.02E-06, and coefficient of determination was 0.99984. Model efficiency predictions and observed values differed by −0.04% on average. This study investigated using artificial intelligence to optimize salvage processes and systems to increase energy efficiency and reduce environmental impact. The results show that a neural network model trained with real data can predict dairy plant productivity.

Machine learning-based life cycle assessment for environmental sustainability optimization of a food supply chain.

Effective resource allocation in the agri-food sector is essential in mitigating environmental impacts and moving toward circular food supply chains. The potential of integrating life cycle assessment (LCA) with machine learning has been highlighted in recent studies. This hybrid framework is valuable not only for assessing food supply chains but also for improving them toward a more sustainable system. Yet, an essential step in the optimization process is defining the optimization boundaries, or minimum and maximum quantities for the variables. Usually, the boundaries for optimization variables in these studies are obtained from the minimum and maximum values found through interviews and surveys. A deviation in these ranges can impact the final optimization results. To address this issue, this study applies the Delphi method for identifying variable optimization boundaries. A hybrid environmental assessment framework linking LCA, multilayer perceptron artificial neural network, the Delphi method, and genetic algorithm was used for optimizing the pomegranate production system. The results indicated that the suggested framework holds promise for achieving substantial mitigation in environmental impacts (potential reduction of global warming by 46%) within the explored case study. Inclusion of the Delphi method for variable boundary determination brings novelty to the resource allocation optimization process in the agri-food sector. Integr Environ Assess Manag 2024;20:1759-1769. © 2024 SETAC.

An analysis of urban sprawl growth and prediction using remote sensing and machine learning techniques

This study examines urban growth to support sustainability by investigating urban sprawl in the Vila Velha Urban Agglomeration (UA). If the spatial area of Vila Velha UA increases by 24.36%, from 22.98 km2 in 1994 to 51.64 km2 in 2024, urban sprawl may affect surrounding areas. Landsat satellite images from 1994, 2004, 2014, and 2024 were analyzed to study the spatiotemporal pattern of urban growth. Shannon's entropy index identified urban sprawl in Vila Velha UA. The QGIS software's MOLUSCE (Modules for Land Use Change Simulations) plug-in, utilizing Cellular Automata (CA) and Artificial Neural Network-Multi Layer Perceptron (ANN-MLP) models, forecasted urban expansion for 2034. The built-up area is projected to increase from 51.64 km2 in 2024 to 62.31 km2 in 2034. Shannon's entropy index indicated a high rate of urban sprawl. Remote sensing and machine learning techniques are crucial for understanding spatial trends, forecasting future expansion, and informing sustainable urban planning. These methods facilitate an accurate assessment of urbanization dynamics, aiding in developing strategies to manage and mitigate sprawl's adverse impacts. This study offers a unique approach to defining urban growth, emphasizing the need for zoning regulations, green space conservation, and infrastructure development to curb urban sprawl and promote sustainable growth in Vila Velha.

Recognition of Intergranular Corrosion in AISI 304 Stainless Steel by Integrating a Multilayer Perceptron Artificial Neural Network and Metallographic Image Processing

The correct management of operations in thermoelectric plants is based on the continuous evaluation of the structural integrity of its components, among which there are elements made of stainless steel that perform water conduction functions at elevated temperatures. The working conditions generate progressive wear that must be monitored from the perspective of the microstructure of the material. When AISI 304 stainless steel is subjected to a temperature range between 450 and 850 °C, it is susceptible to intergranular corrosion. This phenomenon, known as sensitization, causes the material to lose strength and generates different patterns in its microstructure. This research analyzes three different patterns present in the microstructure of stainless steel, which manifest themselves through the following characteristics: the absence of intergranular corrosion, the presence of intergranular corrosion, and the precipitation of chromium carbides. This article shows the development of a methodology capable of recognizing the corrosion patterns generated in stainless steel with an accuracy of 98%, through the integration of a multilayer perceptron neural network and the following digital image processing methods: phase congruence and a gray-level co-occurrence matrix. In this way, an automatic procedure for the analysis of the intergranular corrosion present in AISI 304 stainless steel using artificial intelligence is proposed.

