A novel rule-based evolving Fuzzy System applied to the thermal modeling of power transformers

Alves, Kaike Sa Teles Rocha; Aguiar, Eduardo Pestana de

doi:10.1016/j.asoc.2021.107764

Cited by 22 publications

(10 citation statements)

References 106 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 1 depicts the graphic of the predictions of the proposed model. [18] 0.0005543 0.0000677 0.0004575 --eTS [4] 0.0002608 0.0000319 0.0002108 4 12.65 ± 0.66 Simpl eTS [10] 0.0001561 0.0000191 0.0001103 1 5.75 ± 0.18 exTS [11] 0.0001467 0.0000179 0.0000304 7 23.01 ± 0.45 eMG [14] 0.5598119 0.0683843 0.3872295 56 225.88 ± 1.72 ePL+ [13] 0.00000156 0.0000019 0.0000110 1 9.52 ± 0.33 ePL-KRLS-DISCO [15] 0.0294889 0.0036022 0.0191308…”

Section: A Lorenz Attractor Time Seriesmentioning

confidence: 99%

“…The eTS approach is an evolving Fuzzy System(eFS), a class with its parameters and structure continually updated as new input/output pair enters the system, creating and excluding rules when necessary. The eTS algorithm inspired the proposal of other rule-based eFS approaches, such as Simpl eTS [10], exTS [11], ePL [12], ePL+ [13], eMG [14], ePL-KRLS-DISCO [15], FLEXFIS [16].…”

Section: Introductionmentioning

confidence: 99%

“…It implies that the input vectors will be close or similar but doesn't guarantee that the XVI Brazilian Conference on Computational Intelligence (CBIC 2023), Salvador, October 8th to 11th rules will have outputs with linear curves. Furthermore, the models usually have many hyperparameters [15].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A New Fuzzy Inference System Applied to Time Series Forecasting

Alves,

Ballini,

Aguiar

2023

Anais Do XVI Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

Fuzzy systems are a class of machine learning introduced by Zadeh that combine accuracy and interpretability. This class of models consists of two main parts, the antecedent and the consequent. While the antecedent is responsible for modeling the inputs, the consequent concerns modeling the output. The literature reports two main types of fuzzy systems: Mamadani and Takagi-Sugeno. While Mamdani uses fuzzy sets in the consequent part, Takagi-Sugeno uses polynomial functions. Consequently, Mamdani provides better understandable models and Takagi-Sugeno more accurate ones. In this paper, we propose a new Takagi-Sugeno model. Still, instead of defining the rules based on the input, the proposed model designs the rules based on the output variation to capture linearities in the output and clusters them in the same rule. The model is applied in the regression problems of benchmark series and real datasets of power transformers. The performance of the proposed model is compared with the performance of classical models and evolving Fuzzy Systems. The results are evaluated using error metrics, the number of final rules, and runtime.

show abstract

Section: A Lorenz Attractor Time Seriesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A New Fuzzy Inference System Applied to Time Series Forecasting

Alves,

Ballini,

Aguiar

2023

Anais Do XVI Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

show abstract

“…A zero-order TSK fuzzy regression model is made up of a collection of 𝑅 fuzzy rules based upon a set of input features 𝐅 = (𝐹 1 , … , 𝐹 𝒏 ′ ) 𝑇 ∈ ℝ 𝑛 ′ ×1 , the 𝑟th of which is as follows [15]: (7) in which the consequence of fuzzy rule is 𝑦 𝑟 , 𝐴 𝑗 is the antecedent fuzzy set on the 𝑗th input feature. The widely used Gaussian functions are employed as fuzzy membership functions in this work because their derivatives are easier to compute [16].…”

Section: Tsk Fuzzy Systemmentioning

confidence: 99%

Cutting down high dimensional data with Fuzzy weighted forests (FWF)

