Artificial Neural Network Learning Research Articles

Insight into nettle straw pyrolysis: Multicomponent kinetics, gas emissions and machine learning models

This study aimed to comprehensively investigate the pyrolysis process of nettle straw by analyzing its multicomponent kinetics modeling, kinetics and thermodynamic parameters, pyrolysis performance, reaction mechanisms, gas emissions, correlation and statistical analysis, as well as feasibility for industrial application. The average apparent activation energies for pseudo-hemicellulose (PS-HEC), cellulose (PS-CEL) and lignin (PS-LIG) were estimated as 147.41, 182.66, and 197.12 kJ/mol, respectively, using four model-free methods: Ozawa-Flynn-Wall, Kissinger-Akahira-Sunose, Starink, and Tang. The pyrolysis reaction mechanisms were best described by diffusional and order-based models, with reaction mechanism functions of f(α) = [3/2(1-α)2/3]/[1-(1-α)1/3], f(α) = (1-α)1.5, and f(α) = (1-α)3 for PS-HEC, PS-CEL, and PS-LIG, respectively. TGA-FTIR analysis revealed that the gas emissions during nettle straw pyrolysis followed the order of CO > CO2 > C-O(H) > H2O > SO2 > CO > CH4 > NH3 > HCN. Best learning rate and artificial neural network (ANN) structure model were optimized. The optimal learning rate and ANN model structure were 0.0013 and ANN (5 *9 *1), respectively. NS has the potential to serve as a viable alternative to biofuel due to its certain energy, low price, abundant reserves, good pyrolysis performance, high volatile content, and low nitrogen and sulfur levels.

Prediction of Regional Forest Biomass Using Machine Learning: A Case Study of Beijing, China

Dynamic changes in forest biomass are closely related to the carbon cycle, climate change, forest productivity and biodiversity. However, most previous studies mainly focused on the calculation of current forest biomass, and only a few studies attempted to predict future dynamic changes in forest biomass which obtained uncertain results. Therefore, this study comprehensively considered the effects of multi-stage continuous survey data of forest permanent sample plots, site condition factors and corresponding meteorological factors using Beijing as an example. The geographic detector method was used to screen the key interfering factors that affect the growth of forest biomass. Then, based on the back-propagation artificial neural network (BP-ANN) and support vector machine (SVM) learning methods, 80% of the sample data were extracted to train the model, and thereby verify the prediction accuracy of different modeling methods using different training samples. The results showed that the forest biomass prediction models based on both the machine learning algorithms had good fitting accuracy, and there was no significant difference in the prediction results between the two models. However, the SVM model was better than the BP-ANN. While the BP-ANN model provided more volatile predictions, and the accuracy was above 80%, the prediction results of the SVM model were relatively stable, and the accuracy was above 90%. This study not only provides good technical support for the scientific estimation of regional forest biomass in the future, but also offers reliable basic data for sustainable forest management, planning decisions, forest carbon sequestration and sustainable development.

A long lead time forecast model applying an ensemble approach for managing the great Karun multi-reservoir system

Flow prediction is regarded as a major computational process in strategic water resources planning. Prediction’s lead time has an inverse relationship with results’ accuracy and certainty. This research studies the impact of climate-atmospheric indices on surface runoff predictions with a long lead time. To this end, the correlation of 36 long-distance climate indices with runoff was examined at 10 key nodes of the Great Karun multi-reservoir system in Iran, and indices with higher correlation are divided into 4 different groups. Then, using Artificial Neural Network (ANN) and Ensemble Learning to combine the input variables, flow is predicted in 6-month horizons, and results are compared with observed values. To assess the impact of extending the prediction lead time, results from the proposed model are compared with those of a monthly prediction model. The performed comparison shows that using an ensemble approach improves the final results significantly. Moreover, Tropical Pacific SST EOF, Caribbean SST, and Nino1 + 2 indices are found to be influential parameters to the basin’s inflow. It is observed that the performance of the prediction process varies in different hydrological conditions and the best results are obtained for dry seasons.

