DOI : https://doi.org/10.18517/ijaseit.15.1.20944

Multivariate Variable-Based LSTM-AE Model for Solar Power Prediction

GunHa Park (1), JongChan Kim (2)
(1) Department of Computer Engineering, Sunchon National University, 255, Jungang-ro, Suncheon-si, Republic of Korea
(2) Department of Computer Engineering, Sunchon National University, 255, Jungang-ro, Suncheon-si, Republic of Korea
Fulltext View | Download
How to cite (IJASEIT) :
[1]
G. Park and J. Kim, “Multivariate Variable-Based LSTM-AE Model for Solar Power Prediction”, Int. J. Adv. Sci. Eng. Inf. Technol., vol. 15, no. 1, pp. 293–299, Feb. 2025.
Citation Format :
This study proposes a multivariate-based LSTM-Autoencoder (LSTM-AE) model for short-term photovoltaic power generation prediction. The LSTM-based encoder-decoder structure effectively learns multivariate relationships and time series dependencies between significant environmental and power-related variables. Input variables include DC voltage, DC current, DC power, ambient temperature, solar radiation, and environmental factors, and they are preprocessed through scaling to increase learning efficiency. The encoder compresses multivariate time series data to the latent space, and the decoder restores the corresponding sequence to learn the complex time series patterns of the data. The normalization technique was applied to the algorithm to prevent overfitting and improve the model's generalization performance. The prediction accuracy evaluation was made through mean absolute percentage error (MAPE), mean square root error (RMSE), mean absolute error (MAE), and coefficient of determination (R²). As a result of the experiment, the proposed LSTM-AE model outperformed the existing model in capturing nonlinear and long-term dependence. The results of this study suggest that the LSTM-AE architecture can contribute to the development of renewable energy prediction fields, contributing to the development of more accurate and reliable photovoltaic prediction systems.

M. Lee, P. Lee, D. Jeong, M. Han, and S. P. Jung, “RE100 for 100% renewable electricity: Status and prospects,” J. Korean Soc. Environ. Eng., vol. 45, no. 3, pp. 161–169, Mar. 2023, doi:10.4491/ksee.2023.45.3.161.

D.-H. T. Kim, C. Park, J.-B. Park, and J.-H. Roh, “An analytical approach for transmission use-of-system charges for a renewable energy power purchase agreement,” Trans. Korean Inst. Electr. Eng., vol. 72, no. 7, pp. 801–808, Jul. 2023, doi: 10.5370/kiee.2023.72.7.801.

S. Sengupta, A. Sahay, and R. D. Hisrich, “The social–market convergence in a renewable energy social enterprise,” J. Clean. Prod., vol. 270, p. 122516, Oct. 2020, doi: 10.1016/j.jclepro.2020.122516.

A. Kuś and D. Grego-Planer, “A model of innovation activity in small enterprises in the context of selected financial factors: The example of the renewable energy sector,” Energies, vol. 14, no. 10, p. 2926, May 2021, doi: 10.3390/en14102926.

C. D. Diale, M. G. Kanakana-Katumba, and R. W. Maladzhi, “Ecosystem of renewable energy enterprises for sustainable development: A systematic review,” Adv. Sci. Technol. Eng. Syst. J., vol. 6, no. 1, pp. 401–408, Jan. 2021, doi: 10.25046/aj060146.

H. Choi, “Stock prediction analysis through artificial intelligence using big data,” J. Korea Inst. Inf. Commun. Eng., vol. 25, no. 10, pp. 1435–1440, 2021.

Y. Lim, W. Shin, and H. Kim, “Prediction model for gastric cancer in patients with chronic gastritis using RNN based on medical big data,” in Proc. Korean Inst. Inf. Sci. Eng. Conf., vol. 2020, no. 7, pp. 1289–1291, 2020.

Y. Kim, S. Kim, and H. Yeo, “A study on traffic prediction using hybrid approach of machine learning and simulation techniques,” J. Korea Inst. Intell. Transp. Syst., vol. 20, no. 5, pp. 100–112, Oct. 2021, doi: 10.12815/kits.2021.20.5.100.

H. Kim and M. Lee, “Research trends in machine learning applications in marketing,” Korean J. Mark., vol. 36, no. 1, pp. 1–25, Feb. 2021, doi: 10.15830/kjm.2021.36.1.1.

S. Lee, W. Choi, and M. Kim, “Development of weather forecasting algorithm for improvement of power generation prediction accuracy,” in Proc. Korean Inst. Electr. Eng. (KIEE) Summer Conf., pp. 843–844, Jul. 2021.

H. Jeong et al., “A research of prediction of photovoltaic power using SARIMA model,” J. Korea Multimed. Soc., vol. 25, no. 1, pp. 82–91, 2022, doi: 10.9717/kmms.2022.25.1.082.

