Search

Building automation systems installed in large commercial buildings record sub-hourly measurements from hundreds of sensors. The use of such large datasets are challenging because of missing and erroneous data, which can prevent the development of accurate prediction models of the performance of heating, ventilation, and air-conditioning equipment. The use of the transfer learning (TL) method for building applications attracted researchers to solve the problems created by small and incomplete datasets. This paper verifies the hypothesis that the deep neural network models that are pre-trained for one chiller (called the source chiller) with a small dataset of measurements from July 2013 could be applied successfully, by using TL strategies, for the prediction of the operation performance of another chiller (called the target chiller) with different datasets that were recorded during the cooling season of 2016. Measurements from a university campus are used as a case study. The results show that the initial hypothesis of this paper is confirmed. [ABSTRACT FROM AUTHOR]

This paper presents the development and use of benchmarking grey-box models for the detection and diagnosis of multiple-dependent faults (MDFDD) of a water-cooled centrifugal chiller. Models are developed using data recorded by a Building Automation System (BAS) from a central cooling plant of an institutional building. The forward residual-based fault detection model identifies a fault symptom, when the difference between the measured value of target variable and benchmarking value exceeds the corresponding threshold. For the fault diagnosis, most publications start from a known single fault and establish the impact on following variables in the system. This paper presents a rule-based backward approach. The proposed method identifies if (i) the fault symptom is correct (i.e., a variable has abnormal values), or (ii) the fault symptom is incorrect (i.e., the symptom of target variable is caused by impacts generated by other faulty variables due to the dependency between variables), or (iii) both target and regressor variables are abnormal. For testing the proposed MDFDD model, some artificial faults are inserted into the measurement data file, and results are discussed about the method potential for the application. [ABSTRACT FROM AUTHOR]

The change over time of energy performance of large chillers used in commercial and institutional buildings is seldom presented in the literature. Most studies are based on measurements of small capacities of residential air-conditioners and heat pumps. Authors found in the literature only three papers on this topic. This paper presents the evidence-based assessment of long-term performance of two identical large capacity chillers that serve the cooling system of a university campus. The paper presents the estimation of uncertainty of direct and derived measurements by using as example measurements recorded at 15 min time intervals by the building automation system during the summer of 2016. First, the paper verifies the hypothesis that weekly values of coefficient of performance (COP) of these large chillers do not change significantly over the summer of 2016. Second, the t-test also verifies the hypothesis of no significant change of seasonal COP values from 2009 to 2017. Third, a linear correlation-based model shows a variation of COP values by 1-2% per year between 2009 and 2017. The results show that the seasonal variation of COP is not statistically significant. [ABSTRACT FROM AUTHOR]

Building automation system uses several networks of sensors for continuous monitoring of building control systems for energy-efficient operation. Physical sensors are costly and need frequent calibration. The accurate measurement of supply air temperature from the air handling units (AHUs) has an important effect on the control of cooling coil, supply air temperature and supply airflow rate delivered to rooms. In the case when such a sensor gives erroneous measurements, a virtual sensor can replace temporarily the faulty sensor, and it can also be used for automated fault detection of HVAC systems. This article proposes two different gray-box (models A and B) for predicting the supply air temperature of air handling units that were developed and tested using the measurements from two buildings. The models require the measurement of three variables (mixed air temperature, cooling coil valve signal, and chilled water inlet temperature). The results of both models are discussed and compared. A sliding window approach for cross-validation of the models was also carried out. The developed grayboxmodels A and B could be integrated into BAS for virtual measurement, virtual calibration, and fault detection in HVAC systems. [ABSTRACT FROM AUTHOR]

Heating ventilation, and air-conditioning (HVAC) systems in North America are major users of energy in commercial and institutional buildings. Although modern monitoring strategies allow for energy waste reduction and building performance optimization, the poor quality or absence of measurements may compromise those strategies. Virtual sensors can provide measurements where physical sensors are not installed (virtual measurement) or can replace the faulty sensors (virtual re-calibration). The paper presents three different methods with different goals. Method A that is based on a physically-based model is applied to virtually re-calibrate one faulty temperature sensor of outdoor air temperature. Methods B and C, which are based on data-driven models, require additional short-term measurements with portable calibrated sensors. Method B is used for the virtual measurement of two physical variables (temperature and relative humidity). Method C is used for the re-calibration of one faulty sensor. The three methods use measurements from the building automation systems (BAS) of a university building. Methods A and B effectively correct the faulty measurements of the outdoor air properties as the mean absolute error (MAE) does not exceed the correspondent uncertainty. The virtual re-calibration capability of Method C varies between the training and validation periods as a consequence of the variation of weather conditions. [ABSTRACT FROM AUTHOR]

