Search

Currently, building energy consumption prediction models are usually based on a large amount of historical operational data in high demands of building operating hours and monitoring systems. However, many buildings may lack operational data due to relatively limited monitoring systems, causing the failure to use data-driven methods to characterize the energy profile. In this context, transfer learning is a promising method to establish the knowledge transfer between many high-quality building operation datasets and a small amount of target building data, and to help predict energy consumption in the target building. This paper studies the possibility of employing transfer learning to achieve both short and long-term building energy consumption prediction. Firstly, a similarity analysis method, based on variable modal decomposition and dynamic time warping, is proposed for identifying the source buildings with the most similar energy features to the target building. Then, transfer learning for long-term prediction air-conditioning energy consumption is developed with weather parameters generated by the Morphing method as inputs. For the short-term single-step prediction, the proposed model CV-RMSE improves 81.3% (AEC) and 77.4% (EEC), respectively, compared to the prediction model that does not implement the transfer learning strategy and directly uses the target BEC data. As for the short-term multi-step prediction, the proposed model CV-RMSE improves 62.0% (AEC) and 65.5% (EEC), respectively. For the long-term prediction, the average CV-RMSE for the whole year is 6.62% and 11.15% for the proposed and directly target domain-based model, respectively. The proposed method explores the practicality of transfer learning in building energy forecasting, contributing to the use of existing building operation data for energy management at different timespan. • A similarity analysis method for building energy consumption data based on VMD-DTW is proposed. • The Seq2Seq-Transformer model is applied to short-term multi-step prediction of building energy consumption. • Transfer learning idea is first introduced to the long-term building energy consumption prediction. [ABSTRACT FROM AUTHOR]

This paper addresses the design and implementation of a model predictive control framework for temperature control in buildings zones via direct control of their thermal energy inputs. Comfort-centric approach in ensured by selecting building thermal zones to be equal to the physical building rooms. The framework integrates different identification and estimation technologies, machine learning and model predictive control to assure systematic handling of non-modelled disturbances and offset-free control. It is envisioned as the lowest level in the hierarchical decomposition of building subsystems responsible for comfort and shaping the overall thermal energy consumption in building zones. The paper shows how it is deployed on a full scale occupied skyscraper building. To enable optimization of the whole building behaviour a special focus is put on developing the possibility for interaction and coordination with other building subsystems or energy distribution grids. This ensures the scalability of the approach, computational relaxation, technology independency, cost-effective implementation and enables upscaling towards the smart grid and smart city concepts where buildings play decisive roles. [Display omitted] • Direct control of thermal energy per zone. • Enabled interaction with other building subsystems. • Integral part for upscaling towards smart grid and smart city concepts. • Deployment and verification on a scale of the whole skyscraper building. • Modular service built on top of the existing building automation infrastructure. [ABSTRACT FROM AUTHOR]

Overall energy consumption has expanded over the previous decades because of rapid population, urbanization and industrial growth rates. The high demand for energy leads to higher cost per unit of energy, which, can impact on the running costs of commercial and residential dwellings. Hence, there is a need for more effective predictive techniques that can be used to measure and optimize energy usage of large arrays of connected Internet of Things (IoT) devices and control points that constitute modern built environments. In this paper, we propose a lightweight IoT framework for predicting energy usage at a localized level for optimal configuration of building-wide energy dissemination policies. Autoregressive Integrated Moving Average (ARIMA) as a statistical liner model could be used for this purpose; however, it is unable to model the dynamic nonlinear relationships in nonstationary fluctuating power consumption data. Therefore, we have developed an improved hybrid model based on the ARIMA, Support Vector Regression (SVRs) and Particle Swarm Optimization (PSO) to predict precision energy usage from supplied data. The proposed model is evaluated using power consumption data acquired from environmental actuator devices controlling a large functional space in a building. Results show that the proposed hybrid model out-performs other alternative techniques in forecasting power consumption. The approach is appropriate in building energy policy implementations due to its precise estimations of energy consumption and lightweight monitoring infrastructure which can lead to reducing the cost on energy consumption. Moreover, it provides an accurate tool to optimize the energy consumption strategies in wider built environments such as smart cities. [ABSTRACT FROM AUTHOR]

