Çimento Değirmenlerinde Çizelgeleme ve Dinamik Enerji Optimizasyonu

Öz

Bu çalışma, çimento öğütme operasyonlarında enerji tüketimiyle ilgili zorlukları ele almak amacıyla, üretim belirsizliklerini ve stok yönetimi etkinliğini bütüncül bir çerçevede ele alan bütünleşmiş bir yapı geliştirmeyi amaçlamaktadır. Temel amaç, üretim planlamasını ve çevresel sürdürülebilirliği iyileştirirken enerji maliyetlerini minimize etmektir. Bu kapsamda, elektrik fiyat tahmini için Otoregresif Entegre Hareketli Ortalama (ARIMA), Aşırı Gradyan Artırma (XGBoost) ve Uzun Kısa Süreli Bellek (LSTM) gibi istatistiksel ve makine öğrenmesi teknikleri kullanılmaktadır. Tahmin sonuçlarına dayanarak, üretim çizelgelemesini ve enerji tüketimini optimize etmek amacıyla Karma Tam Sayılı Doğrusal Programlama (MILP) modeli geliştirilmiştir. Önerilen yöntem, üretim faaliyetlerini dalgalanan elektrik fiyatlarına göre uyarlayarak enerji maliyetlerini başarıyla azaltmaktadır. Tahmin yöntemleri arasında, LSTM modeli en düşük hata oranları ile en üstün performansı göstermektedir. Ayrıca, optimizasyon modeli operasyonel kısıtları karşılayan ve kaynak kullanımını iyileştiren etkin üretim planları sunmaktadır. Elde edilen sonuçlar, tahmin tekniklerinin optimizasyon modelleriyle entegre edilmesinin, çimento tesislerinde enerji verimliliğini ve üretim yönetimini önemli ölçüde artırdığını, aynı zamanda çevresel sürdürülebilirliğe katkıda bulunduğunu ve endüstriyel operasyonların rekabet avantajını güçlendirdiğini göstermektedir.

Anahtar Kelimeler

• Fiyat Tahmini
• Çizelgeleme
• Enerji
• Optimizasyon
• Karma Tam Sayılı Doğrusal Programlama

Scheduling and Dynamic Energy Optimization in Cement Mills

Öz

Background/Aim: This study aims to address energy consumption challenges in cement milling operations by developing an integrated framework that incorporates production uncertainties and inventory management efficiency. The main objective is to minimize energy costs while improving production planning and environmental sustainability. Methods: In this context, statistical and machine learning techniques, including Autoregressive Integrated Moving Average (ARIMA), Extreme Gradient Boosting (XGBoost), and Long Short-Term Memory (LSTM), are employed for electricity price forecasting. Based on the forecasting results, a Mixed Integer Linear Programming (MILP) model is developed to optimize production scheduling and energy consumption. The proposed framework incorporates forecasted electricity-price fluctuations into production scheduling to generate cost-oriented production plans while satisfying operational constraints. Results: Among the forecasting methods, the LSTM model demonstrates superior performance with the lowest error rates. Furthermore, the optimization model provides efficient production plans that satisfy operational constraints and improve resource utilization. Conclusion: The results indicate that integrating forecasting techniques with optimization models significantly enhances energy efficiency and production management in cement plants, while also contributing to environmental sustainability and strengthening the competitive advantage of industrial operations.

Anahtar Kelimeler

• Price Forecasting
• Scheduling
• Energy
• Optimization
• Mixed-Integer Linear Programming (MILP)