Measurements from the operation of Heating, Ventilation and Air conditioning (HVAC) system are collected by Building Automation System (BAS) mostly for control purpose. In many air handling units (AHUs) the airflow meter is not installed on the outdoor air stream to reduce the initial costs. To provide such a missing information needed for the ongoing commissioning of HVAC system, a virtual flow meter (VFM) should be implemented in the BAS. This paper presents VFMs for the heating season when the heat recovery coils are used. Two models are presented which predict the outdoor air ratio (factor α), in the absence of such an air flow meter, by using BAS trend data. Two energy balance equations, one for the air mixing box and another for the heat recovery coils, are coupled. The results from a case study building show that the two models predict the daily average outdoor air ratio of 0.89 and 0.87, respectively, compared with the reference daily average ratio of 0.86. [ABSTRACT FROM AUTHOR]

Solar combisystems supplying thermal energy for both domestic hot water and space heating can reduce primary energy consumption for residential buildings. As their overall performance depends on their design, this paper presents the development and use of a multi-objective optimization framework for the selection of configuration and equipment sizing of a residential solar combisystem in Montreal, Quebec, Canada. A generic solar combisystem model, which enables different configurations to be chosen, is first developed, and then coupled with a micro-time variant multi-objective particle swarm optimization, which was developed and used to find the non-dominated combisystem design alternatives. The life cycle cost (LCC), energy use (LCE), and exergy destroyed (LCX) of the solar combisystem are used as objective functions to find the best feasible designs. For the minimum LCC, only one flat-plate collector is required to store energy within one thermal storage tank, whereas seven evacuated-tube collectors and two thermal storage tanks are used for the minimum LCE value. The micro-time variant multi-objective particle swarm optimization (micro-TVMOPSO) algorithm used found a non-dominated solution that reduced the LLC by 29% compared with the initial design solution, with the increase of LCE by 72%. Since the LCC and LCE objective functions are conflicting, another non-dominated solution increased the LCC by 36% and reduced the LCE by 27%. The proposed multi-objective optimization framework can therefore be used to get the most out of solar thermal combisystems given specific economic and environmental conditions. [ABSTRACT FROM AUTHOR]

Solar thermal systems for domestic hot water and space heating, referred to as solar combisystems, can significantly reduce primary energy consumption for residential buildings. In most studies, single-objective optimization algorithms are used for the design and operation strategies of such complex systems. This paper presents the use of two conflicting objective functions, the life cycle cost and life cycle energy, for the optimization of a residential solar combisystem in Montreal, Quebec, Canada. Since the design the solar combisystem is treated as a multi-objective optimization problem, two different approaches for solving such problems are presented and compared: the weighted sum method (WSM) using a hybrid particle swarm optimization/Hooke-Jeeves (PSO/HJ) algorithm, and a multi-objective particle swarm optimization (MOPSO). Finally, a hybrid (MOPSO/HJ) is proposed to enhance the local search of MOPSO. The results show that the WSM was time-consuming for such an optimization problem. MOPSO/HJ was more than six times faster than the WSM. Compared with the base case combisystem, MOPSO/HJ found different design options, where the life cycle cost and life cycle energy were reduced by up to 88.6% and 63.9%, respectively. [ABSTRACT FROM AUTHOR]

This paper presents a new PCA-based method of soft fault detection and identification for the ongoing commissioning of chillers, which is composed of the three main phases: threshold model training, outliers detection, and variables identification. The method is applied to Building Automation System (BAS) trend data from a case study building. The first case used one week of data for training the threshold model, and then applied the model to two months of application data, both data sets being collected during the summer of 2009. The threshold model detected 139 outliers of normal operation, and identified the variables that caused those abnormal operation outcomes. In the second case, the measurements over the full summer of 2009 were used for training the threshold model, which was then applied to measurements of summer seasons of 2010–2015. The 2009 data set was collected from the first year of operation, and partially under commissioning. All subsequent years (2010–2015) present less than 10% of detected outliers over the entire data set for all combinations of retained Principal Components. The reduction of number of detected outliers can be due to the corrections along the first year of operation. There are still a few detected outliers due to sensors and components degradation which usually occurs in HVAC systems after initial commissioning. [ABSTRACT FROM AUTHOR]

