Relevant bibliographies by topics / Sigmoid transfer function

Contents

  1. Journal articles
  2. Dissertations / Theses
  3. Book chapters
  4. Conference papers

Academic literature on the topic 'Sigmoid transfer function'

Author: Grafiati
Published: 4 June 2025
Last updated: 15 July 2025

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles

Select a source type:

Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Sigmoid transfer function.'

Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.

You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.

Journal articles on the topic "Sigmoid transfer function"

1

Belete, Biazen Bezabeh, and Debasu Mengistu Abrham. "The effects of multiple layers feed-forward neural network transfer function in digital based ethiopian soil classification and moisture prediction." International Journal of Electrical and Computer Engineering (IJECE) 10, no. 4 (2020): 4073–79. https://doi.org/10.11591/ijece.v10i4.pp4073-4079.

Full text
Abstract:
In the area of machine learning performance analysis is the major task in order to get a better performance both in training and testing model. In addition, performance analysis of machine learning techniques helps to identify how the machine is performing on the given input and also to find any improvements needed to make on the learning model. Feed-forward neural network (FFNN) has different area of applications, but the epoch convergences of the network differs from the usage of transfer function. In this study, to build the model for classification and moisture prediction of soil, rectified linear units (ReLU), Sigmoid, hyperbolic tangent (Tanh) and Gaussian transfer function of feed-forward neural network had been analyzed to identify an appropriate transfer function. Color, texture, shape and brisk local feature descriptor are used as a feature vector of FFNN in the input layer and 4 hidden layers were considered in this study. In each hidden layer 26 neurons are used. From the experiment, Gaussian transfer function outperforms than ReLU, sigmoid and tanh transfer function. But the convergence rate of Gaussian transfer function took more epoch than ReLU, Sigmoid and tanh.
APA, Harvard, Vancouver, ISO, and other styles
2

Khan, M. Imad, Yakov Frayman, and Saeid Nahavandi. "Knowledge Extraction from a Mixed Transfer Function Artificial Neural Network." Journal of Advanced Computational Intelligence and Intelligent Informatics 10, no. 3 (2006): 295–301. http://dx.doi.org/10.20965/jaciii.2006.p0295.

Full text
Abstract:
One of the main problems with Artificial Neural Networks (ANNs) is that their results are not intuitively clear. For example, commonly used hidden neurons with sigmoid activation function can approximate any continuous function, including linear functions, but the coefficients (weights) of this approximation are rather meaningless. To address this problem, current paper presents a novel kind of a neural network that uses transfer functions of various complexities in contrast to mono-transfer functions used in sigmoid and hyperbolic tangent networks. The presence of transfer functions of various complexities in a Mixed Transfer Functions Artificial Neural Network (MTFANN) allow easy conversion of the full model into user-friendly equation format (similar to that of linear regression) without any pruning or simplification of the model. At the same time, MTFANN maintains similar generalization ability to mono-transfer function networks in a global optimization context. The performance and knowledge extraction of MTFANN were evaluated on a realistic simulation of the Puma 560 robot arm and compared to sigmoid, hyperbolic tangent, linear and sinusoidal networks.
APA, Harvard, Vancouver, ISO, and other styles
3

Lady, Silk Moonlight, Riyanto Trilaksono Bambang, Bagus Harianto Bambang, and Faizah Fiqqih. "Implementation of recurrent neural network for the forecasting of USD buy rate against IDR." International Journal of Electrical and Computer Engineering (IJECE) 13, no. 4 (2023): 4567–81. https://doi.org/10.11591/ijece.v13i4.pp4567-4581.