R. W. Divisekara, G. J. M. S. R. Jayasinghe, and K. W. S. N. Kumari, “Forecasting the red lentils commodity market price using SARIMA models,” SN Bus. Econ., vol. 1, no. 1, Nov. 2020, doi:10.1007/s43546-020-00020-x.

S. Ma, Q. Liu, and Y. Zhang, “A prediction method of fire frequency: Based on the optimization of SARIMA model,” PLoS ONE, vol. 16, no. 8, p. e0255857, Aug. 2021, doi: 10.1371/journal.pone.0255857.

N. Deretić et al., “SARIMA modelling approach for forecasting of traffic accidents,” Sustainability, vol. 14, no. 8, p. 4403, Apr. 2022, doi:10.3390/su14084403.

M. Abumohsen et al., “Hybrid machine learning model combining of CNN-LSTM-RF for time series forecasting of solar power generation,” e-Prime - Adv. Electr. Eng. Electron. Energy, vol. 9, p. 100636, Sep. 2024, doi: 10.1016/j.prime.2024.100636.

H. Munawer Al-Otum, “Classification of anomalies in electroluminescence images of solar PV modules using CNN-based deep learning,” Sol. Energy, vol. 278, p. 112803, Aug. 2024, doi:10.1016/j.solener.2024.112803.

I. Jebli, F.-Z. Belouadha, M. I. Kabbaj, and A. Tilioua, “Deep learning based models for solar energy prediction,” Adv. Sci. Technol. Eng. Syst. J., vol. 6, no. 1, pp. 349–355, Jan. 2021, doi: 10.25046/aj060140.

L. P. Cinelli, M. A. Marins, E. A. Barros da Silva, and S. L. Netto, Variational Methods for Machine Learning with Applications to Deep Networks. Springer International Publishing, 2021. doi:10.1007/978-3-030-70679-1.

V. Suresh et al., “Probabilistic LSTM-autoencoder based hour-ahead solar power forecasting model for intra-day electricity market participation: A Polish case study,” IEEE Access, vol. 10, pp. 110628–110638, 2022, doi: 10.1109/access.2022.3215080.

D. Kim, S.-M. Kim, J. Suh, and Y. Choi, “Anomaly detection of photovoltaic systems installed in renewable energy housing support project sites by analyzing power generation data,” J. Korean Sol. Energy Soc., vol. 42, no. 1, pp. 33–46, Feb. 2022, doi:10.7836/kses.2022.42.1.033.

J. Seo, T. Lee, W. Lee, and J. Park, “A study on the outlier data estimation method for anomaly detection of photovoltaic system,” J. IKEEE (Inst. Korean Electr. Electron. Eng.), vol. 24, no. 2, pp. 403–408, 2020.

Y. Yoon and S. Jeong, “Deterioration and abnormality condition diagnosis through measuring the DC capacitor capacity of PV inverter,” J. Inst. Internet Broadcast. Commun., vol. 24, no. 5, pp. 135–140, 2024.

L. Wang et al., “Accurate solar PV power prediction interval method based on frequency-domain decomposition and LSTM model,” Energy, vol. 262, p. 125592, Jan. 2023, doi:10.1016/j.energy.2022.125592.

A. Djaafari et al., “Hourly predictions of direct normal irradiation using an innovative hybrid LSTM model for concentrating solar power projects in hyper-arid regions,” Energy Rep., vol. 8, pp. 15548–15562, Nov. 2022, doi: 10.1016/j.egyr.2022.10.402.

L. Ren et al., “A data-driven auto-CNN-LSTM prediction model for lithium-ion battery remaining useful life,” IEEE Trans. Ind. Inform., vol. 17, no. 5, pp. 3478–3487, May 2021, doi:10.1109/tii.2020.3008223.

N. L. M. Jailani et al., “Investigating the power of LSTM-based models in solar energy forecasting,” Processes, vol. 11, no. 5, p. 1382, May 2023, doi: 10.3390/pr11051382.

S.-C. Lim et al., “Solar power forecasting using CNN-LSTM hybrid model,” Energies, vol. 15, no. 21, p. 8233, Nov. 2022, doi:10.3390/en15218233.

P. Li, Y. Pei, and J. Li, “A comprehensive survey on design and application of autoencoder in deep learning,” Appl. Soft Comput., vol. 138, p. 110176, May 2023, doi: 10.1016/j.asoc.2023.110176.

C. Zhang et al., “Autoencoder in autoencoder networks,” IEEE Trans. Neural Netw. Learn. Syst., vol. 35, no. 2, pp. 2263–2275, Feb. 2024, doi: 10.1109/tnnls.2022.3189239.

A. Sherstinsky, “Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network,” Physica D: Nonlinear Phenom., vol. 404, p. 132306, Mar. 2020, doi:10.1016/j.physd.2019.132306.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Authors who publish with this journal agree to the following terms:

    1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
    2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
    3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).