Mechanical Strength Estimation of Self-Compacting Geopolymer Concrete using ReLU based Deep Neural Network

ABSTRACT A deep learning artificial neural network model developed in Tensorflow using the ReLU function is presented in this paper to determine the effect of nine different input parameters on the mechanical strength of self-compacting geopolymer concrete (SGC). With the application of the above model, it has been observed that the convergence outcome of our model is predicted with minimum error. The novelty of our model is the ReLU function which improves the accuracy of the results. Our study presents a systematic investigation of the compressive strength, flexural strength, and tensile strengths of SGC depending upon nine vital parameters, such as GFS, Fly ash, NaOH, Na2SiO3 content, Age of specimens, Slump Value, T50, L Box and J Ring value as input. A large number of experiments were conducted to prepare a comprehensive database of the mechanical strength of Flyash-GFS-based SGC. The database has a total of 396 samples which were collected from 132 distinct mixtures, where the minimum and maximum compressive strength value is 7.7 Mpa and 66.7 Mpa, respectively. For the prediction based on deep learning, authors have performed an adequate number of iterations for both training and testing using error estimation parameters such as mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE). The error parameter values for the compressive strength of SGC are 2.42, 2.94 and 1.07, respectively, for MAE, RMSE and MSE. The other mechanical properties and their error parameter have also been thoroughly evaluated in our work.

A data-driven intelligent decision support system that combines predictive and prescriptive analytics for the design of new textile fabrics

In this paper, we propose an Intelligent Decision Support System (IDSS) for the design of new textile fabrics. The IDSS uses predictive analytics to estimate fabric properties (e.g., elasticity) and composition values (% cotton) and then prescriptive techniques to optimize the fabric design inputs that feed the predictive models (e.g., types of yarns used). Using thousands of data records from a Portuguese textile company, we compared two distinct Machine Learning (ML) predictive approaches: Single-Target Regression (STR), via an Automated ML (AutoML) tool, and Multi-target Regression, via a deep learning Artificial Neural Network. For the prescriptive analytics, we compared two Evolutionary Multi-objective Optimization (EMO) methods (NSGA-II and R-NSGA-II) when optimizing 100 new fabrics, aiming to simultaneously minimize the physical property predictive error and the distance of the optimized values when compared with the learned input space. The two EMO methods were applied to design of 100 new fabrics. Overall, the STR approach provided the best results for both prediction tasks, with Normalized Mean Absolute Error values that range from 4% (weft elasticity) to 11% (pilling) in terms of the fabric properties and a textile composition classification accuracy of 87% when adopting a small tolerance of 0.01 for predicting the percentages of six types of fibers (e.g., cotton). As for the prescriptive results, they favored the R-NSGA-II EMO method, which tends to select Pareto curves that are associated with an average 11% predictive error and 16% distance.

Pengembangan Sistem Prediksi Bantuan Program Keluarga Harapan (PKH) Berbasis Machine Learning

The Family Hope Program (PKH) is a poverty alleviation program which is one of the government's strategies in reducing the poverty line. This program provides cash social assistance to poor families who are included in the list of beneficiary families with a focus on education and health. The purpose of implementing the PKH program is not only to reduce poverty and increase human resources but to break the poverty chain. The implementation of PKH in its realization experienced many obstacles that caused the program not to be on target, this was because the data verification process was not yet effective and was still carried out manually. A process is needed to digitize the distribution and realization of the family of hope program. Through this research, a system was developed that can predict the value of PKH beneficiary assistance. The system developed is based on machine learning with a prediction model using Artificial Neural Network (ANN) and Backpropagation learning algorithm. Parameters in the learning system using PKH assessment as many as 8 indicators from the data of PKH beneficiaries in Tabanan Regency. Based on the prediction model testing using two data treatments, namely with and without preprocessing data. Parameters treated with data on numeric attributes and categories provide optimal values with an R2 Score of 0.695824 with a number of hidden layers of 500 and a max epoch of 375

Back Propagation

Abstract: Back Propagation Algorithm research is now very active in the Artificial Neural Network (ANN) and machine learning communities. It has increased a wide range of applications, including image compression, pattern recognition, time series prediction, sequence identification, data filtering, and other intelligent processes carried out by the human brain, have had enormous results. In this paper, we give a quick introduction to ANN and BP algorithms, explain how they operate, and highlight some of the ongoing research projects and the difficulties they face

Evolution of neural network to deep learning in prediction of air, water pollution and its Indian context.

