Eusébio Z.E. Conceição

Neural network PMV estimation for model-based predictive control of HVAC systems

Heating, Ventilating and Air Conditioning (HVAC) systems are used to provide adequate comfort to occupants of spaces within buildings. One important aspect of comfort, the thermal sensation, is commonly assessed by computation of the Predicted Mean Vote (PMV) index. Model-based predictive control may be applied to HVAC systems in existing buildings in order to provide a desired degree of thermal comfort and simultaneously achieve significant energy savings. This control strategy may be formulated as a discrete optimisation problem and solved by means of structured search techniques. Finding the optimal solution depends on the ability of computing many PMV values in a small amount of time. As the PMV formulation involves iterative computations consuming variable time, it is crucial to have a method for fast, possibly constant execution time, computation of the PMV index. In this paper it is experimentally shown that an Artificial Neural Network (ANN) can estimate the PMV index with varying degrees of efficiency over the trade-off of accuracy versus computational speed-up.

Air quality inside a school building: air exchange monitoring, evolution of carbon dioxide and assessment of ventilation strategies

Abstract This paper presents an assessment of indoor air quality and various ventilation strategies inside a school building located in the south of Portugal. In the first phase, ventilation rate was experimentally evaluated using the tracer gas method. In the second part, different airflow typologies were investigated and, after calculating the air exchange and flow rates for each of them, the evolution of metabolic carbon dioxide inside the spaces was numerically estimated. Ventilation measurements were made in classrooms, auditorium, offices, staff and computer rooms. The assessment of ventilation was based on evaluating the carbon dioxide produced by the occupants for three ventilation approaches; these were: one based on cross-flow natural ventilation (in current use) and two based on forced ventilation systems. In the case of the forced systems, one was based on providing a constant flow to meet the required Portuguese ventilation standard in the main occupied rooms while the other was an adjusted constant rate based on a simple calculation procedure that took into consideration the air quality needs of all the spaces including corridors and atria. This approach was developed to produce an efficient yet inexpensive ventilation approach that did not incorporate expensive sensors and control systems. Carbon dioxide evolution predictions were made using software that evaluated the thermal response and the air quality of a building with complex topology. The numerical model used to evaluate air quality, was based on mass conservation integral equations in which the final equations system was solved through the Runge-Kutta-Fehlberg method with error control. A statistical study of the occupation cycle in the school building during the day was developed.

Evaluation of Thermal Comfort in Slightly Warm Ventilated Spaces in Nonuniform Environments

The present work analyzes and evaluates the global thermal comfort and local thermal discomfort levels of an occupant subjected to a symmetric nonuniform airflow, originated in common use ventilators. Several incident airflow directions are studied and their effects are described. The global thermal comfort level is evaluated through a multi-nodal numerical model that simulates human and clothing thermal responses, while the local thermal discomfort level is analyzed using an empirical model that predicts draft risks. The computational model of the human body and clothing thermal systems is based on the energy balance integral equations for human body tissue, blood, and clothing, as well as mass balance integral equations for the blood and transpired water in skin surface and the clothing. The human body is divided into 35 elements, each one in several layers of tissue, which could be protected through some clothing layers. A thermoregulatory system model was adapted to control the human body tissue temperature. The experimental tests were carried out in a test chamber in controlled environmental conditions; a thermal manikin was used to simulate the human posture, an indoor climate analyzer was used to measure the environmental variables around the occupant, and two ventilators were used to produce an airflow field around the occupant. The frontal and ascendant airflows from the ventilators placed in front of the occupant are characterized and their velocities around the occupant are measured for several incident angles. The global thermal comfort conditions of the occupant are evaluated both with and without ventilation, and the local thermal discomfort level is evaluated with ventilation for slightly warm, moderate environments.

