Image of Prediksi Penggunaan Energi Listrik Di Provinsi Riau Menggunakan Metode Long Short-Term Memory
Penanda Bagikan

CD Skripsi

Prediksi Penggunaan Energi Listrik Di Provinsi Riau Menggunakan Metode Long Short-Term Memory

The growth of electricity energy consumption continues to increase every year, requiring a reliable prediction system to support effective energy planning and management. This research aims to predict electricity consumption in Riau Province using the Long Short-Term Memory (LSTM) method with deep learning algorithms and to determine the accuracy of this method in predicting electricity demand. The data used is historical monthly electricity consumption data in Riau Province from October 2022 to December 2024, totaling 27 rows of data, supported by average temperature data from BMKG and monthly electricity customer data. In this study, the dataset division used is 70% for training data and 30% for testing data. This LSTM model is designed by performing hyperparameter tuning to obtain optimal parameters in the form of a two-layer LSTM architecture with LSTM1 (256 units) and LSTM2 (128 units), dense layer (32 units), learning rate (0.01) and number of epochs (68) to obtain the best prediction results. Model performance evaluation uses Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) metrics. The research results show that the RMSE value (16.36) with good accuracy category and MAPE percentage (1.7%) with excellent accuracy category. It can be concluded that the model successfully predicts electricity consumption in Riau Province.

Kata Kunci: Deep Learning, Electricity Consumption Forecasting, hyperparameter tuning, LSTM, Riau Province.

Ketersediaan
#
Perpustakaan Universitas Riau 2103113576
2103113576
Tersedia
Informasi Detail
Judul Seri
-
No. Panggil
2103113576
Penerbit
Pekanbaru : Universitas Riau FMIPA Sistem Informasi.,
Deskripsi Fisik
-
Bahasa
Indonesia
ISBN/ISSN
-
Klasifikasi
2103113576
Tipe Isi
-
Tipe Media
-
Tipe Pembawa
-
Edisi
-
Subjek
SISTEM INFORMASI
Info Detail Spesifik
-
Pernyataan Tanggungjawab
Versi lain/terkait

Tidak tersedia versi lain

Lampiran Berkas
Komentar
Anda harus masuk sebelum memberikan komentar