Optimal Artificial Neural Network Research Articles

Genetic Algorithm-Based Optimization of Artificial Neural Network of Process Parameters and Characterization of Machining Errors in Graphene Mixed Dielectric

Genetic Algorithm-Based Optimization of Artificial Neural Network of Process Parameters and Characterization of Machining Errors in Graphene Mixed Dielectric

An experimental study on enhancing performance and reducing emissions of CRDI engine operated on Co-pyrolysis oil using particle swarm optimization

This study explores the use of Artificial Neural Network and Particle Swarm Optimization to improve the performance and emissions of a CRDI engine fuelled by a mixture of co-pyrolyzed oil and diesel. Experimental results show that the mixed fuels perform better than pure diesel in various engine conditions, achieving higher brake thermal efficiency and lower smoke emissions. The blend D90P10 has the highest BTE of 37.4% working at FIT 30°bTDC and FIP 500 bar. However, NO and CO2 emissions were increased for the same condition. A multi-parametric ANN model is created to build a predictive model based on this experimental data, relating inputs (fuel injection pressure, fuel injection timing, and blend percentages) to outputs (Brake Specific Fuel Consumption, NO, smoke, CO2 emissions). The ANN model attains high accuracy with R2 values of 0.99, 0.98, 0.99, and 0.99 for BSFC, NO, Smoke, and CO2 emission predictions, respectively. The results indicate that for optimal performance, the best conditions are a FIT of 29.71ºbTDC, a FIP of 500 bar, and a co-pyrolysis oil blend percentage of 19.61%. On the other hand, to minimize engine emissions, the best conditions are FIT of 20.28ºbTDC, FIP of 300 bars, and a blend percentage of 25.6%. A balanced approach results in an optimal condition of FIT at 29.5ºbTDC, FIP of 500 bars, and a blend percentage of 6.11%. Experimental validation of these optimal predictions shows a maximum error of only 3.18%, highlighting the reliability and accuracy of the model.

Rapid reactivation of long-term frozen preservation anammox granular sludge by biochar: Performance, mechanism and artificial neural network optimization

Anaerobic ammonia oxidation (Anammox) is an efficient nitrogen removal technology for wastewater, and has been successfully applied in practical wastewater treatment. The preservation and rapid reactivation of anammox bacteria (AnAOB) play an important role in the anammox application. Biochar has been widely used in the water treatment industry due to its stable structure, economic and environmental protection characteristics. Hence, this study used biochar to accelerate anammox granular sludge (AnGS) performance recovery after 60 d preservation under low-temperature frozen (−20 °C). The time of AnGS recovery in the group with biochar addition was shortened by 15 d than the natural recovery group. Meanwhile, biochar addition improved the shock loading resistance of the anammox system. EPS secretion, and the types and quantities of functional groups in EPS increased under biochar stimulation, especially tighter PN. The functional groups in biochar can act as electron shuttles during anammox process, accelerating the electron transfer rate. Biochar addition enhanced the relative abundance of AnAOB, especially Candidatus Brocadia. PICRUSt analysis indicated that biochar promoted key gene related to anammox process and accelerated the recovery of frozen sludge. In addition, the changes in effluent nitrogen removal efficiency were successfully predicted by artificial neural network modeling with R = 0.9895 and MSE = 0.0048. This result may provide a new technological idea for the rapid recovery of AnGS under long time frozen conditions.

Optimizing reverse osmosis desalination from brackish waters: Predictive approach employing response surface methodology and artificial neural network models

A mathematical model frame has been developed to optimize the performance of reverse osmosis (RO) desalination in brackish waters with medium to high salt concentrations using response surface methodology (RSM) and artificial neural network (ANN), specifically catered towards efficiently operating RO desalination systems. The RSM model aims at predicting RO performance with second-order polynomials based on 15 experimental datasets. Inputs for RSM model include feed TDS concentration, feed water flow rate, and concentration-to-desalination ratio, while the outputs consider rejection, membrane flux, performance index, and specific energy consumption. The Genetic Algorithm optimized back-propagation Neural Network was employed to update network weights and biases, leading to an optimal ANN network structure of 3–10-4. A ratio of 70 % training, 15 % validation, and 15 % test data was chosen to establish an ANN model that predicts the performance of RO desalination. The ANN model demonstrated significantly higher prediction accuracy than the RSM model, with impressive correlation coefficients of 0.9936 for the performance index and 0.9932 for SEC prediction, and low RMSE of 0.85 for rejection. To maximize the performance index and minimize specific energy consumption, a two-step optimization method was further set up by applying the RSM model to initially determine optimal process conditions followed by the ANN model to enhance the accuracy of performance forecasts. Optimal process conditions were proposed for specific feed TDS scenarios, which involved setting a feed TDS concentration limit to meet drinking water standards.

