IoT-Based Sensor System for Electricity Consumption Forecasting in Boarding Rooms Using Kalman Filter Algorithm

Dewi Humeira Amriah Dewi; Farid; Muhammad Fathur Rahman N; Riesa Krisna Astuti Sakir; Muh. Erdin; Nurfitri

doi:10.31963/elekterika.v22i2.5760

Authors

Dewi Humeira Amriah Dewi Politeknik Negeri Ujung Pandang
Farid Politeknik Negeri Ujung Pandang
Muhammad Fathur Rahman N Politeknik Negeri Ujung Pandang
Riesa Krisna Astuti Sakir
Muh. Erdin
Nurfitri

DOI:

https://doi.org/10.31963/elekterika.v22i2.5760

Keywords:

Electricity, Kalman Filter, PZEM-004T, Forecasting, Accuracy

Abstract

Humans demand electricity to conduct their daily tasks. Indonesia undergoes a yearly escalation in its electricity use. Moreover, customers encounter the difficulty of overestimating their electrical energy consumption, since they remain unaware of the power utilization linked to each frequently utilized electrical load and possess limited control over their electricity expenditure. The Kalman Filter Algorithm was utilized to estimate electricity usage via the execution of a research system. The Kalman Filter can forecast future states using minimal information. This system incorporates an IoT framework with a network communication module, the Raspberry Pi, which relays data to the database. The PZEM-004T sensor is utilized to gather data on electrical parameters from loads, including voltage, current, active power, and energy consumption. The electrical consumption was documented every 15 minutes over a duration of 60 days. The dataset was divided in an 80:20 ratio, allocating 80% for training and 20% for testing. RMSE, MSE, and MAPE are utilized to determine the accuracy metrics of each test. Additionally, the fan load is assessed in one evaluation, yielding an error percentage of 0.077% for the training data and 0.076% for the test data, determined using RMSE. The error percentage calculated using the MSE equation is 0.006% for the training data and 0.005% for the test data. The error percentage calculated by MAPE is 0.789% for the training dataset and 0.202% for the testing dataset. The findings indicate that the Kalman Filter prediction method is exceptionally proficient in forecasting electrical load consumption

References

[1] D. A. Regina et al., “Penggunaan Beban Listrik Rumah Berdasarkan Pola Penggunaan Beban Listrik”.

[2] B. A. Tengger and R. Ropiudin, “Pemanfaatan Metode Kalman Filter Diskrit untuk Menduga Suhu Udara,” Square: Journal of Mathematics and Mathematics Education, vol. 1, no. 2, p. 127, 2019, doi: 10.21580/square.2019.1.2.4202.

[3] P. Ozoh, S. Abd-Rahman, and J. Labadin, “Predicting electricity consumption: A comparative analysis of the accuracy of various computational techniques,” 2015 9th International Conference on IT in Asia: Transforming Big Data into Knowledge, CITA 2015 - Proceedings, no. August 2015, doi: 10.1109/CITA.2015.7349819.

[4] F. satya Purnomo, “Penggunaan metode ARIMA untuk perkiraan konsumsi listrik jangka pendek,” Journal Teknik, p. 136, 2015.

[5] G. Cao and L. Wu, “Support vector regression with fruit fly optimization algorithm for seasonal electricity consumption forecasting,” Energy, vol. 115, pp. 734–745, 2016, doi: 10.1016/j.energy.2016.09.065.

[6] A. González-Briones, G. Hernandez, J. M. Corchado, S. Omatu, and M. S. Mohamad, “Machine Learning Models for Electricity Consumption Forecasting: A Review,” 2nd International Conference on Computer Applications and Information Security, ICCAIS 2019, 2019, doi: 10.1109/CAIS.2019.8769508.

[7] E. K. Korir, J. Aduda, and T. Mageto, “Forecasting Electricity Prices Using Ensemble Kalman Filter,” vol. 9, no. 1, pp. 27–45, 2020.

[8] W. R. Danu, “Model Estimasi Daya Listrik Gedung Universitas Pembangunan Nasional ‘Veteran’ Jakarta Menggunakan Metode Kalman Filter,” Universitas Pembangunan Nasioal Veteran Jakarta, 2015. [Online]. Available: http://repository.upnvj.ac.id/3348/

[9] A. D. dan A. S. R. M. S. M. N. Abdillah, “Sistem Monitoring Konsumsi Energi Listrik Real-Time Berbasis IoT,” Jurnal Teknologi Elekterika, vol. 18, pp. 24–30, May 2021.

[10] R. S. W. dan H. S. M. Rusli, “Pengaturan Pencahayaan dan Pendingin Ruangan Otomatis Berbasis Mikrokontroler untuk Efisiensi Energi,” Jurnal Teknologi Elekterika, vol. 19, no. 2, pp. 115–121, Nov. 2022.

[11] M. Kafil, “Penerapan Metode K-Nearest Neighbors Untuk Prediksi Penjualan Berbasis Web Pada Boutiq Dealove Bondowoso,” JATI (Jurnal Mahasiswa Teknik Informatika), vol. 3, no. 2, pp. 59–66, 2019, doi: 10.36040/jati. v3i2.860.

[12] X. Lai, T. Yang, Z. Wang, and P. Chen, “IoT implementation of Kalman Filter to improve accuracy of air quality monitoring and prediction,” Applied Sciences (Switzerland), vol. 9, no. 9, May 2019, doi: 10.3390/app9091831.

[13] L. Wang, S. X. Lv, and Y. R. Zeng, “Effective sparse adaboost method with ESN and FOA for industrial electricity consumption forecasting in China,” Energy, vol. 155, pp. 1013–1031, 2018, doi: 10.1016/j.energy.2018.04.175.

[14] C. Fan Shuxiang and Li, “Sensor Fusion,” in Encyclopedia of Smart Agriculture Technologies, Q. Zhang, Ed., Cham: Springer International Publishing, 2022, pp. 1–15. doi: 10.1007/978-3-030-89123-7_142-1.

[15] D. H. Amriah, F. Dewanta, and M. Abdillah, “Implementasi Kalman Filter Untuk Prediksi Pemakaian Listrik Pada Kamar Kost Dengan Smart Energy Meter (Electricity Load Forecasting Using Kalman Filter in Boarding House with Smart Energy Meter),” vol. 8, no. 5, p. 5009, 2021.