This paper presents the application of a deep learning based model for the short-term forecasting of the electric demand in a heating, ventilation, and air conditioning system (HVAC) for the demand response programs of utility companies. The deep learning model is applied through two different approaches comparing their merits. The approaches consist of: (i) a monolithic approach that applies a single large model to forecast the target variables, and (ii) a sequential approach that consists of multiple deep learning models coupled together each targeting a specific energy load within the HVAC system. The model accuracy of both approaches is explored over two case studies applied to the same institutional building; however, the case studies differ in their data source. The first case study uses synthetic data obtained from an eQuest simulation, while the second case study uses measurement data obtained from the building automation system. Results show that the difference in forecasting error of these approaches is negligible; however, the monolithic approach required the least amount of calibration time. Next, this paper explores the application of off-site weather data applied to a building model calibrated with on-site data. The experiments demonstrated that the off-site weather data can be applied with a slight reduction in forecasting performance. [ABSTRACT FROM AUTHOR]

This article presents a virtual flow meter that estimates the chilled and condenser water mass flow rates for chillers using trend data commonly found in a building automation system. The quasi-steady-state thermodynamic mathematical model uses six different scenarios of available sensors that could be available in a building automation system. Two case studies of chillers installed in an ice rink and an institutional campus are used to demonstrate the use of the virtual flow meter tool. The estimates of virtual flow meter compared well with measurements, as the coefficient of variance of the root mean squared error and the normal mean bias error, two goodness-of-fit statistics of virtual flow meter models, are between 3.3% and 9.8% and between –6.9% and 7.8%, respectively, depending on the number of sensors available. The article also presents a discussion on the uncertainty of estimates and measurements. [ABSTRACT FROM PUBLISHER]

A very large number of measurement points are required to calibrate building energy models for the purpose of ongoing commissioning in HVAC systems. Building automation systems (BASs), common in many commercial and institutional buildings, can provide a large fraction of the required data. To reduce the time for BAS trend data analysis and export, a proof-of-concept prototype, called the Automatic Assisted Calibration Tool (AACT) was developed. The tool is a first step toward developing software that can automatically couple trend data for use in ongoing commissioning and calibrating building energy models. The AACT was tested using a case study institutional building, which proved its capacity to extract useful data and calculate indices of energy performance. [ABSTRACT FROM AUTHOR]

Detection and diagnosis of the malfunction of the heating, ventilation, and air conditioning (HVAC) systems result in more energy efficient systems with a higher level of indoor comfort. The information from the system combined with the artificial intelligence methods contributes to powerful fault detection and diagnosis. The paper presents a novel method for the detection and diagnosis of multiple dependent faults in an air handling unit (AHU) of HVAC system of an institutional building during heating season. The proposed method guided the search for faults, by using the information and operation flow between sensors. Support vector regression (SVR) models, developed from building automation system (BAS) trend data, predicted air temperature of two target sensors, under normal operation conditions without known problems. The fault symptom was detected when the residual of measured and predicted values exceeded the threshold. The recurrent neural network (RNN) models predicted the normal operation values of regressor sensors, which were compared with measurements, as the first step for the identification of fault symptoms. Rule-based models were used for fault diagnosis of sensors or equipment. Results from a case study of an existing building showed the quality of proposed method for the detection and diagnosis of the multiple dependent faults. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents the development of benchmarking models for the ongoing commissioning of heat recovery process in a cooling and heating plant that provides chilled water and heating water to HVAC (heating, ventilation and air conditioning) systems on a university campus. The heat recovered from the chillers in the summer was used for the heating water loop. The proposed ongoing commissioning approach and GUI (graphical user interface) are presented. The benchmarks were developed using measurements from the BAS (Building Automation System). The results indicated that the performance indices should be analyzed at both levels: the heat exchanger and the whole heat recovery system. Although the heat exchanger (HX3) effectiveness was reduced from an average value of 0.85 in 2008 to 0.60 in 2010, the effectiveness was greater than the benchmarks limits of 0.13–0.15. That reduction did not translate into a significantly lower performance of the whole heat recovery system. The BAS was able to control the system adaptation to this degradation of performance of HX3, by increasing the heating water flow rate, with an increase of electric power of pumps of only 6%. As a result, additional heat input to the system was never required. The whole heat recovery system works within the benchmarking limits. [Copyright &y& Elsevier]

