Prediksi Tinggi Muka Air (TMA) Untuk Deteksi Dini Bencana Banjir Menggunakan SVR-TVIWPSO

Penulis

  • Arief Andy Soebroto
  • Imam Cholissodin
  • Randy Cahya Wihandika
  • Maria Tenika Frestantiya
  • Ziya El Arief

DOI:

https://doi.org/10.25126/jtiik.201522126

Abstrak

Abstrak

Banjir merupakan salah satu jenis bencana alam yang tidak dapat diprediksi kedatangannya, salah satu penyebabnya adalah adanya hujan yang terus – menerus(dari peristiwa alam). Faktor penyebab banjir dari segi meteorologi yaitu curah hujan yang tinggi dan air laut yang sedang pasang sehingga mengakibatkan tinggi permukaan air meningkat. Analisis terhadap data curah hujan serta tinggi permukaan air setiap periodenya dirasa masih belum dapat menyelesaikan permasalahan yang ada. Oleh karena itu, pada penelitian ini diusulkan teknik integrasi metode Time Variant Inertia Weight Particle Swarm Optimization(TVIWPSO) dan Support Vector Regression(SVR). Implementasi memadukan metode Regresi yaitu SVR untuk forecasting TMA, sedangkan TVIWPSO digunakan untuk mengoptimalisasi parameter – parameter yang digunakan di dalam SVR untuk memperoleh kinerja yang maksimal dan hasil yang akurat. Harapannya sistem ini akan dapat membantu mengatasi permasalahan untuk pendeteksian dini bencana banjir karena faktor cuaca yang tidak menentu. Hasil pengujian yang didapat dari 10 data bulanan yang berbeda menunjukkan bahwa didapatkan nilai error terkecil sebesar 0.00755 dengan menggunakan Mean Absolute Error untuk data Juni 2007 dengan menggunakan integrasi metode SVR-TVIWPSO.

Kata Kunci : Support Vector Regression, Tinggi Muka Air, Time Variant Inertia Weight Particle Swarm Optimization.

Abstract

Flood is one type of natural disaster that can not be predicted its arrival, one reason is the rain that constantly occurs (from natural events). Factors that cause flooding in terms of meteorology are high rainfall and sea water was high, resulting in high water level increases. Analysis of rainfall data and water level in each period it is still not able to solve existing problems. Therefore, in this study the method proposed integration techniques Time Variant Inertia Weight Particle Swarm Optimization (TVIWPSO) and Support Vector Regression (SVR). Implementation combines regression method for forecasting TMA is SVR, while TVIWPSO used to optimize parameters that used in the SVR to obtain maximum performance and accurate results. Hope this system will be able to help solve the problems for the early detection of floods due to erratic weather. The result of forecasting experiment in water level forecasting from 10 monthly different data show that the smallest error rate is amount to 0.00755 using Mean Absolute Error for June 2007 with the integration method SVR-TVIWPSO.

Keywords: Support Vector Regression, water level, Time Variant Inertia Weight Particle Swarm Optimization.

Downloads

Download data is not yet available.

Referensi

ALWEE, R., MARIYAM, S. & SALLEHUDDIN R. 2013. Hybrid Support Vector Regression and Autoregressive Integrated Moving Average Models Improved by Particle Swarm Optimization for Property Crime Rates Forecasting with Economic Indicators. Hindawi Publishing Corporation The Scientific World Journal 2013.

BU-HAI, S., LING, X.,QI-PENG, W. & YOU LIANG, Z., 2014. Active Heave Compensation Prediction Research for Deep Sea Homework Crane based on SVR-KPSO. Proceedings of the 33rd Chinese Control Conference, Nanjing, China 2014.

CHAI, T. & DRAXLER R.R. 2014. Root Mean Square Error (RMSE) or Mean Absolute Error (MAE)? – Arguments Against Avoiding RMSE in The Literature. Published by Copernicus Publications on behalf of the European Geosciences Union, Geosci. Model Dev., 7, 1247–1250, 2014.

CHEN, H., YANG, B. & WANG, G. 2011. A Novel Bankruptcy Prediction Model Based on an Adaptive Fuzzy K-Nearest Neighbor Method. College of Computer Science and Technology, Jilin University, Changchun 130012, China.

GUO, Y., 2009. An Integrated PSO for Parameter Determination and Feature Selection of SVR and its Application in SLTF. Proceedings of the Eighth International Conference on Machine Learning and Cybernetics, Baoding, 12-15 July 2009.