Airflow around a passenger seated in a bus

The thermal comfort conditions perceived by the occupants of a bus during a typical summer are evaluated through the mapping of the flow field in the zone occupied by passengers, in terms of mean air velocity, turbulence intensity, and temperature. A full scale bus section was used in laboratory tests, with the passenger presence simulated by a thermally-regulated mannequin and the solar radiation by a panel of lamps with a spectrum similar to that of the sun. Given the symmetry of the vehicle, the only situations reproduced were those where the vehicle was subjected to radiation from the left-hand side. Measurements were performed both with and without a passenger seated in the window seat and in the aisle seat. In each case, two situations were considered, one with the solar protection curtains up and the other with them down.

Thermal behaviour simulation of the passenger compartment of vehicles

In this work, a calculus program developed with the objective of simulating the thermal behaviour in the passenger compartment of vehicles is presented. The model is based on the space-integral energy balance equations for the inside air and for the main vehicle body and surfaces. It can solve two kinds of problems. In the first one, it calculates the heat stress that the air conditioning system must equilibrate in order to satisfy predefined permanent regimen project specifications. In the second one, once imposed a particular air conditioning system and given the ambient conditions, it computes the different temperatures and heat fluxes, either in transient or steady regimens. The validation of this model was done with a railway car, in a summer situation, when it was immobilised and running. The model reproduced well the experimentally determined temperature and heat flux evolutions. However, the numeric simulation showed best agreement with the experimental results when used with the convection heat transfer coefficients, determined experimentally in this work.

Evaluation of indoor air quality in classrooms equipped with cross-flow ventilation

Abstract In this work the evaluation of indoor air quality in a classroom equipped with cross-flow ventilation is presented. A numerical methodology, based on comparison with experimental data, used in the evaluation of the air exchange rate, airflow rate and the age of the air, was applied in the first phase of this work. The evolution of carbon dioxide inside spaces, with different airflow typologies, was then predicted in the second part. The study was based on a school located in the South of Portugal. In the experimental methodology the tracer gas decay method was applied. In order to reduce the experimental time, the first minutes of the test were measured, while the remaining decay was obtained using a numerical exponential regression. Natural and forced cross-flow ventilation topologies were analyzed. In the case of forced ventilation, fresh air from the external environment was driven into the classroom through an air inlet using a supply fan. An extract strategy was also used in which stale air was mechanically extracted from the classroom. Natural ventilation consisted of opening perimeter and above-door windows. The forecast of carbon dioxide evolution was made using software that evaluates the thermal response of and air quality in a building with complex topology. The numerical model used to evaluate internal air quality was based on energy and mass conservation integral equations. These were solved using the Runge-Kutta-Fehlberg method with error control.

Study of Airflow around Occupants Seated in Desks Equipped with Upper and Lower Air Terminal Devices for Slightly Warm Environments

This paper will discuss the study of turbulent and mean airflow exiting air terminal devices and surrounding occupants seated in classroom desks for slightly warm environments equipped with personalized ventilation systems with upper and lower air terminal devices. In the turbulent airflow analysis the air root mean square, the air turbulence intensity, and the air velocity fluctuations frequencies are calculated, while in the mean airflow analysis the mean air velocity and temperature, the human body skin temperature, and the thermal comfort indexes are evaluated using a multi-node thermal regulation model for two different airflow rates. In the experimental tests made in a wood chamber a manikin, a ventilated desk, and two interior climate analyzers are used. The fluctuations of air velocity and temperature are measured in the air terminal devices and in 15 human body sections around the manikin, while the mean value of air relative humidity and mean radiant temperature are evaluated inside the experimental chamber. The mean air temperature in the air terminal devices is around 28°C (82.4°F), while the mean radiant temperature in the occupation area, the mean air temperature far from the occupation area, and the internal mean air relative humidity in the occupation area are around 28°C (82.4°F), 28°C (82.4°F), and 50%, respectively. The airflow rate in tests I and II are 25.75 m3/h (15.16 ft3/min) and 48.04 m3/h (28.27 ft3/min), respectively. The mean air velocity, root mean square, and turbulence intensity for test I are 0.59 m/s (1.94 ft/s), 0.13 m/s (0.43 ft/s), and 22.4%, in the upper air terminal device, and 0.9 m/s (2.96 ft/s), 0.15 m/s (0.49 ft/s), and 16.7%, in the lower air terminal device; while, for test II they are 1.72 m/s (5.64 ft/s), 0.16 m/s (0.52 ft/s), and 9.4%, in the upper air terminal device, and 1.06 m/s (3.48 ft/s), 0.16 m/s (0.52 ft/s), and 14.9%, in the lower air terminal device. In test I the mean air velocity and the airflow rate are higher in the lower exit air terminal device than in the upper exit air terminal device; while in test II, the opposite is true. It is also true that the skin temperature is slightly lower in test II than in test I, mainly in human body sections near the air terminal devices, such as the chest, arms, and legs. The occupant in test I conditions is thermally uncomfortable; however, in test II conditions, the obtained results are near the comfort recommendations.

