Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks
Temperature forecasting has been a consistent research topic owing to its significant effect on daily lives and various industries. However, it is an ever-challenging task because temperature is affected by various climate factors. Research on temperature forecasting has taken one of two directions:...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/5/1609 |
| _version_ | 1811330415874539520 |
|---|---|
| author | Sungjae Lee Yung-Seop Lee Youngdoo Son |
| author_facet | Sungjae Lee Yung-Seop Lee Youngdoo Son |
| author_sort | Sungjae Lee |
| collection | DOAJ |
| description | Temperature forecasting has been a consistent research topic owing to its significant effect on daily lives and various industries. However, it is an ever-challenging task because temperature is affected by various climate factors. Research on temperature forecasting has taken one of two directions: time-series analysis and machine learning algorithms. Recently, a large amount of high-frequent climate data have been well-stored and become available. In this study, we apply three types of neural networks, multilayer perceptron, recurrent, and convolutional, to daily average, minimum, and maximum temperature forecasting with higher-frequency input features than researchers used in previous studies. Applying these neural networks to the observed data from three locations with different climate characteristics, we show that prediction performance with highly frequent hourly input data is better than forecasting performance with less-frequent daily inputs. We observe that a convolutional neural network, which has been mostly employed for processing satellite images rather than numeric weather data for temperature forecasting, outperforms the other models. In addition, we combine state of the art weather forecasting techniques with the convolutional neural network and evaluate their effects on the temperature forecasting performances. |
| first_indexed | 2024-04-13T16:02:26Z |
| format | Article |
| id | doaj.art-b86b4540e5e04e27afea3dfebf041231 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-04-13T16:02:26Z |
| publishDate | 2020-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-b86b4540e5e04e27afea3dfebf0412312022-12-22T02:40:31ZengMDPI AGApplied Sciences2076-34172020-02-01105160910.3390/app10051609app10051609Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural NetworksSungjae Lee0Yung-Seop Lee1Youngdoo Son2Department of Industrial and Systems Engineering, Dongguk University—Seoul, Seoul 04620, KoreaDepartment of Statistics, Dongguk University—Seoul, Seoul 04620, KoreaDepartment of Industrial and Systems Engineering, Dongguk University—Seoul, Seoul 04620, KoreaTemperature forecasting has been a consistent research topic owing to its significant effect on daily lives and various industries. However, it is an ever-challenging task because temperature is affected by various climate factors. Research on temperature forecasting has taken one of two directions: time-series analysis and machine learning algorithms. Recently, a large amount of high-frequent climate data have been well-stored and become available. In this study, we apply three types of neural networks, multilayer perceptron, recurrent, and convolutional, to daily average, minimum, and maximum temperature forecasting with higher-frequency input features than researchers used in previous studies. Applying these neural networks to the observed data from three locations with different climate characteristics, we show that prediction performance with highly frequent hourly input data is better than forecasting performance with less-frequent daily inputs. We observe that a convolutional neural network, which has been mostly employed for processing satellite images rather than numeric weather data for temperature forecasting, outperforms the other models. In addition, we combine state of the art weather forecasting techniques with the convolutional neural network and evaluate their effects on the temperature forecasting performances.https://www.mdpi.com/2076-3417/10/5/1609convolution neural networkdeep learninglong short term memorytemperature forecasting |
| spellingShingle | Sungjae Lee Yung-Seop Lee Youngdoo Son Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks Applied Sciences convolution neural network deep learning long short term memory temperature forecasting |
| title | Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks |
| title_full | Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks |
| title_fullStr | Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks |
| title_full_unstemmed | Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks |
| title_short | Forecasting Daily Temperatures with Different Time Interval Data Using Deep Neural Networks |
| title_sort | forecasting daily temperatures with different time interval data using deep neural networks |
| topic | convolution neural network deep learning long short term memory temperature forecasting |
| url | https://www.mdpi.com/2076-3417/10/5/1609 |
| work_keys_str_mv | AT sungjaelee forecastingdailytemperatureswithdifferenttimeintervaldatausingdeepneuralnetworks AT yungseoplee forecastingdailytemperatureswithdifferenttimeintervaldatausingdeepneuralnetworks AT youngdooson forecastingdailytemperatureswithdifferenttimeintervaldatausingdeepneuralnetworks |