Full text
Abstract:
This study implements a recurrent neural network (RNN) by comparing two RNN network structures, namely Elman and Jordan using the backpropagation through time (BPTT) programming algorithm in the training and forecasting process in foreign exchange forecasting cases. The activation functions used are the linear transfer function, the tan-sigmoid transfer function (Tansig), and the log-sigmoid transfer function (Logsig), which are applied to the hidden and output layers. The application of the activation function results in the log-sigmoid transfer function being the most appropriate activation function for the hidden layer, while the linear transfer function is the most appropriate activation function for the output layer. Based on the results of training and forecasting the USD against IDR currency, the Elman BPTT method is better than the Jordan BPTT method, with the best iteration being the 4000 th iteration for both. The lowest root mean square error (RMSE) values for training and forecasting produced by Elman BPTT were 0.073477 and 122.15 the following day, while the Jordan backpropagation RNN method yielded 0.130317 and 222.96 also the following day.
APA, Harvard, Vancouver, ISO, and other styles
4

Moonlight, Lady Silk, Bambang Riyanto Trilaksono, Bambang Bagus Harianto, and Fiqqih Faizah. "Implementation of recurrent neural network for the forecasting of USD buy rate against IDR." International Journal of Electrical and Computer Engineering (IJECE) 13, no. 4 (2023): 4567. http://dx.doi.org/10.11591/ijece.v13i4.pp4567-4581.

Full text
Abstract:
<span lang="EN-US">This study implements a recurrent neural network (RNN) by comparing two RNN network structures, namely Elman and Jordan using the backpropagation through time (BPTT) programming algorithm in the training and forecasting process in foreign exchange forecasting cases. The activation functions used are the linear transfer function, the tan-sigmoid transfer function (Tansig), and the log-sigmoid transfer function (Logsig), which are applied to the hidden and output layers. The application of the activation function results in the log-sigmoid transfer function being the most appropriate activation function for the hidden layer, while the linear transfer function is the most appropriate activation function for the output layer. Based on the results of training and forecasting the USD against IDR currency, the Elman BPTT method is better than the Jordan BPTT method, with the best iteration being the 4000<sup>th</sup> iteration for both. The lowest root mean square error (RMSE) values for training and forecasting produced by Elman BPTT were 0.073477 and 122.15 the following day, while the Jordan backpropagation RNN method yielded 0.130317 and 222.96 also the following day.</span><p> </p>
APA, Harvard, Vancouver, ISO, and other styles
5

Md, Mainul Islam, Shareef Hussain, Nagrial Mahmood, Rizk Jamal, Hellany Ali, and Nizam Khalid Saiful. "Performance comparison of various probability gate assisted binary lightning search algorithm." International Journal of Artificial Intelligence (IJ-AI) 8, no. 3 (2019): 228–36. https://doi.org/10.11591/ijai.v8.i3.pp228-236.

Full text
Abstract:
Recently, many new nature-inspired optimization algorithms have been introduced to further enhance the computational intelligence optimization algorithms. Among them, lightning search algorithm (LSA) is a recent heuristic optimization method for resolving continuous problems. It mimics the natural phenomenon of lightning to find out the global optimal solution around the search space. In this paper, a suitable technique to formulate binary version of lightning search algorithm (BLSA) is presented. Three common probability transfer functions, namely, logistic sigmoid, tangent hyperbolic sigmoid and quantum bit rotating gate are investigated to be utilized in the original LSA. The performances of three transfer functions based BLSA is evaluated using various standard functions with different features and the results are compared with other four famous heuristic optimization techniques. The comparative study clearly reveals that tangent hyperbolic transfer function is the most suitable function that can be utilized in the binary version of LSA.
APA, Harvard, Vancouver, ISO, and other styles
6

Pignon, D., M. van Daalen, J. Shawe-Taylor, et al. "Sigmoid neural transfer function realized by percolation." Optics Letters 21, no. 3 (1996): 222. http://dx.doi.org/10.1364/ol.21.000222.

Full text
APA, Harvard, Vancouver, ISO, and other styles
7

Jabrayili, Sharokh, Vahid Farzaneh, Zahra Zare, et al. "Modelling of mass transfer kinetic in osmotic dehydration of kiwifruit." International Agrophysics 30, no. 2 (2016): 185–91. http://dx.doi.org/10.1515/intag-2015-0091.

