IDEAS home Printed from https://ideas.repec.org/a/gam/jsusta/v15y2023i8p6951-d1128502.html
   My bibliography  Save this article

RAID: Robust and Interpretable Daily Peak Load Forecasting via Multiple Deep Neural Networks and Shapley Values

Author

Listed:
  • Joohyun Jang

    (Department of AI and Big Data, Soonchunhyang University, Asan 31538, Republic of Korea
    These authors contributed equally to this work.)

  • Woonyoung Jeong

    (Department of AI and Big Data, Soonchunhyang University, Asan 31538, Republic of Korea
    These authors contributed equally to this work.)

  • Sangmin Kim

    (Department of AI and Big Data, Soonchunhyang University, Asan 31538, Republic of Korea)

  • Byeongcheon Lee

    (Department of AI and Big Data, Soonchunhyang University, Asan 31538, Republic of Korea)

  • Miyoung Lee

    (Department of Software, Sejong University, Seoul 05006, Republic of Korea)

  • Jihoon Moon

    (Department of AI and Big Data, Soonchunhyang University, Asan 31538, Republic of Korea)

Abstract

Accurate daily peak load forecasting (DPLF) is crucial for informed decision-making in energy management. Deep neural networks (DNNs) are particularly apt for DPLF because they can analyze multiple factors, such as timestamps, weather conditions, and historical electric loads. Interpretability of machine learning models is essential for ensuring stakeholders understand and trust the decision-making process. We proposed the RAID (robust and interpretable DPLF) model, which enhances DPLF accuracy by recognizing daily peak load patterns and building separate DNN models for each day of the week. This approach was accessible for energy providers with limited computational resources, as the DNN models could be configured without a graphics processing unit (GPU). We utilized scikit-learn’s MLPRegressor for streamlined implementation, Optuna for hyperparameter optimization, and the Shapley additive explanations (SHAP) method to ensure interpretability. Applied to a dataset from two commercial office buildings in Richland, Washington, RAID outperformed existing methods like recurrent neural networks, Cubist, and HYTREM, achieving the lowest mean absolute percentage error values: 14.67% for Building 1 and 12.74% for Building 2. The kernel SHAP method revealed the influence of the previous day’s peak load and temperature-related variables on the prediction. The RAID model substantially improved energy management through enhanced DPLF accuracy, outperforming competing methods, providing a GPU-free configuration, and ensuring interpretable decision-making, with the potential to influence energy providers’ choices and promote overall energy system sustainability.

Suggested Citation

  • Joohyun Jang & Woonyoung Jeong & Sangmin Kim & Byeongcheon Lee & Miyoung Lee & Jihoon Moon, 2023. "RAID: Robust and Interpretable Daily Peak Load Forecasting via Multiple Deep Neural Networks and Shapley Values," Sustainability, MDPI, vol. 15(8), pages 1-27, April.
  • Handle: RePEc:gam:jsusta:v:15:y:2023:i:8:p:6951-:d:1128502
    as

    Download full text from publisher

    File URL: https://www.mdpi.com/2071-1050/15/8/6951/pdf
    Download Restriction: no