Highlights: [•] We present results from a reduced-scale physical model of revolving door. [•] Air displaced by door rotation is two orders of magnitude inferior to previous data. [•] Increasing the door rotation speed does not increase significantly the air displaced. [•] Correlation-based models to be used for on-site estimation of the air displacement. [ABSTRACT FROM AUTHOR]

Highlights: [•] Optimized configurations have objective functions of 19–34% smaller than base case. [•] All optimal configurations have energy payback times between 5.8 and 6.6years. [•] Optimal combisystem based on cost or exergy (technical) has one solar collector. [•] Optimal combisystem based on energy or exergy (physical) has 8–9 solar collectors. [Copyright &y& Elsevier]

Abstract: This paper proposes the development of the benchmarking model for the ongoing commissioning of the operation of the refrigeration system of an indoor ice rink. The paper presents (1) the development of models for chillers, ice-concrete slab and controller, (2) the training phase in which the parameters of the benchmarking model are identified based on measurements under normal operating conditions, and (3) the comparison between measured and predicted energy performances. The proposed approach for the development of a benchmarking model can be implemented as application software in the Building Energy Monitoring System of any ice skating rink, if measurements of relevant independent variables are available. [Copyright &y& Elsevier]

Highlights: [•] We propose an information-driven evolutionary algorithm for solar building design. [•] We present a case study on a Net Zero Energy House. [•] We find design trade-offs in passive solar, renewable energy and energy efficiency. [•] Algorithm performance is compare to GenOpt using three criteria. [ABSTRACT FROM AUTHOR]

The reduction of electricity use for heating and domestic hot water in cold climates can be achieved by: (1) reducing the heating loads through the improvement of the thermal performance of house envelopes, and (2) using solar energy through a residential solar-based thermal storage system. First, this paper presents the life cycle energy and cost analysis of a typical one-storey detached house, located in Montreal, Canada. Simulation of annual energy use is performed using the TRNSYS software. Second, several design alternatives with improved thermal resistance for walls, ceiling and windows, increased overall air tightness, and increased window-to-wall ratio of South facing windows are evaluated with respect to the life cycle energy use, life cycle emissions and life cycle cost. The solution that minimizes the energy demand is chosen as a reference house for the study of long-term thermal storage. Third, the computer simulation of a solar heating system with solar thermal collectors and long-term thermal storage capacity is presented. Finally, the life cycle cost and life cycle energy use of the solar combisystem are estimated for flat-plate solar collectors and evacuated tube solar collectors, respectively, for the economic and climatic conditions of this study. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes two different types of benchmark models for the comparison of energy performance of water-cooled electric chillers: correlation-based models and Artificial Neural Network (ANN) models. Different techniques are proposed to establish the models and are evaluated with data collected from two chillers installed in an existing central cooling and heating plant. Both chillers have identical capacity and performance characteristics; however, they have quite different operating hours. The results show that models developed in this case study with 7days of data monitored at the beginning of the summer season provide accurate results over the remaining of the summer and for the following summer. The proposed Multivariable Polynomial (MP) models for chillers provide the most accurate prediction with CV(RMSE) below 7% over the remaining of the summer season, and below 8% for the following summer season. [Copyright &y& Elsevier]

Abstract: The Net Zero Energy House (NZEH) presented in this paper is an energy efficient house that uses available solar technologies to generate at least as much primary energy as the house uses over the course of the year. The computer simulation results show that it is technically feasible to reach the goal of NZEH in the cold climate of Montreal. In terms of the life cycle energy use, which considers the operating and embodied energy of the house, the energy payback time is 8.4–8.7years, when the NZEH is compared with an average house that complies with the provincial code. The energy payback ratio of the combisystem is 3.5–3.8 compared with the heating system of conventional house. By converting solar energy, the combisystem supplies at least 3.5 times more energy than the energy invested for manufacturing and shipping the system. The life cycle cost analysis of the NZEH shows, however, that due to the high cost of the solar technologies and the low cost of electricity in Montreal, financial payback is never achieved. [ABSTRACT FROM AUTHOR]