The scenario of developed and developing countries nowadays is disturbed due to modern living style which affects environment, wildlife and natural habitat. Environmental quality has become or is a subject of major concern as it is responsible for health hazard of mankind and animals. Measurements and prediction of hazardous parameters in different fields of environment is a recent research topic for safety and betterment of people as well as nature. Pollution in nature is an after-effect of civilization. To combat the damage already happened, some processes should be evolved for measurement and prediction of pollution in various fields. Researchers of all over the world are active to find out ways of predicting such hazard. In this paper, application of neural network and deep learningalgorithms is chosen for air pollution and water pollution cases. The purpose of this review isto reveal how family of neural network algorithms has applied on these two pollution parameters. In this paper, importance is given on algorithm, and datasets used for air and water pollution as well as the predicted parameters have also been noted for ease of future development. One major concern of this paper is Indian context of air and water pollution research, and the research potential presents in this area using Indian dataset. Another aspect for including both air and water pollutions in one review paper is to generate an idea of artificial neural network and deep learning techniques which can be cross applicable for future purpose.

Synergy-Space Recurrent Neural Network for Transferable Forearm Motion Prediction from Residual Limb Motion.

Transhumeral amputees experience considerable difficulties with controlling a multifunctional prosthesis (powered hand, wrist, and elbow) due to the lack of available muscles to provide electromyographic (EMG) signals. The residual limb motion strategy has become a popular alternative for transhumeral prosthesis control. It provides an intuitive way to estimate the motion of the prosthesis based on the residual shoulder motion, especially for target reaching tasks. Conventionally, a predictive model, typically an artificial neural network (ANN), is directly trained and relied upon to map the shoulder-elbow kinematics using the data from able-bodied subjects without extracting any prior synergistic information. However, it is essential to explicitly identify effective synergies and make them transferable across amputee users for higher accuracy and robustness. To overcome this limitation of the conventional ANN learning approach, this study explicitly combines the kinematic synergies with a recurrent neural network (RNN) to propose a synergy-space neural network for estimating forearm motions (i.e., elbow joint flexion-extension and pronation-supination angles) based on residual shoulder motions. We tested 36 training strategies for each of the 14 subjects, comparing the proposed synergy-space and conventional neural network learning approaches, and we statistically evaluated the results using Pearson's correlation method and the analysis of variance (ANOVA) test. The offline cross-subject analysis indicates that the synergy-space neural network exhibits superior robustness to inter-individual variability, demonstrating the potential of this approach as a transferable and generalized control strategy for transhumeral prosthesis control.

Automatic Hybrid Access Control in SCADA-Enabled IIoT Networks Using Machine Learning.

The recent advancements in the Internet of Things have made it converge towards critical infrastructure automation, opening a new paradigm referred to as the Industrial Internet of Things (IIoT). In the IIoT, different connected devices can send huge amounts of data to other devices back and forth for a better decision-making process. In such use cases, the role of supervisory control and data acquisition (SCADA) has been studied by many researchers in recent years for robust supervisory control management. Nevertheless, for better sustainability of these applications, reliable data exchange is crucial in this domain. To ensure the privacy and integrity of the data shared between the connected devices, access control can be used as the front-line security mechanism for these systems. However, the role engineering and assignment propagation in access control is still a tedious process as its manually performed by network administrators. In this study, we explored the potential of supervised machine learning to automate role engineering for fine-grained access control in Industrial Internet of Things (IIoT) settings. We propose a mapping framework to employ a fine-tuned multilayer feedforward artificial neural network (ANN) and extreme learning machine (ELM) for role engineering in the SCADA-enabled IIoT environment to ensure privacy and user access rights to resources. For the application of machine learning, a thorough comparison between these two algorithms is also presented in terms of their effectiveness and performance. Extensive experiments demonstrated the significant performance of the proposed scheme, which is promising for future research to automate the role assignment in the IIoT domain.

