General Multi-Fidelity Framework for Training Artificial Neural Networks With Computational Models

Abstract The development of computational models for analysis of the operation of water supply systems requires the calibration of pipes' roughness, among other parameters. Inadequate values of this parameter can result in inaccurate solutions, compromising the applicability of the model as a decision-making tool. This paper presents a metamodel to estimate the pressure at all nodes of a distribution network based on artificial neural networks (ANNs), using a set of field data obtained from strategically located pressure sensors. This approach aims to increase the available pressure data, reducing the degree of freedom of the calibration problem. The proposed model uses the inlet flow of the district metering area and pressure data monitored in some nodes, as input data to the ANN, obtaining as output, the pressure values for nodes that were not monitored. Two case studies of real networks are presented to validate the efficiency and accuracy of the method. The results ratify the efficiency of ANN as state forecaster, showing the high applicability of the metamodel tool to increase a database or to identify abnormal events during an operation.

Download Full-text

Evolutionary Approaches for ANNs Design

Encyclopedia of Artificial Intelligence ◽

10.4018/978-1-59904-849-9.ch088 ◽

2011 ◽

pp. 575-580

Author(s):

Antonia Azzini ◽

Andrea G.B. Tettamanzi

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Evolutionary Algorithms ◽

Computational Models ◽

Artificial Neurons ◽

Important Research Topic ◽

Recent Developments ◽

Evolving System ◽

Artificial Neural ◽

Ann Application

Artificial neural networks (ANNs) are computational models, loosely inspired by biological neural networks, consisting of interconnected groups of artificial neurons which process information using a connectionist approach. ANNs are widely applied to problems like pattern recognition, classification, and time series analysis. The success of an ANN application usually requires a high number of experiments. Moreover, several parameters of an ANN can affect the accuracy of solutions. A particular type of evolving system, namely neuro-genetic systems, have become a very important research topic in ANN design. They make up the so-called Evolutionary Artificial Neural Networks (EANNs), i.e., biologicallyinspired computational models that use evolutionary algorithms (EAs) in conjunction with ANNs. Evolutionary algorithms and state-of-the-art design of EANN were introduced first in the milestone survey by Xin Yao (1999), and, more recently, by Abraham (2004), by Cantu-Paz and Kamath (2005), and then by Castellani (2006). The aim of this article is to present the main evolutionary techniques used to optimize the ANN design, providing a description of the topics related to neural network design and corresponding issues, and then, some of the most recent developments of EANNs found in the literature. Finally a brief summary is given, with a few concluding remarks.

Download Full-text

Hybrid Computational Models for Software Cost Prediction: An Approach Using Artificial Neural Networks and Genetic Algorithms

Enterprise Information Systems - Lecture Notes in Business Information Processing ◽

10.1007/978-3-642-00670-8_7 ◽

2009 ◽

pp. 87-100 ◽

Cited By ~ 2

Author(s):

Efi Papatheocharous ◽

Andreas S. Andreou

Keyword(s):

Neural Networks ◽

Genetic Algorithms ◽

Artificial Neural Networks ◽

Computational Models ◽

Cost Prediction ◽

Software Cost ◽

Artificial Neural

Download Full-text

Enhancing the reusability and interoperability of artificial neural networks with DEVS modeling and simulation

International Journal of Modeling Simulation and Scientific Computing ◽

10.1142/s1793962316500057 ◽

2016 ◽

Vol 07 (02) ◽

pp. 1650005 ◽

Cited By ~ 1

Author(s):

David Ifeoluwa Adelani ◽

Mamadou Kaba Traoré

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Modeling And Simulation ◽

Computational Models ◽

Learning Algorithm ◽

Discrete Event ◽

Neural Network Modeling ◽

System Specification ◽

Ann Models ◽

Artificial Neural

Artificial neural networks (ANNs), a branch of artificial intelligence, has become a very interesting domain since the eighties when back-propagation (BP) learning algorithm for multilayer feed-forward architecture was introduced to solve nonlinear problems. It is used extensively to solve complex nonalgorithmic problems such as prediction, pattern recognition and clustering. However, in the context of a holistic study, there may be a need to integrate ANN with other models developed in various paradigms to solve a problem. In this paper, we suggest discrete event system specification (DEVS) be used as a model of computation (MoC) to make ANN models interoperable with other models (since all discrete event models can be expressed in DEVS, and continuous models can be approximated by DEVS). By combining ANN and DEVS, we can model the complex configuration of ANNs and express its internal workings. Therefore, we are extending the DEVS-based ANN proposed by Toma et al. [A new DEVS-based generic artficial neural network modeling approach, The 23rd European Modeling and Simulation Symp. (Simulation in Industry), Rome, Italy, 2011] for comparing multiple configuration parameters and learning algorithms and also to do prediction. The DEVS models are described using the high level language for system specification (HiLLS), [Maïga et al., A new approach to modeling dynamic structure systems, The 29th European Modeling and Simulation Symp. (Simulation in Industry), Leicester, United Kingdom, 2015] a graphical modeling language for clarity. The developed platform is a tool to transform ANN models into DEVS computational models, making them more reusable and more interoperable in the context of larger multi-perspective modeling and simulation (MAS).

