Author(s)

Mengyang Li ¹, Yang Wang ², Yehui Bo ², Huiyi Zheng ², Xuanyao Yu ³

Affiliation(s)

¹ Kunlun Digital Technology Co., Ltd., Beijing, China
² Petrochina Kunlun Gas Co., Ltd., Shandong Branch, Jinan, China
³ China University of Petroleum (Beijing), Beijing, China

Corresponding Author

Mengyang Li

ABSTRACT

This study used indicators such as dispersion coefficient, entropy value, skewness, etc. to determine the data fluctuation of users' natural gas consumption. We collected 2736 daily natural gas consumption data from 12 users in a certain region, calculated the accuracy of natural gas prediction for each user using the LSTM model, and analyzed the correlation between various volatility indicators and MAPE. The results showed that the greater the volatility, the greater the LSTM prediction error, indicating a positive correlation between the volatility value and MAPE. Adding a filtering algorithm to the original data can effectively reduce the volatility of the original data, but the MAPE values of users have not all decreased. For example, users with smaller MAPE values, such as User 5, have increased prediction errors after adding the filtering algorithm. Users with larger MAPE values, such as User 2, have reduced prediction errors after adding the filtering algorithm. The filtering algorithm reduces volatility while partially missing the original data features, making it difficult for the prediction results to be completely accurate and ultimately maintain around 5%. However, adding missing filtering algorithms can effectively reduce some user prediction errors. The specific use of this algorithm depends on the prediction results of the model itself.

KEYWORDS

Data volatility, Entropy value, LSTM, filtering algorithm

CITE THIS PAPER

Mengyang Li, Yang Wang, Yehui Bo, Huiyi Zheng, Xuanyao Yu, Fluctuation analysis of natural gas forecast. Industrial Engineering and Innovation Management (2024) Vol. 7: 164-172. DOI: http://dx.doi.org/10.23977/ieim.2024.070323.

REFERENCES

[1] Simsek Y E. Electric energy demands of Turkey in residential and industrial sectors[J]. Renewable and Sustainable Energy Reviews, 2012.
[2] Alvarez F M, Troncoso A, Riquelme J C, et al. Energy Time Series Forecasting Based on Pattern Sequence Similarity[J]. IEEE Transactions on Knowledge & Data Engineering, 2011, 23(8):1230-1243.
[3] Beyca O F, Ervural B C, Tatoglu E, et al. Using machine learning tools for forecasting natural gas consumption in the province of Istanbul[J]. Energy Economics, 2019, 80(MAY):937-949.
[4] BAI Y, LI C. Daily natural gas consumption forecasting based on a structure-calibrated support vector regression approach [J]. Energy and Buildings, 2016, 127: 571-579.
[5] Taşpinar F, Çelebi N, Tutkun N. Forecasting of daily natural gas consumption on regional basis in Turkey using various computational methods[J]. Energy and Buildings, 2013, 56: 23-31.
[6] Potočnik P, Thaler M, Govekar E, et al. Forecasting risks of natural gas consumption in Slovenia[J]. Energy Policy, 2007, 35(8): 4271-4282.
[7] RIGAMONTI M, BARALDI P, ZIO E, et al. Ensemble of optimized echo state networks for remaining useful life prediction [J]. Neurocomputing, 2018, 281: 121-138.
[8] Kamilaris A, Prenafeta-Boldú F X. Deep learning in agriculture: A survey[J]. Computers and Electronics in Agriculture, 2018, 147: 70-90.
[9] Hirafuji Neiva D, Zanchettin C. Gesture recognition: A review focusing on sign language in a mobile context [J]. Expert Systems with Applications, 2018, 103: 159-183. 
[10] NAGY A M, SIMON V. Survey on traffic prediction in smart cities[J]. Pervasive and Mobile Computing, 2018, 50: 148-163. 
[11] ERDOGDU E. Natural gas demand in Turkey[J]. Applied Energy, 2010, 87(1): 211-219.
[12] Laib O, Khadir M T, Mihaylova L. Toward efficient energy systems based on natural gas consumption prediction with LSTM Recurrent Neural Networks[J]. Energy, 2019, 177: 530-542.

Sponsors, Associates, and Links

---

• Information Systems and Economics
  
  
• Accounting, Auditing and Finance
  
  
• Tourism Management and Technology Economy
  
  
• Journal of Computational and Financial Econometrics
  
  
• Financial Engineering and Risk Management
  
  
• Accounting and Corporate Management
  
  
• Social Security and Administration Management
  
  
• Population, Resources & Environmental Economics
  
  
• Statistics & Quantitative Economics
  
  
• Agricultural & Forestry Economics and Management
  
  
• Social Medicine and Health Management
  
  
• Land Resource Management
  
  
• Information, Library and Archival Science
  
  
• Journal of Human Resource Development
  
  
• Manufacturing and Service Operations Management
  
  
• Operational Research and Cybernetics