Decision models under uncertainty can be used as an alternative method for assessing the embodied energy and greenhouse gas missions of different heating equipment when detailed information is not available. The method can help in the estimation of life-cycle energy use of equipment by removing obstacles in the evaluation of embodied energy. The paper presents the estimation of environmental impacts from manufacturing a noncondensing, natural gas-fired hot-water boiler with a heat capacity of about 12 kW for residential applications, in the absence of accurate data about the materials used. Results from a case study found in the literature are used for the verification of the proposed method. The embodied energy in a hot-water boiler with the heat exchanger in aluminum, for instance, is about 20,000 MJ or about 30% of the total embodied energy of a house and in the case of stainless steel, it is about 3,000 MJ or about 5%. [ABSTRACT FROM AUTHOR]

The component-based simulation of the energy performance of refrigeration systems of an indoor ice rink requires mathematical models for several of the components including chillers, heat exchangers, the ice sheet cooling load, and the concrete slab. This paper presents two mathematical models of the thermal response of the concrete slab. Those models predict the brine temperature leaving the concrete slab and returning to the evaporator of the refrigeration system in terms of variables that can be measured in the refrigeration system: the brine temperature leaving the evaporator and entering in the ice-concrete slab and the ice temperature. Both models are calibrated and compared with measurements from an existing indoor ice rink located in Montreal. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents the exergy analysis of variable air volume (VAV) systems used in office buildings for air-conditioning. The mathematical models are developed using the engineering equations solver (EES) environment. The following indicators are presented: energy efficiency, exergy efficiency, and equivalent-CO2 emissions due to the generation of electricity used by the VAV system. [Copyright &y& Elsevier]

Abstract: The dome-covered house is an example of designing buildings by learning from the optimized biological forms from the nature in order to reduce the energy use for heating in cold climates. This paper presents the three-dimensional thermal and airflow (3D-TAF) model that predicts the impact of such dome on the heating load of the protected house. The emphasis is on the airflow model, the formulation of the system of equations, the calculation procedure, and the comparison with a CFD model. A case study of a dome located in Montreal (Canada) is presented. A linear correlation model is developed that predicts the variation of the dimensionless air temperature inside the dome with the dimensionless height. [Copyright &y& Elsevier]

Abstract: A dome-covered house is an example of designing sustainable buildings by learning from the optimized biological forms from the nature. The paper presents a 3-D thermal and air flow (3D-TAF) model that estimates the energy needs of a dome-covered house. The model is verified with CFD simulations under the COMSOL Multiphysics environment, experimental measurements and simulation results from similar structures published by other researchers. The use of the dome is expected to reduce the annual heating load of the house by 62.6%, compared with that of an isolated house located in Montreal at latitude. [Copyright &y& Elsevier]

Abstract: The paper brings new evidence on the impact of separate mechanical ventilation system on the annual energy and exergy performance of several design alternatives of residential heating systems, when they are designed for a house in Montreal. Mathematical models of residential heating, ventilation and domestic hot water (HVAC–DHW) systems, which are needed for this purpose, are developed and furthermore implemented in the Engineering Equation Solver (EES) environment. The Coefficient of Performance and the exergy efficiency are estimated as well as the entropy generation and exergy destruction of the overall system. The equivalent greenhouse gas emissions due to the on-site and off-site use of primary energy sources are also estimated. The addition of a mechanical ventilation system with heat recovery to any HVAC–DHW system discussed in the paper increases the energy efficiency; however, it decreases the exergy efficiency, which indicates a potential long-term damaging impact on the natural environment. Therefore, the use of a separate mechanical ventilation system in a house should be considered with caution, and recommended only when other means for controlling the indoor air quality cannot be applied. [Copyright &y& Elsevier]

This paper presents several correlation based models to be used for estimation of the energy performance of vapor compression screw chillers. They were developed using a detail thermodynamic model, contained in the ASHRAE Toolkit-I, for (1) identification of chiller parameters along with data from the manufacturer’s catalogue and (2) simulation of the energy performance of screw chillers. The new models were compared with those currently used for centrifugal chillers. The energy performance of screw chillers from two different manufacturers was also compared. [Copyright &y& Elsevier]

