Scholarlink

<p>Demand Response (DR) is a critical strategy for managing the integration of renewable energy sources into the power grid, addressing the challenges posed by their intermittent and unpredictable nature. This study introduces a rapid evaluation method for assessing the DR potential of large-scale Heating, Ventilation, and Air Conditioning (HVAC) systems, focusing on the significant role these systems play in energy consumption and grid flexibility. Firstly, the methodology involves constructing a simulation model library that encompasses three dimensions including room type, room location, and internal heat gain mode to reflect the dynamic characteristics of cooling load. Additionally, batch simulations generate DR profiles under various typical weather conditions, and surrogate models are trained for each simulation model, leveraging feature engineering and cross-validation to enhance accuracy. The Multi-Layer Perceptron (MLP) surrogate models achieve high accuracy in predicting DR potential under various scenarios, with R2 values exceeding 0.95. This study provides a robust framework that enables load aggregators to accurately estimate the demand response potential of large-scale HVAC systems. It supports the quantification of response capabilities and facilitates participation in bidding processes. Furthermore, it highlights the potential of data-driven models to enable rapid and scalable energy management.</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

Scalable evaluation of demand response potential of HVAC systems: Establishing comprehensive room-centric model library and surrogate models

  • 期刊论文

  • 发表日期：

    2025年7月2日

查看全文
  • Ziliang Wei,
  • Zhuofan Tang,
  • Shuyi Chen,
  • Yihu Zhang,
  • Zhenyu Wang,
  • Yang Geng,
  • Borong Lin

摘要

Demand Response (DR) is a critical strategy for managing the integration of renewable energy sources into the power grid, addressing the challenges posed by their intermittent and unpredictable nature. This study introduces a rapid evaluation method for assessing the DR potential of large-scale Heating, Ventilation, and Air Conditioning (HVAC) systems, focusing on the significant role these systems play in energy consumption and grid flexibility. Firstly, the methodology involves constructing a simulation model library that encompasses three dimensions including room type, room location, and internal heat gain mode to reflect the dynamic characteristics of cooling load. Additionally, batch simulations generate DR profiles under various typical weather conditions, and surrogate models are trained for each simulation model, leveraging feature engineering and cross-validation to enhance accuracy. The Multi-Layer Perceptron (MLP) surrogate models achieve high accuracy in predicting DR potential under various scenarios, with R2 values exceeding 0.95. This study provides a robust framework that enables load aggregators to accurately estimate the demand response potential of large-scale HVAC systems. It supports the quantification of response capabilities and facilitates participation in bidding processes. Furthermore, it highlights the potential of data-driven models to enable rapid and scalable energy management.