Partial discharge localization on dry-type transformers

Szczechowski, Janusz; Siodła, Krzysztof

doi:10.1109/ichve.2014.7035432

Cited by 1 publication

(4 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unfortunately, the presented model allows only to foresee the temperature of a single point of the transformer. However, when performing a test on the transformer, for example, a short‐circuit test, one can verify that the temperature of the transformer is not homogeneous, existing in higher‐temperature zones . Therefore, in the training of the neural network, it is important not to consider an average temperature.…”

Section: Implementation Of the Proposed Toolmentioning

confidence: 99%

“…However, owing to their similarities, R HV and R LV were considered equal. 1 In turn, the output vector of the neural network has a single output, L T , which represents the total loss. Finally, based on the Kolmogorov method, considering the number of inputs, the maximum number of neurons in the hidden layer was set to 11.…”

Section: Total Losses Networkmentioning

confidence: 99%

“…Figure 6 shows the neural network used to estimate the temperature. The input variables of this neural network are as follows: the cross-sectional area of the conductors used in the windings (in mm 2 ), referred to as Sections; thickness of the winding sheet (in mm), referred to as Thickness; 1 It is possible because the resistances of the windings are balanced per phase in both high and low voltages, since it is required by standard. The transformer leaves the factory with balanced windings.…”

Section: Total Losses Networkmentioning

confidence: 99%

“…Also, they have advantages such as a higher mechanical robustness; the possibility of installation close to the load point; does not require transformation wells for oil collection, fire fighting systems, fireguard walls, and oil supervision accessories; and the lowest level of internal partial discharges (owing to its vacuum encapsulation). 1 Therefore, they are typically used where there is the presence of people, for example, in industries, residential buildings, and hospitals. 2 However, dry-type transformers are more expensive and have power and voltage limitations.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Neural networks applied to the design of dry-type transformers: an example to analyze the winding temperature and elevate the thermal quality

Finocchio

Lopes

França

et al. 2016

Int. Trans. Electr. Energ. Syst.

View full text Add to dashboard Cite

Summary Much of the cost of manufacturing transformers is related to the complexity in the development of its project, which involves many variables, such as different types of materials, methods, and manufacturing processes. Also, it is notoriously difficult to establish standards that relate the transformer characteristics with these design variables, since each, almost always, variable is calculated empirically. One of the most important design variables is the internal temperature of the transformers, which directly influences the lifetime of such equipments. So, a computational tool is under development whose purpose is to automate the design of cast resin dry‐type transformers and minimize the time taken for its completion, by means of artificial neural networks. In this article, we present some results of this tool, which relate some geometrical parameters of the specific transformer with the temperature of its windings. For this, the winding losses and total losses are also estimated. The training of the neural network was done with the test data of about 300 dry‐type transformers from the same manufacturer, so, with the same constructive features. Nevertheless, our results show that the technique is promising because the neural networks were well fitted and its results present errors lower than 1% when compared with data from tests with real transformers. Evidently, the proposed methodology is dependent on the constructive technique; that is, once trained, the neural network can be used only in the design of dry‐type transformers of the same constructive technique as those whose results of the tests were used to train the network. Nevertheless, even if only used to provide the initial parameters for a more complex design tool, the proposed method is useful since it is rapid and has low computational cost.

show abstract

Section: Implementation Of the Proposed Toolmentioning

confidence: 99%

Section: Total Losses Networkmentioning

confidence: 99%

Section: Total Losses Networkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations