TEKSTİL ENDÜSTRİSİNDE DERİN ÖĞRENME KULLANARAK AŞIRI ELEKTRİK TÜKETİMİNİN ÖNLENMESİNE YÖNELİK BİR VAKA ÇALIŞMASI

Abstract

Endüstrinin en kritik girdileri kaynaklardır ve bu nedenle kaynak tüketimi endüstriyel süreçlerde en aza indirilmesi gereken önemli bir konudur. Öte yandan, kaynak tüketimi birçok parametreye bağlı olduğu için tahmin edilmesi zordur. Son dönemlerde, Makine Öğrenmesi (MÖ) ve Derin Öğrenme (DÖ) kavramları, herhangi bir alanda gelecek tahmini için kullanılan güçlü Yapay Zeka alt alanlarıdır. Bu çalışmada tekstil endüstrisi için bir vaka çalışması olarak, makinelerin bekleme durumunda aşırı kaynak tüketimini önlemek amacıyla DÖ destekli bir elektrik tahmin modeli tasarlanmıştır. Bu yöntem, makinelerin karar verme süreçlerini içeren ve aşırı tüketime nedeniyle üretimi kesintiye uğratmasına yardımcı olan Uzun-Kısa Süreli Bellek (UKSB) tabanlı kayan pencere tekniği sayesinde elektrik tüketiminin saatlik dinamik eşik değerlerini tahminlemektedir. Hesaplanan eşik değerleri, Tekrarlayan Sinir Ağları (TSA) ve Kapılı Tekrarlayan Birimler (KTB) gibi diğer Derin Öğrenme yöntemleri ve geleneksel bir yöntem olan Otomatik Regresif Entegre Hareketli Ortalama (ARIMA) yöntemi ile karşılaştırılmış, elde edilen sonuçların makinelerin bekleme durumundaki gerçek zamanlı elektrik tüketim verilerine ne kadar yaklaştığı analiz edilmiştir. Elde edilen sonuçlara göre, UKSB modeli elektrik tüketim seviyelerini başarılı bir şekilde tahmin etmekte, tüketim seviyeleri eşiğe ulaştığında Programlanabilir Mantık Denetleyicisi (PMD) ünitesine durma sinyali göndermekte ve bu sayede aşırı kaynak tüketimini engellemektedir.

Keywords

• Derin Öğrenme
• UKSB
• Kayan Pencere Tekniği
• Elektrik Tüketimi
• Tekstil Endüstrisi

Supporting Institution

Menderes Tekstil A.Ş., Denizli, Türkiye

Thanks

Menderes Tekstil A.Ş., Denizli, Türkiye

A CASE STUDY FOR PREVENTING ELECTRICITY OVER-CONSUMPTION USING DEEP LEARNING IN TEXTILE INDUSTRY

Abstract

Resources are the most critical input in the manufacturing industry therefore, resource consumption is an essential issue to be minimized. On the other hand, consumption depends on several parameters thus, it is difficult to estimate. Recently, Machine Learning (ML) and Deep Learning (DL) are powerful Artificial Intelligence (AI) subdomains for future prediction in any area. In this paper, a DL-supported electricity prediction method is designed for the textile industry as a case study in order to prevent resource over-consumption while the machines are in the standby state. This method provides dynamic consumption thresholds of electricity consumption by sliding window technique based Long-Short Term Memory (LSTM) model that helps the machines to interrupt manufacturing in their decision. These calculated thresholds are also compared with the results of Recurrent Neural Networks (RNN) and Gated Recurrent Units (GRU) as the other DL methods and Automated Regressive Integrated Moving Average (ARIMA) as a traditional method and then the results have been analyzed how close they are to real-time electricity consumption data at standby. According to the results, the LSTM model successfully predicts electricity consumption levels, sends an interrupt signal to Programmable Logic Controller (PLC) unit when the consumption levels reach the threshold and therefore prevents resource over-consumption.