Kaynakça

1. Wang, J., Dai, Y., Gao, L. 2009. Exergy analyses and parametric optimizations for different cogeneration power plants in cement industry. Applied Energy, 86(6), 941–948.
2. Emir, T., Güler, M. G. 2018. Production Planning Using Day-Ahead Prices in a Cement Plant. pp. 149–166. Dincer, I., Colpan, C. O., Kizilkan, O., ed. 2018. Exergetic, Energetic and Environmental Dimensions, Academic Press, 1106p.
3. Hadera, H., Harjunkoski, I. 2013. Continuous-time Batch Scheduling Approach for Optimizing Electricity Consumption Cost. Computer Aided Chemical Engineering, 32, 403–408.
4. Castro, P. M., Harjunkoski, I., Grossmann, I. E. 2009. New continuous-time scheduling formulation for continuous plants under variable electricity cost. Industrial and Engineering Chemistry Research, 48(14), 6701–6714.
5. Swanepoel, J. A., Mathews, E. H., Vosloo, J., Liebenberg, L. 2014. Integrated energy optimisation for the cement industry: A case study perspective. Energy Conversion and Management, 78, 765–775.
6. Zhao, X. Z., He, B., Xu, F. Y., Lai, L. L., Yang, C., Lu, S., Li, D. 2014. A model of demand response scheduling for cement plant. IEEE International Conference on Systems, Man and Cybernetics, October 5-8, San Diego, USA, 3042–3047.
7. Kis, T., Kovács, A. 2012. A cutting plane approach for integrated planning and scheduling. Computers & Operations Research, 39(2), 320–327.
8. Petkov, S. B., Maranas, C. D. 1997. Multiperiod Planning and Scheduling of Multiproduct Batch Plants under Demand Uncertainty. Industrial and Engineering Chemistry Research, 36(11), 4864–4881.

9. Mahdieh, M., Clark, A., Bijari, M. 2018. A novel flexible model for lot sizing and scheduling with non-triangular, period overlapping and carryover setups in different machine configurations. Flexible Services and Manufacturing Journal, 30(4), 884–923.
10. Hadera, H., Harjunkoski, I., Grossmann, I. E., Sand, G., Engell, S. 2014. Steel Production Scheduling Optimization under Time-sensitive Electricity Costs. Computer Aided Chemical Engineering, 33, 373–378.
11. Salahi, N., Jafari, M. A. 2016. Energy-Performance as a driver for optimal production planning. Applied Energy, 174, 88–100.
12. Weron, R., Misiorek, A. 2008. Forecasting spot electricity prices: A comparison of parametric and semiparametric time series models. International Journal of Forecasting, 24(4), 744–763.
13. Dooley, G., Lenihan, H. 2005. An assessment of time series methods in metal price forecasting. Resources Policy, 30(3), 208–217.
14. Tularam, G. A., Saeed, T. 2016. Oil-price forecasting based on various univariate time-series models. American Journal of Operations Research, 6(3), 226-235.
15. Windler, T., Busse, J., Rieck, J. 2019. One month-ahead electricity price forecasting in the context of production planning. Journal of Cleaner Production, 238, 117910.
16. Zhang, Q., Cremer, J. L., Grossmann, I. E., Sundaramoorthy, A., Pinto, J. M. 2016. Risk-based integrated production scheduling and electricity procurement for continuous power-intensive processes. Computers & Chemical Engineering, 86, 90–105.
17. Zakeri, S. A. H. 2015. Time series analysis and forecasting electricity prices in Turkey. Middle East Technical University, Graduate School of Natural and Applied Sciences, Master Thesis, 105p, Ankara.
18. Jan, F., Shah, I., Ali, S. 2022. Short-Term Electricity Prices Forecasting Using Functional Time Series Analysis. Energies, 15(9), 3423.
19. Sirisha, U. M., Belavagi, M. C., Attigeri, G. 2022. Profit Prediction Using ARIMA, SARIMA and LSTM Models in Time Series Forecasting: A Comparison. IEEE Access, 10, 124715–124727.
20. Zhang, L., Bian, W., Qu, W., Tuo, L., Wang, Y. 2021. Time series forecast of sales volume based on XGBoost. Journal of Physics: Conference Series, 1873(1), 012067.
21. Paliari, I., Karanikola, A., Kotsiantis, S. 2021. A comparison of the optimized LSTM, XGBOOST and ARIMA in Time Series forecasting. 12th International Conference on Information, Intelligence, Systems and Applications (IISA), July 12-14, Chania, Greece.