Optimization of a photovoltaic/wind/battery energy-based microgrid in distribution network using machine learning and fuzzy multi-objective improved Kepler optimizer algorithms

In this study, a fuzzy multi-objective framework is performed for optimization of a hybrid microgrid (HMG) including photovoltaic (PV) and wind energy sources linked with battery energy storage (PV/WT/BES) in a 33-bus distribution network to minimize the cost of energy losses, minimizing the voltage oscillations as well as power purchased minimization from the HMG incorporated forecasted data. The variables are microgrid optimal location and capacity of the HMG components in the network which are determined through a multi-objective improved Kepler optimization algorithm (MOIKOA) modeled by Kepler’s laws of planetary motion, piecewise linear chaotic map and using the FDMT. In this study, a machine learning approach using a multilayer perceptron artificial neural network (MLP-ANN) has been used to forecast solar radiation, wind speed, temperature, and load data. The optimization problem is implemented in three optimization scenarios based on real and forecasted data as well as the investigation of the battery's depth of discharge in the HMG optimization in the distribution network and its effects on the different objectives. The results including energy losses, voltage deviations, and purchased power from the HMG have been presented. Also, the MOIKOA superior capability is validated in comparison with the multi-objective conventional Kepler optimization algorithm, multi-objective particle swarm optimization, and multi-objective genetic algorithm in problem-solving. The findings are cleared that microgrid multi-objective optimization in the distribution network considering forecasted data based on the MLP-ANN causes an increase of 3.50%, 2.33%, and 1.98%, respectively, in annual energy losses, voltage deviation, and the purchased power cost from the HMG compared to the real data-based optimization. Also, the outcomes proved that increasing the battery depth of discharge causes the BES to have more participation in the HMG effectiveness on the distribution network objectives and affects the network energy losses and voltage deviation reduction.

Emulator-Based Bayesian Calibration of the CISNET Colorectal Cancer Models.

To calibrate Cancer Intervention and Surveillance Modeling Network (CISNET)'s SimCRC, MISCAN-Colon, and CRC-SPIN simulation models of the natural history colorectal cancer (CRC) with an emulator-based Bayesian algorithm and internally validate the model-predicted outcomes to calibration targets. We used Latin hypercube sampling to sample up to 50,000 parameter sets for each CISNET-CRC model and generated the corresponding outputs. We trained multilayer perceptron artificial neural networks (ANNs) as emulators using the input and output samples for each CISNET-CRC model. We selected ANN structures with corresponding hyperparameters (i.e., number of hidden layers, nodes, activation functions, epochs, and optimizer) that minimize the predicted mean square error on the validation sample. We implemented the ANN emulators in a probabilistic programming language and calibrated the input parameters with Hamiltonian Monte Carlo-based algorithms to obtain the joint posterior distributions of the CISNET-CRC models' parameters. We internally validated each calibrated emulator by comparing the model-predicted posterior outputs against the calibration targets. The optimal ANN for SimCRC had 4 hidden layers and 360 hidden nodes, MISCAN-Colon had 4 hidden layers and 114 hidden nodes, and CRC-SPIN had 1 hidden layer and 140 hidden nodes. The total time for training and calibrating the emulators was 7.3, 4.0, and 0.66 h for SimCRC, MISCAN-Colon, and CRC-SPIN, respectively. The mean of the model-predicted outputs fell within the 95% confidence intervals of the calibration targets in 98 of 110 for SimCRC, 65 of 93 for MISCAN, and 31 of 41 targets for CRC-SPIN. Using ANN emulators is a practical solution to reduce the computational burden and complexity for Bayesian calibration of individual-level simulation models used for policy analysis, such as the CISNET CRC models. In this work, we present a step-by-step guide to constructing emulators for calibrating 3 realistic CRC individual-level models using a Bayesian approach. We use artificial neural networks (ANNs) to build emulators that surrogate complex individual-based models to reduce the computational burden in the Bayesian calibration process.ANNs showed good performance in emulating the CISNET-CRC microsimulation models, despite having many input parameters and outputs.Using ANN emulators is a practical solution to reduce the computational burden and complexity for Bayesian calibration of individual-level simulation models used for policy analysis.This work aims to support health decision scientists who want to quantify the uncertainty of calibrated parameters of computationally intensive simulation models under a Bayesian framework.