• We decompose the building energy consumption into a global part and a local part through domain knowledge analysis. • A hybrid prediction model, combining static model and time series model, is proposed. • Proposed hybrid prediction model outperformed other machine learning models for a large margin. • The types of buildings have an influence on the model performance. The task of building energy prediction (BEP) is essential to several emerging research domains, including energy management, control optimization and fault detection. An accuracy prediction model contributes significantly to the improvement of building energy efficiency and flexibility. However, most existing prediction models are either based on single time series model or static model, making them perform poorly for long-term predictions. To solve this problem, this paper proposes a hybrid prediction model, which combines the deep ensemble (DE) model and autoregressive (AR) model together. Its motivation comes from the domain knowledge analysis of building energy consumption, where the energy consumption is decomposed into a global part and a local part. The deep ensemble model is adopted to predict its global part, and the AR model is employed to predict its local part. Comprehensive data experiments are conducted based on 50 real buildings across five building types to validate the model performance. The prediction horizon is the 24-h ahead sliding window prediction for one-year energy consumption. The results indicate that the hybrid prediction model outperforms the LSTM, DE-only, and ARIMA-only model, where its relative improvements of CV-RMSE are 28.7%, 35.98% and 18.47%, respectively. The experimental results also reveal that the types of buildings, i.e., office, lodging, will affect the model performance, which is attributed to their different user-behaviors. Based on the experimental results, it is demonstrated that the integration of static model and time series model is robust to the varying of building types and prediction steps, which should be the preferred choice of BEP task. [ABSTRACT FROM AUTHOR]

In recent decades there has been a growing awareness regarding energy consumption in buildings. Unfortunately, at a time when all building actors should get involved in the challenge to reduce energy consumption, designers cannot rely on effective tools to help them in their decision making process concerning airtightness. This literature review allows the identification of two important issues: new airtightness predictive models are complex to develop and existing airtightness predictive models do not meet the needs of designers and contractors. This paper is divided into three main parts in addition to the introduction and the conclusion. The first part deals with the key concepts of infiltration and airtightness, the second part with influencing factors and the third part with airtightness predictive models. These different chapters highlight a need for standardization regarding the metrics used for data presentation, parameters definition and statistical quantification. The lack of standardization hinders the development of a new airtightness predictive tool for designers and contractors. Along with the problem of standardization, supervision and workmanship are parameters that are difficult to model. Their significant impact can explain why designers and contractors find some existing models unreliable. This paper concludes that none of the existing models can be used in their present form as design tools. Further work should focus on the standardization of data presentation and on the development of a new airtightness predictive model. The first step in the development of such a model is to draw an appropriate classification of “air paths.” [ABSTRACT FROM AUTHOR]

This paper reviews occupancy based model predictive control (MPC) for building indoor climate control. Occupancy behavior in buildings is stochastic and complex in nature. With better understanding of occupancy presence in rooms and spaces, advanced controls, such as MPC, can be designed to achieve a more energy efficient operation, compared to more traditional control methods, while comfort is maintained. This paper starts with an overview of traditional controls implemented in buildings, and importance of occupancy based controls. Various control-oriented building modeling methods including physics-based and data-driven models are reviewed. Later on, a comprehensive review of MPC in terms of control theory, objective functions, constrains, optimization methods, system characteristics and various types of MPC is presented conducted. In principle, MPC finds an optimal sequence of control commands to optimize an objective function, considering system model, disturbances, predictions and actuation constraints. Lastly, occupancy based controls including commonly used rule-based and latest model-based controls are reviewed. In addition, a few experimental studies are presented and discussed. The paper presents a holistic overview of occupancy-based MPC for building heating, ventilation, and air conditioning (HVAC) systems, and discusses current status and future challenges. The purpose of this paper is to provide a guideline forresearchers who would like to conduct similar studies to have a better understanding of established research methods. [ABSTRACT FROM AUTHOR]

Abstract This article introduces a parameter estimation and state prediction technique called constrained dual Extended Kalman Filter (EKF) that can be used in model prediction controls (MPC) and fault detection and diagnostics (FDD) in building systems. The proposed method is an improvement over the existing nonlinear filter-based methods such as the joint EKF or Unscented Kalman Filter, which is widely adopted in previous building controls research. Case studies using both simulation and real measurements are conducted to demonstrate the proposed algorithm when used for a building thermal zone. Data from parametric simulations are used to test the thermal model's ability to capture parameter variations. Measured data from five identical offices is collected to test the performance of parameter estimates and state predictions under real-life operation. Overall, the proposed algorithm is 25% faster than a conventional EKF with improved numerical stability. It can also help mitigate numerical instabilities seen in previous EKF research. The reduced thermal model used in this paper is also capable of detecting most of the parametric changes to the thermal zone and providing reliable temperature predictions. [ABSTRACT FROM AUTHOR]

Smart building systems are opening up new markets, nevertheless the implementation of these novel technologies still lacks suitable and proven whole engineering solutions in complex buildings. This paper presents a detailed approach for the ventilation control of an underground space, as an example of application of the developed solution to a very harsh environment but also highly demanding in terms of energy consumption. The underground spaces are characterized by a particular thermal behavior, because of the continuous and huge thermal exchange they have with the outside, via the openings and the ground surrounding the majority of the building. The main objective of the developed methodology is to reduce energy consumption of ventilation control while maintaining acceptable comfort levels: succeeding in achieving this twofold goal in a real station and the generalization of the approach are the most relevant contributions of the paper. The developed solution is based on a Model-based Predictive Control algorithm used together with a proper monitoring platform. The model predictive control is based on a Bayesian environmental prediction model, which works in cooperation with a weather forecast web service, schedule-based predictions about trains and external fans and an occupancy detection system to appraise the real amount of people. The prediction model develops scenarios useful to allow the controller acting in advance in order to adapt the system to the current and future conditions of use, taking profit of the knowledge of the real ventilation demand. Finally, the proposed control architecture is applied to the Passeig de Gràcia metro station in Barcelona as a case study, validating the usefulness of the proposed approach and obtaining more than 30% of energy savings in the ventilation system, while maintaining the pre-existing comfort levels. The saving percentage values estimated by simulation are confirmed by the direct measures continuously taken on site through energy-meters. [ABSTRACT FROM AUTHOR]

Clustering methods are often used to model energy consumption for two reasons. First, clustering is often used to process data and to improve the predictive accuracy of subsequent energy models. Second, stable clusters that are reproducible with respect to non-essential changes can be used to group, target, and interpret observed subjects. However, it is well known that clustering methods are highly sensitive to the choice of algorithms and variables. This can lead to misleading assessments of predictive accuracy and mis-interpretation of clusters in policymaking. This paper therefore introduces two methods to the modeling of energy consumption in buildings: clusterwise regression, also known as latent class regression, which integrates clustering and regression simultaneously; and cluster validation methods to measure stability. Using a large dataset of multifamily buildings in New York City, clusterwise regression is compared to common two-stage algorithms that use K -means and model-based clustering with linear regression. Predictive accuracy is evaluated using 20-fold cross validation, and the stability of the perturbed clusters is measured using the Jaccard coefficient. These results show that there seems to be an inherent tradeoff between prediction accuracy and cluster stability. This paper concludes by discussing which clustering methods may be appropriate for different analytical purposes. [ABSTRACT FROM AUTHOR]

Buildings represent a large portion of total US energy usage. Religious facilities, which consume a significant percentage of the total floorspace and energy usage in the commercial sector, have generally not been the focus of efficiency studies or building energy audits. Religious facilities are characterized by unique patterns of occupancy and energy use. This paper presents the results of a long-term, in-depth energy study of architecturally similar church buildings in different climates in an effort to identify energy efficiency opportunities in religious buildings. Specifically, the paper details the experiment design and general usage trends from the buildings in different climates. A clear relationship between energy use and climate is evident, with HVAC and lighting systems consuming the majority of energy. An analysis on the prediction capabilities of using only a small subset of sensors to achieve similar results reveals the uncertainty of results by not metering all loads. [ABSTRACT FROM AUTHOR]

The data-driven method is a promising way to predict the energy consumption of buildings, however suffering from the data shortage problem in various scenarios. Even though transfer learning can improve the few-shot prediction performance by utilizing other buildings' data, the centralized approach poses potential privacy risks. To tackle this issue, the paper proposes a privacy-preserving knowledge sharing framework to facilitate the few-shot building energy prediction based on federated learning. First, a private data aggregation scheme is established to encrypt the sensitive data with shared random masks and guarantee the privacy of the data preprocessing and model optimization. Then, to alleviate the intrinsic data heterogeneity, a dynamical clustering federated learning algorithm is proposed to implement the intra-cluster and inter-cluster knowledge sharing along with the iterative clustering process for participating buildings. Finally, the network-based transfer learning approach is incorporated into the distributed framework to establish the customized model based on trained cluster models and further boost the prediction performance for each building. Extensive experiments on the Building Data Genome Project 2 (BDGP2) dataset indicate that the federated approach witnesses a desirable prediction performance while preserving the privacy of building occupants. • Propose a privacy-preserving knowledge sharing method for building energy prediction. • Establish a private data aggregation scheme for the sensitive information. • Develop a dynamical clustering federated learning algorithm for knowledge sharing. • Integrate transfer learning into the framework to build customized prediction models. [ABSTRACT FROM AUTHOR]

Energy consumption and air quality index (AQI) prediction is important for efficient heating, ventilation, and air conditioning (HVAC) system operation and management. A data-mining approach is presented in this paper for modeling and short-term prediction of the complicated non-linear system. The multilayer perceptron (MLP) ensemble performs best among the data mining algorithms discussed in this paper. A clustering-based method from preprocessing input data to construct the prediction models is proposed to decreases the prediction errors and the computational cost. The effectiveness of the proposed method is validated through a practical case study with both modeling and short-term prediction. The analytical results showed that the method was capable of reducing the prediction errors for modeling and short-term prediction by 11.05% and 12.21%, respectively, comparing with the models built without clustering method. [ABSTRACT FROM AUTHOR]

