Estimating morning ramp-up duration for the cooling season in a smart building using machine learning - Determining most promising features

Aug 5, 2026

Upload by Quinn

Nighttime setback strategies typically rely on fixed morning schedules without accounting for each room’s thermal behavior, which leads to conditioning before occupants arrive. This work builds machine learning-based pipelines to estimate ramp-up duration for multiple indoor spaces: the time required for the HVAC system to reach a target setpoint. A monitored cooling-season case study at a northern Italy medical center evaluates experimentally deployed ramp-up procedures across climatic conditions. Six algorithms generate just-in-time starts for fan coil units, identify influential features, and reduce both computational cost and prediction error to under 5 minutes on average.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Introduction

Cooling consumption of buildings

Smart buildings and IoT technologies

Sustainable Energy Technologies and Assessments 69 (2024) 103911 | Estimating morning ramp-up duration for the cooling season in a smart building using machine learning: Determining most promising features Farzad Dadras Javana, Italo Aldo Campodonico Avendano b, Behzad Najafia,∗, Michele Rossic, Fabio Rinaldia
a Dipartimento di Energia, Politecnico di Milano, Via Lambruschini 4, Milano 20156, Italy
b Department of Ocean Operations and Civil Engineering, Faculty of Engineering, NTNU, 6009 Ålesund, Norway c SIRAM VEOLIA S.p.A., Via Anna Maria Mozzoni 12, Milano, Italy | | | | | --- | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords:
Machine learning Cooling ramp-up duration Smart building
Feature selection
Smart cooling | | The nighttime setback approaches commonly consider fixed morning schedules regardless of the rooms’ thermal behavior, resulting in rooms being conditioned before occupants’ arrival. The present work is focused on developing machine learning-based pipelines for estimating the ramp-up duration in different indoor spaces, which is the time the HVAC system needs to bring the space temperature to the desired setpoint. A medical center located in northern Italy, with an HVAC system monitored during the cooling season, is employed asthe case study, where the ramp-up procedure was experimentally deployed and investigated under various climatic conditions. Accordingly, machine learning-based pipelines with six different algorithms for estimating the ramp-up of each space are developed, permitting just-in-time start-up of the fan coil units for each day. Next, the most influential features are specified for each room’s best-performing model, aiming for performance improvement and computational cost reduction. The obtained results indicated an average Mean Absolute Error of less than 5 min for all cases. Finally, the predicted ramp-ups are compared with the existing fixed schedule, where significant saving windows of up to 80 min are achieved. | | Introduction The operational activities of buildings make up 30% of global final energy usage and contribute to 26% of global energy-related emissions [1]. Heating, ventilation, and air conditioning (HVAC) systems account for nearly half of a building’s energy consumption during its operational phase [2]. Additionally, building operations’ direct and indirect emissions, in general, have surged by around 2% compared to 2019 and approximately 5% compared to 2020, reaching around 10 Gt in 2021 [3]. A continued rise in energy demand is expected owing to population growth, higher demand for building services and comfort levels, as well as the increase in the amount of time spent indoors. This growing trend brings about issues regarding meeting the demand, depletion of energy reservoirs, and significant environmental impacts [4]. Consequently, energy optimization in buildings is of the essence for reducing global consumption. Despite the numerous attempts toward energy-efficient technologies, aspects such as operation and maintenance, occupant behavior, and indoor environmental conditions are yet to be further investigated [5]. ∗ Corresponding author. E-mail address: [behzad.najafi@polimi.it](behzad.najafi@polimi.it) (B. Najafi). Cooling consumption of buildings Among the energy-related applications in the building sector, cooling load is extremely important. The International Energy Agency (IEA) estimates a doubled cooling energy usage since 2000, making this demand the fastest growing in buildings. Elevated living standards lead to higher demand for cooling units, and the trend will continue to grow due to economic growth in hotter countries. This trend is supported by the IEA baseline scenarios, which estimate a triple need for cooling by 2050. Additionally, the increase in temperature, the escalating intensity and prolonged duration of hot weather, and a notable increase in extreme weather occurrences such as heat wa

FAQs

What is the “ramp-up duration” estimated in this study?

Ramp-up duration is the time the HVAC system needs to bring an indoor space temperature to the desired setpoint.

How is the proposed method tested?

It is evaluated using a case study at a medical center in northern Italy, where the HVAC system was monitored during the cooling season under various climatic conditions.

What is the main benefit over existing fixed morning schedules?

The approach enables just-in-time fan coil unit startup and achieves significant savings windows, reported as up to 80 minutes.

Aug 5, 2026

Upload by Quinn