This work presents the development and experimental validation of a system for monitoring and managing indoor comfort in workplaces, designed to operate under real-world operating conditions and without invasive interventions on the systems. The system integrates the measurement of environmental parameters, space occupancy information, and comfort models within a single application architecture, based on a mobile robotic platform with supervision and information interaction functions. Thermohygrometric parameters, air quality, and lighting are acquired using fixed sensors, while room occupancy is estimated indirectly from workstation electricity consumption. Thermal comfort is assessed using the sPMV index and analyzed using KPIs based on the temporal distribution of comfort classes. The analysis distinguishes the periods of occupant presence and absence within working hours; The robotic supervision and coaching system is active only when the room is estimated to be occupied, while periods of absence constitute the unsupervised reference configuration. The validation was conducted in four real offices during an experimental campaign lasting approximately one month. In addition to thermal comfort, air quality and lighting comfort were also monitored; for these parameters, the suggested interventions showed average return times of approximately 50 minutes for CO2 concentration and 28 minutes for illumination, respectively. The results show that, compared to the baseline configuration in the absence of occupancy, the presence of the system with active coaching during occupied periods is associated with a systematic reduction in the persistence of thermal discomfort conditions. The proportion of time classified as uncomfortable decreased between 4% and 9% in the different environments analyzed. Finally, the analysis highlights that environments characterized by greater occupant compliance with coaching interventions show more marked reductions in discomfort conditions, suggesting a significant role for informational interaction in the improvement process. To support this evidence, a seasonal comparison was conducted between two monthly periods characterized by comparable external climate conditions. The analysis compared the month of February 2025, in which the robotic system was not present and the environment operated in unsupervised mode, with the month of December 2025, during which the robot was active and could provide informational interventions in the presence of estimated occupancy of the spaces. The comparison, extended to the entire monthly time horizon, highlights an approximately 9% reduction in the proportion of time spent in critical conditions in the supervised period. Overall, the study shows how a robotic supervision system, integrated with thermal comfort indicators and occupancy information, can contribute tangibly to the management of indoor comfort in real-world work contexts.
Il lavoro presenta lo sviluppo e la validazione sperimentale di un sistema per il monitoraggio e la gestione del comfort indoor in ambienti di lavoro, progettato per operare in condizioni di esercizio reale e senza introdurre interventi invasivi sugli impianti. Il sistema integra la misura delle grandezze ambientali, l’informazione di occu- pazione degli spazi e modelli di comfort all’interno di un’unica architettura applicativa, basata su una piattaforma robotica mobile con funzioni di supervisione e interazione informativa. I parametri termoigrometrici, la qualità dell’aria e l’illuminamento sono acquisiti mediante sensoristica fissa, mentre lo stato di occupazione degli ambienti è stimato in modo indiretto a partire dai consumi elettrici delle postazioni di lavoro. Il comfort termico è valutato tramite l’indice sPMV e analizzato attraverso KPI basati sulla distribuzione temporale delle classi di comfort. L’analisi distingue i periodi di presenza e assenza degli occupanti all’interno dell’orario di lavoro; il sistema di supervisione e coaching robotico risulta attivo esclusivamente in presenza di occupazione stimata dell’ambiente, mentre i periodi di assenza costituiscono la configurazione di riferimento non supervisionata. La validazione è stata condotta in quattro uffici reali durante una campagna sperimentale di circa un mese. Oltre al comfort termico, sono stati monitorati anche la qualità dell’aria e il comfort luminoso; per tali parametri, gli interventi suggeriti mostrano tempi medi di rientro rispettivamente pari a circa 50 minuti per la concentrazione di CO2 e 28 minuti per l’illuminamento. I risultati mostrano che, rispetto alla configurazione di baseline in assenza di occupazione, la presenza del sistema con coaching attivo durante i periodi occupati è associata a una riduzione sistematica della persistenza delle condizioni di discomfort termico. La quota di tempo classificata come non confortevole diminuisce tra il 4% e il 9% nei diversi ambienti analizzati. L’analisi evidenzia infine che ambienti caratterizzati da una maggiore adesione degli occupanti agli interventi di coaching presentano riduzioni più marcate delle condizioni di discomfort, suggerendo un ruolo rilevante dell’interazione informativa nel processo di miglioramento. A supporto di tali evidenze, è stato condotto un confronto stagionale tra due periodi mensili caratterizzati da condizioni climatiche esterne comparabili. L’analisi ha messo a confronto il mese di febbraio 2025, in cui il sistema robotico non era presente e l’ambiente operava in modalità non supervisionata, e il mese di dicembre 2025, durante il quale il robot era attivo e poteva fornire interventi informativi in presenza di occupazione stimata degli ambienti. Il confronto, esteso all’intero orizzonte temporale mensile, evidenzia una riduzione di circa 9 % della quota di tempo trascorsa in condizioni critiche nel periodo supervisionato. Nel complesso, il lavoro mostra come un sistema di supervisione robotica, integrato con indicatori di comfort termico e informazioni di occupazione, possa contribuire in modo tangibile alla gestione del comfort indoor in contesti di lavoro reali.
Sviluppo di un sistema di monitoraggio multidominio basato su un robot mobile per la misurazione del benessere indoor e del consumo energetico
ZACCONI, ALESSANDRO
2024/2025
Abstract
This work presents the development and experimental validation of a system for monitoring and managing indoor comfort in workplaces, designed to operate under real-world operating conditions and without invasive interventions on the systems. The system integrates the measurement of environmental parameters, space occupancy information, and comfort models within a single application architecture, based on a mobile robotic platform with supervision and information interaction functions. Thermohygrometric parameters, air quality, and lighting are acquired using fixed sensors, while room occupancy is estimated indirectly from workstation electricity consumption. Thermal comfort is assessed using the sPMV index and analyzed using KPIs based on the temporal distribution of comfort classes. The analysis distinguishes the periods of occupant presence and absence within working hours; The robotic supervision and coaching system is active only when the room is estimated to be occupied, while periods of absence constitute the unsupervised reference configuration. The validation was conducted in four real offices during an experimental campaign lasting approximately one month. In addition to thermal comfort, air quality and lighting comfort were also monitored; for these parameters, the suggested interventions showed average return times of approximately 50 minutes for CO2 concentration and 28 minutes for illumination, respectively. The results show that, compared to the baseline configuration in the absence of occupancy, the presence of the system with active coaching during occupied periods is associated with a systematic reduction in the persistence of thermal discomfort conditions. The proportion of time classified as uncomfortable decreased between 4% and 9% in the different environments analyzed. Finally, the analysis highlights that environments characterized by greater occupant compliance with coaching interventions show more marked reductions in discomfort conditions, suggesting a significant role for informational interaction in the improvement process. To support this evidence, a seasonal comparison was conducted between two monthly periods characterized by comparable external climate conditions. The analysis compared the month of February 2025, in which the robotic system was not present and the environment operated in unsupervised mode, with the month of December 2025, during which the robot was active and could provide informational interventions in the presence of estimated occupancy of the spaces. The comparison, extended to the entire monthly time horizon, highlights an approximately 9% reduction in the proportion of time spent in critical conditions in the supervised period. Overall, the study shows how a robotic supervision system, integrated with thermal comfort indicators and occupancy information, can contribute tangibly to the management of indoor comfort in real-world work contexts.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_Zacconi_Alessandro_PDFA1b.pdf
embargo fino al 21/08/2027
Dimensione
8.48 MB
Formato
Adobe PDF
|
8.48 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.12075/25594