Download Full-text

Research on the Development and Applications of Artificial Neural Networks

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.556-562.6011 ◽

2014 ◽

Vol 556-562 ◽

pp. 6011-6014 ◽

Cited By ~ 1

Author(s):

Xiao Guang Li

Keyword(s):

Machine Learning ◽

Neural Networks ◽

Artificial Neural Networks ◽

Intelligent Control ◽

Computational Models ◽

Bayesian Probability ◽

Control Techniques ◽

Central Nervous Systems ◽

Artificial Neural ◽

The Brain

Intelligent control is a class of control techniques that use various AI computing approaches like neural networks, Bayesian probability, fuzzy logic, machine learning, evolutionary computation and genetic algorithms. In computer science and related fields, artificial neural networks are computational models inspired by animals’ central nervous systems (in particular the brain) that are capable of machine learning and pattern recognition. They are usually presented as systems of interconnected “neurons” that can compute values from inputs by feeding information through the network. Like other machine learning methods, neural networks have been used to solve a wide variety of tasks that are hard to solve using ordinary rule-based programming, including computer vision and speech recognition.

Download Full-text

Recent Advancements in Study of Effects of Nano Micro Additives on Solid Propellants Combustion by Means of the Data Science Methods

Defence Science Journal ◽

10.14429/dsj.69.12948 ◽

2019 ◽

Vol 69 (1) ◽

pp. 20-26 ◽

Cited By ~ 4

Author(s):

Victor S. Abrukov ◽

Alexander N. Lukin ◽

Darya A. Anufrieva ◽

Charlie Oommen ◽

V. R. Sanalkumar ◽

...

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Kinetic Parameters ◽

Data Science ◽

Computational Models ◽

Science Methods ◽

Solid Propellants ◽

Rocket Propellants ◽

Artificial Neural ◽

Solid Rocket

The efforts of Russian-Indian research team for application of the data science methods, in particular, artificial neural networks for development of the multi-factor computational models for studying effects of additive’s properties on the solid rocket propellants combustion are presented. The possibilities of the artificial neural networks (ANN) application in the generalisation of the connections between the variables of combustion experiments as well as in forecasting of “new experimental results” are demonstrated. The effect of particle size of catalyst, oxidizer surface area and kinetic parameters like activation energy and heat release on the final ballistic property of AP-HTPB based propellant composition has been modelled using ANN methods. The validated ANN models can predict many unexplored regimes, like pressures, particle sizes of oxidiser, for which experimental data are not available. Some of the regularly measured kinetic parameters extracted from non-combustion conditions could be related to properties at combustion conditions. Results predicted are within desirable limits accepted in combustion conditions.

Download Full-text

Prediction of sea ice concentration using artificial neural networks

10.32920/ryerson.14646963 ◽

2021 ◽

Author(s):

Mohammed Kandil El-Emam El-Diasty

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Networks ◽

Computational Models ◽

Ocean Model ◽

Performance Comparison ◽

Eastern Canada ◽

Sea Ice Concentration ◽

Ice Concentration ◽

Artificial Neural

Artificial neural networks are computational models capable of solving complex problems through learning, or training, and then generalizing the network solution for other inputs. This thesis examines the performance of two neural network-based models, which were developed for predicting the ice concentration in the Gulf of St. Lawrence in Eastern Canada. The first is a batch model which uses time to predict future ice concentration, while the second model predicts the ice concentration sequentially. It is shown that the performance of the two models is almost identical, as long as no abrupt changes occur in the ice conditions. If, however, the ice condition changes suddenly, only the sequential model is proved to be capable of predicting the ice condition without noticeable accuracy degradation. A performance comparison is made between the developed neural network model and coupled ice-ocean model for ice concentration prediction to further validate the model.