Buildings account for a significant portion of our overall energy usage and associated greenhouse gas emissions. With the increasing concerns regarding climate change, there are growing needs for energy reduction and increasing our energy efficiency. Forecasting energy use plays a fundamental role in building energy planning, management and optimization. The most common approaches for building energy forecasting include physics and data-driven models. Among the data-driven models, deep learning techniques have begun to emerge in recent years due to their: improved abilities in handling large amounts of data, feature extraction characteristics, and improved abilities in modelling nonlinear phenomena. This paper provides an extensive review of deep learning-based techniques applied to forecasting the energy use in buildings to explore its effectiveness and application potential. First, we present a summary of published literature reviews followed by an overview of deep learning-based definitions and techniques. Next, we present a breakdown of current trends identified in published research along with a discussion of how deep learning-based models have been applied for feature extraction and forecasting. Finally, the review concludes with current challenges faced and some potential future research directions. [ABSTRACT FROM AUTHOR]

Based on the monitoring and analysis of an existing house, additional evidence is brought forth on the major impact of people's behavior on the energy performance of a house. The following items are discussed: (1) The impact of renovations and people's behavior on the history of energy performance of a house; and (2) the contribution of major electric appliances on the total electricity consumption. Finally, the paper recommends the installation of energy monitoring and management systems in houses to continuously record the energy performance of the whole house and of the major end-uses, to increase the awareness of energy performance, and to provide helpful information to the house's owners for improving its performance. [ABSTRACT FROM AUTHOR]

[Display omitted] • Carbon neutrality feasibility of buildings was reviewed. • Carbon neutrality of buildings is feasible, but complexity varies. • Multi-building carbon neutrality case studies are limited. • Climate type affects neutrality; high-rise research is deficient. • Long-term lifecycle impacts of carbon measures remain unclear. Achieving carbon–neutral building stock by 2050 contributes to coping with the detrimental impacts of global warming since buildings account for almost 37% of final energy-related CO 2 emissions. This paper reviews the publications on carbon neutrality (CN) feasibility at both single and multi-building complex scales. Publications are retrieved from the Scopus database, and the snowball method is used to find the most relevant studies. The paper addresses the primary question: "Is it possible to reach the life cycle carbon neutrality of buildings?". Various information such as building life cycle carbon assessment, building characteristics (usage type and number of stories), climate type, mitigation measures, and simulation results are extracted, classified, and analyzed. Technically, reaching CN is feasible, but it is challenging given the need to implement multiple mitigation measures simultaneously, especially in the regions with low emissions intensity of the electricity grid, and may not always be economically feasible. A lack of successful case studies for large multi-building complexes, such as cities, is evident in the literature. It can be attributed to the limited availability of input data for carbon assessments as well as the complexity of simulations. Knowledge gaps in the literature and suggestions for future works are also discussed in detail. Due to the small number of studies on the topic, conclusive paths to reach the CN of buildings, specific to different climates and types of buildings, remain unclear, and thus, further research is necessary. [ABSTRACT FROM AUTHOR]

In most cases, the benchmarking models of energy use in houses are developed based on current and past data, and they continue to be used without any update. This paper proposes the method of retraining of benchmarking models by applying machine learning techniques when new measurements are made available. The method uses as a case study the measurements of heating energy demand from two semi-detached houses of Northern Canada. The results of the prediction of heating energy demand using static or augmented window techniques are compared with measurements. The daily energy signature is used as a benchmarking model due to its simplicity and performance. However, the proposed retraining method can be applied to any form of benchmarking model. The method should be applied in all possible situations, and be an integral part of intelligent building automation and control systems (BACS) for the ongoing commissioning for building energy-related applications. [ABSTRACT FROM AUTHOR]

During the past century, energy consumption and associated greenhouse gas emissions have increased drastically due to a wide variety of factors including both technological and population-based. Therefore, increasing our energy efficiency is of great importance in order to achieve overall sustainability. Forecasting the building energy consumption is important for a wide variety of applications including planning, management, optimization, and conservation. Data-driven models for energy forecasting have grown significantly within the past few decades due to their increased performance, robustness and ease of deployment. Amongst the many different types of models, artificial neural networks rank among the most popular data-driven approaches applied to date. This paper offers a review of the studies published since the year 2000 which have applied artificial neural networks for forecasting building energy use and demand, with a particular focus on reviewing the applications, data, forecasting models, and performance metrics used in model evaluations. Based on this review, existing research gaps are identified and presented. Finally, future research directions in the area of artificial neural networks for building energy forecasting are highlighted. [ABSTRACT FROM AUTHOR]