Full text
Abstract:
Abstract Osmotic dehydration characteristics of kiwifruit were predicted by different activation functions of an artificial neural network. Osmotic solution concentration (y1), osmotic solution temperature (y2), and immersion time (y3) were considered as the input parameters and solid gain value (x1) and water loss value (x2) were selected as the outlet parameters of the network. The result showed that logarithm sigmoid activation function has greater performance than tangent hyperbolic activation function for the prediction of osmotic dehydration parameters of kiwifruit. The minimum mean relative error for the solid gain and water loss parameters with one hidden layer and 19 nods were 0.00574 and 0.0062% for logarithm sigmoid activation function, respectively, which introduced logarithm sigmoid function as a more appropriate tool in the prediction of the osmotic dehydration of kiwifruit slices. As a result, it is concluded that this network is capable in the prediction of solid gain and water loss parameters (responses) with the correlation coefficient values of 0.986 and 0.989, respectively.
APA, Harvard, Vancouver, ISO, and other styles
8

Bezabeh, Belete Biazen, and Abrham Debasu Mengistu. "The effects of multiple layers feed-forward neural network transfer function in digital based Ethiopian soil classification and moisture prediction." International Journal of Electrical and Computer Engineering (IJECE) 10, no. 4 (2020): 4073. http://dx.doi.org/10.11591/ijece.v10i4.pp4073-4079.

Full text
Abstract:
In the area of machine learning performance analysis is the major task in order to get a better performance both in training and testing model. In addition, performance analysis of machine learning techniques helps to identify how the machine is performing on the given input and also to find any improvements needed to make on the learning model. Feed-forward neural network (FFNN) has different area of applications, but the epoch convergences of the network differs from the usage of transfer function. In this study, to build the model for classification and moisture prediction of soil, rectified linear units (ReLU), Sigmoid, hyperbolic tangent (Tanh) and Gaussian transfer function of feed-forward neural network had been analyzed to identify an appropriate transfer function. Color, texture, shape and brisk local feature descriptor are used as a feature vector of FFNN in the input layer and 4 hidden layers were considered in this study. In each hidden layer 26 neurons are used. From the experiment, Gaussian transfer function outperforms than ReLU, sigmoid and tanh transfer function. But the convergence rate of Gaussian transfer function took more epoch than ReLU, Sigmoid and tanh.
APA, Harvard, Vancouver, ISO, and other styles
9

Islam, Md Mainul, Hussain Shareef, Mahmood Nagrial, Jamal Rizk, Ali Hellany, and Saiful Nizam Khalid. "Performance comparison of various probability gate assisted binary lightning search algorithm." IAES International Journal of Artificial Intelligence (IJ-AI) 8, no. 3 (2019): 299. http://dx.doi.org/10.11591/ijai.v8.i3.pp299-306.

Full text
Abstract:
<div style="’text-align: justify;">Recently, many new nature-inspired optimization algorithms have been introduced to further enhance the computational intelligence optimization algorithms. Among them, lightning search algorithm(LSA) is a recent heuristic optimization method for resolving continuous problems. It mimics the natural phenomenon of lightning to find out the global optimal solution around the search space. In this paper, a suitable technique to formulate binary version of lightning search algorithm(BLSA) is presented. Three common probability transfer functions, namely, logistic sigmoid, tangent hyperbolic sigmoid and quantum bit rotating gate are investigated to be utilized in the original LSA. The performances of three transfer functions based BLSA is evaluated using various standard functions with different features and the results are compared with other four famous heuristic optimization techniques. The comparative study clearly reveals that tangent hyperbolic transfer function is the most suitable function that can be utilized in the binary version of LSA.</div>
APA, Harvard, Vancouver, ISO, and other styles
10

Srinivas, Kankanala, and Ashish Kumar Bhandari. "Low light image enhancement with adaptive sigmoid transfer function." IET Image Processing 14, no. 4 (2020): 668–78. http://dx.doi.org/10.1049/iet-ipr.2019.0781.

Full text
APA, Harvard, Vancouver, ISO, and other styles
More sources

Dissertations / Theses on the topic "Sigmoid transfer function"

1

Brown, Marvin Lane. "The Impact of Data Imputation Methodologies on Knowledge Discovery." Cleveland State University / OhioLINK, 2008. http://rave.ohiolink.edu/etdc/view?acc_num=csu1227054769.

