Cite this DOI

Varun Gupta, V. G. (2023). Bi-Modal Oil Temperature Forecasting in Electrical Transformers using a Hybrid of Transformer, CNN and Bi-LSTM. *Journal of Science & Technology*, *8*(5), 49–56. https://doi.org/10.46243/jst.2023.v8.i05.pp49-56

@article{varungupta2023bimodal,
  author    = {Varun Gupta, Varun Gupta},
  title     = {{Bi-Modal Oil Temperature Forecasting in Electrical Transformers using a Hybrid of Transformer, CNN and Bi-LSTM}},
  journal   = {Journal of Science \& Technology},
  year      = {2023},
  month     = {oct},
  volume    = {8},
  number    = {5},
  pages     = {49--56},
  publisher = {Longman Publishers},
  issn      = {2456-5660},
  doi       = {10.46243/jst.2023.v8.i05.pp49-56},
  url       = {https://doi.org/10.46243/jst.2023.v8.i05.pp49-56},
  language  = {en},
  abstract  = {Power consumption prediction is a tough task because of its fluctuating nature. If the expected demand is excessively high in comparison to the existing demand, the transformer may damage. Predicting the temperature of transformer oil is an efficient approach to verify the transformer’s safety status. As a result, in this study, we offer a bimodal architecture for predicting oil temperature given a sequence of prior temperatures. Our model was tested using the Ettm1, Ettm2, and Etth1 datasets and achieved an RMSE of 0.41375, MAE of 0.3031 and MAPE of 8.292\% on Ettm1 test dataset, an RMSE of 0.4105, MAE 0.3090 and MAPE of 6.678\% on Ettm2 test dataset and an RMSE of 0.6762, MAE 0.4690 and MAPE of 11.23\% on Etth1 test dataset}
}

TY  - JOUR
TI  - Bi-Modal Oil Temperature Forecasting in Electrical Transformers using a Hybrid of Transformer, CNN and Bi-LSTM
AU  - Varun Gupta, Varun Gupta
JO  - Journal of Science & Technology
PY  - 2023
DA  - 2023/10/05/
VL  - 8
IS  - 5
SP  - 49
EP  - 56
PB  - Longman Publishers
SN  - 2456-5660
LA  - en
AB  - Power consumption prediction is a tough task because of its fluctuating nature. If the expected demand is excessively high in comparison to the existing demand, the transformer may damage. Predicting the temperature of transformer oil is an efficient approach to verify the transformer’s safety status. As a result, in this study, we offer a bimodal architecture for predicting oil temperature given a sequence of prior temperatures. Our model was tested using the Ettm1, Ettm2, and Etth1 datasets and achieved an RMSE of 0.41375, MAE of 0.3031 and MAPE of 8.292% on Ettm1 test dataset, an RMSE of 0.4105, MAE 0.3090 and MAPE of 6.678% on Ettm2 test dataset and an RMSE of 0.6762, MAE 0.4690 and MAPE of 11.23% on Etth1 test dataset
DO  - 10.46243/jst.2023.v8.i05.pp49-56
UR  - https://doi.org/10.46243/jst.2023.v8.i05.pp49-56
ER  -

{
    "type": "article-journal",
    "id": "10.46243/jst.2023.v8.i05.pp49-56",
    "DOI": "10.46243/jst.2023.v8.i05.pp49-56",
    "URL": "https://doi.org/10.46243/jst.2023.v8.i05.pp49-56",
    "title": "Bi-Modal Oil Temperature Forecasting in Electrical Transformers using a Hybrid of Transformer, CNN and Bi-LSTM",
    "source": "Smart Scholars DOI Registry",
    "container-title": "Journal of Science & Technology",
    "author": [
        {
            "family": "Varun Gupta",
            "given": "Varun Gupta"
        }
    ],
    "issued": {
        "date-parts": [
            [
                2023,
                10,
                5
            ]
        ]
    },
    "volume": "8",
    "issue": "5",
    "page": "49-56",
    "publisher": "Longman Publishers",
    "language": "en",
    "abstract": "Power consumption prediction is a tough task because of its fluctuating nature. If the expected demand is excessively high in comparison to the existing demand, the transformer may damage. Predicting the temperature of transformer oil is an efficient approach to verify the transformer’s safety status. As a result, in this study, we offer a bimodal architecture for predicting oil temperature given a sequence of prior temperatures. Our model was tested using the Ettm1, Ettm2, and Etth1 datasets and achieved an RMSE of 0.41375, MAE of 0.3031 and MAPE of 8.292% on Ettm1 test dataset, an RMSE of 0.4105, MAE 0.3090 and MAPE of 6.678% on Ettm2 test dataset and an RMSE of 0.6762, MAE 0.4690 and MAPE of 11.23% on Etth1 test dataset",
    "ISSN": "2456-5660"
}

From the record as registered (version 2) — the record and its history. Programs: https://registry.smartscholars.in/api.php?action=cite&doi=10.46243%2Fjst.2023.v8.i05.pp49-56 gives all four in one JSON answer.