Numerical Study of the Thermal Efficiency of a School Building with Complex Topology for Different Orientations

In this work a numerical model that simulates the thermal behavior of a building with complex topology and evaluates the indoor thermal and air quality, in transient conditions, is used for a school building thermal project. The program calculates the building surfaces solar radiation field, the buildings temperatures, the internal environmental variables, and the occupants comfort levels. Initially, after the numerical model is validated, the software is used to evaluate the school buildings thermal response for four different orientations, either in winter or summer conditions. The work then aims to identify uncomfortable spaces in order to propose, as an example, several solutions that could be introduced for each orientation, that would improve the thermal comfort and air quality levels to which the occupants are subjected, and decrease the buildings energy consumption levels. The information obtained from this study could be used to help a designer choose which thermal systems and solutions function best for a preferred school building orientation.

Evaluation of Local Thermal Discomfort in a Classroom Equipped with Cross Flow Ventilation

Abstract This paper presents an evaluation of the local thermal discomfort level in a classroom equipped with cross ventilation, for a typical moderate summer day in Portugal. Three different ventilation configurations based on window and door opening were considered. In each, the thermal comfort, air quality and acoustical comfort conditions were also evaluated. This experimental study was made in the South of Portugal, exposed to a Mediterranean climate. Thermal comfort was based on the PMV index, the air quality was based on the air renovation rate and acoustical comfort levels were based on the reverberation time, voice clarity, definition and early reflection ratio. The detailed local thermal discomfort analysis was based on draught risk and uncomfortable air velocity fluctuations. Other measurements included relative humidity, the radiative mean temperature, carbon dioxide concentration (tracer gas decay), and noise level decay of impulsive response. Results showed that for the warmest of weather open windows and classroom door gave the best air quality and comfort conditions.

Application of a School Building Thermal Response Numerical Model in the Evolution of the Adaptive Thermal Comfort Level in the Mediterranean Environment

Abstract In this paper, a review is made of the adaptive thermal comfort model. This is then applied and compared with the performance of the conventional thermal comfort model for a school located in a Mediterranean weather environment. Measurement data, combined with a building thermal response numerical model, are used to define the comfort performance under ambient natural ventilation and passive conditions for various classrooms. These results can then be used to identify the locations that require further measures to improve comfort, such as extra passive heat load and shading measures. The school design is based on that of an actual school and consists of three buildings, with 94 rooms. Envelope construction consists of opaque panels, 307 glazed window units and concrete floors and ceilings. The adaptive method uses external and internal environmental variables. Input data include occupation pattern and ventilation strategies. External environmental variables include air temperature, relative humidity, wind velocity and wind direction. Internal parameters include occupancy cycle, occupant activity level, clothing level, airflow rate and flow velocity. Indoor ventilation conditions are based on the airflow rate and the air velocity values measured in real classrooms. Environmental thermal comfort conditions were evaluated, in all occupied spaces, using the PMV index method of the Fanger model corrected with the adaptive model.