RSM versus ANN for modeling and optimization of magnetic adsorbent based on montmorillonite and CoFe2O4

A highly resourceful, environmentally benign, and recyclable magnetic montmorillonite composite (MMT/CF) was obtained through a simple one-step hydrothermal method and exhibited excellent Pb (II) removal. The as-synthesized adsorbent was then characterized by XRD, SEM–EDX, FTIR, BET, and TGA-DTA. The operating parameters including adsorbent dosage, initial Pb (II) concentration, solution pH, and time were studied. Also, a comparative approach was formed between response surface methodology (RSM) and artificial neural network (ANN) to optimize and model the removal efficiency of Pb (II) by MMT/CF. The results indicated that the ANN model was more precise and quite trusted optimization tool than RSM in consideration of its higher correlation coefficient (R2 = 0.998) and lower prediction errors (RMSE = 0.851 and ADD = 0.505). Langmuir isotherm provided the best fit to the experimental data, and the maximum adsorption capacity was 101.01 mg/g. Additionally, the kinetic studies showed that the pseudo-second-order model fitted well with the experimental data. The magnetic MMT/CF composite possesses high adsorption capacity and is suitable for reuse. Therefore, this study shows that MMT/CF composite can be a potential adsorbent in Pb (II) uptake from aqueous media.

Assessing Viscosity in Sustainable Deep Eutectic Solvents and Cosolvent Mixtures: An Artificial Neural Network-Based Molecular Approach.

Deep eutectic solvents (DESs) are gaining recognition as environmentally friendly solvent alternatives for diverse chemical processes. Yet, designing DESs tailored to specific applications is a resource-intensive task, which requires an accurate estimation of their physicochemical properties. Among them, viscosity is crucial, as it often dictates a DES's suitability as a solvent. In this study, an artificial neural network (ANN) is introduced to accurately describe the viscosity of DESs and their mixtures with cosolvents. The ANN utilizes molecular parameters derived from σ-profiles, computed using the conductor-like screening model for the real solvent segment activity coefficient (COSMO-SAC). The data set comprises 1891 experimental viscosity measurements for 48 DESs based on choline chloride, encompassing 279 different compositions, along with 1618 data points of DES mixtures with cosolvents as water, methanol, isopropanol, and dimethyl sulfoxide, covering a wide range of viscosity measurements from 0.3862 to 4722 mPa s. The optimal ANN structure for describing the logarithmic viscosity of DESs is configured as 9-19-16-1, achieving an overall average absolute relative deviation of 1.6031%. More importantly, the ANN shows a remarkable extrapolation capacity, as it is capable of predicting the viscosity of systems including solvents (ethanol) and hydrogen bond donors (2,3-butanediol) not considered in the training. The ANN model also demonstrates an extensive applicability domain, covering 94.17% of the entire database. These achievements represent a significant step forward in developing robust, open source, and highly accurate models for DESs using molecular descriptors.

Optimization design of an innovative francis draft tube: Insight into improving operational flexibility

Hydropower has transitioned from base-load generation into an energy regulator in the hydro-wind-solar integrated systems, resulting in prolonged turbine operation under off-design conditions. Inspired by the new requirements of operation mode, an innovative Francis draft tube with inclined conical diffuser is presented to mitigate the pressure fluctuations and fatigue damage induced by flow instabilities. The design is optimized and comprehensively investigated to improve the operational flexibility. Computational fluid dynamics combined with design of experiments, artificial neural network and multi-objective optimization process lead to two optimal designs denoted as Opt-1 and Opt-2 from the Pareto fronts. Corresponding maximum pressure fluctuation amplitude is reduced by 15 % and 20 % compared to the traditional, respectively. An 81 % decrease of the fluctuating pressure recovery for the modified draft tube is obtained, leading to a steadier runner torque and consequently lower power swing. The asymmetric design of inclined daft tube is mainly targeted at the mitigation of the synchronous component as well as the attenuation of interaction between the swirling flow and elbow, accompanied with the mitigated energy of first-order vortex structure. The proposed innovative draft tube and optimization technique are found useful to broaden the operating range of hydraulic turbine in a computationally affordable way.