    File URL: https://www.mdpi.com/2071-1050/15/8/6951/
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Lee, Juyong & Cho, Youngsang, 2022. "National-scale electricity peak load forecasting: Traditional, machine learning, or hybrid model?," Energy, Elsevier, vol. 239(PD).
    2. Mukund Sundararajan & Amir Najmi, 2019. "The many Shapley values for model explanation," Papers 1908.08474, arXiv.org, revised Feb 2020.
    3. Bilal Naji Alhasnawi & Basil H. Jasim & Pierluigi Siano & Josep M. Guerrero, 2021. "A Novel Real-Time Electricity Scheduling for Home Energy Management System Using the Internet of Energy," Energies, MDPI, vol. 14(11), pages 1-29, May.
    4. Stefano Frizzo Stefenon & Laio Oriel Seman & Viviana Cocco Mariani & Leandro dos Santos Coelho, 2023. "Aggregating Prophet and Seasonal Trend Decomposition for Time Series Forecasting of Italian Electricity Spot Prices," Energies, MDPI, vol. 16(3), pages 1-18, January.
    5. Ahmad Alzahrani & Senthil Kumar Ramu & Gunapriya Devarajan & Indragandhi Vairavasundaram & Subramaniyaswamy Vairavasundaram, 2022. "A Review on Hydrogen-Based Hybrid Microgrid System: Topologies for Hydrogen Energy Storage, Integration, and Energy Management with Solar and Wind Energy," Energies, MDPI, vol. 15(21), pages 1-32, October.
    6. Geun-Cheol Lee, 2022. "Regression-Based Methods for Daily Peak Load Forecasting in South Korea," Sustainability, MDPI, vol. 14(7), pages 1-19, March.
    7. Noor Shakir Mahmood & Ahmed Ali Ajmi & Shamsul Bin Sarip & Hazilah Mad Kaidi & Khairur Rijal Jamaludin & Hayati Habibah Abdul Talib, 2022. "Modeling the Sustainable Integration of Quality and Energy Management in Power Plants," Sustainability, MDPI, vol. 14(4), pages 1-19, February.
    8. Dong-Hoon Kim & Eun-Kyu Lee & Naik Bakht Sania Qureshi, 2020. "Peak-Load Forecasting for Small Industries: A Machine Learning Approach," Sustainability, MDPI, vol. 12(16), pages 1-19, August.
    9. Saxena, Harshit & Aponte, Omar & McConky, Katie T., 2019. "A hybrid machine learning model for forecasting a billing period’s peak electric load days," International Journal of Forecasting, Elsevier, vol. 35(4), pages 1288-1303.
    10. Saif Jamal & Nadia M. L. Tan & Jagadeesh Pasupuleti, 2021. "A Review of Energy Management and Power Management Systems for Microgrid and Nanogrid Applications," Sustainability, MDPI, vol. 13(18), pages 1-31, September.
    11. Bibi Ibrahim & Luis Rabelo, 2021. "A Deep Learning Approach for Peak Load Forecasting: A Case Study on Panama," Energies, MDPI, vol. 14(11), pages 1-26, May.
    12. Husein Ali Zeini & Duaa Al-Jeznawi & Hamza Imran & Luís Filipe Almeida Bernardo & Zainab Al-Khafaji & Krzysztof Adam Ostrowski, 2023. "Random Forest Algorithm for the Strength Prediction of Geopolymer Stabilized Clayey Soil," Sustainability, MDPI, vol. 15(2), pages 1-15, January.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Luca Gugliermetti & Fabrizio Cumo & Sofia Agostinelli, 2024. "A Future Direction of Machine Learning for Building Energy Management: Interpretable Models," Energies, MDPI, vol. 17(3), pages 1-27, February.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Bingjie Jin & Guihua Zeng & Zhilin Lu & Hongqiao Peng & Shuxin Luo & Xinhe Yang & Haojun Zhu & Mingbo Liu, 2022. "Hybrid LSTM–BPNN-to-BPNN Model Considering Multi-Source Information for Forecasting Medium- and Long-Term Electricity Peak Load," Energies, MDPI, vol. 15(20), pages 1-20, October.
    2. Roman V. Klyuev & Irbek D. Morgoev & Angelika D. Morgoeva & Oksana A. Gavrina & Nikita V. Martyushev & Egor A. Efremenkov & Qi Mengxu, 2022. "Methods of Forecasting Electric Energy Consumption: A Literature Review," Energies, MDPI, vol. 15(23), pages 1-33, November.
    3. Amara-Ouali, Yvenn & Fasiolo, Matteo & Goude, Yannig & Yan, Hui, 2023. "Daily peak electrical load forecasting with a multi-resolution approach," International Journal of Forecasting, Elsevier, vol. 39(3), pages 1272-1286.
