Scholarlink

<p>Accurate power consumption prediction is a crucial part of energy management. Some of the machine learning models that are the focus of this study for the prediction of power use include Support Vector Regression, Adaptive Boosting, and Decision Tree Regression. These models have been improved with the use of some novel optimizers-namely, the Trochoid Search Optimization, Red-Tailed Hawk, and Giant Armadillo Optimization methods-for hyper-parameter tuning to enhance prediction accuracy. When tested against real data, DTGA outperformed with <i>R</i><sup>2</sup> values of 0.9918, 0.9924, and 0.9934 for three zones. This work extends the study on the forecast of power consumption by integrating machine learning and optimization techniques that provide effective energy management strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Application of Metaheuristic Algorithms with Supervised Machine Learning for Accurate Power Consumption Prediction

  • 期刊论文

  • 发表日期：

    2025年1月30日

查看全文
  • Mengxia Wang,
  • Chaoyang Zhu,
  • Yunxiang Zhang,
  • Jinxin Deng,
  • Yiwei Cai,
  • Wei Wei,
  • Mengxing Guo

摘要

Accurate power consumption prediction is a crucial part of energy management. Some of the machine learning models that are the focus of this study for the prediction of power use include Support Vector Regression, Adaptive Boosting, and Decision Tree Regression. These models have been improved with the use of some novel optimizers-namely, the Trochoid Search Optimization, Red-Tailed Hawk, and Giant Armadillo Optimization methods-for hyper-parameter tuning to enhance prediction accuracy. When tested against real data, DTGA outperformed with R2 values of 0.9918, 0.9924, and 0.9934 for three zones. This work extends the study on the forecast of power consumption by integrating machine learning and optimization techniques that provide effective energy management strategies.