JAVED, F., GREGORY, S.H.C. & SAVKIN A.V. 2009. RBF Kernel Based Support Vector Regression to Estimate The Blood Volume and Heart Rate Responses During Hemodialysis. 31st Annual International Conference of the IEEE EMBS Minneapolis, Minnesota, USA, September 2-6, 2009.

LIU, H. 2010. Internet Public Opinion Hotspot Detection and Analysis Based on KMEANS and SVM Algorithm. 2010 International Conference of Information Science and Management Engineering.

LI, Y. & CHEN, P.2012. A Parallel SVR Model for Short Term Load Forecasting Based on Windows Azure Platform. School of Control and Computer Engineering North China Electric Power University Bei Jing, China.

LU, C., LEE T. & LIAN, C. 2010. Sales Forecasting of IT Products using A Hybrid MARS and SVR MoDel. 2010 IEEE International Conference on Data Mining Workshops.

MA, D., DING, N., WANG, J. & CUI J. 2010. Research on Flood Submergence Analysis System Based on ArcEngine Component Library. Ministry of Construction P.R.China Science & Technology Planning Project(2010-K9-24.

MEESUK, V., VOJINOVIC, Z. & MYNETT, A.E. 2012. Using Multidimensional Views of Photographs for Flood Modelling. Hydro and Agro Informatics Institute, Ministry of Science and Technology, 978-1-4673-1975-1/12/$31.00 ©2012 IEEE.

NING, A. & ZHANG, X. 2010. A Speech Recognition System Based on Fuzzy Neural Network Optimized by Time Variant PSO. 2010 International Conference on Computational Aspects of Social Networks.

SAHAK, R., MANSOR, Y.K.L, ZABIDI, A. & YASSIN A.I.M. 2011. Orthogonal Least Square and Optimized Support Vector Machine with Polynomial Kernel for Classifying Asphyxiated Infant Cry. 2011 IEEE International Conference on Signal and Image Processing Applications (ICSIPA 2011).

SAUD, I. 2007. Kajian Penanggulangan Banjir di Wilayah Pematusan Surabaya Barat. Jurnal Aplikasi Vol.3, No.1, ISSN : 1937-753X.

SENGUPTA, A. & MISHRA, V.K. 2014. Time Varying vs. Fixed Acceleration Coefficient PSO Driven Exploration during High Level Synthesis: Performance and Quality Assessment. 2014 13th International Conference on Information Technology.

SUTIKNO, BEKTI, R., SUSANTI, P. & ISTRIANA. 2010. Prakiraan Cuaca Dengan Metode Autoregressive Integrated Moving Average, Neural Network, dan Adaptive Splines Threshold Autoregression di Stasiun Juanda Surabaya. Jurnal SAINS Dirgantara.

VIJAYAKUMAR, S. & WU S. 1999. Sequential Support Vector Classifiers and Regression. RIKEN Brain Science Institute 1999, pp.610 – 619.

TONDE, C. & ELGAMMAL, A. 2014. Simultaneous Twin Kernel Learning using Polynomial Transformations for Structured Prediction. 2014 IEEE Conference on Computer Vision and Pattern Recognition.

WADEY, M.P., COPE, S.N., NICHOLLS, R.J. & MCHUGH, K. 2015. Coastal Flood Analysis and Visualisation for A Small Town. Ocean & Coastal Management 116 (2015) 237e247.

YOUNGJOO, K., JONGGEOL, P., ATSUHIRO Y. & KAZUHIKO F. 2012. Estimation Of Flood Volume in Chao Phraya River Basin, Thailand, From Modis Images Couppled with Flood Inundation Level. International Centre for Water Hazard and Risk Management (ICHARM) under the auspices of UNESCO, Public Works Research Institute (PWRI), 978-1-4673-1159-5/12/$31.00 ©2012 IEEE.

ZHAO, B., DING, X. & CAO, H. 2011. The Recognition of River Channel Flood Infiltration Capacity Based on Numerical Simulation. Xi'an Research Institute of China Coal Technology & Engineering Group Corp Xi’an , China 978-1-61284-34 0-7/11/$26.00 ©2011 IEEE.

ZHENG, W. 2012. The Flood Monitoring Information System Framework Based on Multi-source Satellite Remote Sensing Data. 2012 International Conference on System Science and Engineering June 30-July 2, 2012, Dalian, China.

Unduhan

Diterbitkan

22-07-2015

Terbitan

Bagian

Teknologi Informasi

Cara Mengutip

Prediksi Tinggi Muka Air (TMA) Untuk Deteksi Dini Bencana Banjir Menggunakan SVR-TVIWPSO. (2015). Jurnal Teknologi Informasi Dan Ilmu Komputer, 2(2), 79-86. https://doi.org/10.25126/jtiik.201522126