Time in range prediction using the experimental mobile application in type 1 diabetes

BACKGROUND: Time in range (TIR) is a promising indicator of glycemic control used for evaluation of continuous glucose monitoring (CGM) for patients with diabetes mellitus (DM). The current problem is the assessment and prediction of TIR for patients who use self-monitoring of blood glucose (SМBG) corresponding low CGM availability for the majority of diabetic patients.AIM: To develop a predictive model of TIR for patients with T1DM based on data of the experimental mobile application.MATERIALS AND METHODS: An analysis of 1253 professional CGM profiles of patients with T1DM was performed. On the base of included records, TIR(CGM) was calculated and training models of 7-point SMBG profiles were generated. SMBG profiles’re loaded into the developed experimental mobile application that calculated standard glycemic control parameters. The dataset was divided into main and test samples (80 and 20%). For the main sample, the following methods’re used to develop predictive models: simple linear regression (SLR), multiple linear regression (MLR), artificial neural network (ANN). The effectiveness of the developed models was assessed on the test sample with the calculation of the mean absolute error (MAE), the root mean square error (RMSE).RESULTS: The 568 CGM profiles’re included in the study. TIR in the main group (n=454) — 45 [33; 65]%, in the test group (n=114) — 43 [33; 58]%. The most significant predictors of the regression models were the derived TIR (dTIR), p<0,001; derived time below range level 1 (dTBR1), p<0,001; standard deviation of blood glucose (SD), p=0,007. Determination coefficient for SLR (predictor: dTIR) — 0,844; for MLR (predictors: dTIR, dTBR1, SD) — 0,907. ANN multilayer perceptron models with two and one hidden layers’re developed, with the RMSE on the validation set 4,617 and 6,639%, respectively. The results of the forecast efficiency on the test sample were: dTIR: MAE — 6,82%, RMSE — 8,60%; SLR: MAE — 5,66%, RMSE — 7,34%; MLR: MAE — 4,18%, RMSE — 5,28%; ANN (2 layers): MAE — 4,14%, RMSE — 5,19%; ANN (1 layer): MAE — 4,44%, RMSE — 5,52%.CONCLUSION: ANN with two hidden layers and MLR demonstrated the best ability for TIR prediction. Further studies are required for clinical validation of developed prognostic models.

Optimization of resources in intelligent electronic health systems based on internet of things to predict heart diseases via artificial neural network

As a paradigm shift in tandem with the expansion of ICT, smart electronic health systems hold great promise for enhancing healthcare delivery and illness prevention efforts. These systems acquire an in-depth understanding of patient health states through the real-time collection and analysis of medical data enabled by the Internet of Things (IoT) and machine learning. With the widespread use of cutting-edge artificial intelligence and machine learning techniques, predictive analytics in medicine can assist in making the shift from a reactive to a proactive healthcare strategy. With the ability to rapidly and precisely evaluate massive amounts of data, draw intelligent conclusions, and solve difficult issues, artificial neural networks could revolutionize several industries. Two cardiac illnesses were assessed in this study using a multilayer perceptron artificial neural network that incorporated a genetic algorithm and an error-back propagation mechanism. The ability of artificial neural networks to handle consecutive time series data is crucial for optimizing resources in smart electronic health systems, especially with the increasing volume of patient information and the broad use of electronic clinical records. This requires the creation of more accurate predictive models. Through the use of Internet of Things (IoT) sensors, the proposed system gathers data, which is then used to do predictive analytics on patient history-related electronic clinical data saved in the cloud. A smart healthcare system that uses Mu-LTM (multidirectional long-term memory) to accurately monitor and predict the risk of heart disease has a coverage error of 97.94 %, an accuracy of 97.89 %, a sensitivity of 97.96 %, and a specificity of 97.99 %. In comparison to other smart heart disease prediction systems, the F1-score of 97.95 % and precision of 97.71 % is very good.

