Research Article

PREDICTION OF GRID POLLUTION RATES IN RIVER-TYPE POWER PLANTS USING MACHINE LEARNING METHODS

Volume: 28 Number: 2 June 3, 2025
EN TR

PREDICTION OF GRID POLLUTION RATES IN RIVER-TYPE POWER PLANTS USING MACHINE LEARNING METHODS

Abstract

The grid pollution value is obtained from the pressure values measured by sensors placed at different locations. Deterioration of the calibration of the sensors or incorrect measurements due to environmental factors cause the units to operate inefficiently and unhealthily. In this study, firstly, a dataset consisting of samples taken from the SCADA system of Karkamış hydroelectric power plant are collected. This dataset consists of flow, power, head and grid pollution measurements taken from the sensors. The collected dataset is divided into training, validation and test subsets to perform grid pollution prediction. Training study is realized for machine learning methods with training data. Then, the performances of machine learning methods for grid pollution prediction are compared on different units on the testing dataset. From the studies, the best correlation coefficient (R) values are provided with the thin tree model as 0.9833 for unit-1, 0.9716 for unit-2, 0.9792 for unit-4 and 0.9810 for unit-5. It is clearly observed from the obtained results that grid pollution prediction can be done effectively with machine learning methods.

Keywords

References

  1. Acar, G. (2022). Zaman serisi yöntemleri ve xgboost algoritması ile kömür satış tahmini: türkiye tarım kredi kooperatifleri uygulaması.Yüksek Lisans Tezi. Sakarya Üniversitesi Fen Bilimleri Enstitüsü Bilişim Sistemleri Mühendisliği Ana Bilim Dalı, Sakarya 72s.
  2. Alatas, M.,Susilowati E.,Theresia M., Budiastuti S., Gunawan T., Setyono P. (2022). Horizontal Trash Rack Diverter Trash (HTDT) to Minimize Trash Clogging at the Intake of Micro-Hydro Power Plant. International Journalof Sustainable Developmentand Planning, Vol.17,No.6,October.1713-1720. https://doi.org/10.18280/ijsdp.170604
  3. Artuğer, F & Özkaynak, F (2022). Görüntü Sıkıştırma Algoritmalarının Performans Analizi İçin Değerlendirme Rehberi.Munzur Üniversitesi Dergisi, 8(1).102-110. https://doi.org/10.29132/ijpas.1012013
  4. Alpaydin, E. (2014). Introduction to machine learning . The MIT press. (4th ed.)
  5. Arı, A & Önder, H. (2013). Farklı Veri Yapılarında Kullanılabilecek Regrasyon Yöntemleri. Anadolu Tarım Bilim Dergisi, 28 (3), 168-174. https://doi.org/10.7161/anajas.2013.28.3.168
  6. Bragalli C. , Micocci D. , Naldi G. (2023).On the influence of net head and efficiency fluctuations over the performance of existing run-of-river hydropower plants, Volume 206, April 2023. 1170-1179.
  7. Başeşme, Hidayet (2003). Hidroelektrik santrallar ve hidroelektrik santral esisleri, EÜAŞ Genel Müdürlüğü Hidrolik Santraller Dairesi Başkanlığı.
  8. Bekçioğulları, M. F., Dikici, B., Açıkgöz, H., Keçecioğlu, Ö. F. (2021). Güneş Enerjisinin Kısa-Dönem Tahmininde Farklı Makine Öğrenme Yöntemlerinin Karşılaştırılması. EMO Bilimsel Dergi, 11(22), 37-45.

Details

Primary Language

Turkish

Subjects

Electrical Energy Generation (Incl. Renewables, Excl. Photovoltaics)

Journal Section

Research Article

Publication Date

June 3, 2025

Submission Date

October 20, 2024

Acceptance Date

January 17, 2025

Published in Issue

Year 1970 Volume: 28 Number: 2

APA
Konu, K., Bilgin, O., & Açıkgöz, H. (2025). NEHİR TİPİ SANTRALLERDEKİ IZGARA KİRLİLİK ORANLARININ MAKİNE ÖĞRENME YÖNTEMLERİ İLE TAHMİNİ. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 28(2), 613-629. https://doi.org/10.17780/ksujes.1570840

Cited By