Window systems play a key role in establishing both the thermal and luminous environments within buildings, as well as the consequent energy required to maintain these for the comfort of their occupants. Various strategies have been employed to improve the thermal and optical performance of window systems. Some of these approaches result in products with relatively complex structures. Thus, it becomes difficult to characterise their optical and thermal properties for use in building performance prediction. This review discusses the experimental and numerical methods used to predict the thermal and optical behaviour of complex window systems. Following a discussion of thermal characterisation methods available in the literature that include experimental test methods, theoretical calculation methods and Computational Fluid Dynamic methods, sophisticated optical methods, such as use of Bidirectional Scatter Distribution Functions (BSDF) to optically characterise complex window systems, are introduced. The application of BSDF allows advanced daylight assessment metrics along with daylight evaluation tools to be used to realise dynamic annual prediction of the luminous environment. Finally, this paper reviews methods that permit the prediction of the combined thermal, daylight and energy behaviour of buildings that make use of complex window systems. [ABSTRACT FROM AUTHOR]

Large scale smart meter deployments have resulted in popularization of sensor-based electricity forecasting which relies on historical sensor data to infer future energy consumption. Although those approaches have been very successful, they require significant quantities of historical data, often over extended periods of time, to train machine learning models and achieve accurate predictions. New buildings and buildings with newly installed meters have small historical datasets that are insufficient to create accurate predictions. Transfer learning methods have been proposed as a way to use cross-domain datasets to improve predictions. However, these methods do not consider the effects of seasonality within domains. Consequently, this paper proposes Hephaestus, a novel transfer learning method for cross-building energy forecasting based on time series multi-feature regression with seasonal and trend adjustment. This method enables energy prediction with merged data from similar buildings with different distributions and different seasonal profiles. Thus, it improves energy prediction accuracy for a new building with limited data by using datasets from other similar buildings. Hephaestus works in the pre- and post- processing phases and therefore can be used with any standard machine learning algorithm. The case study presented here demonstrates that the proposed approach can improve energy prediction for a school by 11.2% by using additional data from other schools. [ABSTRACT FROM AUTHOR]

This paper presents a comprehensive review of the open literature on motivations, methods and applications of linking stratified airflow simulation to building energy simulation (BES). First, we reviewed the motivations for coupling prediction models for building energy and indoor environment. This review classified various exchanged data in different applications as interface data and state data , and found that choosing different data sets may lead to varying performance of stability, convergence, and speed for the co-simulation. Second, our review shows that an external coupling scheme is substantially more popular in implementations of co-simulation than an internal coupling scheme. The external coupling is shown to be generally faster in computational speed, as well as easier to implement, maintain and expand than the internal coupling. Third, the external coupling can be carried out in different data synchronization schemes, including static coupling and dynamic coupling. In comparison, the static coupling that performs data exchange only once is computationally faster and more stable than the dynamic coupling. However, concerning accuracy, the dynamic coupling that requires multiple times of data exchange is more accurate than the static coupling. Furthermore, the review identified that the implementation of the external coupling can be achieved through customized interfaces, middleware, and standard interfaces. The customized interface is straightforward but may be limited to a specific coupling application. The middleware is versatile and user-friendly but usually limited in data synchronization schemes. The standard interface is versatile and promising, but may be difficult to implement. Current applications of the co-simulation are mainly energy performance evaluation and control studies. Finally, we discussed the limitations of the current research and provided an overview for future research. [ABSTRACT FROM AUTHOR]

Abstract: Nowadays, heating energy demand has become a significant estimator used during the design stage of any new building. In this paper we are proposing a model to predict the heating energy demand, based on the main factors that influence a building''s heat consumption. It was found out that these factors are: the building global heat loss coefficient (G), the south equivalent surface (SES) and the difference between the indoor set point temperature and the sol-air temperature. In the second part of this paper, multiple dynamic simulations were carried out in order to determine the values of the inputs and output data of the future prediction model. Using the obtained database, a multiple regression prediction model was further used to develop the prediction model. In the last part of this paper the model results was validated with the measured data from 17 blocks of flats. Moreover, in this article it is also shown the comparison with the results calculated using the building''s energy certification methodology. A detailed error analysis showed that the model presents a very good accuracy (correlation coefficient of 0.987). In conclusion, the proposed model presents the following characteristics: three inputs and one output, simplicity, large applicability, good match with the simulations and with the energy certification calculations, human behavior correction. [Copyright &y& Elsevier]

Abstract: Recent results show that a predictive building automation can be used to operate buildings in an energy and cost effective manner with only a small retrofitting requirements. In this approach, the dynamic models are of crucial importance. As industrial experience has shown, modeling is the most time-demanding and costly part of the automation process. Many papers devoted to this topic actually deal with modeling of building subsystems. Although some papers identify a building as a complex system, the provided models are usually simple two-zones models, or extremely detailed models resulting from the use of building simulation software packages. These are, however, not suitable for predictive control. The objective of this paper is to share the years-long experience of the authors in building modeling intended for predictive control of the building''s climate. We provide an overview of identification methods for buildings and analyze their applicability for subsequent predictive control. Moreover, we propose a new methodology to obtain a model suitable for the use in a predictive control framework combining the building energy performance simulation tools and statistical identification. The procedure is based on the so-called co-simulation that has appeared recently as a feature of various building simulation software packages. [Copyright &y& Elsevier]