Intelligent assessment of quality characteristics of miscanthus fiber-reinforced polypropylene for sustainable products development

This study aims at determining the optimal material and process parameters that will maximize the quality characteristics of miscanthus fiber-reinforced polypropylene (MFRPP) composite. Taguchi L16 was utilized in the design of experiment and larger-the-better signal-to-noise ratio was used in the analysis of the impact strength of MFRPP. Optimal combination of material and process parameters and the main effects were determined, and the significant variables were identified using analysis of variance. Artificial neural network (ANN) and extreme learning machine (ELM) were utilized on the Taguchi experimental data and used in the prediction of the impact strength of MFRPP. The results showed the optimum impact strength occurred at 25 wt% miscanthus fiber loading. The results of the predictions made revealed that both ANN and ELM are very efficient in predicting the impact strength of MFRPP as shown by the relative errors when compared with the experimental data. Both the performance metrics assessments and the quality of prediction show that ELM is a better machine learning technique than ANN. The high impact strength of MFRPP composite is a confirmation that MFRPP is a very useful material for industrial applications.

Weight versus Node Perturbation Learning in Temporally Extended Tasks: Weight Perturbation Often Performs Similarly or Better

Biological constraints often impose restrictions on plasticity rules such as locality and reward-based rather than supervised learning. Two learning rules that comply with these restrictions are weight (WP) and node (NP) perturbation. NP is often used in learning studies, in particular, as a benchmark; it is considered to be superior to WP and more likely neurobiologically realized, as the number of weights and, therefore, their perturbation dimension typically massively exceed the number of nodes. Here, we show that this conclusion no longer holds when we take two properties into account that are relevant for biological and artificial neural network learning: First, tasks extend in time and/or are trained in batches. This increases the perturbation dimension of NP but not WP. Second, tasks are (comparably) low dimensional, with many weight configurations providing solutions. We analytically delineate regimes where these properties let WP perform as well as or better than NP. Furthermore, we find that the changes in weight space directions that are irrelevant for the task differ qualitatively between WP and NP and that only in WP gathering batches of subtasks in a trial decreases the number of trials required. This may allow one to experimentally distinguish which of the two rules underlies a learning process. Our insights suggest new learning rules which combine for specific task types the advantages of WP and NP. If the inputs are similarly correlated, temporally correlated perturbations improve NP. Using numerical simulations, we generalize the results to networks with various architectures solving biologically relevant and standard network learning tasks. Our findings, together with WP’s practicability, suggest WP as a useful benchmark and plausible model for learning in the brain.2 MoreReceived 9 December 2021Revised 9 September 2022Accepted 23 January 2023DOI:https://doi.org/10.1103/PhysRevX.13.021006Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.Published by the American Physical SocietyPhysics Subject Headings (PhySH)Research AreasEvolving networksFluctuations & noiseNeuronal networksNeuroplasticityPhysical SystemsArtificial neural networksBrain networkLearningNetworksStatistical PhysicsBiological Physics

Response of rectangular footing resting on reinforced silty sand treated with lime using experimental and computational approach

ABSTRACT The usage of geosynthetics can be employed widely in the field and geotechnical engineering. The ground improvement approaches using geosynthetics as reinforcing materials are widely used to overcome the limitations of other methods. Considering these aspects, a sequence of model load versus settlement experiments with variance in the position of geotextile and percentage of lime was executed to observe the toughness as well as settlement properties of the silty sand with single layer of geotextile. There are several aspects of this paper like the computational techniques such as artificial neural network (ANN) and extreme learning machine (ELM). The implementation of computational models gives an advanced resolution for forecasting the load sustaining capability of the rectangular footing in a simple and economic way. These computational models could be judged by employing several well-approved statistical indices. It could also be verified through the experimental results achieved from the studies. The upshot shows that the developed ELM model has a remarkable potential to evaluate the ultimate bearing capacity (UBC) of the rectangular footing. It can be set forth as a predictive tool for the initial designing process.