Full text
APA, Harvard, Vancouver, ISO, and other styles

Book chapters on the topic "Sigmoid transfer function"

1

Yusof, Norfadzlia Mohd, Azah Kamilah Muda, Satrya Fajri Pratama, Ramon Carbo-Dorca, and Ajith Abraham. "Binary Whale Optimization Algorithm with Logarithmic Decreasing Time-Varying Modified Sigmoid Transfer Function for Descriptor Selection Problem." In Lecture Notes in Networks and Systems. Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-27524-1_65.

Full text
APA, Harvard, Vancouver, ISO, and other styles
2

Amitab, Khwairakpam, Debdatta Kandar, and Arnab K. Maji. "Speckle Noise Filtering Using Back-Propagation Multi-Layer Perceptron Network in Synthetic Aperture Radar Image." In Research Advances in the Integration of Big Data and Smart Computing. IGI Global, 2016. http://dx.doi.org/10.4018/978-1-4666-8737-0.ch016.

Full text
Abstract:
Synthetic Aperture Radar (SAR) are imaging Radar, it uses electromagnetic radiation to illuminate the scanned surface and produce high resolution images in all-weather condition, day and night. Interference of signals causes noise and degrades the quality of the image, it causes serious difficulty in analyzing the images. Speckle is multiplicative noise that inherently exist in SAR images. Artificial Neural Network (ANN) have the capability of learning and is gaining popularity in SAR image processing. Multi-Layer Perceptron (MLP) is a feed forward artificial neural network model that consists of an input layer, several hidden layers, and an output layer. We have simulated MLP with two hidden layer in Matlab. Speckle noises were added to the target SAR image and applied MLP for speckle noise reduction. It is found that speckle noise in SAR images can be reduced by using MLP. We have considered Log-sigmoid, Tan-Sigmoid and Linear Transfer Function for the hidden layers. The MLP network are trained using Gradient descent with momentum back propagation, Resilient back propagation and Levenberg-Marquardt back propagation and comparatively evaluated the performance.
APA, Harvard, Vancouver, ISO, and other styles
3

Amitab, Khwairakpam, Debdatta Kandar, and Arnab K. Maji. "Speckle Noise Filtering Using Back-Propagation Multi-Layer Perceptron Network in Synthetic Aperture Radar Image." In Deep Learning and Neural Networks. IGI Global, 2020. http://dx.doi.org/10.4018/978-1-7998-0414-7.ch028.

Full text
Abstract:
Synthetic Aperture Radar (SAR) are imaging Radar, it uses electromagnetic radiation to illuminate the scanned surface and produce high resolution images in all-weather condition, day and night. Interference of signals causes noise and degrades the quality of the image, it causes serious difficulty in analyzing the images. Speckle is multiplicative noise that inherently exist in SAR images. Artificial Neural Network (ANN) have the capability of learning and is gaining popularity in SAR image processing. Multi-Layer Perceptron (MLP) is a feed forward artificial neural network model that consists of an input layer, several hidden layers, and an output layer. We have simulated MLP with two hidden layer in Matlab. Speckle noises were added to the target SAR image and applied MLP for speckle noise reduction. It is found that speckle noise in SAR images can be reduced by using MLP. We have considered Log-sigmoid, Tan-Sigmoid and Linear Transfer Function for the hidden layers. The MLP network are trained using Gradient descent with momentum back propagation, Resilient back propagation and Levenberg-Marquardt back propagation and comparatively evaluated the performance.
APA, Harvard, Vancouver, ISO, and other styles
4

Mishra, Prakash Chandra, and Anil Kumar Giri. "Prediction of Biosorption Capacity Using Artificial Neural Network Modeling and Genetic Algorithm." In Handbook of Research on Manufacturing Process Modeling and Optimization Strategies. IGI Global, 2017. http://dx.doi.org/10.4018/978-1-5225-2440-3.ch013.