The location of neutral plane level ( NPL) of a high-rise shaft is an important factor for the evaluation of risks of smoke spreads in high-rise buildings, where the shaft internal pressure is equal to that of the building floor at the same height. The current method to determine the location of NPL assumes uniform temperature distribution inside a shaft, which causes concerns over accuracy of the predicted NPL for high-rise shafts with non-uniformly distributed temperatures during fires. To address the effect of temperature distribution on NPL location, this paper introduces a method to calculate temperature distribution and its associated NPL location based on a coupled model of smoke temperature profile, flow rate and pressure distribution inside a shaft. The measured data from a 1/3 scale experiment is used to validate the method and used to develop two empirical equations for NPL locations in terms of dimensionless numbers: one empirical equation based on shaft top temperature and the second based on shaft bottom temperature. A sensitivity study of the empirical equations is then conducted to evaluate the applicability of the developed equations when compared to the existing NPL method. It was found that for the existing NPL equation based on uniform temperature assumption may under-/overestimate the NPL locations, and the NRMSE would be over 50%, while the NRMSE of the empirical equation is only around 6% based on the results of coupled equations. For non-adiabatic shafts, the effects of non-uniform temperature distribution on NPL should be considered and the suggested empirical equations can predict NPL locations with a reasonable accuracy. [ABSTRACT FROM AUTHOR]

This article presents the development of a novel solar heat pump system using ice slurry as a latent storage material. First, the proposed system integration into a high-performance building is introduced and discussed from an operations perspective. Then, a test bench of the proposed concept is developed and used to validate a detailed simulation model. The validated simulation model is then integrated with energy models of single-family homes in Montreal, Toronto, and Vancouver to assess the potential of the system across Canada. Results demonstrate annual heating and hot water energy savings of over 60% in all three regions. [ABSTRACT FROM PUBLISHER]

Abstract: Fire protection in high-rise buildings requires a good understanding of the physics of smoke spread so that control measures can be properly undertaken. The problem is often complicated by the coupled heat and mass transfer phenomena, especially when smoke spread through vertical shafts far from a fire origin. Numerical analysis is often challenging due to limited computer resources for such large structures. This study aims to develop an analytical model of the smoke movement through a high-rise shaft under two ventilation conditions: the shaft with a given constant smoke flow rate, and with the smoke purely driven by stack effect. A hand-calculation procedure is proposed to obtain the solution to the analytical model, and demonstrated in a case of a 40-storey building with a fire located at the 1st floor. The accuracy of the analytical model is confirmed by comparisons to a numerical simulation and three experiments in the literature. It was found that the calculated profiles of smoke temperatures and shaft wall temperatures depend on the temperature attenuation coefficient α, a non-dimensional parameter associated with the geometrical and thermal properties of the smoke and the shaft. The analytical solutions of the smoke temperatures and smoke flow rates were plotted at different fire floor temperatures in non-dimensional forms, which can be used for the design of shaft smoke controls. The effect of radiation heat transfer on the calculation results was also discussed through a sensitivity study of the analytical model. It was found that the calculated smoke and shaft wall temperatures seem not quite sensitive to the radiation heat transfer in the case being studied. [Copyright &y& Elsevier]

Abstract: This paper presents the modeling and optimization of a solar assisted heat pump using ice slurry. Solar collectors are used as the primary source of thermal energy, with two distinct loops allowing the collectors to operate in series with an ice tank, or a warm water tank. Thermal energy stored in the ice tank is transferred to a warm water distribution tank via a heat pump. First, a new mathematical model of an ice slurry storage tank is presented. Validation of the model with experimental data confirms its ability to predict the ice mass and tank fluid temperatures during the charging and discharging modes of operation. The developed ice tank model is combined with the TRNSYS energy simulation program to formulate a complete model of the proposed heat pump system. This computer model then serves as a base for a mathematical optimization with the objective to minimize the energy use for heating and DHW over a single heating season. Simulated results demonstrate the potential of the optimized system in reducing the heating operating energy use of a high performance home in Montreal, QC. [Copyright &y& Elsevier]

Abstract: In situ determination of global air leakage of the building envelope is presently done with the fan depressurization test. During such test, infrared thermography could also be used to dimension unintentional small openings (cracks). In this study, thermography was used to measure in laboratory the surface temperature of single-layer walls subjected to air flow through surrogates of cracks. Two image-processing methods were developed and applied to a dataset of 36 thermograms recorded in laboratory. First, using the edge detection technique, the opening length and large width (more than 4mm) can be graphically estimated with an error of less than 8%. Second, for smaller openings, correlations for two image-processing characteristics, peak height and missing attenuation, were established. These relationships result in estimation with a relative error of less than 4% of the widths of small cracks on thermograms. The development of correlations for the spectrum of conditions found on site could be a step towards in situ quantification of air leakage areas. [Copyright &y& Elsevier]

