Prediksi Kebutuhan Energi Litrik di Provinsi Jawa Tengah Menggunakan Metode Adaptive Neuro-Fuzzy Inference system (ANFIS)

Ramadhan, Iqbal (2023) Prediksi Kebutuhan Energi Litrik di Provinsi Jawa Tengah Menggunakan Metode Adaptive Neuro-Fuzzy Inference system (ANFIS). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02311940000167_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 2 February 2026. Download (1MB) | Request a copy

Abstract

Energi listrik sangat penting untuk berbagai fungsi, seperti rumah tangga, industri, dan komersial. PT. PLN (persero) memproduksi energi untuk memenuhi kebutuhan listrik Indonesia, yang terus meningkat setiap tahunnya, termasuk di Jawa Tengah. Oleh karena itu, PT. PLN (persero) perlu melakukan perencanaan produksi energi yang tepat agar dapat memenuhi kebutuhan pelanggan dengan tepat. Metode Adaptive Neuro-Fuzzy Inference System (ANFIS) digunakan untuk memprediksi kebutuhan energi listrik wilayah Jawa Tengah di masa depan. ANFIS adalah teknik optimasi yang menggabungkan gagasan neural network dengan fuzzy logic. Dengan menggabungkan keduanya, ANFIS dapat mengenali pola dan menyesuaikannya dengan perubahan lingkungan serta menggunakan pengetahuan manusia untuk membuat keputusan. Metode ini memungkinkan untuk membuat prediksi yang lebih akurat. Setelah dilakukan training terhadap data konsumsi energi listrik pada tahun 2010-2022 didapatkan hasil RMSE sebesar 2,147 %. Setelah dilakukan perhitungan nilai MAPE untuk mengetahui keakuratan hasil prediksi didapatkan hasil nilai MAPE sebesar 11,3169 % dimana hasil masih bisa dikatakan akurat dikarenakan nilai MAPE 10% - 20%.
====================================================================================================================================
Electricity is crucial for various functions, including household, industrial, and commercial sectors. PT. PLN (persero) is responsible for supplying electricity to meet the demands of Indonesia, which continuously increase each year, including in the region of Central Java. Therefore, PT. PLN (persero) needs to accurately plan the production of electricity to fulfill the customers' needs effectively. The Adaptive Neuro-Fuzzy Inference System (ANFIS) method is employed to predict the future electricity requirements for electric vehicles. ANFIS is an optimization technique that combines neural network concepts with fuzzy logic. By merging these two approaches, ANFIS can recognize patterns and adapt them to environmental changes, while utilizing human knowledge to make decisions. This method enables the generation of more accurate predictions. After training the data on electricity consumption in 2010-2022, an RMSE (Root Mean Square Error) of 2,147 %. was obtained. The calculation of MAPE (Mean Absolute Percentage Error) was then conducted to assess the accuracy of the prediction results, yielding a MAPE value of 11,3169 %. This result can still be considered
accurate as the MAPE value is 10% - 20%

Item Type:	Thesis (Other)
Additional Information:	RSF 333.793 2 IQB p 2023
Uncontrolled Keywords:	Kebutuhan energi listrik, Prediksi, ANFIS, MAPE
Subjects:	Q Science T Technology > T Technology (General)
Divisions:	Faculty of Industrial Technology > Physics Engineering
Depositing User:	Iqbal Ramadhan
Date Deposited:	12 Feb 2024 02:38
Last Modified:	04 Nov 2024 06:57
URI:	http://repository.its.ac.id/id/eprint/106882

Actions (login required)

View Item