Peramalan Data Kualitas Udara Menggunakan Multivariat LSTM di Wilayah Kota Surabaya
Abstract
Abstrak — Peningkatan polusi udara di wilayah perkotaan, termasuk Kota Surabaya, mendorong perlunya pengembangan model peramalan kualitas udara yang akurat dan adaptif. Penelitian ini bertujuan untuk menganalisis hubungan antara parameter meteorologi dan kualitas udara, serta membangun model peramalan menggunakan metode Long Short-Term Memory (LSTM) berbasis data multivariat. Data yang digunakan diperoleh dari Stasiun Pemantauan Kualitas Udara (SPKU) Kebonsari periode Januari 2022– Desember 2024, dengan parameter suhu udara, kelembapan, dan kecepatan angin sebagai input, serta PM10 dan CO sebagai target.Analisis korelasi dilakukan untuk meng identifikasi pengaruh antar parameter, dan hasilnya menunjukkan hubungan signifikan yang dapat dimanfaatkan dalam peramalan. Model LSTM dibangun dengan pendekatan time series dan dilatih menggunakan arsitektur jaringan yang mampu menangkap pola temporal antar variabel. Evaluasi kinerja mo del dilakukan dengan metrik Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), dan Symmetric Me an Absolute Percentage Error (SMAPE). Hasil evaluasi menunjukkan bahwa model menghasilkan MSE sebesar 113.6211, RMSE sebesar 10.6593, MAPE sebesar 49.45%, dan SMAPE sebesar 28.17%, yang mengindikasikan performa peramalan yang cukup baik. Dengan hasil tersebut, model multivariat LSTM memiliki potensi untuk digunakan sebagai alat bantu dalam pemantauan dan pengendalian kualitas udara oleh instansi terkait di Kota Surabaya.
Kata kunci— CO, Kualitas Udara, LSTM, Multivariat, PM10, Peramalan
Downloads
References
[1] A. Amalia, A. Zaidiah, and I. N. Isnainiyah, “Prediksi Kualitas Udara Menggunakan Algoritma K-Nearest Neighbor,” JIPI: J. Ilm. Penelit. dan Pembelajaran Inform., vol. 7, no. 2, pp. 496–507, May 2022, doi: 10.29100/jipi.v7i2.2843.
[2] T. V. Damayanti and R. E. Handriyono, “Monitoring Kualitas Udara Ambien Melalui Stasiun Pemantau Kualitas Udara Wonorejo, Kebonsari dan Tandes Kota Surabaya,” Environ. Eng. J. ITATS, vol. 2, no. 1, pp. 11–18, Mar. 2022, doi: 10.31284/j.envitats.2022.v2i1.2897.
[3] M. N. Atalla and F. Agustina, “Penentuan Daerah Rawan Polusi Udara Menggunakan Logika Fuzzy Metode Mamdani.”
[4] J. A. Ramírez-Montañez, J. D. J. Rangel-Magdaleno, M. A. Aceves-Fernández, and J. M. Ramos-Arreguín, “Modeling of Particulate Pollutants Using a Memory-Based Recurrent Neural Network Implemented on an FPGA,” Micromachines, vol. 14, no. 9, p. 1804, Sep. 2023, doi: 10.3390/mi14091804.
[5] V. I. Kontopoulou, A. D. Panagopoulos, I. Kakkos, and G. K. Matsopoulos, “A Review of ARIMA vs. Machine Learning Approaches for Time Series Forecasting in Data Driven Networks,” Future Internet, vol. 15, no. 8, p. 255, Jul. 2023, doi: 10.3390/fi15080255.
[6] G. Cican, A.-N. Buturache, and R. Mirea, “Applying Machine Learning Techniques in Air Quality Prediction—A Bucharest City Case Study,” Sustainability, vol. 15, no. 11, p. 8445, May 2023, doi: 10.3390/su15118445.
[7] Y. Chen, S. Cui, P. Chen, Q. Yuan, P. Kang, and L. Zhu, “An LSTM-based neural network method of particulate pollution forecast in China,” Environ. Res. Lett., vol. 16, no. 4, p. 044006, Apr. 2021, doi: 10.1088/1748-9326/abe1f5.
[8] A. Azzouni and G. Pujolle, “A Long Short-Term Memory Recurrent Neural Network Framework for Network Traffic Matrix Prediction,” Jun. 08, 2017, arXiv: arXiv:1705.05690, doi: 10.48550/arXiv.1705.05690.
[9] M. A. A. Bakar, N. M. Ariff, M. S. Mohd Nadzir, O. L. Wen, and F. N. A. Suris, “Prediction of Multivariate Image Air Quality Time Series Data using Long Short-Term Memory Network,” Malays. J. Fundam. Appl. Sci., vol. 18, no. 1, pp. 52–59, Feb. 2022, doi: 10.11113/mjfas.v18n1.2393.
[10] Muh. A. Faishol, E. Endroyono, and A. N. Irfansyah, “Predict Urban Air Pollution in Surabaya Using Recurrent Neural Network – Long Short Term Memory,” JUTI: J. Ilm. Teknol. Inf., pp. 102–114, Jul. 2020, doi: 10.12962/j24068535.v18i2.a988.
[11] R. T. D. Kurniawati, R. Rahmawati, and Y. Wilandari, “Pengelompokan Kualitas Udara Ambien Menurut Kabupaten/Kota di Jawa Tengah Menggunakan Analisis Klaster.”