Wang

Gault

Greer

2022

2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

View full text Add to dashboard Cite

Takagi-Sugeno-Kang (TSK) rule-based fuzzy systems struggle to deal with high dimensional data and suffer from the curse of dimensionality. As the number of input features increases, the number of rules increases exponentially, which reduces the model's interpretability rapidly. This paper presents a novel fuzzy weighted forest aggregation method to effectively model high dimensional data by reducing the number of fuzzy rules, without sacrificing accuracy. The fuzzy weighted forest is comprised of several fuzzy weighted trees. Each tree is created based on a subset of features captured across different parts of the input space. Given 𝒏 input features and 𝑵 samples, every fuzzy tree is randomly assigned 𝒏 ′ features and 𝑵 ′ samples, where 𝒏 ′ and 𝑵 ′ are significantly smaller than 𝒏 and 𝑵 respectively. Each path within a tree, from root to leaf, forms a fuzzy rule. The non-leaf nodes represent the antecedents of rules, and the leaf node represents the consequents. This study shows how the proposed method utilizes pruning to significantly reduce the number of fuzzy rules. This method therefore creates a less complex model whilst achieving high accuracy comparable with, and sometimes better than, existing state-ofthe-art TSK-based fuzzy models.

show abstract

“…Hot spot temperature refers to the temperature at the transformer's hottest point, representing the most severe thermal ageing of the transformer insulation. Therefore, monitoring and computing HST is crucial for determining transformer conditions and life assessment [6][7][8]. In the preset service environment, the temperature rise of the winding must comply with the relevant standards and must not exceed the specified limits [9].…”

mentioning

confidence: 99%

A fast calculation method of transient hot spot temperature for the design of dry‐type on‐board transformers

Guo

Cai

Zhu

et al. 2022

IET Electric Power Appl

View full text Add to dashboard Cite

The design of the dry-type on-board traction transformer (OBTT), a new lightweight method in the traction system of electric multiple units, needs to consider the transient hot spot temperature (HST) under different railroad line operating conditions. In order to quickly calculate the transient HST of dry-type OBTT when operating under different railroad line conditions, a calculation method (hereinafter referred to as GP-HST method) by combining an enhanced genetic programming algorithm (GP) with computational fluid dynamics (CFD) is presented. First, intercept a section from the railroad line to be calculated and use the CFD method to calculate the transient HST of the section railroad line. Then, the enhanced GP modelling is driven by this section of railroad line measured and simulated data. Next, a calculation model with an implicit mathematical correlation between railroad line operating conditions and the dry-type OBTT transient HST is established. Finally, the calculation model quickly calculates the transient HST of the remaining railroad line in operation. To validate the presented method, the calculation model obtained using the presented method is used to calculate the transient HST for another railroad line with significantly different operating conditions, and it is compared with the direct use of the CFD method. The calculation results show that the GP-HST method reduces the simulation time of the CFD method from 950 to 0.25 min with a maximum error, mean absolute error, and mean relative error of 1.856 K, 0.778 K, and 0.22%, respectively, for a railroad line with a running time of about half an hour. This shows that the GP-HST method can effectively help to improve the design efficiency of dry-type OBTT. K E Y W O R D Scalculation model, genetic programing (GP), hot-spot temperature (HST), on-board traction transformer (OBTT) | INTRODUCTIONElectric multiple units (EMU) are the high-speed train of electric power supply. The heart of its power supply systemthe on-board traction transformer (OBTT), is responsible for stepping down the voltage of 25 kV in the supply network to the rated operating voltage of the equipment. In addition, OBTT is the heaviest single piece of equipment in the power supply system, accounting for about 12%-18% of the mass of vehicle weight [1][2][3]. Currently, all OBTTs in EMUs are of the oil-cooled type. The insulating oil required for a single oilimmersed OBTT alone is about 1.5 tons, about 30% of the total mass of OBTT. The excessive mass of the OBTT will not only limit train operating speed but also increase the operating energy consumption. Dry-type OBTT is currently the primary lightweight method of OBTT, which changes the traditional oilThis is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

show abstract

A novel rule-based evolving Fuzzy System applied to the thermal modeling of power transformers

Cited by 22 publications

References 106 publications

A New Fuzzy Inference System Applied to Time Series Forecasting

A New Fuzzy Inference System Applied to Time Series Forecasting

Cutting down high dimensional data with Fuzzy weighted forests (FWF)

A fast calculation method of transient hot spot temperature for the design of dry‐type on‐board transformers

Contact Info

Product

Resources

About