Forecasting of commercial building thermal loads can be achieved using data from Building Energy Management (BEM) systems. Experience in building load prediction using historical data has shown that data analysis is a key factor in order to produce accurate results. This paper examines the selection of appropriate input variables, for data-driven predictive models, from wider datasets obtained from BEM systems sensors, as well as from weather data. To address the lack of available complete datasets from actual commercial buildings BEM systems, detailed representation of reference buildings using EnergyPlus were implemented. Different types of commercial buildings in various climates are examined to investigate the existence of patterns in the selection of input variables. Data analysis of the simulated results is used to detect the correlation between thermal loads and possible input variables. The selection process is validated by comparing the performance of predictive models when the full or the pre-selected set of variables is introduced as inputs. [ABSTRACT FROM AUTHOR]

Building energy simulators such as TRNSYS, EnergyPlus, and Esp-r offer an excellent opportunity for detailed design of house thermal model and its Heating, Ventilating, and Air Conditioning (HVAC) system and provide very accurate simulation results useful for performance analysis and optimization process. In contrast, these energy simulators do not include sub-models of advanced devices/strategies for control of HVAC system operation and suffer from poor control mechanism. In addition to lack of an advance controller, they inherently offer no mechanism for estimating the future state of their process models based on forecast weather dataset. Hence, no predictive controller can be designed and implemented within these simulators. This paper discusses the development of a Matlab-TRNSYS co-simulator in order to control/manage a TRNSYS program, which was previously developed and calibrated based on the characteristics of a real case study house, with an advanced predictive controller. This co-simulator investigates the effectiveness of different predictive strategy planning models, including Load Shifting (LSH), Smart Dual Fuel Switching System (SDFSS), and LSHSDFSS, as the integration of fuel switching and load shifting strategy planning models on 24 h ahead energy cost saving of the case study house HVAC system. Simulation results of different consecutive sample days indicate that SDFSS could bring significant energy cost saving. However, LSH and LSHSDFSS effectiveness is sensitive to the outdoor temperature. [ABSTRACT FROM AUTHOR]

Abstract: Many systems used in buildings for heating, ventilating, and air-conditioning waste energy because of the way they are operated or controlled. This paper explores the application of model predictive control (MPC) to air-conditioning units and demonstrates that the closed-loop performance and energy efficiency can be improved over conventional approaches. This work focuses on the problem of controlling the vapor compression cycle (VCC) in an air-conditioning system, containing refrigerant which is used to provide cooling. The VCC considered in this work has two manipulated variables that affect operation: compressor speed and the position of an electronic expansion valve. The system is subject to constraints, such as the range of permissible superheat, and also needs to regulate temperature variables to set points. An MPC strategy is developed for this type of system based on linear models identified from data obtained from a first-principles model of the VCC. The MPC strategy incorporates economic measures in the objective function as well as control objectives. Tests are carried out on a simulated VCC system that is linked to a simulation of a realistic building that is developed in the U.S. Department of Energy Computer Simulation Program, EnergyPlus. The MPC demonstrated significantly better tracking control relative to conventional approaches (a reduction of 70% in terms of the integral of squared error for step changes in the temperature set-point), while reducing the VCC energy requirements by 16%. The paper describes the control approach in detail and presents results from the tests. [Copyright &y& Elsevier]

A great challenge associated with urban growth is to design for energy efficient and healthy built environments. Exploiting the potential for natural ventilation in buildings might improve pedestrian comfort and lower cooling loads, particularly in warm and tropical climates. As a result, predicting wind behavior around naturally ventilated buildings has become important and one of the most common prediction approaches is computational fluid dynamics (CFD) simulation. While accurate wind prediction is essential, simulation is complex and predictions are often inconsistent with field measurements. Discrepancies are due to the large uncertainties associated with modeling assumptions, as well as the high spatial and temporal climatic variability that influences sensor data. This paper proposes metrics to estimate the expected predictive performance of sensor configurations and assesses their usefulness in improving simulation predictions. The evaluations are based on the premise that measurement data are best used for falsifying model instances whose predictions are inconsistent with the data. The potential of the predictive performance metrics is demonstrated using full-scale high-rise buildings in Singapore. The metrics are applied to assess previously proposed sensor configurations. Results show that the performance metrics successfully evaluate the robustness of sensor configurations with respect to reducing uncertainty of wind predictions at other unmeasured locations. [ABSTRACT FROM AUTHOR]