[12] F. M. Ottaviani and A. D. Marco, “Multiple Linear Regression Model for Improved Project Cost Forecasting,” Procedia Comput. Sci., vol. 196, pp. 808–815, 2022, doi: 10.1016/j.procs.2021.12.079.
[13] R. T. Alegado and G. M. Tumibay, “Statistical and Machine Learning Methods for Vaccine Demand Forecasting: A Comparative Analysis,” J. Comput. Commun., vol. 08, no. 10, pp. 37–49, 2020, doi: 10.4236/jcc.2020.810005.
[14] V. Papastefanopoulos, P. Linardatos, T. Panagiotakopoulos, and S. Kotsiantis, “Multivariate Time-Series Forecasting: A Review of Deep Learning Methods in Internet of Things Applications to Smart Cities,” Smart Cities, vol. 6, no. 5, pp. 2519–2552, Sep. 2023, doi: 10.3390/smartcities6050114.
[15] M. E. S. Wicaksono, G. M. Arya Sasmita, and I. P. A. Eka Pratama, “Peramalan Kualitas Udara di Kota Jakarta Pusat dengan Metode Long Short-Term Memory dan Support-Vector Regression,” JITTER: J. Ilm. Teknol. dan Komput., vol. 4, no. 1, p. 1576, Jan. 2023, doi: 10.24843/JTRTI.2023.v04.i01.p03.
[16] S. D. Pasupuleti and S. Ludwig, “Forecasting Air Quality: A Comparative Study of Time Series Approaches,” Sep. 07, 2023, Preprints, doi: 10.22541/au.169406670.05499290/v1.
[17] K. Fahim, S. A. Sholichah, L. Hanafi, S. D. Surjanto, I. Herisman, and W. F. Doctorina, “Prediksi Indeks Standar Pencemar Udara (ISPU) di Kota Surabaya Menggunakan Rantai Markov Waktu Diskrit,” J. Ilm. Soulmath: J. Edukasi Pendidik. Mat., vol. 12, no. 2, pp. 121–134, Aug. 2024, doi: 10.25139/smj.v12i2.8684.
[18] G. Naresh and D. B. Indira, “Air Pollution Prediction using Multivariate LSTM Deep Learning Model.”
[19] A. P. Baluk, H. Yasin, and Sugito, “Peramalan Tinggi Gelombang Laut Dengan Metode Vector Autoregressive-Radial Basis Function Network (VAR-RBFN),” J. Stat.: J. Ilm. Teori dan Apl. Stat., vol. 13, no. 2, pp. 39–46, Dec. 2020, doi: 10.36456/jstat.vol13.no2.a3270.
[20] H. Hajmohammadi and B. Heydecker, “Multivariate Time Series Modelling for Urban Air Quality,” Urban Clim., vol. 37, p. 100834, May 2021, doi: 10.1016/j.uclim.2021.100834.
[21] R. Espinosa, J. Palma, F. Jiménez, J. Kamińska, G. Sciavicco, and E. Lucena-Sánchez, “A Time Series Forecasting Based Multi-Criteria Methodology for Air Quality Prediction,” Appl. Soft Comput., vol. 113, p. 107850, Dec. 2021, doi: 10.1016/j.asoc.2021.107850.
[22] B. S. Freeman, G. Taylor, B. Gharabaghi, and J. Thé, “Forecasting Air Quality Time Series Using Deep Learning,” J. Air Waste Manag. Assoc., vol. 68, no. 8, pp. 866–886, Aug. 2018, doi: 10.1080/10962247.2018.1459956.
[23] T. Mathonsi and T. L. Van Zyl, “A Statistics and Deep Learning Hybrid Method for Multivariate Time Series Forecasting and Mortality Modeling,” Forecasting, vol. 4, no. 1, pp. 1–25, Dec. 2021, doi: 10.3390/forecast4010001.
[24] R. Navares and J. L. Aznarte, “Predicting Air Quality with Deep Learning LSTM: Towards Comprehensive Models,” Ecol. Inform., vol. 55, p. 101019, Jan. 2020, doi: 10.1016/j.ecoinf.2019.101019.
[25] Stavelin Abhinandithe K, Madhu B, Somanathan Balasubramanian, and Sridhar Ramachandran, “Forecasting Multivariate Time-Series Data Using LSTM Neural Network in Mysore District, Karnataka,” Indian J. Public Health Res. Dev., vol. 13, no. 4, pp. 284–289, Oct. 2022, doi: 10.37506/ijphrd.v14i4.18631.
[26] A. G. Salman, Y. Heryadi, E. Abdurahman, and W. Suparta, “Single Layer & Multi-layer Long Short-Term Memory (LSTM) Model with Intermediate Variables for Weather Forecasting,” Procedia Comput. Sci., vol. 135, pp. 89–98, 2018, doi: 10.1016/j.procs.2018.08.153.
[27] Y. Yin and P. Shang, “Forecasting Traffic Time Series with Multivariate Predicting Method,” Appl. Math. Comput., vol. 291, pp. 266–278, Dec. 2016, doi: 10.1016/j.amc.2016.07.017.
[28] A. W. Saputra, A. P. Wibawa, U. Pujianto, A. B. Putra Utama, and A. Nafalski, “LSTM-based Multivariate Time-Series Analysis: A Case of Journal Visitors Forecasting,” Ilk. J. Ilm., vol. 14, no. 1, pp. 57–62, Apr. 2022, doi: 10.33096/ilkom.v14i1.1106.57-62.