Download Full-text

Ecophysiology modeling by artificial neural networks for different spacings in eucalypt

Comunicata Scientiae ◽

10.14295/cs.v9i3.2741 ◽

2018 ◽

Vol 9 (3) ◽

pp. 438-448

Author(s):

Bruno Oliveira Lafetá ◽

Reynaldo Campos Santana ◽

Gilciano Saraiva Nogueira ◽

Tamires Mousslech Andrade Penido ◽

Diego Dos Santos Vieira

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Leaf Area ◽

Computational Models ◽

Experimental Unit ◽

Wood Biomass ◽

Sample Tree ◽

Production Models ◽

Artificial Neural ◽

Multilayer Perceptron Networks

Growth and production models are widely used to predict yields and support forestry decisions. Artificial Neural Networks (ANN) are computational models that simulate the brain and nervous system human functions, with a memory capable of establishing mathematical relationships between independent variables to estimate the dependent variables. This work aimed to evaluate the efficiency of eucalypt biomass modeling under different spacings using Multilayer Perceptron networks, trained through the backpropagation algorithm. The experiment was installed in randomized block, and the effect of five planting spacings was studied in three blocks: T1 – 3.0 x 0.5 m; T2 – 3.0 x 1.0 m; T3 – 3.0 x 1.5 m; T4 – 3.0 x 2.0 m e T5 – 3.0 x 3.0 m. A continuous forest inventory was carried out at the ages of 48, 61, 73, 85 and 101 months. The leaf area, leaf perimeter and specific leaf area were measured at 101 months in one sample tree per experimental unit. Two thousand ANN were trained, using all inventoried trees, to estimate the eco-physiological attributes and the prognosis of the wood biomass. The artificial neural networks modeling was adequate to estimate eucalypt wood biomass, according to age and under different spacings, using the diameter-at-breast-height and leaf perimeter as predictor variables.

Download Full-text

Analysis of Transportation Companies in the Czech Republic by the Kohonen Networks – Identification of Industry Leaders

Communications - Scientific letters of the University of Zilina ◽

10.26552/com.c.2021.1.a32-a43 ◽

2020 ◽

Vol 23 (1) ◽

pp. A32-A43

Author(s):

Jakub Horák ◽

Petr Šuleř ◽

Jaromír Vrbka

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Czech Republic ◽

Computational Models ◽

Data File ◽

Transportation Industry ◽

The Czech Republic ◽

Kohonen Networks ◽

Artificial Neural ◽

Accuracy Of Results

Computational models of artificial neural networks are currently used in different areas. Accuracy of results exceeds the performance of traditional statistical techniques. Artificial neural networks as the Kohonen map may be used e.g. to identify industry leaders, thus replacing the traditional cluster analysis and other methods. The aim of this contribution is to analyse the transportation industry in the Czech Republic by the Kohonen networks and identify industry leaders. The data file contains results - division of companies into a total of 100 clusters. Each cluster is subjected to analysis of absolute indicators and several parameters, average, as well as absolute, are examined. In total, 88 firms may be considered as industry leaders. Consequently, a fairly small group of companies has a strong influence on development of the whole transportation industry in the Czech Republic.

Download Full-text

Artificial Neural Networks and Cognitive Modelling

Encyclopedia of Artificial Intelligence ◽

10.4018/978-1-59904-849-9.ch025 ◽

2011 ◽

pp. 161-166

Author(s):

Amanda J.C. Sharkey

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Computational Models ◽

Cognitive Models ◽

Cognitive Modelling ◽

Connectionist Models ◽

Category Formation ◽

Connectionist Approach ◽

Artificial Neural ◽

The Impact

In their heyday, artificial neural networks promised a radically new approach to cognitive modelling. The connectionist approach spawned a number of influential, and controversial, cognitive models. In this article, we consider the main characteristics of the approach, look at the factors leading to its enthusiastic adoption, and discuss the extent to which it differs from earlier computational models. Connectionist cognitive models have made a significant impact on the study of mind. However connectionism is no longer in its prime. Possible reasons for the diminution in its popularity will be identified, together with an attempt to identify its likely future. The rise of connectionist models dates from the publication in 1986 by Rumelhart and McClelland, of an edited work containing a collection of connectionist models of cognition, each trained by exposure to samples of the required tasks. These volumes set the agenda for connectionist cognitive modellers and offered a methodology that subsequently became the standard. Connectionist cognitive models have since been produced in domains including memory retrieval and category formation, and (in language) phoneme recognition, word recognition, speech perception, acquired dyslexia, language acquisition, and (in vision) edge detection, object and shape recognition. More than twenty years later the impact of this work is still apparent.

Download Full-text