    4. Zhigang Liu & Jin Wang & Tao Tao & Ziyun Zhang & Siyi Chen & Yang Yi & Shuang Han & Yongqian Liu, 2023. "Wave Power Prediction Based on Seasonal and Trend Decomposition Using Locally Weighted Scatterplot Smoothing and Dual-Channel Seq2Seq Model," Energies, MDPI, vol. 16(22), pages 1-17, November.
    5. Saif Jamal & Jagadeesh Pasupuleti & Nur Azzammudin Rahmat & Nadia M. L. Tan, 2022. "Energy Management System for Grid-Connected Nanogrid during COVID-19," Energies, MDPI, vol. 15(20), pages 1-20, October.
    6. Yang, Weijia & Sparrow, Sarah N. & Wallom, David C.H., 2024. "A comparative climate-resilient energy design: Wildfire Resilient Load Forecasting Model using multi-factor deep learning methods," Applied Energy, Elsevier, vol. 368(C).
    7. Alireza Rezazadeh & Yasamin Jafarian & Ali Kord, 2022. "Explainable Ensemble Machine Learning for Breast Cancer Diagnosis Based on Ultrasound Image Texture Features," Forecasting, MDPI, vol. 4(1), pages 1-13, February.
    8. Hu'e Sullivan & Hurlin Christophe & P'erignon Christophe & Saurin S'ebastien, 2022. "Measuring the Driving Forces of Predictive Performance: Application to Credit Scoring," Papers 2212.05866, arXiv.org, revised Jun 2023.
    9. Hugh Chen & Scott M. Lundberg & Su-In Lee, 2022. "Explaining a series of models by propagating Shapley values," Nature Communications, Nature, vol. 13(1), pages 1-15, December.
    10. Jonathan Berrisch & Micha{l} Narajewski & Florian Ziel, 2022. "High-Resolution Peak Demand Estimation Using Generalized Additive Models and Deep Neural Networks," Papers 2203.03342, arXiv.org, revised Nov 2022.
    11. Zhewei Huang & Yawen Yi, 2024. "Short-Term Load Forecasting for Regional Smart Energy Systems Based on Two-Stage Feature Extraction and Hybrid Inverted Transformer," Sustainability, MDPI, vol. 16(17), pages 1-25, September.
    12. Winita Sulandari & Yudho Yudhanto & Sri Subanti & Crisma Devika Setiawan & Riskhia Hapsari & Paulo Canas Rodrigues, 2023. "Comparing the Simple to Complex Automatic Methods with the Ensemble Approach in Forecasting Electrical Time Series Data," Energies, MDPI, vol. 16(22), pages 1-16, November.
    13. Masayoshi Mase & Art B. Owen & Benjamin B. Seiler, 2021. "Cohort Shapley value for algorithmic fairness," Papers 2105.07168, arXiv.org.
    14. Pesantez, Jorge E. & Li, Binbin & Lee, Christopher & Zhao, Zhizhen & Butala, Mark & Stillwell, Ashlynn S., 2023. "A Comparison Study of Predictive Models for Electricity Demand in a Diverse Urban Environment," Energy, Elsevier, vol. 283(C).
    15. Wenguang Zhang & Ting Lei & Yu Gong & Jun Zhang & Yirong Wu, 2022. "Using Explainable Artificial Intelligence to Identify Key Characteristics of Deep Poverty for Each Household," Sustainability, MDPI, vol. 14(16), pages 1-21, August.
    16. Akash Kumar & Bing Yan & Ace Bilton, 2022. "Machine Learning-Based Load Forecasting for Nanogrid Peak Load Cost Reduction," Energies, MDPI, vol. 15(18), pages 1-23, September.
    17. Salma Hamad Almuhaini & Nahid Sultana, 2023. "Forecasting Long-Term Electricity Consumption in Saudi Arabia Based on Statistical and Machine Learning Algorithms to Enhance Electric Power Supply Management," Energies, MDPI, vol. 16(4), pages 1-28, February.
    18. Khairul Eahsun Fahim & Liyanage C. De Silva & Fayaz Hussain & Sk. A. Shezan & Hayati Yassin, 2023. "An Evaluation of ASEAN Renewable Energy Path to Carbon Neutrality," Sustainability, MDPI, vol. 15(8), pages 1-27, April.
    19. Masayoshi Mase & Art B. Owen & Benjamin B. Seiler, 2022. "Variable importance without impossible data," Papers 2205.15750, arXiv.org, revised Apr 2023.
    20. Aleksandr N. Grekov & Elena V. Vyshkvarkova & Aleksandr S. Mavrin, 2024. "Forecasting and Anomaly Detection in BEWS: Comparative Study of Theta, Croston, and Prophet Algorithms," Forecasting, MDPI, vol. 6(2), pages 1-14, May.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:gam:jsusta:v:15:y:2023:i:8:p:6951-:d:1128502. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: MDPI Indexing Manager (email available below). General contact details of provider: https://www.mdpi.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.