Three-dimensional simulation and artificial neural network optimization of a flat plate heat pipe for the cooling of telecommunication equipment

To address the significantly increased heat dissipation requirements of 5G base stations, a flat plate heat pipe (FPHP) of 500 × 200 × 3 mm3 comprising of 19 independent channels was developed. To solve the problem of simulation distortion due to the high superheat of the initial boiling of the working fluid in the ultra-long channel of 490 × 5 × 1.5 mm3, this paper adopts the Lee model considering the superheat, and simulates based on the Volume of fluid method, and the simulation results are compared with the experiments, and the average relative error is less than 5 %. The FPHP was explored at 100–500 W heat power and filling ratio of 30 %-70 %, depending on the application scenarios of the base station. By partitioning the 19 channels and calculating the thermal resistance of the FPHP based, an Artificial neural network (ANN) model with 10 hidden layer nodes is established based on 675 data sets. The results showed that the mean square error and correlation coefficient were 1.30 × 10-5 and 0.9990. The ANN prediction results are validated, the mean relative error is 5.4 %, demonstrating that the ANN model can be an effective tool to optimize the FPHP for the telecommunication equipment.

An optimal framework for natural English translate processing: On the application of artificial neural network and Adam optimization algorithm

Traditional rule-based and statistical methods have limitations when dealing with complex language structures and semantics. In neural network machine translation algorithms, the objective function is usually to improve the accuracy of n-ary words. However, this does not guarantee a more natural and accurate translation. To overcome these challenges, this paper proposes an optimization algorithm for English natural translation processing based on neural networks, which combines Generative Adversarial Network (GAN) and Transformer models. In GAN, the generative model uses the Transformer model to generate false samples, while the discriminative model uses a binary classifier based on convolutional neural networks and attention mechanisms to distinguish between true and false samples. During the training process, reinforcement learning algorithms are added to evaluate and adjust the generated sentences, and the parameters of the generated model are updated. The classification results of the discriminative model are used together with the Bilingual Evaluation Basis Value (BLEU) objective function to evaluate false samples, and the results are fed back to the generating model to guide parameter updates and optimization. Extensive experiments were conducted on a standard English-Chinese machine translation dataset to evaluate our method. Compared with the benchmark model that only uses supervised learning methods, our neural network-based optimization algorithm for English natural translation processing has achieved significant improvements in translation quality. According to statistical comparison, compared with the Transformer model (BLUE = 33.63 and AP = 90%) and the deep learning model based on long-term and short-term memory (BLUE = 30.26 and AP = 83%), the GAN and Transformer models proposed as the best framework exhibit better performance in bilingual evaluation deficiency (BLEU) (34.35) and accuracy (AP = 95%).

A Two-stage Method for Damage Detection in Z24 Bridge Based on K-nearest Neighbor and Artificial Neural Network

In this paper, we propose an effective approach for identifying damages in the Z24 bridge, a large-scale bridge in Switzerland. The dataset of the Z24 bridge is evaluated as a benchmark, reflecting the behavior of a real structure that has been used for numerous studies. However, most of the previous studies have only addressed the issues of updating the model or estimating the location and severity of damages. The core idea behind our proposed approach is to leverage the strengths of two effective Machine Learning (ML) algorithms: K-Nearest Neighbor (KNN) and Artificial Neural Network (ANN), to assess both the location and severity of damages in the Z24 bridge. First, we employ KNN, an unsupervised learning algorithm, for pinpointing the damage location. This strategy proves highly efficient, significantly reducing computation time by circumventing the need for a loss function during KNN training. By adopting this approach, KNN effectively mitigates the risk of encountering local minima in the ANN optimization process. Subsequently, we deploy ANN to determine the damage severity. When compared to previous studies on the Z24 bridge, our proposed method (KNN-ANN) exhibits promising results. Furthermore, our results illustrate that KNN-ANN consistently outperforms traditional ANN methodologies.