Keywords

• Deep Learning
• LSTM
• Sliding Window Technique
• Electricity Consumption
• Textile Industry

References

1. Agga, A., Abbou, A., Labbadi, M., El Houm, Y., and Ali, I. H. O., 2022. CNN-LSTM: An efficient hybrid deep learning architecture for predicting short-term photovoltaic power production. Electric Power Systems Research, 208, 107908.
2. Alazab, M., Khan, S., Krishnan, S. S. R., Pham, Q. V., Reddy, M. P. K., Gadekallu, T. R., 2020. A multidirectional LSTM model for predicting the stability of a smart grid. IEEE Access, 8, 85454-85463.
3. Albuquerque, P. C., Cajueiro, D. O., Rossi, M. D., 2022. Machine learning models for forecasting power electricity consumption using a high dimensional dataset. Expert Systems with Applications, 187, 115917.
4. Aparna, S., 2018. Long short term memory and rolling window technique for modeling power demand prediction. Second International Conference on Intelligent Computing and Control Systems (ICICCS), 1675-1678.
5. Arora, S., and Majumdar, A., 2022. Machine learning and soft computing applications in textile and clothing supply chain: Bibliometric and network analyses to delineate future research agenda. Expert Systems with Applications, 117000.
6. Awan, M. R., González Rojas, H. A., Hameed, S., Riaz, F., Hamid, S., & Hussain, A. (2022). Machine learning-based prediction of specific energy consumption for cut-off grinding. Sensors, 22(19), 7152.
7. Bhatt, A., Ongsakul, W., and Singh, J. G. (2022). Sliding window approach with first-order differencing for very short-term solar irradiance forecasting using deep learning models. Sustainable Energy Technologies and Assessments, 50, 101864.
8. Chen, C., Zhang, Q., Kashani, M. H., Jun, C., Bateni, S. M., Band, S. S., ... & Chau, K. W. (2022). Forecast of rainfall distribution based on fixed sliding window long short-term memory. Engineering Applications of Computational Fluid Mechanics, 16(1), 248-261.