Full text
Abstract:
Artificial neural network model is applied for the prediction of the biosorption capacity of living cells of Bacillus cereus for the removal of chromium (VI) ions from aqueous solution. The maximum biosorption capacity of living cells of Bacillus cereus for chromium (VI) was found to be 89.24% at pH 7.5, equilibrium time of 60 min, biomass dosage of 6 g/L, and temperature of 30 ± 2 °C. The biosorption data of chromium (VI) ions collected from laboratory scale experimental set up is used to train a back propagation (BP) learning algorithm having 4-7-1 architecture. The model uses tangent sigmoid transfer function at input to hidden layer whereas a linear transfer function is used at output layer. The data is divided into training (75%) and testing (25%) sets. Comparison between the model results and experimental data gives a high degree of correlation R2 = 0.984 indicating that the model is able to predict the sorption efficiency with reasonable accuracy. Bacillus cereus biomass is characterized using AFM and FTIR.
APA, Harvard, Vancouver, ISO, and other styles
5

Mishra, Prakash Chandra, and Anil Kumar Giri. "Prediction of Biosorption Capacity Using Artificial Neural Network Modeling and Genetic Algorithm." In Deep Learning and Neural Networks. IGI Global, 2020. http://dx.doi.org/10.4018/978-1-7998-0414-7.ch010.

Full text
Abstract:
Artificial neural network model is applied for the prediction of the biosorption capacity of living cells of Bacillus cereus for the removal of chromium (VI) ions from aqueous solution. The maximum biosorption capacity of living cells of Bacillus cereus for chromium (VI) was found to be 89.24% at pH 7.5, equilibrium time of 60 min, biomass dosage of 6 g/L, and temperature of 30 ± 2 °C. The biosorption data of chromium (VI) ions collected from laboratory scale experimental set up is used to train a back propagation (BP) learning algorithm having 4-7-1 architecture. The model uses tangent sigmoid transfer function at input to hidden layer whereas a linear transfer function is used at output layer. The data is divided into training (75%) and testing (25%) sets. Comparison between the model results and experimental data gives a high degree of correlation R2 = 0.984 indicating that the model is able to predict the sorption efficiency with reasonable accuracy. Bacillus cereus biomass is characterized using AFM and FTIR.
APA, Harvard, Vancouver, ISO, and other styles
6

Punys Vytenis and Maknickas Ramunas. "Cell Edge Detection in JPEG2000 Wavelet Domain – Analysis on Sigmoid Function Edge Model." In Studies in Health Technology and Informatics. IOS Press, 2011. https://doi.org/10.3233/978-1-60750-806-9-470.

Full text
Abstract:
Big virtual microscopy images (80K x 60K pixels and larger) are usually stored using the JPEG2000 image compression scheme. Diagnostic quantification, based on image analysis, might be faster if performed on compressed data (approx. 20 times less the original amount), representing the coefficients of the wavelet transform. The analysis of possible edge detection without reverse wavelet transform is presented in the paper. Two edge detection methods, suitable for JPEG2000 bi-orthogonal wavelets, are proposed. The methods are adjusted according calculated parameters of sigmoid edge model. The results of model analysis indicate more suitable method for given bi-orthogonal wavelet.
APA, Harvard, Vancouver, ISO, and other styles
7

El Hamzaoui, Y., N. Pérez Morga, B. F. Tejero Campos, M. S. Ramos Ocampo, M. López Oliveros, and J. A. de la Cruz Koyoc. "Predicción de la Demanda Química de Oxígeno Durante el Tratamiento de Herbicidas Comerciales por Redes Neuronales Artificiales." In Tópicos Selectos de Contaminación Ambiental. EPOMEX-UAC, 2025. https://doi.org/10.26359/epomex01202506.

