Model Deteksi Pemborosan Energi Listrik Rumah Tangga Berbasis Algoritma Autoencoder dengan Konsep Internet of Things: Household Electricity Waste Detection Model Based on Autoencoder Algorithm with Internet of Things Concept

Sriwahyuningsih Piu A; Muhammad Rizal; Maya Itasari

doi:10.57152/malcom.v6i1.2338

Authors

Sriwahyuningsih Piu A Universitas Dipa Makassar
Muhammad Rizal Politeknik Negeri Ujung Pandang
Maya Itasari Politeknik Negeri Ujung Pandang

DOI:

https://doi.org/10.57152/malcom.v6i1.2338

Keywords:

Autoencoder, Internet of Things, Listrik Rumah Tangga, Pemborosan Energi, Unsupervised

Abstract

Penelitian ini menyajikan pendekatan unsupervised untuk mendeteksi pemborosan energi listrik rumah tangga dari fitur jendela waktu berbasis IoT menggunakan Autoencoder (AE). Pipeline meliputi imputasi median, normalisasi RobustScaler, pemisahan berbasis waktu (70/15/15), pelatihan AE dengan AdamW dan early stopping, serta ambang adaptif per-rumah dari persentil 99,5% error rekonstruksi. Evaluasi dilakukan pada dataset energi listrik (23.028 window; dua rumah, dua kanal) dengan fitur statistik tegangan, arus, daya, faktor daya (rata-rata/deviasi standar/p95), slope, dan transien. Grid search memilih AE dangkal (depth=1, width=128, bottleneck=16, dropout=0, ReLU, lr=1e-4, wd=1e-5). Pada test set (n=3.455), rerata error rekonstruksi sebesar 3,17×10?³. Dengan treshold per-rumah (homeA=0,0634, homeB=0,0859), sistem menandai ~0,52% window sebagai pemborosan/anomali, dengan insidensi lebih tinggi pada homeA (0,81%) dibanding homeB (0,23%). Atribusi error per-fitur menonjolkan metrik faktor daya serta slope/variabilitas tegangan, arus, dan daya, pola yang konsisten dengan beban tidak efisien atau switching mendadak. Metode ini siap dioperasikan pada ekosistem IoT dan memberikan indikator yang dapat ditafsirkan untuk tindakan penghematan tanpa memerlukan label.

Downloads

Download data is not yet available.

References

Z. Chuncheng, Z. Lili, and L. Qing, “Research on Electricity Demand Forecasting Model and Application of China Adapting to Supply-Side Structural Reform,” in 2019 2nd Asia Conference on Energy and Environment Engineering (ACEEE), 2019, pp. 69–73. doi: 10.1109/ACEEE.2019.8816907.

M. A. McNeil, N. Karali, and V. Letschert, “Forecasting Indonesia’s electricity load through 2030 and peak demand reductions from appliance and lighting efficiency,” Energy for Sustainable Development, vol. 49, pp. 65–77, Apr. 2019, doi: 10.1016/j.esd.2019.01.001.

A. Copiaco, Y. Himeur, A. Amira, W. Mansoor, F. Fadli, and S. Atalla, “Exploring Deep Time-Series Imaging for Anomaly Detection of Building Energy Consumption,” in 2022 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), 2022, pp. 1–5. doi: 10.1109/CSDE56538.2022.10089265.

Z. K. Shahid, S. Saguna, and C. Åhlund, “Variational Autoencoders for Anomaly Detection and Transfer Knowledge in Electricity and District Heating Consumption,” IEEE Trans Ind Appl, vol. 60, no. 5, pp. 7437–7450, 2024, doi: 10.1109/TIA.2024.3425805.

H.-A. Dang, V.-D. Dao, C.-D. Dang, N. H. Nguyen, and A. Hoang, “Identifying Abnormal Energy Consumption Patterns in Industrial Settings: Application of Local Outlier Factor Algorithm for a Processing Factory in Vietnam,” in 2023 Asia Meeting on Environment and Electrical Engineering (EEE-AM), 2023, pp. 1–5. doi: 10.1109/EEE-AM58328.2023.10395078.

S. Gupta and B. Gupta, “Energy Anomaly Detection and Modelling on Smart Premises using SDAR,” in 2021 International Conference on Industrial Electronics Research and Applications (ICIERA), 2021, pp. 1–5. doi: 10.1109/ICIERA53202.2021.9726730.