A hybrid machine learning and embedded IoT-based water quality monitoring system

Since access to clean and safe water is one of life's most basic needs, the exponential growth of the world's population makes it vital to guarantee a workable framework. Manually collecting samples and sending them to a research center for discovery and analysis is the typical approach for gathering information on water qualities. However, this approach is illogical in the long run because it requires a lot of time and labor. This paper aimed to construct, test, and evaluate the usefulness of machine learning and the Internet of Things (IoT) at water storage stations. First, we developed a system prototype and assessed its performance using classifier and reliability matrices. This study considers water's physical and chemical parameters to evaluate the level of water pollutants present in drinking water. The parameters measured include temperature, pH, turbidity, Dissolved Oxygen (DO), Total Dissolved Solids (TDS), Oxidation Reduction Potential (ORP), and electrical conductivity. After analyzing the sensor data, we used Artificial Neural Network (ANN) and Support Vector Machine (SVM) machine learning algorithms to forecast the impurity level of the water measured. The performance showed that the ANN models used have the highest accuracy and are the most suitable to predict water source and status. We also introduced a water treatment method to provide an automated corrective measure based on a specific amount of water contamination. Based on the system's results, we concluded that AI and IoT are more efficient in remotely monitoring safe and harmful water conditions.

Functional Responses of Autaptic Neural Circuits to Acoustic Signals

It is important for functional neurons of animals or human beings to adapt to external stimuli, such as sound, pressure, and light. Regarding this aspect, autaptic neuron enables itself to utilize historical information to modulate its instant dynamics, such that it may be able to behave adaptively. In this paper, a FitzHugh–Nagumo based autaptic neuron is employed to investigate the capability of a sound-sensitive neural circuit’s adaptation and filtering to analog acoustic signals. Extensive simulations are performed for excitatory and inhibitory types of autaptic neurons. The results show that the time-delayed feedback of the excitatory chemical autapse can be tuned to play the role of a narrow-band filter in response to a broadband acoustic signal. While the excitatory chemical autaptic neuron cannot saturate its response amplitude due to its positive feedback gain, the inhibitory chemical autapse can drive the neuron’s amplitude to converge as the intensity of external drive increases, which reveals the capability of adaptation. What’s more, the inhibitory chemical autaptic neuron can also exhibit a novel bursting adaptation, in which the number of spikings contained in one bursting changes as the electrical activity evolves. For electrical autaptic neurons, it is also found that both time-delay feedback gains can effectively modulate the response of neuron to acoustic signal. While the variation of time-lags mainly changes the spiking rates of the excitatory electrical autaptic neuron, the feedback gain alters its response amplitude. Lastly, by carefully tuning the time-lags, the expected subthreshold dynamics for larger inhibitory feedback gains can be switched to nearby quasi-periodic firings, which implies a competing relation between the time-delays and the feedback gains in the spiking dynamics of the inhibitory electrical autaptic neurons. The diverse emerging phenomena are expected to facilitate the design of online or interactive learning artificial neural networks with these functional autaptic neurons.

Modeling of standardized groundwater index of Bihar using machine learning techniques

Groundwater is the most preferred source of water resource for the human needs. Over-exploitation of the groundwater has been led to the tremendous effect on the groundwater drought. Assessment of groundwater drought is difficult to understand due to its complexity, non-linearity feature. In this study, groundwater drought indices of the state of Bihar, India, have been modeled using machine learning technique. The prediction of SGI was done by using Artificial Neural Network (ANN) and Random Forest (RF) machine learning models. The best input combinations of Gravity Recovery and Climate experiment (GRACE) satellite water equivalent data, rainfall data and groundwater level data was used to predicted the SGI. In this study, SGI of 38 districts of Bihar was calculated using groundwater data from 2002 to 2019. The accuracy and efficiency of the RF and ANN models were measured based on the mean square error (MSE) and correlation coefficient value (r). Compared to two models are shown the RF model is a performs superior as compare to ANN model, which model superior is decided based on the correlation coefficient value (r) as 0.95 and MSE value of 0.11. The results show both ML models such as ANN and RF is showed best results with the input combination of GRACE satellite water equivalent data, rainfall and groundwater data.