Occupant behavior is one of the major factors influencing building energy consumption and contributing to uncertainty in building energy use prediction and simulation. Currently the understanding of occupant behavior is insufficient both in building design, operation and retrofit, leading to incorrect simplifications in modeling and analysis. This paper introduced the most recent advances and current obstacles in modeling occupant behavior and quantifying its impact on building energy use. The major themes include advancements in data collection techniques, analytical and modeling methods, and simulation applications which provide insights into behavior energy savings potential and impact. There has been growing research and applications in this field, but significant challenges and opportunities still lie ahead. [ABSTRACT FROM AUTHOR]

In this paper a building occupancy prediction method is presented, which is based on the spatio-temporal analysis of historical data (occupancy modelling) and further relies heavily on current contextual information, being therefore suitable for providing real-time prediction. Two different algorithmic approaches are proposed, based on Markov models, revealing how context awareness adds the capability of rapidly adjusting to current conditions and capturing unexpected events, as opposed to capturing only typical occupancy fluctuation expected on a regular basis. Both proposed approaches are evaluated against accurate real-life data collected from a tertiary building, achieving notable results which outperform currently used methods. [ABSTRACT FROM AUTHOR]

As a popular data driven method, artificial neural networks (ANNs) have been widely applied in building energy prediction field for decades. To improve the short term prediction accuracy, this paper presents a kind of optimized ANN model for hourly prediction of building electricity consumption. An improved Particle Swarm Optimization algorithm (iPSO) is applied to adjust ANN structure's weights and threshold values. The principal component analysis (PCA) is used to select the significant modeling inputs and simplify the model structure. The investigation utilizes two different historical data sets in hourly interval, which are collected from the Energy Prediction Shootout contest I and a campus building located in East China. For performance comparison, another two prediction models, ANN model and hybrid Genetic Algorithm-ANN (GA-ANN) model are also constructed in this study. The comparison results reveal that both iPSO-ANN and GA-ANN models have better accuracy than that of ANN ones. From the perspective of time consuming, the iPSO-ANN model has shorter modeling time than GA-ANN method. The proposed prediction model can be thought as an alternative technique for online prediction tasks of building electricity consumption. [ABSTRACT FROM AUTHOR]

Natural ventilation is a promising approach to reducing building energy use if designed properly. Most of the previous design models for calculating airflow due to single-sided natural ventilation have been based on the assumption of simple openings. Since, most windows are not simple openings, but rather can create flow obstructions when opened; the impact of window structure on ventilation needs to be accounted for in order to accurately predict the ventilation rate in buildings. This paper presents an experimental and numerical evaluation of the impact of three types of windows—hopper, awning, and casement—on airflow in the case of single-sided natural ventilation. Semiempirical models for predicting the ventilation rate were developed for these window types and validated by both large-eddy simulations and full-scale measurements. In general, the predictions agreed with the measured results within an error of 25%, and the new models can be used for the design of natural ventilation systems. [ABSTRACT FROM AUTHOR]

Model predictive control (MPC) is an advanced control that can be used for dynamic optimization of HVAC equipment. Although the benefits of this technology have been shown in numerous research papers, currently there is no commercially or publicly available software that allows the analysis of building systems that employ MPC. The lack of detailed and robust tools is preventing more accurate analysis of this technology and the identification of factors that influence its energy saving potential. The modeling environment (ME) presented here is a simulation tool for buildings that employ MPC. It enables a systematic study of primary factors influencing dynamic controls and the savings potential for a given building. The ME is highly modular to enable easy future expansion, and sufficiently fast and robust for implementation in a real building. It uses two commercially available computer programs, with no need for source code modifications or complex connections between programs. A simplified building model is used during the optimization, whereas a more complex building model is used after the optimization. It is shown that a simplified building model can adequately replace a more complex model, resulting in significantly shorter computational times for optimization than those found in the literature. [ABSTRACT FROM AUTHOR]

This paper provides a comparative study of state-of-the-art means of predicting occupancy for smart heating control applications. We focus on approaches that predict the occupancy state of a home using occupancy schedules – that is, past records of the occupancy state. We ran our analysis on actual occupancy schedules covering several months for 45 homes. Our results show that state-of-the-art, schedule-based occupancy prediction algorithms achieve an overall prediction accuracy of over 80%. We also show that the performance of these algorithms is close to the theoretical upper bound expressed by the predictability of the input schedules. Building upon these results, we used ISO 13790-standard modelling techniques to analyse the energy savings that can be achieved by smart heating controllers that use occupancy predictors. Furthermore, we investigated the trade-off between achievable savings (typically 6–17% on average) and the risk of comfort loss for household residents. [ABSTRACT FROM AUTHOR]

This paper presents an occupancy-predicting control algorithm for heating, ventilation, and air conditioning (HVAC) systems in buildings. It incorporates the building's thermal properties, local weather predictions, and a self-tuning stochastic occupancy model to reduce energy consumption while maintaining occupant comfort. Contrasting with existing approaches, the occupancy model requires no manual training and adapts to changes in occupancy patterns during operation. A prediction-weighted cost function provides conditioning of thermal zones before occupancy begins and reduces system output before occupancy ends. Simulation results with real-world occupancy data demonstrate the algorithm's effectiveness. [ABSTRACT FROM AUTHOR]