L. G and R. Gopal, “Design and Development of IoT Prototype for Real-Time Theft Detection and Optimization of Electricity Using Machine Learning Techniques,” in 2024 2nd International Conference on Computing and Data Analytics (ICCDA), 2024, pp. 1–4. doi: 10.1109/ICCDA64887.2024.10867342.

S. Madakam, R. Ramaswamy, and S. Tripathi, “Internet of Things (IoT): A Literature Review,” Journal of Computer and Communications, vol. 03, no. 05, pp. 164–173, 2015, doi: 10.4236/jcc.2015.35021.

J. Corona, M. Antunes, and R. L. Aguiar, “Eco-Friendly Intrusion Detection: Evaluating Energy Costs of Learning,” in 2023 IEEE 9th World Forum on Internet of Things (WF-IoT), 2023, pp. 1–6. doi: 10.1109/WF-IoT58464.2023.10539426.

S. Piu, A. Arifin, and M. Rizal, “Optimasi Penggunaan Energi Listrik Bagi Pelanggan Rumah Tangga Berbasis Machine Learning dan Internet of Things,” MALCOM: Indonesian Journal of Machine Learning and Computer Science, vol. 5, no. 1, pp. 84–92, Nov. 2024, doi: 10.57152/malcom.v5i1.1694.

N. S. Krishna, Y. V. P. Kumar, K. P. Prakash, and G. P. Reddy, “Machine Learning and Statistical Techniques for Outlier Detection in Smart Home Energy Consumption,” in 2024 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), 2024, pp. 1–4. doi: 10.1109/eStream61684.2024.10542609.

A. Takiddin, M. Ismail, U. Zafar, and E. Serpedin, “Deep Autoencoder-Based Anomaly Detection of Electricity Theft Cyberattacks in Smart Grids,” IEEE Syst J, vol. 16, no. 3, pp. 4106–4117, 2022, doi: 10.1109/JSYST.2021.3136683.

D. Wang, T. Enlund, J. Trygg, M. Tysklind, and L. Jiang, “Toward Delicate Anomaly Detection of Energy Consumption for Buildings: Enhance the Performance From Two Levels,” IEEE Access, vol. 10, pp. 31649–31659, 2022, doi: 10.1109/ACCESS.2022.3160170.

Y. Choi, T. Y. Ku, and W.-K. Park, “Handling anomaly in residential energy consumption data,” in 2023 14th International Conference on Information and Communication Technology Convergence (ICTC), 2023, pp. 1309–1312. doi: 10.1109/ICTC58733.2023.10392520.

W. Zhuang, W. Zhang, and H. Yuan, “Abnormal Analysis System and Analysis of Residential Electricity Consumption Based on Electricity Data,” in 2024 International Conference on Optimization Computing and Wireless Communication (ICOCWC), 2024, pp. 1–4. doi: 10.1109/ICOCWC60930.2024.10470748.

Arwansyah et al., “Deep Sequence Models for Time Series Data: A Comparative Study and Parameter Fine-Tuning Approach,” in 2024 11th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), 2024, pp. 703–709. doi: 10.1109/EECSI63442.2024.10776052.

T. ?mia?kowski and A. Czy?ewski, “Autoencoder Application for Anomaly Detection in Power Consumption of Lighting Systems,” IEEE Access, vol. 11, pp. 124150–124162, 2023, doi: 10.1109/ACCESS.2023.3330135.

H. Çimen, Y. Wu, Y. Wu, Y. Terriche, J. C. Vasquez, and J. M. Guerrero, “Deep Learning-Based Probabilistic Autoencoder for Residential Energy Disaggregation: An Adversarial Approach,” IEEE Trans Industr Inform, vol. 18, no. 12, pp. 8399–8408, 2022, doi: 10.1109/TII.2022.3150334.

G. Li, S. Peng, C. Wang, J. Niu, and Y. Yuan, “An energy-efficient data collection scheme using denoising autoencoder in wireless sensor networks,” Tsinghua Sci Technol, vol. 24, no. 1, pp. 86–96, 2019, doi: 10.26599/TST.2018.9010002.

C. Wang, S. H. Tindemans, and P. Palensky, “Improved Anomaly Detection and Localization Using Whitening-Enhanced Autoencoders,” IEEE Trans Industr Inform, vol. 20, no. 1, pp. 659–668, 2024, doi: 10.1109/TII.2023.3268685.