Full text
Abstract:
A forward and inverse artificial neural network (rna and rnai) approach was developed to predict chemical oxygen demand (cod) removal during the degradation of commercial herbicides alazine and gesaprim under various experimental conditions. The network had a 9-9-1 configuration (9 inputs, 9 hidden neurons, and 1 output neuron) and showed a high level of agreement (R2 = 0.9913) between the experimental and simulated cod values. This was achieved using hyperbolic and linear tangent sigmoid transfer functions in the hidden and output layers. The sensitivity analysis revealed that all the input variables studied (reaction time, pH, herbicide concentration, contaminant, US ultrasound, UV light intensity, [TiO2]o,[K2S2O8]o, solar radiation sr) strongly influence the degradation of the commercial herbicide in terms of cod removal. Among these variables, the reaction time was found to be the most influential, with a relative importance of 33.49%, followed by the initial herbicide concentration. The inverse neural network (rnai) was then used to calculate the optimal reaction time needed to achieve the desired cod removal. This makes the methodology attractive for online control of the Advanced Oxidation Process (aop) for the degradation of commercial herbicides due to its low error percentage and efficient calculation. Keywords: Artificial neural networks, chemical oxygen demand, sono-photocatalysis, sensitivity analysis.
APA, Harvard, Vancouver, ISO, and other styles

Conference papers on the topic "Sigmoid transfer function"

1

Colorado, D., S. Serna-Barquera, J. A. Hernández, Y. Barrera-Rojas, M. Lucio-García, and B. Campillo. "Neural Network for Dispersion Strengthened Microalloyed Steel Sour Corrosion from Electrochemical Impedance Spectroscopy Laboratory Measurements." In CORROSION 2010. NACE International, 2010. https://doi.org/10.5006/c2010-10279.

Full text
Abstract:
Abstract Microalloyed pipeline steels mechanical resistance can be improved by dispersion strengthening. The enhancement of steel dispersion strengthening by tempering at a suitable temperature has been studied at various holding times at 3, 6, 8 and 10 hours. Depending on the elapsed time, microalloying elements that were still located within steel iron lattice can be re-diffused, thus developing different nanoparticle sizes, densities and distribution. The steel yield strength and sulphide stress cracking resistance were significantly improved under sour environment. A systematic electrochemical impedance spectroscopy (EIS) corrosion study was carried out. The objective of the present work was to predict corrosion results from EIS collected data from the different steel tempering times and exposure temperatures to sour environment (room temperature and 50 °C) by means of an artificial neural network (ANN). For the ANN, an approach based on Levenberg–Marquardt learning algorithm, hyperbolic tangent sigmoid transfer function, and a linear transfer function was used. The model takes into account of the variations of the real impedance, time and steel exposure temperature. The developed model can be used for prediction at short simulation times illustrating the utility of the ANN. On the validation data set, the simulations and the theoretical data tests were in good agreement with R2 > 0.98 for all experimental databases. These results suggest that ANN may play a key role in making lifetime predictions for components based on laboratory measurements.
APA, Harvard, Vancouver, ISO, and other styles
2

Liu, Fucai, Shu'en Wang, and Jinmei Dou. "An improved fuzzy identification method based on Sigmoid data transfer function." In 2012 10th World Congress on Intelligent Control and Automation (WCICA 2012). IEEE, 2012. http://dx.doi.org/10.1109/wcica.2012.6358385.

Full text
APA, Harvard, Vancouver, ISO, and other styles
3

Gang Yang, Yunfeng Zhou, and Shixiang Qian. "Using hyperbolic tangent sigmoid transfer function for companding transform in OFDM system." In 2007 International Symposium on Communications and Information Technologies. IEEE, 2007. http://dx.doi.org/10.1109/iscit.2007.4391990.

Full text
APA, Harvard, Vancouver, ISO, and other styles
4

Pacheco-Vega, Arturo, Mihir Sen, K. T. Yang, and Rodney L. McClain. "Prediction of Humid Air Heat Exchanger Performance Using Artificial Neural Networks." In ASME 1999 International Mechanical Engineering Congress and Exposition. American Society of Mechanical Engineers, 1999. http://dx.doi.org/10.1115/imece1999-1087.