Predicting Bulk Density for Agglomerated Raspberry Ketone via Integrating Morphological and Size Metrics Using Artificial Neural Networks

The bulk density of the particles, which is directly related to transportation and storage costs, is an important basic characteristic of products as well as an important parameter in many processing systems. This work quantified the relationship between the tapped bulk density of raspberry ketone with different degrees of agglomeration and morphological metrics (particle shape descriptors and roughness descriptors) and size metrics (size descriptors) and developed an artificial neural network (ANN) prediction model for the tapped bulk density of raspberry ketone. Samples prepared under different conditions were sieved and remixed, the tapped bulk density of the particles was then measured, and the descriptor features of the particles were obtained by combining them with image processing. The dimensions of the variables were decreased by principal component analysis and variance processing. To overcome the hyperparameter estimation of the heuristic-based artificial neural networks, the network model architectures were optimized by a neural architecture search strategy combining two-objective optimization. The results demonstrated that the tapped bulk density of raspberry ketone products is not only related to the descriptors of particle size and shape but also has a non-negligible relationship with particle roughness descriptors. The performance of the optimal ANN model demonstrated that the model can well predict the tapped bulk density of raspberry ketone with different degrees of agglomeration. The ANN model obtained by extracting morphology and size metrics through online image analysis can be used to measure the tapped bulk density in real-time and has the potential to be used for developing model-based online process monitoring.

Chaos Game Optimization-Hybridized Artificial Neural Network for Predicting Blast-Induced Ground Vibration

In this study, we introduced the chaos game optimization-artificial neural network (CGO-ANN) model as a novel approach for predicting peak particle velocity (PPV) induced by mine blasting. The CGO-ANN model is compared with other established methods, including the particle swarm optimization-artificial neural network (PSO-ANN), the genetic algorithm-artificial neural network (GA-ANN), single ANN, and the USBM empirical model. The aim is to demonstrate the superiority of the CGO-ANN model for PPV prediction. Utilizing a dataset comprising 180 blasting events from the Tonglushan Copper Mine in China, we investigated the performance of each model. The results showed that the CGO-ANN model outperforms other models in terms of prediction accuracy and robustness. This study highlights the effectiveness of the CGO-ANN model as a promising tool for PPV prediction in mining operations, contributing to safer and more efficient blasting practices.

Multi-performance collaborative optimization of existing residential building retrofitting in extremely arid and hot climate zone: A case study in Turpan, China

Multi-dimensional performance optimization of buildings is crucial to achieving comfort and realizing green and sustainable development. This study developed a multi-performance collaborative optimization framework (MPC_BPO) to analyze modern residential retrofitting in extremely arid and hot climate zone. The parametric environmental simulation (PES) demonstrated high consistency with actual operating results (R2 > 0.92), and the multi-objective optimized artificial neural network (OPT-ANN) effectively established a high precision multi-performance prediction model (R2 > 0.96). Various multi-objective algorithms exhibited a high convergence level in the high-dimensional complex optimization process. The entropy weight and subjective weight methods ranked the Pareto frontier solution set, where the optimization scheme dominated by energy use intensity (EUI) increased daylighting index (DLI) and thermal comfort hours (TCH) by 12.32 % and 6.67 %, respectively, while reducing EUI by over 8.47 %. Applying different scenarios to other household buildings verified the generalization characteristics, showing overall high consistency despite variations in performance indicators. This research provides a scientific basis for improving residential performance in special climate zones through an appropriate framework. Using scientific methods, it analyzes the intrinsic logic between high-dimensional parameters and multiple building performances.

A cluster-based human resources analytics for predicting employee turnover using optimized Artificial Neural Networks and data augmentation

This study presents an innovative methodology to predict employee turnover by integrating Artificial Neural Networks (ANN) with clustering techniques. We focus on hyperparameter tuning with various input parameters to obtain optimal ANN models. By segmenting data, the study identifies critical turnover predictors, allowing targeted interventions to be implemented to improve the efficiency and effectiveness of retention policies. Data augmentation using Conditional Generative Adversarial Networks (CTGAN) is performed on clusters with imbalanced data. Following this, the optimized ANN models are applied to these augmented clusters, leading to a notable improvement in their performance. We evaluate our optimized ANN models against five other ANN variants and four traditional machine learning models to demonstrate their superior accuracy and recall. The proposed approach achieves operational advantages by shifting away from generalized strategies to more focused, cluster-based policies, which can optimize resource utilization and reduce costs. Because of its practicality and enhanced ability to predict and manage employee turnover, this method, supported by empirical evidence, is a significant advancement in human resource (HR) analytics