Advancements in daily precipitation forecasting: A deep dive into daily precipitation forecasting hybrid methods in the Tropical Climate of Thailand

Climate change and increasing water demands underscore the importance of water resource management. Precise precipitation forecasting is critical to effective management. This study introduced a Daily Precipitation Forecasting Hybrid (DPFH) technique for central Thailand, which uses three different input-based models to improve prediction accuracy.•The proposed methods precisely combine the biorthogonal wavelet transformation (BWT) function through BWT-RBFNN (Radial Basis Function Neural Networks) and (BWT-LSTM-RNN)Long Short-Term Memory Recurrent Neural Networks. Comparative analyses reveal that hybrid models perform better than conventional deep LSTM-RNN and Multilayer Perceptron Artificial Neural Networks (MLP-ANN). Although MLP-ANN showed moderate effectiveness, LSTM-RNN displayed notable enhancements, particularly evidenced by an impressive R2 (0.96) in Model M-2.•The combination of BWT-LSTM-RNN yielded substantial enhancements, constantly surpassing standalone models. Specifically, DPFH-3 exhibited superior performance across multiple observation stations.•The findings emphasize the efficiency of the BWT-LSTM-RNN models in capturing varied precipitation patterns, highlighting their potential to significantly improve the accuracy of precipitation forecasts, particularly in the context of water resource management in central Thailand.

Numerical ellipsometry: Artificial intelligence for rapid analysis of indium tin oxide films on silicon

Ellipsometry is a well-known material analytical method widely used to measure thickness and optical properties of thin films and surfaces across a wide range of industrial and research applications including critical dimensions in chipmaking. The method employs the fact that light undergoes a change in polarization state upon reflection from or transmission through a material. The desired properties of the surface structure are related to measurements by the electromagnetic models expressed by Maxwell’s equations as well as models of material properties. The work here demonstrates the use of artificial intelligence in the form of a multilayer perceptron artificial neural network to apply the electromagnetic model. The reflecting surface examined here is composed of indium tin oxide (ITO) films approximately 400 nm in thickness deposited on silicon substrates. Solutions are provided by 299 artificial neural networks, one per wavelength from 210 to 1700 nm across which ITO exhibits transparent as well as absorbing characteristics. Thus, it serves as a proxy for a wide range of other materials. To train the network, simulated measurements are computed at two thicknesses which differ randomly by 1–6 nm and at three different incidence angles of 55°, 65°, and 75°. Following training, results are obtained in less than one second on a conventional desktop computer.

Decoding green food safety information dependency in the digital era: An intelligent validation using SEM-ANN framework

Dependence on mobile phone information in the Internet era is a crucial issue, and psychological perception and personalized and accurate pushing significantly impact this field. This introduction of rational and planned behaviour combined with the theoretical framework of Structural Equation Modelling (SEM) and Artificial Neural Networks (ANN) are intricacy analytical tools to explore the key factors affecting the information dependence on green food safety. The key factors affecting the reliance on green food safety information were investigated. The survey covered 630 participants aged 17 to 70 in Guangdong Province, China. The study showed that perceived risk, regulation, price volatility, innovativeness, and anticipatory confirmation significantly influence the perceived usefulness of green food information, affecting satisfaction with green food. The study also highlights the significant impact of the degree of anticipatory confirmation by users on satisfaction and perceived usefulness of green food. Model path analysis revealed that anticipatory confirmation and perceived usefulness explained 61% and 74% of the variance in financial sustainability, respectively. The application of deep ANN multilayer perceptron improved the prediction accuracy of perceived usefulness with 87.54% training accuracy and 90.34% testing accuracy, further deepening the understanding of the mechanism of relying on green food safety information. This study provides strong support for a deeper understanding of the relationship between consumer behaviour and green food safety information dependence, which helps promote enterprises and market camps to more effectively formulate product or service promotion strategies and provides a solid theoretical and empirical foundation for future green food safety and personalized and accurate push thus enhancing information dependence.