Алгоритмизация задачи видеоскопического контроля состояния стальных канатов

The paper considers the problem of monitoring the technical state of steel ropes used in industrial and civil installations, such as elevators, cable cars, lifts, etc. To date, the assessment of the state of steel ropes and the safety of their operation is carried out mainly through visual-instrumental control, which is long in time and subject to the influence of the human factor, which determines the relevance of the task of automating this process using computer technology and artificial intelligence methods in order to improve the quality of the assessment, reducing the check interval, detecting defects in steel ropes at an early stage. The article proposes two approaches to the algorithmizing of the problem of videoscopic control of the state of steel ropes using computer vision technologies, based on software modeling of the logic of visual analysis and machine learning methods, such as deep learning artificial neural networks. The authors have developed algorithms for automatic detection of the main types of defects in images of the surface of steel ropes: increase/decrease in diameter, deformation in the form of rope, breaks in the outer wires, thermal and electric damage. The paper presents the results of preliminary testing of the developed algorithms and programs, the composition and methods for determining the main indicators for evaluating their effectiveness are proposed.

Online Machine Vision-Based Modeling during Cantaloupe Microwave Drying Utilizing Extreme Learning Machine and Artificial Neural Network

Online microwave drying process monitoring has been challenging due to the incompatibility of metal components with microwaves. This paper developed a microwave drying system based on online machine vision, which realized real-time extraction and measurement of images, weight, and temperature. An image-processing algorithm was developed to capture material shrinkage characteristics in real time. Constant-temperature microwave drying experiments were conducted, and the artificial neural network (ANN) and extreme learning machine (ELM) were utilized to model and predict the moisture content of materials during the drying process based on the degree of material shrinkage. The results demonstrated that the system and algorithm operated effectively, and ELM provided superior predictive performance and learning efficiency compared to ANN.

The Use of Artificial Neural Networks in the Public Sector

Artificial intelligence (AI) is an extensive scientific field, part of which is the concept of deep learning, belonging to broader family of machine learning (ML) methods, based on artificial neural networks (ANNs). ANNs are active since the 1940s and are applied in many fields. There have been actions around the world for the digital transformation of the public sector and the use of new innovative technologies, but the trajectory and degree of adoption of artificial intelligence technologies in the public sector have been unsatisfactory. Similar issues must be handled, and these problems must be classified. In the present work, preparatory searches were made on Scopus and IEEE bibliographic databases in order to obtain information for the progress of the adoption of ANNs in the public sector starting from the year 2019. Then, a systematic review of published scientific articles was conducted using keywords. Among the 2412 results returned by the search and the application of the selection/rejection criteria, 10 articles were chosen for analysis. The conclusion that emerged after reading the articles was that while the scientific community has a lot of suggestions and ideas for the implementation of ANNs and their financial effects, in practice, there is no appropriate use of them in the public sector. Occasionally, there are cases of implementation funded by state or non-state bodies without a systematic application and utilization of these technologies. The ways and methods of practical application are not further specified, so there are no indications for the systematic application of specialized deep learning techniques and ANNs. The legal framework for the development of artificial intelligence applications, at least in the European Union (EU), is under design, like the necessary ISO standards from an international perspective, and the economic impact of the most recent AI-based technologies has not been fully assessed.

A two-dimensional MoS2 array based on artificial neural network learning for high-quality imaging

A two-dimensional MoS2 array based on artificial neural network learning for high-quality imaging