Multivariate modelling of AA6082-T6 drilling performance using RSM, ANN and response optimization

The AA6082-T6 was experimentally studied in the present research with respect to the drilling performance. Drill diameter, cutting speed and feed rate were examined, using a full factorial design. Mathematical modelling of the process was carried out using the Response Surface Methodology (RSM) as well as the Artificial Neural Network (ANN) techniques. The output results in terms of cutting force, torque and surface roughness, revealed high levels of correlation between the experimental and the predicted data. Specifically, the Mean Absolute Percentage Error (MAPE) values using RSM compared to the ones of the experiments, were equal to 2.14%, 3.49% and 6.16% for Fz, Mz and Ra respectively. The equivalent MAPE between the ANN and the experiments were found to be 2.19%, 1.82% and 2.85% accordingly. Moreover, the most significant terms were revealed, being the interaction D × f for the thrust force and the torque with contribution percentages equal to approximately 44% and 42% respectively, and the term D2 for the surface roughness with 51%. The evaluation of the machining parameters, identified their significance, enabling the selection of the optimal cutting parameters, which were obtained by the desirability function, taking into account the importance of the generated surface quality and the reduction of cost. The solutions given by this approach, pointed out the Ø9 tool, coupled with Vc = 50 m/min and f = 0.15mm/rev as a well-balanced combination, whereas the Ø9.9 tool used under the same conditions, yielded the best possible surface quality (appr. 0.2 μm).

Enhancing urban blue-green landscape quality assessment through hybrid genetic algorithm-back propagation (GA-BP) neural network approach: a case study in Fucheng, China

This study employs an artificial neural network optimization algorithm, enhanced with a Genetic Algorithm-Back Propagation (GA-BP) network, to assess the service quality of urban water bodies and green spaces, aiming to promote healthy urban environments. From an initial set of 95 variables, 29 key variables were selected, including 17 input variables, such as water and green space area, population size, and urbanization rate, six hidden layer neurons, such as patch number, patch density, and average patch size, and one output variable for the comprehensive value of blue-green landscape quality. The results indicate that the GA-BP network achieves an average relative error of 0.94772%, which is superior to the 1.5988% of the traditional BP network. Moreover, it boasts a prediction accuracy of 90% for the comprehensive value of landscape quality from 2015 to 2022, significantly outperforming the BP network’s approximate 70% accuracy. This method enhances the accuracy of landscape quality assessment but also aids in identifying crucial factors influencing quality. It provides scientific and objective guidance for future urban landscape structure and layout, contributing to high-quality urban development and the creation of exemplary living areas.

Optimizing PCF-SPR sensor design through Taguchi approach, machine learning, and genetic algorithms

Designing Photonic Crystal Fibers incorporating the Surface Plasmon Resonance Phenomenon (PCF-SPR) has led to numerous interesting applications. This investigation presents an exceptionally responsive surface plasmon resonance sensor, seamlessly integrated into a dual-core photonic crystal fiber, specifically designed for low refractive index (RI) detection. The integration of a plasmonic material, namely silver (Ag), externally deposited on the fiber structure, facilitates real-time monitoring of variations in the refractive index of the surrounding medium. To ensure long-term functionality and prevent oxidation, a thin layer of titanium dioxide (TiO2) covers the silver coating. To optimize the sensor, five key design parameters, including pitch, air hole diameter, and silver thickness, are fine-tuned using the Taguchi L8(25) orthogonal array. The optimal results obtained present spectral and amplitude sensitivities that reach remarkable values of 10,000 nm/RIU and 235,882 RIU-1, respectively. In addition, Artificial Neural Network (ANN) optimization techniques, specifically Multi-Layer Perceptron (MLP) and Particle Swarm Optimization (PSO), are used to predict a critical optical property of the sensor confinement loss (αloss). These predictions are derived from the same input structure parameters that are present in the full L32(25) design experiment. A genetic algorithm (GA) is then applied for optimization with the goal of maximizing the confinement loss. Our results highlight the effectiveness of training PSO artificial neural networks and demonstrate their ability to quickly and accurately predict results for unknown geometric dimensions, demonstrating their significant potential in this innovative context. The proposed sensor design can be used for various applications including pharmaceutical inspection and detection of low refractive index analytes.

