Journals / Politeknik Dergisi / 2021 / Cilt: 24 - Sayı: 4

Development of Ice Prediction Mobile Application

Buzlanma Tahmini Yapan Mobil Uygulama Geliştirilmesi

Pages
1543–1555
Publication Date
2021-12-01
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

However, in the developed prediction algorithm, although the classification accuracy is lower compared to others, it has been observed that the number of samples used in training increases, the icing prediction accuracy increases in direct proportion.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Cold weather and heavy winter conditions cause icing on the roads, and therefore many fatal, injured and materially damaged accidents occur every year.
📊
Key Findings & Data:After the training of the system is completed, weather forecast information is obtained from the meteorology and icing forecast is made for the next 12 hours on the developed mobile application.
💡
Conclusion & Impact:However, in the developed prediction algorithm, although the classification accuracy is lower compared to others, it has been observed that the number of samples used in training increases, the icing prediction accuracy increases in direct proportion.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Makine öğrenmesi · mobil uygulama · buzlanma

Abstract

Cold weather and heavy winter conditions cause icing on the roads, and therefore many fatal, injured and materially damaged accidents occur every year. In this study, an icing prediction algorithm and mobile application has been developed to prevent accidents caused by icing on the roads. With the developed application, it is aimed to give preliminary information about the formation of icing in line with the routes of the drivers. In the study, the temperature, dew point, sensed temperature, wind intensity, wind direction, relative humidity, wind speed input parameters were taken from the road condition sensor and weather stations. At the exit, double classification was made with icing information. After the training of the system is completed, weather forecast information is obtained from the meteorology and icing forecast is made for the next 12 hours on the developed mobile application. In addition, in order to measure and compare the accuracy of the developed system, the multi-layer perceptron (MLP) neural network model and linear and nonlinear support vector machines (SVM) methods are used. Considering the classification accuracy of the algorithms used in the study, based on the total number of correctly classified samples, it was seen that the model of the MLP performed best with 87,26% accuracy rate, followed by the linear SVM model with 86,32% and our proposed model with 75,47% accuracy rate. However, in the developed prediction algorithm, although the classification accuracy is lower compared to others, it has been observed that the number of samples used in training increases, the icing prediction accuracy increases in direct proportion.

Özet

Soğuk hava ve ağır kış şartları, yollarda buzlanmaya sebep olmakta ve bu nedenle her yıl birçok ölümlü, yaralanmalı ve maddi hasarlı kaza meydana gelmektedir. Bu çalışmada yollardaki buzlanmadan kaynaklı kazaların önlenmesine yönelik bir buzlanma tahmin algoritması ve mobil uygulama geliştirilmiştir. Geliştirilen uygulama ile sürücülerin güzergâhları doğrultusunda buzlanma oluşumu ile ilgili ön bilgi verilmesi amaçlanmaktadır. Çalışmada yol durum sensörü ve hava istasyonlarından alınan sıcaklık, çiğ noktası, hissedilen sıcaklık, rüzgâr şiddeti, rüzgâr yönü, bağıl nem, rüzgâr hızı giriş parametreleri olarak kullanılmıştır. Çıkışta ise buzlanma bilgisi ile ikili sınıflandırma yapılmıştır. Sistemin eğitimi tamamlandıktan sonra meteorolojiden hava durumu tahmin bilgisi alınarak, geliştirilen mobil uygulama üzerinde gelecek 12 saat için buzlanma tahmini yapılmaktadır. Ayrıca geliştirilen sistemin doğruluğunu ölçmek ve karşılaştırma yapabilmek için sınıflandırma alanında en çok kullanılan yöntemlerden çok katmanlı algılayıcı (ÇKA) sinir ağı modeli ile doğrusal ve doğrusal olmayan destek vektör makineleri (DVM) yöntemleri kullanılmıştır. Çalışmada kullanılan algoritmaların sınıflandırma doğruluğuna bakıldığında, toplam doğru sınıflandırılan örnek sayısı temel alındığında ÇKA modelinin %87,26 doğruluk oranı ile en iyi sonucu verdiği, ardından %86,32 ile doğrusal DVM modelinin geldiği önerilen modelimizin ise %75,47 doğruluk oranına sahip olduğu görülmüştür. Ancak geliştirilen tahmin algoritmasında sınıflandırma doğruluğu diğerlerine kıyasla daha az olmasına rağmen eğitimde kullanılan örnek sayısı arttıkça, buzlanma tahmin doğruluğunun da doğru orantılı olarak arttığı gözlemlenmiştir.

🔖

Cite This Article

KABAOĞLU, H., UÇAR, E., DURAN, F. (2021). Development of Ice Prediction Mobile Application. Politeknik Dergisi, Cilt: 24 - Sayı: 4, 1543–1555. https://ajindex.com/articles/422192

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (13)

  1. Balbay, A., Esen, M., “Yollardaki kar ve buzu önleyici sistemler." Fırat Üniversitesi Doğu Araştırmaları Dergisi, 6.1: 169-174, (2007)
  2. Joshi, A., Kamble, B., Joshi, V., Kajale, K. and Dhange, N., “Weather Forecasting and Climate Changing Using Data Mining Application”. International Journal of Advanced Research in Computer and Communication Engineering, 4(3): 19-21, (2015)
  3. Drezga, I. and Rahman, S., “Short-term load forecasting with local ANN predictors”, IEEE Transactions on Power Systems, 14(3): 844-850, (1999)
  4. Chen, S. M. and Hwang, J. R., “Temperature prediction using fuzzy time series”, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 30(2): 263-275, (2000)
  5. De Giorgi, M. G., Ficarella, A. and Tarantino, M., “Assessment of the benefits of numerical weather predictions in wind power forecasting based on statistical methods”, Department of Engineering for Innovation, 36(7): 3968-3978, (2011)
  6. Lee, M., & Kim, M., “Development of Real-Time Road Surface Condition Determination Algorithm Using an Automatic Weather System”, In 2016 6th International Conference on IT Convergence and Security (ICITCS), Prague, Czech Republic, pp. 1-2, (2016)
  7. Yang C. H., Park M. S.,Yun D. G., “A Road Surface Temperature Prediction Modeling for Road Weather Information System”, Journal of Korean Society of Transportation, Vol. 29, No. 2, pp. 123-131, (2011)
  8. Krsmanc R., SajnSlak A., Carman S., Korosec M., “Metro Model Testing at Slovenian Road Weather Stations and Suggestions for Further Improvements”, 16th International Road Weather Conference in Helsinki, (2012)
  9. Bogren, J., Gustavsson, T., Nordin, L., Ekström, P., & Sjölander, P. O., “SRIS—Slippery Road Information System”, In Proc., of the 14th Standing International Road Weather Commission—SIRWEC Conf, Stockholm, Sweden, (2008)
  10. Yang C. H., Yun D. G., Sung J. G., “Validation of a Road Surface Temperature Prediction Model Using Real-time Weather Forecasts”, KSCE Journal of Civil Engineering, 16, pp. 1289-1294, (2012)
  11. Jokela, M., Kutila, M., & Le, L., “Road condition monitoring system based on a stereo camera”, In 2009 IEEE 5th International conference on intelligent computer communication and processing, Romania, pp. 423-428, (2009
  12. Yamada, M., Oshima, T., Ueda, K., Horiba, I., & Yamamoto, S., “A study of the road surface condition detection technique for deployment on a vehicle”, JSAE review, 24(2): 183-188, (2003)
  13. Dregza, I. and Rahman,