Full text
Abstract:
Abstract In the present study we apply an artificial neural network to predict the operation of a humid air-water fin-tube compact heat exchanger. The network configuration is of the feedforward type with a sigmoid activation function and a backpropagation algorithm. Published experimental data, corresponding to humid air flowing over the heat exchanger tubes and water flowing inside them, are used to train the neural network. After training with known experimental values of the humid-air flow rates, dry-bulb and wet-bulb inlet temperatures for various geometrical configurations, the j-factor and heat transfer rate predictions of the network were tested against the experimental values. Comparisons were made with published predictions of power-law correlations which were obtained from the same data. The results demonstrate that the neural network is able to predict the performance of this heat exchanger much better than the correlations.
APA, Harvard, Vancouver, ISO, and other styles
5

Ghassemi, Payam, Kaige Zhu, and Souma Chowdhury. "Optimal Surrogate and Neural Network Modeling for Day-Ahead Forecasting of the Hourly Energy Consumption of University Buildings." In ASME 2017 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. American Society of Mechanical Engineers, 2017. http://dx.doi.org/10.1115/detc2017-68350.

Full text
Abstract:
This paper presents the development and evaluation of Artificial Neural Networks (ANN) based models and optimally selected surrogate models to provide the day-ahead forecast of the hourly-averaged energy load of buildings, by relating it to eight weather parameters as well as the hour of the day. Although ANN and other surrogate models have been used to predict building energy loads in the past, there is a limited understanding of what type of model prescriptions impact their performance as well as how un-recorded impact factors (e.g., human behavior and building repair work) should be accounted for. Here, the recorded energy data of three university buildings, from 9/2013–12/2015, is cleaned and synchronized with the local weather data. The data is then classified into eight classes; weekends and weekdays of Fall/Winter/Spring/Summer semesters. Both Multi-Layer Perceptron (MLP) and Radial Basis Function (RBF) NNs are explored. Differing number of hidden layers and transfer function choices are also explored, leading to the choice of the hyperbolic-tangent-sigmoid transfer function and 60 hidden layers. Similarly, an automated surrogate modeling framework is used to select the best models from among a pool of Kriging, RBF, and SVR models. A baseline concept, that uses energy information from the previous day as an added input to the ANN, helps to account for otherwise unrecorded recent changes in the building behavior, leading to improvement in fidelity of up to 30%.
APA, Harvard, Vancouver, ISO, and other styles
6

A´lvarez del Castillo, A., E. Santoyo, and O. Garci´a-Valladares. "Development of a New Void Fraction Correlation for Modeling Two-Phase Flow in Producing Geothermal Wells Using Artificial Neural Networks." In ASME 2010 International Mechanical Engineering Congress and Exposition. ASMEDC, 2010. http://dx.doi.org/10.1115/imece2010-40444.

Full text
Abstract:
An artificial neural network (ANN) was used to develop a new empirical correlation to estimate void fractions for modeling two-phase flows in geothermal wells. Flowing pressure, wellbore diameter, steam quality, fluid density and viscosity, and Reynolds numbers were used as input data. An explicit relationship among the input data was obtained from an ANN model. A computational architecture based on, the Levenberg-Marquardt optimization algorithm, the hyperbolic tangent sigmoid transfer-function, and the linear transfer-function, was designed. A geothermal database containing thirty-two data sets logged in production well tests were used both to train and to validate the ANN. The best training results were obtained for an ANN architecture of five neurons in the hidden layer, which made possible to predict void fractions with a satisfactory efficiency (R2 = 0.992). From this ANN training pattern, a new empirical correlation was developed and coupled into a wellbore simulator for modeling two-phase flows in other geothermal wells (to avoid bias). Four well-known engineering correlations for calculating the void fraction were simultaneous evaluated. The simulated results (obtained with the five void fraction correlations) were statistically compared with measured field data. A better agreement between simulated and field data was systematically obtained for the new ANN correlation with matching errors less than 3%. These results suggest that the new empirical correlation can be reliable used to estimate void fractions in two-phase geothermal wellbores.
APA, Harvard, Vancouver, ISO, and other styles
7

Pacheco-Vega, Arturo, Mihir Sen, and Rodney L. McClain. "Analysis of Fin-Tube Evaporator Performance With Limited Experimental Data Using Artificial Neural Networks." In ASME 2000 International Mechanical Engineering Congress and Exposition. American Society of Mechanical Engineers, 2000. http://dx.doi.org/10.1115/imece2000-1466.