School Electricity Consumption in a Small Island Country: The Case of Fiji

Electricity consumption in buildings is one of the major causes of energy usage and knowledge of this can help building owners and users increase energy efficiency and conservation efforts. For Pacific Island countries, building electricity demand data is not readily accessible or available for constructing models to predict electricity demand. This paper starts to fill this gap by studying the case of schools in Fiji. The aim of the paper is to assess the factors affecting electricity demand for grid-connected Fijian schools and use this assessment to build mathematical models (multiple linear regression (MLR) and artificial neural network (ANN)) to predict electricity consumption. The average grid-connected electricity demand in kWh/year was 1411 for early childhood education schools, 5403 for primary schools, and 23,895 for secondary schools. For predicting electricity demand (ED) for all grid-connected schools, the stepwise MLR model shows that taking logarithm transformations on both the dependent variable and independent variables (number of students, lights, and air conditioning systems) yields statistically significant independent variables with an R2 value of 73.3% and RMSE of 0.2248. To improve the predicting performance, ANN models were constructed on both the natural form of variables and transformed variables. The optimum ANN model had an R2 value of 95.3% and an RMSE of 59.4 kWh/year. The findings of this study can assist schools in putting measures in place to reduce their electricity demand, associated costs, and carbon footprint, as well as help government ministries make better-informed policies.

Comparative Performance Evaluation of ANN and Hybrid ANN-HBA Models for Temporal and Spatial Modeling of Precipitation in East Azarbaijan Province of Iran

In this study, temporal and spatial modeling of precipitation was performed in East Azarbaijan province of Iran using 27-year (1996-2022) data by employing two neural network models, i.e., stand-alone artificial neural network (ANN) and hybrid ANNHoney badger optimization algorithm (ANN-HBA). In temporal modeling, one-month and two-month precipitation delay steps were used as input parameters, and monthly precipitation was considered as output parameter. On the other hand, in spatial modeling, longitude, latitude, and altitude were used as input parameters and average monthly precipitation was considered as output parameter. Comparative performance of both the employed models was evaluated using three statistical criteria, i.e., root mean square error (RMSE), correlation coefficient (R) and Nash- Sutcliffe efficiency (NSE). In addition, precipitation in ungauged areas was estimated based on rainfall data of the nearby rain gauges through spatial interpolation methods such as isohyet and Thiessen polygon. Results of the temporal precipitation modeling revealed better performance of the hybrid ANN-HBA model over the stand-alone ANN model. Furthermore, the hybrid ANN-HBA model, during testing phase of the temporal modeling, could resemble the precipitation most accurately at Maraghe station (with values of R as 0.95, RMSE as 1 mm, and NSE as 0.85). Similarly, the hybrid ANN-HBA model during testing phase (with values of R as 0.95, RMSE as 1 mm, and NSE as 0.96) outperformed the stand-alone ANN model in depicting the precipitation adequately at spatial scale.

A neural network framework for optimizing parallel computing in cloud servers

Energy efficiency has become a major focus in optimizing hardware resource usage for Cloud servers. One approach widely employed to enhance the execution of parallel applications is thread-level parallelism (TLP) exploitation. This technique leverages multiple threads to improve computational efficiency and performance. However, the increasing heterogeneity of resources in cloud environments and the complexity of selecting the optimal configuration for each application pose a significant challenge to cloud users due to the massive number of possible configurations and the need to effectively harness TLP in diverse hardware setups to achieve optimal energy efficiency and performance. To address this challenge, we propose TLP-Allocator, an artificial neural network (ANN) optimization strategy that uses hardware and software metrics to build and train an ANN model. It predicts worker node and thread count combinations that provide optimal energy-delay product (EDP) results. In experiments using ten well-known applications on a private cloud with heterogeneous resources, we show that TLP-Allocator predicts combinations that yield EDP values close to the best achieved by an exhaustive search. It also improves the overall EDP by 38.2% compared to state-of-the-art workloads scheduling on cloud environments.