Energy in buildings is a topic that is being widely studied due to its high impact on global energy demand. This problem involves the performance of an adequate management of the energy demand, combining both convectional and renewable sources. To this end, the use of control strategies is an important tool. These control strategies can take advantage of knowledge of variables that act as disturbances in the closed loop scheme. Thus, it is of great importance the development of predictions of such variables. The main objective of this paper is to develop and assess a short-term predictive neural network model of the electricity demand for the CIESOL bioclimatic building, located in the southeast of Spain. The performed experiments show a quick prediction with acceptable final results for real data with a short-term prediction horizon equal to 60 min and with a mean error of 11.48%. One-step ahead predictions and dynamic modeling simulations have also been evaluated. [ABSTRACT FROM AUTHOR]

Implementing energy conservation measures in buildings can reduce energy costs and environmental impacts, but such measures cost money to implement so intelligent investment strategies require the ability to quantify the energy savings by comparing actual energy used to how much energy would have been used in absence of the conservation measures (known as the "baseline" energy use). Methods exist for predicting baseline energy use, but a limitation of most statistical methods reported in the literature is inadequate quantification of the uncertainty in baseline energy use predictions. However, estimation of uncertainty is essential for weighing the risks of investing in retrofits. Most commercial buildings have, or soon will have, electricity meters capable of providing data at short time intervals. These data provide new opportunities to quantify uncertainty in baseline predictions, and to do so after shorter measurement durations than are traditionally used. In this paper, we show that uncertainty estimation provides greater measurement and verification (M&V) information and helps to overcome some of the difficulties with deciding how much data is needed to develop baseline models and to confirm energy savings. We also show that cross-validation is an effective method for computing uncertainty. In so doing, we extend a simple regression-based method of predicting energy use using short-interval meter data. We demonstrate the methods by predicting energy use in 17 real commercial buildings. We discuss the benefits of uncertainty estimates which can provide actionable decision making information for investing in energy conservation measures. [ABSTRACT FROM AUTHOR]

In recent years, a large number of building energy consumption prediction models, with various intended uses, have been proposed. The majority of these models have either taken a data-driven or a physical modeling approach, with each approach having its own strengths and limitations. Towards leveraging the strengths and reducing the limitations of each approach for improved prediction performance, this paper presents a hybrid machine-learning approach for occupant-behavior-sensitive energy consumption prediction. The proposed approach is composed of three constituent models: (1) a machine-learning model that learns the impact of outdoor weather conditions from simulation-generated data, (2) a machine-learning model that learns the impact of occupant behavior from real data, and (3) an ensemble model that predicts cooling energy consumption based on the outputs of the first two models. The simulation-generated data were created through simulating a set of reference buildings in EnergyPlus. The real data were collected from an office building in Pennsylvania. The proposed hybrid model was validated on an unseen real dataset. It achieved 0.73 kWh RMSE and 9.07% CV in hourly cooling energy consumption prediction, which indicates that the proposed approach is promising. [ABSTRACT FROM AUTHOR]

This paper proposes a comparison between an online and offline Deep Reinforcement Learning (DRL) formulation with a Model Predictive Control (MPC) architecture for energy management of a cold-water buffer tank linking an office building and a chiller subject to time-varying energy prices, with the objective of minimizing operating costs. The intrinsic model-free approach of DRL is generally lost in common implementations for energy management, as they are usually pre-trained offline and require a surrogate model for this purpose. Simulation results showed that the online-trained DRL agent, while requiring an initial 4 weeks adjustment period achieving a relatively poor performance (160% higher cost), it converged to a control policy almost as effective as the model-based strategies (3.6% higher cost in the last month). This suggests that the DRL agent trained online may represent a promising solution to overcome the barrier represented by the modelling requirements of MPC and offline-trained DRL approaches. • Online- and offline-trained DRL agents were benchmarked against MPC and RBC. • A DRL agent trained online can perform near-optimally in a few weeks. • Both MPC and DRL trained offline achieved a similar performance. • DRL was capable to learn patterns beyond the horizon provided to the agent. • Online-trained DRL proved to be a viable approach for energy management. [ABSTRACT FROM AUTHOR]

Highlights: [•] This paper presents a novel methodology for modeling building energy performance. [•] The novelty lies in the use of lead-lag dynamics to present building energy system. [•] The developed model is simple and accurate with minimal physical parameters. [•] Comparison with benchmarks and measurements shows good results. [Copyright &y& Elsevier]

