Prabowo, Hendri (2019) Peramalan Kualitas Udara di Kota Surabaya untuk Menentukan Kategori Indeks Standar Pencemar Udara. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211540000068-Undergraduate_Theses.pdf Download (10MB) | Preview |
Abstract
Pencemaran udara merupakan salah satu permasalahan yang dihadapi Kota Surabaya. Untuk melihat kualitas udara digunakan Indeks Standar Pencemar Udara (ISPU) yang didapatkan dari alat AQMS yang ada di stasiun SUF. Terdapat tiga stasiun SUF yang masih aktif di Kota Surabaya. Tujuan dari penelitian ini adalah meramalkan lima parameter kualitas udara ISPU yaitu CO, NO2, O3, PM10 dan SO2 di tiga stasiun SUF Kota Surabaya sehingga didapatkan nilai ISPU dan kategori kualitas udara Kota Surabaya. Peramalan yang dilakukan menggunakan beberapa metode yaitu time series regression, ARIMA, FFNN, LSTM, time series regression dengan AR error, time series regression dengan ARMA error, hibrida time series regression dan FFNN serta hibrida time series regression dan LSTM. Data parameter kualitas udara memiliki pola double seasonal. Hasil analisis menunjukan perbedaan model terbaik dalam meramalkan parameter kualitas udara. ARIMA, LSTM, hibrida time series regression dan FFNN, hibrida time series regression dan LSTM masing-masing baik untuk meramalkan tiga variabel. Sedangkan FFNN, time series regression dengan AR error, time series regression dengan ARMA error masing-masing baik untuk satu variabel. Hasil peramalan menunjukan bahwa nilai ISPU secara umum tinggi pada saat jam berangkat kerja dan pulang kerja.
================================================================================================================================
Air pollution is one of the problems in Surabaya. To see the air quality, air pollution index (API) was obtained from the AQMS tools in the SUF station. There are three SUF stations that are still active in Surabaya. The purpose of this research is to forecast five air quality parameters API in the three SUF stations of Surabaya. From the result of forecast, the API value and the air quality category of Surabaya are obtained. The forecast use several methods that is time series regression, ARIMA, FFNN, LSTM, time series regression with AR error, time series regression with ARMA error, hybrid time series regression-FFNN and hybrid time series regression-LSTM. Air quality parameters data has double seasonal pattern. The result of analysis show difference of the best model in forecasting air quality parameters. ARIMA, LSTM, hybrid time series regression-FFNN and hybrid time series regression-LSTM respectively good for predicting three variables. While FFNN, time series regression with AR error and time series regression with ARMA error respectively good for predicting one variable. Forecasting results show that the API value is generally high at the time of leaving work and returning from work.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.535 Pra p-1 2019 |
| Uncontrolled Keywords: | ARIMA, FFNN, Hibrida, ISPU, LSTM, Time Series Regression |
| Subjects: | H Social Sciences > HA Statistics > HA30.3 Time-series analysis Q Science > QA Mathematics > QA280 Box-Jenkins forecasting T Technology > TD Environmental technology. Sanitary engineering > TD883 Air quality management. |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Hendri Prabowo |
| Date Deposited: | 20 Jun 2023 08:29 |
| Last Modified: | 20 Jun 2023 08:29 |
| URI: | http://repository.its.ac.id/id/eprint/64039 |
Actions (login required)
 |
View Item |