2 minute read

Forecasting Natural Gas Demand/Supply

Author: Baekeun Park

Problem

South Korea relies on foreign imports for 92.8 percent of its energy resources as of the first half of 2020 1. Among the energy resources, the Korea Gas Corporation imports Liquified Natural Gas(LNG) from around world and supplies it to power generation plants, gas-utility companies and city gas companies throughout the country 2. It produces and supplies Natural Gas(NG), in order to ensure stable gas supply for the nation. And it operates LNG storage tanks at LNG acquisition bases which can store LNG during the season when city gas demand is low and replenish LNG during winter when demand is higher than supply 3.

The wholesale charges consist of raw material costs (LNG introduction and incidental costs) and gas supply costs 4. Therefore, the forecasting NG demand/supply will not only help establish an optimized mid-to-long-term plan for the introduction of LNG, but also stable NG supply and economic effects.

Dataset

There is a NG supply dataset5 from public data portal in South Korea. It includes four years of regional(nine different cities) monthly NG supply in South Korea. In addition, climate data6 for the same period can be obtained from the Korea Meteorological Administration. Similarly, data on the price of four types of oil7 and various types of coal price dataset per month8 are also available through corresponding agencies.

Data Preprocessing

Python has a famous package, Pandas9, which can be used for the data preprocessing to deal with various type and shape of data. There may also be situations in which one-hot encoding10 is used, where some transformation can be performed to better train the network.

Deep Learning Algorithm

Tensorflow11 is an end-to-end open source platform for machine learning and deep learning. It can be combined with Keras12 to develop the model.

Timeline

Week1: Find more related dataset, AI algorithms and works

Week2: Perform data preprocessing

Week3: Design network model

Week4-Week5: Forecast NG demand/supply and Complite project

References

  1. 2020 Monthly Energy Statistics, http://www.keei.re.kr/keei/download/MES2009.pdf, Sep. 2020 ↩︎

  2. Online Resource, https://www.kogas.or.kr:9450/eng/contents.do?key=1498 ↩︎

  3. Online Resource, https://www.kogas.or.kr:9450/portal/contents.do?key=2014 ↩︎

  4. Online Resource, https://www.kogas.or.kr:9450/portal/contents.do?key=2026 ↩︎

  5. Natural Gas Supply dataset, https://www.data.go.kr/data/15049904/fileData.do ↩︎

  6. Regional Climate dataset, https://data.kma.go.kr/climate/RankState/selectRankStatisticsDivisionList.do?pgmNo=179 ↩︎

  7. Crude Oil Price dataset, https://www.petronet.co.kr/main2.jsp ↩︎

  8. Bituminous Coal Price dataset, https://www.kores.net/komis/price/mineralprice/ironoreenergy/pricetrend/baseMetals.do?mc_seq=3030003&mnrl_pc_mc_seq=506 ↩︎

  9. pandas, https://github.com/pandas-dev/pandas ↩︎

  10. One-Hot Encoding, https://towardsdatascience.com/what-is-one-hot-encoding-and-how-to-use-pandas-get-dummies-function-922eb9bd4970 ↩︎

  11. Tensorflow, https://github.com/tensorflow/tensorflow ↩︎

  12. Keras, https://github.com/keras-team/keras ↩︎