Abstract: Shape is an important consideration in green building design due to its significant impact on energy performance and construction costs. This paper presents a methodology to optimize building shapes in plan using the genetic algorithm. The building footprint is represented by a multi-sided polygon. Different geometrical representations for a polygon are considered and evaluated in terms of their potential problems such as epistasis, which occurs when one gene pair masks or modifies the expression of other gene pairs, and encoding isomorphism, which occurs when chromosomes with different binary strings map to the same solution in the design space. Two alternative representations are compared in terms of their impact on computational effectiveness and efficiency. An optimization model is established considering the shape-related variables and several other envelope-related design variables such as window ratios and overhangs. Life-cycle cost and life-cycle environmental impact are the two objective functions used to evaluate the performance of a green building design. A case study is presented where the shape of a typical floor of an office building defined by a pentagon is optimized with a multi-objective genetic algorithm. [Copyright &y& Elsevier]

Abstract: While most of the existing artificial neural networks (ANN) models for building energy prediction are static in nature, this paper evaluates the performance of adaptive ANN models that are capable of adapting themselves to unexpected pattern changes in the incoming data, and therefore can be used for the real-time on-line building energy prediction. Two adaptive ANN models are proposed and tested: accumulative training and sliding window training. The computational experiments presented in the paper use both simulated (synthetic) data and measured data. In the case of synthetic data, the accumulative training technique appears to have an almost equal performance with the sliding window training approach, in terms of training time and accuracy. In the case of real measurements, the sliding window technique gives better results, compared with the accumulative training, if the coefficient of variance is used as an indicator. [Copyright &y& Elsevier]

Abstract: Simulation-based optimization can assist green building design by overcoming the drawbacks of trial-and-error with simulation alone. This paper presents an object-oriented framework that addresses many particular characteristics of green building design optimization problems such as hierarchical variables and the coupling with simulation programs. The framework facilitates the reuse of code and can be easily adapted to solve other similar optimization problems. Variable types supported include continuous variables, discrete variables, and structured variables, which act as switches to control a number of sub-level variables. The framework implements genetic algorithms to solve (1) unconstrained and constrained single objective optimization problems, and (2) unconstrained multi-objective optimization problems. The application of this framework is demonstrated through a case study which uses a multi-objective genetic algorithm to explore the trade-off relationship between life-cycle cost and life-cycle environmental impacts for a green building design. [Copyright &y& Elsevier]

• A framework is developed to assess overheating in existing building in whole summer. • Calibration simulation model is used using field measurements of indoor temperature. • Overheating risk in existing school building is assessed. • Climate change effect is studied using current and future extreme RSWY years. • Effective passive mitigation measures are provided for future extreme climates. Canadian buildings have been primarily designed to withstand cold and long winters, not hot summers. With climate change and the increase in the intensity and severity of heatwaves, it has become important to investigate overheating in buildings. However, there are limited studies on assessing and mitigating the overheating risk in existing buildings that house vulnerable populations in cold climates, especially in Canada. This paper provides a framework for the systematic assessment of overheating risks and the development of passive mitigation strategies to reduce the overheating risks without increasing cooling energy consumption under current and future climates with simulation models that are calibrated based on measured indoor air temperatures and outdoor weather conditions. The framework is applied to a school building built in 1958. The calibrated building model achieved an RMSE of less than 0.6 °C compared with measurements, a Maximum-Absolute-Difference of less than 1.9 °C, and a 1 °C Percentage-Error of less than 10 %. The simulation results from the calibrated model predicted 110 overheating hours during the year 2020. The use of exterior blind roll or a combination of night cooling and other mitigation measures that reduce solar heat gain can achieve acceptable thermal conditions. In 2044 (the future extreme midterm year), night cooling with the exterior blind roll shading would be required during extreme heat events. Whereas during 2090 (the future long-term extreme year), additional mitigation measures such as a cool roof may be required to achieve an acceptable level of thermal conditions in the school. [ABSTRACT FROM AUTHOR]