Can machine learning efficiently predict symmetry breaking in physical problems like Bose-Einstein condensates?

In this work, our objective is to evaluate whether machine learning algorithms combined with computational methods used in physical problems such as spontaneous symmetry breaking in Bose-Einstein condensates are capable of efficiently predicting results obtained from solutions of nonlinear equations. Thus, we use K-means algorithms, support vector machines (SVM) and artificial neural network (ANN) multilayer perceptron to evaluate the efficiency of each algorithm in the process of predicting theoretical results in regions where computational calculations cannot be performed. either by numerical inaccuracy or computational cost. The results show that sophisticated machine learning algorithms can predict the behavior of represented physical systems with a satisfactory quality, where we especially highlight the correct interpretation of the physical problem in border regions and/or transitions.

Regression Modeling of Daily PM2.5 Concentrations with a Multilayer Perceptron

Various types of energetic fuel combustion processes emit dangerous pollutants into the air, including aerosol particles, marked as PM10. Routine air quality monitoring includes determining the PM10 concentration as one of the basic measurements. At some air monitoring stations, the PM10 measurement is supplemented by the simultaneous determination of the concentration of PM2.5 as a finer fraction of suspended particles. Since the PM2.5 fraction has a significant share in the PM10 fraction, the concentrations of both types of particles should be strongly correlated, and the concentrations of one of these fractions can be used to model the concentrations of the other fraction. The aim of the study was to assess the error of predicting PM2.5 concentration using PM10 concentration as the main predictor. The analyzed daily concentrations were measured at 11 different monitoring stations in Poland and covered the period 2010–2021. MLP (multilayer perceptron) artificial neural networks were used to approximate the daily PM2.5 concentrations. PM10 concentrations and time variables were tested as predictors in neural networks. Several different prediction errors were taken as measures of modeling quality. Depending on the monitoring station, in models with one PM10 predictor, the RMSE error values were in the range of 2.31–6.86 μg/m3. After taking into account the second predictor D (date), the corresponding RMSE errors were lower and were in the range of 2.06–5.54 μg/m3. Our research aimed to find models that were as simple and universal as possible. In our models, the main predictor is the PM10 concentration; therefore, the only condition to be met is monitoring the measurement of PM10 concentrations. We showed that models trained at other air monitoring stations, so-called foreign models, can be successfully used to approximate PM2.5 concentrations at another station.

Predicting reservoir sedimentation using multilayer perceptron – Artificial neural network model with measured and forecasted hydrometeorological data in Gibe-III reservoir, Omo-Gibe River basin, Ethiopia

The estimation and prediction of the amount of sediment accumulated in reservoirs are imperative for sustainable reservoir sedimentation planning and management and to minimize reservoir storage capacity loss. The main objective of this study was to estimate and predict reservoir sedimentation using multilayer perceptron–artificial neural network (MLP-ANN) and random forest regressor (RFR) models in the Gibe-III reservoir, Omo-Gibe River basin. The hydrological and meteorological parameters considered for the estimation and prediction of reservoir sedimentation include annual rainfall, annual water inflow, minimum reservoir level, and reservoir storage capacity. The MLP-ANN and RFR models were employed to estimate and predict the amount of sediment accumulated in the Gibe-III reservoir using time series data from 2014 to 2022. ANN-architecture N4-100-100-1 with a coefficient of determination (R2) of 0.97 for the (80, 20) train-test approach was chosen because it showed better performance both in training and testing (validation) the model. The MLP-ANN and RFR models' performance evaluation was conducted using MAE, MSE, RMSE, and R2. The models’ evaluation result revealed that the MLP-ANN model outperformed the RFR model. Regarding the train data simulation of MLP-ANN and RFR shown R2 (0.99) and RMSE (0.77); and R2 (0.97) and RMSE (1.80), respectively. On the other hand, the test data simulation of MLP-ANN and RFR demonstrated R2 (0.98) and RMSE (1.32); and R2 (0.96) and RMSE (2.64), respectively. The MLP-ANN model simulation output indicates that the amount of sediment accumulation in the Gibe-III reservoir will increase in the future, reaching 110 MT in 2030–2031, 130 MT in 2050–2051, and above 137 MTin 2071–2072.