Full text
Abstract:
Abstract In the current study we consider the problem of accuracy in heat rate estimations from artificial neural network models of heat exchangers used for refrigeration applications. The network configuration is of the feedforward type with a sigmoid activation function and a backpropagation algorithm. Limited experimental measurements from a manufacturer are used to show the capability of the neural network technique in modeling the heat transfer in these systems. Results from this exercise show that a well-trained network correlates the data with errors of the same order as the uncertainty of the measurements. It is also shown that the number and distribution of the training data are linked to the performance of the network when estimating the heat rates under different operating conditions, and that networks trained from few tests may give large errors. A methodology based on the cross-validation technique is presented to find regions where not enough data are available to construct a reliable neural network. The results from three tests show that the proposed methodology gives an upper bound of the estimated error in the heat rates.
APA, Harvard, Vancouver, ISO, and other styles
8

Gao, Zheng-Ming, Juan Zhao, and Su-Ruo Li. "The Binary Equilibrium Optimization Algorithm with Sigmoid Transfer Functions." In ICMLC 2020: 2020 12th International Conference on Machine Learning and Computing. ACM, 2020. http://dx.doi.org/10.1145/3383972.3384064.

Full text
APA, Harvard, Vancouver, ISO, and other styles
9

Mourgias-Alexandris, George, Apostolos Tsakyridis, Theonitsa Alexoudi, Konstantinos Vyrsokinos, and Nikos Pleros. "Optical Thresholding Device with a Sigmoidal Transfer Function." In 2018 Photonics in Switching and Computing (PSC). IEEE, 2018. http://dx.doi.org/10.1109/ps.2018.8751393.

Full text
APA, Harvard, Vancouver, ISO, and other styles
10

Tripathy, Srinibas, Sridhar Sahoo, and Dhananjay Kumar Srivastava. "Development of an Artificial Neural Network Model for the Performance Prediction of a Variable Compression Ratio Gasoline Spark Ignition Engine." In ASME 2019 Internal Combustion Engine Division Fall Technical Conference. American Society of Mechanical Engineers, 2019. http://dx.doi.org/10.1115/icef2019-7118.

Full text
Abstract:
Abstract The purpose of this study is to develop an artificial neural network (ANN) model for performance prediction of a variable compression ratio gasoline port fuel injection spark ignition engine. For ANN modeling, a large experimental data set was generated in which at random 85% was assigned for training the network, and 15% that are not included during the training process was used for testing the network. A multilayer perception feed forward neural network was used to predict the correlation between input and output layer. The input layer consists of engine speed, throttle position, spark timing, and compression ratio. Whereas, the output layer consists of torque, brake power and indicated mean effective pressure (IMEP). Neurons in the hidden layer were varied and optimized based on a specified goal error. A standard supervised back propagation learning algorithm was used in which the error between the target and network output was calculated and minimized. In the hidden and output layers, a non-linear tan-sigmoid and a linear transfer function were used, respectively, for input-output mapping. The performance of the network was evaluated by statistical parameters like correlation coefficient (R), mean relative error (MRE) and root mean square error (RMSE). It was found from the test data that the R and MRE values are lies in between 0.99853 to 0.99875 and 0.42% to 0.58%, respectively. Whereas, RMSE value for all performance parameters was found to be very low. Hence, this study reveals that the application of ANN modeling has the ability to predict the performance of a variable compression ratio gasoline engine and is the best alternative tool over all classical modeling techniques.
APA, Harvard, Vancouver, ISO, and other styles
You might also be interested in the extended bibliographies on the topic 'Sigmoid transfer function' for particular source types:
Journal articles
We offer discounts on all premium plans for authors whose works are included in thematic literature selections. Contact us to get a unique promo code!

To the bibliography