Participatory Learning in Power Transformers Thermal Modeling

Hell, Michel; Costa, Paula Cristina Pereira da; Gomide, Fernando

doi:10.1109/tpwrd.2008.923994

Cited by 35 publications

(24 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Figure 3 shows the structure of BP Network for prediction of hot winding resistance. Using a given input/output data set, the ANFIS constructs a Fuzzy Inference System (FIS) whose membership function parameters are adjusted using a back propagation algorithm method (Pylvanainen et al, 2007;Hell et al, 2008). This allows fuzzy systems to learn from the data they are modeling.…”

Section: Soft Computing Techniquesmentioning

confidence: 99%

Hot Resistance Estimation for Dry Type Transformer Using Multiple Variable Regression, Multiple Polynomial Regression and Soft Computing Techniques

M.¹

2012

American Journal of Applied Sciences

View full text Add to dashboard Cite

Problem statement:This study presents a novel method for the determination of average winding temperature rise of transformers under its predetermined field operating conditions. Rise in the winding temperature was determined from the estimated values of winding resistance during the heat run test conducted as per IEC standard. Approach: The estimation of hot resistance was modeled using Multiple Variable Regression (MVR), Multiple Polynomial Regression (MPR) and soft computing techniques such as Artificial Neural Network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS). The modeled hot resistance will help to find the load losses at any load situation without using complicated measurement set up in transformers. Results: These techniques were applied for the hot resistance estimation for dry type transformer by using the input variables cold resistance, ambient temperature and temperature rise. The results are compared and they show a good agreement between measured and computed values. Conclusion: According to our experiments, the proposed methods are verified using experimental results, which have been obtained from temperature rise test performed on a 55 kVA dry-type transformer.

show abstract

Section: Soft Computing Techniquesmentioning

confidence: 99%

Hot Resistance Estimation for Dry Type Transformer Using Multiple Variable Regression, Multiple Polynomial Regression and Soft Computing Techniques

M.¹

2012

American Journal of Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Furthermore, these models have a neural network topology which enables the utilization of a large variety of existing machine learning algorithms for structure identification and parameter estimation. Fuzzy neural networks have already been used to solve several distinct problems including pattern classification [10] and [60], time series prediction [4] [27] [8] and dynamic system modeling [26] [19] [41] [42]. Examples of fuzzy neurons are and and or neurons [54].…”

Section: Introductionmentioning

confidence: 99%

“…This step involves defining fuzzy sets for each input variable, selecting a suitable number of neurons and defining network connections. The most commonly used methods for structure definition are clustering [10], [4], [8], [26], [41], [42] and evolutionary optimization [51] [52] [43]. Once the network structure is defined, free parameters are estimated.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Neural Networks based on Fuzzy Logic Neurons Regularized by Resampling Techniques and Regularization Theory for Regression Problems

Souza

Guimarães²,

Araújo³

et al. 2018

INTARTIF

View full text Add to dashboard Cite

This paper presents a novel learning algorithm for fuzzy logic neuron based on neural networks and fuzzy systems able to generate accurate and transparent models. The learning algorithm is based on ideas from Extreme Learning Machine [36], to achieve a low time complexity, and regularization theory, resulting in sparse and accurate models. A compact set of incomplete fuzzy rules can be extracted from the resulting network topology. Experiments considering regression problems are detailed. Results suggest the proposed approach as a promising alternative for pattern recognition with a good accuracy and some level of interpretability.

show abstract

“…The main feature of these networks is its transparency, enabling the utilization of a priori information to define initial network topology and the extraction of valuable information from the resulting topology after training in the form of a set of fuzzy rules [1]. Fuzzy neural networks based on logic neurons, called and and or [2], have been used to solve a large variety of problems, such as pattern classification [1], time series prediction [3] and dynamic system modelling [4], [5]. These type of logic based neurons are nonlinear mappings of the form [0, 1] n → [0, 1] whose standard implementation of fuzzy connectives involves triangular norms.…”

Section: Introductionmentioning

confidence: 99%

“…Some of most recent trend in the neurofuzzy systems area involves the development of new operators (logic connectives) [4], [6]- [12]. Apart from traditional operators several modifications and extensions have been addressed in literature.…”

Section: Introductionmentioning

confidence: 99%

Universal approximation with uninorm-based fuzzy neural networks

Lemos

Kreinovich

Caminhas

et al. 2011

2011 Annual Meeting of the North American Fuzzy Information Processing Society

View full text Add to dashboard Cite

Abstract-Fuzzy neural networks are hybrid models capable to approximate functions with high precision and to generate transparent models, enabling the extraction of valuable information from the resulting topology. In this paper we will show that the recently proposed fuzzy neural network based on weighted uninorms aggregations uniformly approximates any real functions on any compact set. We will describe the network topology and inference mechanism and show that the universal approximation property of this network is valid for a given choice of operators.

show abstract

Participatory Learning in Power Transformers Thermal Modeling

Cited by 35 publications

References 13 publications

Hot Resistance Estimation for Dry Type Transformer Using Multiple Variable Regression, Multiple Polynomial Regression and Soft Computing Techniques

Hot Resistance Estimation for Dry Type Transformer Using Multiple Variable Regression, Multiple Polynomial Regression and Soft Computing Techniques

Fuzzy Neural Networks based on Fuzzy Logic Neurons Regularized by Resampling Techniques and Regularization Theory for Regression Problems

Universal approximation with uninorm-based fuzzy neural networks

Contact Info

Product

Resources

About