Improving pressure drop predictions for R134a evaporation in corrugated vertical tubes using a machine learning technique trained with the Levenberg–Marquardt method

The present investigation utilized a machine learning structure to ascertain the pressure drop in vertically positioned, corrugated copper tubes during the evaporation process of R134a. The evaporator was a counter-flow heat exchanger, in which R134a flowed in the inner corrugated tube and hot water flowed in the smooth annulus. Different evaporation mass fluxes (195–406 kg m-2 s-1) and heat fluxes (10.16–66.61 kW m-2) were used with artificial neural networks at different corrugation depths. A multilayer perceptron artificial neural network model with 13 neurons in the hidden layer was proposed. Tan-Sig and Purelin transfer functions were used in the network model developed with the Levenberg–Marquardt training algorithm. The dataset, which consisted of 252 data points, related to the evaporation process, was divided into training (70%), validation (15%), and testing (15%) groups in an arbitrary manner. The artificial neural network model has been demonstrated to effectively forecast the pressure drop that occurs during evaporation. The mean squared error was computed for the ΔP values observed during the evaporation processes, yielding a value of 1.96E-03. The artificial neural network exhibited a high correlation coefficient value of 0.94479. The estimation fluctuations exhibited a range of ± 10%, whereas the experimental and anticipated ΔP data demonstrated a divergence of ± 10.3%.

Supported heterogeneous catalyst of the copper oxide nanoparticles and nanozeolite for binary dyes mixture degradation: Machine learning and experimental design

The present work aims to evaluate the photocatalytic activity of an alternative supported nanocatalyst (nANA@CuO-NPs) for a binary dye mixture (Crystal Violet and Methylene Blue, labeled CV:MB) under visible light and study to propose a model to predict the reaction pathway for the process. An experimental design (Central Composite Rotational Design with 3 factors and 2 levels – CCRD 23) was proposed to evaluate the ideal condition of the CV/MB photodegradation and four machine learning algorithms (Decision Tree, Random Forest, Xtreme Gradient Boosting and Artificial Neural Network-Multilayer Perceptron) were used to predict the main degradation products. The supported nanocatalyst was characterized by X-ray diffraction (XRD), Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) spectroscopy, N2 porosimetry, Diffuse Reflectance Spectroscopy (DRS), Field Emission Gun – Scanning Electron Microscope (FEG-SEM), Scanning electron microscopy (SEM) with energy dispersive X-ray spectroscopy (EDX), zeta potential (ZP) and zero charge point (pHZCP). XRD diffractogram of the nANA@CuO-NPs showed the tenorite and analcime phases with crystallite size and microstrain ranging from 0.319 to 22.35 nm and 6.48–454 × 10−3. ATR-FTIR spectra indicated the presence of functional groups characteristic of copper oxide nanoparticles (CuO-NPs) and nanozeolite analcime (nANA) on the supported catalyst, confirming the impregnation method. SEM micrographs indicated a heterogeneous surface with clusters (CuO-NPs and nANA@CuO-NPs) and homogenous surface for nANA, while the EDX showed the nANA with Si/Al < 2 (low-silica nanozeolite) and nANA@CuO-NPs with the presence in the elemental composition of the catalytic support (nANA) and the photoactive phase (CuO-NPs). Supported nanocatalyst showed physio-chemical stability due to the electrostatic interactions (ZP = 51.5 ± 0.5 mV) and band gap energy (Eg) of the 1.67 eV allowing its applicability under visible radiation. XGB algorithm resulted in the best predictive model (R2 equals 0.8106 and 0.9880 for training and testing, RMSE < 6.0), confirming the obtention of carbon dioxide (m/z = 44), water (m/z = 18) and low-molar mass compounds at the final of the binary dye mixture degradation reaction. The ideal condition for the by CCRD 23 for the degradation of the dye mixture was [CV:MB] = 135 mg/L, [nANA@CuO-NPs] = 1 g/L and pH = 7 with 82.84 % (k = 0.0089 min−1). Therefore, the synthesized nanocatalyst proved its good photocatalytic activity toward a cationic dye mixture of CV and MB dyes. Furthermore, this work confirms the potentiality of machine learning algorithms to develop predictive models for the elucidation of the degradation reaction pathway of organic dyes.

