Reliability of regression based hybrid machine learning models for the prediction of solar photovoltaics power generation

被引:1
作者
Ahmed, Sina Ibne [1 ]
Bhuiyan, Kaiser Ahmed [2 ]
Rahman, Irin [3 ]
Salehfar, Hossein [1 ]
Selvaraj, Daisy Flora [4 ]
机构
[1] Univ North Dakota, Sch Elect Engn & Comp Sci, Grand Forks, ND 58202 USA
[2] Univ North Dakota, Inst Energy Studies, Grand Forks, ND USA
[3] North Dakota State Univ, Dept Stat, Fargo, ND USA
[4] Univ North Dakota, Energy & Environm Res Ctr, Grand Forks, ND USA
关键词
Solar photovoltaics; Machine learning; Hybrid models; MARS; Reliability analysis; ARMA; MARS;
D O I
10.1016/j.egyr.2024.10.060
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Solar Photovoltaics (PVs) have become widespread as micro-distributed electric power generators in urban residential and commercial areas due to their affordability and minimal maintenance requirements. Despite these advantages, PV generation is intermittent, necessitating the implementation of robust predictive algorithms to capture power generation trends effectively. Accurate PV generation prediction is crucial for balancing energy demand and supply. Precise predictions help utilities use other grid resources efficiently, enhance market participation, and make informed planning decisions. Therefore, ensuring the precise performance of predictive models is imperative in the power generation industry. However, the performance of the predictive models typically fluctuates over time, necessitating a reliability assessment of their performance. This paper develops a hybrid machine learning model by integrating the support vector machine (SVM), random forest (RF), and gradient boosting machine (GBM) with multivariate adaptive regression spline (MARS) to predict the hourly PV power generation for 3-day and 7-day periods in three urban areas of North Dakota. The performance of the models has been recorded over a year, and a probability distribution model is proposed to demonstrate the hybrid algorithm's reliability and effectiveness. Results show the proposed prediction model is more reliable for shorter prediction periods. The hybrid-MARS models significantly improve prediction reliability compared to the stand-alone MARS model. Among these models, MARS-SVM exhibits the highest reliability and consistently better accuracy than other hybrid models.
引用
收藏
页码:5009 / 5023
页数:15
相关论文
共 45 条
  • [1] Forecasting of Mid- and Long-Term Wind Power Using Machine Learning and Regression Models
    Ahmed, Sina Ibne
    Ranganathan, Prakash
    Salehfar, Hossein
    [J]. 2021 IEEE KANSAS POWER AND ENERGY CONFERENCE (KPEC), 2021,
  • [2] Predictive models development using gradient boosting based methods for solar power plants
    Aksoy, Necati
    Genc, Istemihan
    [J]. JOURNAL OF COMPUTATIONAL SCIENCE, 2023, 67
  • [3] Solar power generation forecasting using ensemble approach based on deep learning and statistical methods
    AlKandari, Mariam
    Ahmad, Imtiaz
    [J]. APPLIED COMPUTING AND INFORMATICS, 2024, 20 (3/4) : 231 - 250
  • [4] Short-term photovoltaic power production forecasting based on novel hybrid data-driven models
    Alrashidi, Musaed
    Rahman, Saifur
    [J]. JOURNAL OF BIG DATA, 2023, 10 (01)
  • [5] A Hybrid Approach of Solar Power Forecasting Using Machine Learning
    Bajpai, Arpit
    Duchon, Markus
    [J]. 2019 3RD INTERNATIONAL CONFERENCE ON SMART GRID AND SMART CITIES (ICSGSC 2019), 2019, : 108 - 113
  • [6] Biau G, 2012, J MACH LEARN RES, V13, P1063
  • [7] Performance Analysis of Short and Mid-Term Wind Power Prediction using ARIMA and Hybrid Models
    Biswas, Ashoke Kumar
    Ahmed, Sina Ibne
    Bankefa, Temitope
    Ranganathan, Prakash
    Salehfar, Hossein
    [J]. 2021 IEEE POWER AND ENERGY CONFERENCE AT ILLINOIS (PECI), 2021,
  • [8] Load forecasting using support vector machines: A study on EUNITE competition 2001
    Chen, BJ
    Chang, MW
    Lin, CJ
    [J]. IEEE TRANSACTIONS ON POWER SYSTEMS, 2004, 19 (04) : 1821 - 1830
  • [9] Chi-Jie Lu, 2010, Proceedings 2010 10th IEEE International Conference on Data Mining Workshops (ICDMW 2010), P593, DOI 10.1109/ICDMW.2010.11
  • [10] Chih-Feng Yen, 2018, E3S Web of Conferences, V69, DOI 10.1051/e3sconf/20186901004