9. Drewil, G. I., Al-Bahadili, R. J., 2022. Air pollution prediction using LSTM deep learning and metaheuristics algorithms. Measurement: Sensors, 24, 100546.
10. Fagerström, J., Bång, M., Wilhelms, D., Chew, M. S., 2019. LiSep LSTM: a machine learning algorithm for early detection of septic shock. Scientific reports, 9(1), 15132.
11. Fang, Y., Zou, Y., Xu, J., Chen, G., Zhou, Y., Deng, W., ... & Chen, J. (2021). Ambulatory cardiovascular monitoring via a machine‐learning‐assisted textile triboelectric sensor. Advanced Materials, 33(41), 2104178.
12. Forootan, M. M., Larki, I., Zahedi, R., and Ahmadi, A., 2022. Machine learning and deep learning in energy systems: A review. Sustainability, 14(8), 4832.
13. Garg, D., and Alam, M., 2020. Deep learning and IoT for agricultural applications. Internet of Things (IoT) Concepts and Applications, 273-284.
14. González García, C., Núñez Valdéz, E. R., García Díaz, V., Pelayo García-Bustelo, B. C., and Cueva Lovelle, J. M., 2019. A review of artificial intelligence in the internet of things. International Journal Of Interactive Multimedia And Artificial Intelligence, 5.
15. Güven, I., and Şimşir, F. (2020). Demand forecasting with color parameter in retail apparel industry using artificial neural networks (ANN) and support vector machines (SVM) methods. Computers & Industrial Engineering, 147, 106678.
16. Ikeda, S., and Nagai, T., 2021. A novel optimization method combining metaheuristics and machine learning for daily optimal operations in building energy and storage systems. Applied Energy, 289, 116716.
17. Kulanuwat, L., Chantrapornchai, C., Maleewong, M., Wongchaisuwat, P., Wimala, S., Sarinnapakorn, K., & Boonya-aroonnet, S. (2021). Anomaly detection using a sliding window technique and data imputation with machine learning for hydrological time series. Water, 13(13), 1862.
18. Lee, M. H. L., Ser, Y. C., Selvachandran, G., Thong, P. H., Cuong, L., Son, L. H., Nguyen, T. T., and Gerogiannis, V. C., 2022. A comparative study of forecasting electricity consumption using machine learning models. Mathematics, 10(8), 1329.
19. Majumdar, A., Jindal, A., Arora, S., and Bajya, M. (2022). Hybrid neuro-genetic machine learning models for the engineering of ring-spun cotton yarns. Journal of Natural Fibers, 19(16), 15164-15175.
20. Malakouti, S. M., Ghiasi, A. R., Ghavifekr, A. A., Emami, P., 2022. Predicting wind power generation using machine learning and CNN-LSTM approaches. Wind Engineering, 46(6), 1853-1869.
21. Medina, H., Peña, M., Siguenza-Guzman, L., and Guamán, R. (2022, April). Demand Forecasting for Textile Products Using Machine Learning Methods. In Applied Technologies: Third International Conference, ICAT 2021, Quito, Ecuador, October 27–29, 2021, Proceedings (pp. 301-315). Cham: Springer International Publishing.
22. Milić, S. D., Đurović, Ž., and Stojanović, M. D. (2023). Data science and machine learning in the IIoT concepts of power plants. International Journal of Electrical Power & Energy Systems, 145, 108711.
23. Mirandola, I., Berti, G. A., Caracciolo, R., Lee, S., Kim, N., & Quagliato, L. (2021). Machine learning-based models for the estimation of the energy consumption in metal forming processes. Metals, 11(5), 833.
24. Oprea, S. V., Bâra, A., Puican, F. C., and Radu, I. C., 2021. Anomaly detection with machine learning algorithms and big data in electricity consumption. Sustainability, 13(19), 10963.
25. Shine, P., Murphy, M. D., Upton, J., and Scully, T., 2018. Machine-learning algorithms for predicting on-farm direct water and electricity consumption on pasture based dairy farms. Computers and electronics in agriculture, 150, 74-87.
26. Sun, R., Huang, W., Dong, Y., Zhao, L., Zhang, B., Ma, H., ... & Li, X. (2022). Dynamic Forecast of Desert Locust Presence Using Machine Learning with a Multivariate Time Lag Sliding Window Technique. Remote Sensing, 14(3), 747.
27. Sundar, G., and Patchaiammal, P. (2022, February). Comprehensive Deep Recurrent Artificial Neural Network (CDRANN): Evolutionary Model for Future Prediction. In International Conference on Computing, Communication, Electrical and Biomedical Systems (pp. 217-234). Cham: Springer International Publishing.
28. Wang, J. Q., Du, Y., Wang, J., 2020. LSTM based long-term energy consumption prediction with periodicity. Energy, 197, 117197.
29. Vafaeipour, M., Rahbari, O., Rosen, M. A., Fazelpour, F., Ansarirad, P., 2014. Application of sliding window technique for prediction of wind velocity time series. International Journal of Energy and Environmental Engineering, 5, 1-7.
30. Vu, C. C., and Kim, J. (2018). Human motion recognition by textile sensors based on machine learning algorithms. Sensors, 18(9), 3109.
31. Yang, W., Sun, S., Hao, Y., and Wang, S., 2022. A novel machine learning-based electricity price forecasting model based on optimal model selection strategy. Energy, 238, 121989.
32. Yasir, M., Ansari, Y., Latif, K., Maqsood, H., Habib, A., Moon, J., Rho, S., 2022. Machine learning–assisted efficient demand forecasting using endogenous and exogenous indicators for the textile industry. International Journal of Logistics Research and Applications, 1-20.
33. Yucesan, M., Pekel, E., Celik, E., Gul, M., Serin, F., 2021. Forecasting daily natural gas consumption with regression, time series and machine learning based methods. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 1-16.