Predicting total biogas potential of food waste using the initial output of biogas potential tests as input data to train an artificial neural network

Quantification of biogas potential is important for predicting anaerobic digestion operability and price. This study uses data from 446 biogas potential tests to train and test a multilayer perceptron artificial neural network (ANN) to forecast total biogas production using the evolution at the start of the experiment (3–14 days) as input data. ANN architecture (training algorithm, activation function, hidden nodes, regularisation, and input data) was optimised using response surface methodology. Best conditions (accuracy/computational speed) were obtained using adaptive moment estimation (adam) training algorithm and rectified linear unit (ReLU) activation function. When using three days of biogas production data, the accuracy of the model was reasonable (r2test = 0.881, r2validation = 0.879), although this increased significantly for 7 days (r2test = 0.953, r2validation = 0.925), or 14 days (r2test = 0.971, r2validation = 0.953). The highest accuracy was reported for readily digestible substrates (sugars and carbohydrates) and macronutrient mixtures. The methodology could be used to shorten prediction times of biogas potential tests.

Predicting the deep drawing process of TRIP steel grades using multilayer perceptron artificial neural networks

Predicting the deep drawing process of TRIP steel grades using multilayer perceptron artificial neural networks

Floating-Point Quantization Analysis of Multi-Layer Perceptron Artificial Neural Networks

The impact of quantization in Multi-Layer Perceptron (MLP) Artificial Neural Networks (ANNs) is presented in this paper. In this architecture, the constant increase in size and the demand to decrease bit precision are two factors that contribute to the significant enlargement of quantization errors. We introduce an analytical tool that models the propagation of Quantization Noise Power (QNP) in floating-point MLP ANNs. Contrary to the state-of-the-art approach, which compares the exact and quantized data experimentally, the proposed algorithm can predict the QNP theoretically when the effect of operation quantization and Coefficient Quantization Error (CQE) are considered. This supports decisions in determining the required precision during the hardware design. The algorithm is flexible in handling MLP ANNs of user-defined parameters, such as size and type of activation function. Additionally, a simulation environment is built that can perform each operation on an adjustable bit precision. The accuracy of the QNP calculation is verified with two publicly available benchmarked datasets, using the default precision simulation environment as a reference.

Optimization of 2024-T3 Aluminum Alloy Friction Stir Welding Using Random Forest, XGBoost, and MLP Machine Learning Techniques.

This study optimized friction stir welding (FSW) parameters for 1.6 mm thick 2024T3 aluminum alloy sheets. A 3 × 3 factorial design was employed to explore tool rotation speeds (1100 to 1300 rpm) and welding speeds (140 to 180 mm/min). Static tensile tests revealed the joints' maximum strength at 87% relative to the base material. Hyperparameter optimization was conducted for machine learning (ML) models, including random forest and XGBoost, and multilayer perceptron artificial neural network (MLP-ANN) models, using grid search. Welding parameter optimization and extrapolation were then carried out, with final strength predictions analyzed using response surface methodology (RSM). The ML models achieved over 98% accuracy in parameter regression, demonstrating significant effectiveness in FSW process enhancement. Experimentally validated, optimized parameters resulted in an FSW joint efficiency of 93% relative to the base material. This outcome highlights the critical role of advanced analytical techniques in improving welding quality and efficiency.