Research Article

QUALITY DETERMINATION OF LIGNITE COAL IN THE REGIONS INCLUDING THE WHOLE OF TURKEY USING WAVELET TRANSFORM BASED NAIVE BAYES CLASSIFIER

Volume: 28 Number: 1 March 3, 2025
EN TR

QUALITY DETERMINATION OF LIGNITE COAL IN THE REGIONS INCLUDING THE WHOLE OF TURKEY USING WAVELET TRANSFORM BASED NAIVE BAYES CLASSIFIER

Abstract

Today, the process of determining whether the complex lignite coal obtained from Turkey's drilling fields is of acceptable quality or relatively poor quality is of vital importance in power plants, other areas and energy production. In particular, the accurate determination of multiple complex parameters according to their quality has gained great importance in investment decisions. In this article, the moisture, ash, sulfur and calorie contents of 96 drilling regions of 96 lignite coals covering the whole of Turkey in the lignite inventory published by MTA, received from the General Directorate of Turkish Coal Enterprises, are discussed. The 3 parameters mentioned, such as moisture, ash and sulfur, affect the caloric value, which is the dependent variable. The higher the calories, the better the coal quality. First, by applying the K-Means clustering algorithm to the data, the caloric values of coal were divided into 2 groups. These values, divided into 2 groups, were used as validation data. Feature values obtained from wavelet coefficients were classified with the Naive Bayes classifier. When the classification performances were compared, it was determined that db4, which is from the Daubechies wavelet family, showed higher performance than other wavelet families and CA wavelet coefficients compared to CH, CV and CD coefficients. Wavelet Transform coefficients of moisture, ash and sulfur contents were calculated and classification performances were compared with the Naive Bayes method. It has been determined that the quality detection rate of coal obtained with the approximation coefficients of db4, which is from the Daubechies family, is the highest and is 100%.

Keywords

Thanks

Yazar İnönü üniversitesinde öğretim üyesi olan Prof. Dr. Bülent TÜTMEZ’e teşekkür etmektedir.

References

  1. Aytac Korkmaz, S. (2020). Grade level of lignite coal datas in the different areas with decison tree, random forest, and discriminant analysis methods. Applied Artificial Intelligence, 34(11), 755-776.
  2. Cheng, Y., Xu, L., Li, X., & Guo, Z. (2012, July). Online coal calorific value prediction from mutiband coal/air combustion radiation characteristics. In Instrumentation and Control Technology (ISICT), 2012 8th IEEE International Symposium on,309-313,IEEE.
  3. Chelgani S C, Mesroghli S H, Hower J C. (2010). Simultaneous prediction of coal rank parameters based on ultimate analysis using regression and artificial neural network. International Journal of Coal Geology, 83(1), 31-34.
  4. Doğan, G. & Ergen, B. (2022). Karayollarındaki Asfalt Çatlaklarının Tespiti İçin Yeni Bir Konvolüsyonel Sinir Ağı Tabanlı Yöntem . Fırat Üniversitesi Mühendislik Bilimleri Dergisi,34 (2) , 485-494 . DOI: 10.35234/fumbd.1014951.
  5. Fırat M., Dikbaş F., Koç AC., ve Güngör M. (2012). K-Ortalamalar Yöntemi ile Yıllık Yağışların Sınıflandırılması ve Homojen Bölgelerin Belirlenmesi. İMO Teknik Dergi, 383, 6037-6050.
  6. Galetakis M J, Theodoridis K, Kouridou O. (2002). Lignite quality estimation using ANN and adaptive neuro-fuzzy inference systems (ANFIS). APPCOM: 425-431.
  7. IEA, (2000). International energy annual. France. International Journal of Coal Science & Engineering (China) Energy Agency.
  8. Karhan Z., Ergen B. (2016). Content based medical image classification using discrete wavelet and cosine transforms. (2015). 23nd Signal Processing and Communications Applications Conference (SIU), (pp:1445-1448).

Details

Primary Language

Turkish

Subjects

Decision Support and Group Support Systems , Semi- and Unsupervised Learning

Journal Section

Research Article

Publication Date

March 3, 2025

Submission Date

October 24, 2024

Acceptance Date

December 10, 2024

Published in Issue

Year 1970 Volume: 28 Number: 1

APA
Aytaç, S. (2025). DALGACIK DÖNÜŞÜMÜ TABANLI NAİVE BAYES SINIFLANDIRICI İLE TÜRKİYE’NİN TAMAMINI İÇEREN BÖLGELERDEKİ LİNYİT KÖMÜRÜNÜN KALİTE TESPİTİ. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 28(1), 403-413. https://doi.org/10.17780/ksujes.1572893