Abstract: The field of building energy management, which monitors and analyses the energy use of buildings with the aim to control and reduce energy expenditure, is seeing a rapid evolution. Automated meter reading approaches, harvesting data at hourly or even half-hourly intervals, create a large pool of data which needs analysis. Computer analysis by means of machine learning techniques allows automated processing of this data, invoking expert analysis where anomalies are detected. However, machine learning always requires a historical dataset to train models and develop a benchmark to define what constitutes an anomaly. Computer analysis by means of building performance simulation employs physical principles to predict energy behaviour, and allows the assessment of the behaviour of buildings from a pure modelling background. This paper explores how building simulation approaches can be fused into energy management practice, especially with a view to the production of artificial bespoke benchmarks where historical profiles are not available. A real accommodation block, which is subject to monitoring, is used to gather an estimation of the accuracy of this approach. The findings show that machine learning from simulation models has a high internal accuracy; comparison with actual metering data shows prediction errors in the system (20%) but still achieves a substantial improvement over industry benchmark values. [Copyright &y& Elsevier]

Abstract: Building energy performance simulation tools such as EnergyPlus, Ecotect, and eQuest are widely used to model energy performance of existing buildings and assess retrofit alternatives. Nevertheless, predictions from simulations typically deviate from actual measurements. Monitoring actual performance and measuring deviations from simulated data in 3D can help improve simulation accuracy through model calibrations, and in turn facilitate identification of energy performance problem. To do that, this paper presents Energy Performance Augmented Reality (EPAR) modeling that leverages collections of unordered digital and thermal imagery, in addition to computational fluid dynamics (CFD) models. First, users collect large numbers of digital and thermal imagery from the building under inspection using a single thermal camera. Through an image-based reconstruction pipeline, actual 3D spatio-thermal models are automatically generated and are superimposed with expected building energy performance models generated using CFD analysis through a user-driven process. The outcomes are EPAR models which visualize actual and expected models in a common 3D environment. Within the EPAR models, actual measurements and simulated results can be systematically compared and analyzed. The method is validated on typical residential and instructional buildings. The results demonstrate that EPAR models facilitate calibration of building energy performance models and support detection and analysis of building performance deviations. [Copyright &y& Elsevier]

Abstract: Building managers have specific duties and certain outputs that are required of them. Without the necessary data, information, tools, and time, they are unable to adequately meet their organisational goals. Scenario modelling enables explicit and unambiguous coupling of building functions with other pivotal aspects of building operation in a method that specifically considers the education and technical expertise of building managers. This new method captures, transforms, and communicates the complex interdependencies of environmental and energy management in buildings through an easily navigable, holistic, and reproducible checking mechanism that compares actual performance with predicted performance and completes the “plan-do-check-act” cycle for building managers. Most important, the structured nature of this method caters to the diverse profile of building managers, making it applicable for widespread deployment. This paper demonstrates the benefit of using the new method by examining its application to a performance analysis of two existing buildings. [Copyright &y& Elsevier]

Abstract: Energy consumption of buildings depends significantly on the criteria used for the indoor environment and building design and operation. The environmental factors that define the indoor environmental quality are: thermal comfort, indoor air quality, acoustic comfort and visual comfort. This paper approaches the numerical prediction of thermal comfort in closed spaces on the basis of PMV–PPD model, and its testing to asymmetric or nonuniform thermal radiation, as well as the indoor air quality simulation and control. The following are developed: a computation and testing model of thermal comfort in buildings, a computational model for indoor air quality numerical simulation, as well as a methodology to determine the outside airflow rate and to verify the indoor air quality in rooms, according to the European Standard CEN 1752. Also, the thermal comfort criteria for design of heating systems, the relationship between thermal environment and human performance, as well as the influence of carbon dioxide on human performance and productivity are presented. The performance of the developed computational models, implemented in two computer programs, is illustrated by using some numerical comparative applications for two building construction types. [Copyright &y& Elsevier]

Abstract: An archetypal simulation model of office building representing variability in UK office building stock by parameterising built form, construction elements, occupancy/usage and operational/control strategy has been developed thus enabling detailed energy performance simulation to be used for stock modelling and parametric studies. The paper discusses the building characteristics needed to be considered for energy performance simulation, their values, and how they can be used in parametric studies. These parameters include built forms, fabrics (including thermal mass and insulation positioning), glazing percentages and characteristics, daylight and solar control measures and activity and operational related parameters (heating and cooling set points, ventilation rate, occupancy density and metabolic rate, equipment and lighting gain). The default parameter values suggested for the archetypal simulation model reflect typical existing and currently proposed UK office building stock. An archetypal model, combined with parametric studies, can be used in assessing energy performance of building stock and evaluating adaptation/retrofitting strategies. [Copyright &y& Elsevier]

Abstract: This paper presents the upgrading of a method for predicting short-term building energy consumption that was previously developed by the authors (EUs method). The upgrade uses a time temperature curve (TTC) forecast model. The EUs method involves the use of artificial neural networks (ANNs) for predicting each independent process – end-uses (EUs). End-uses consume energy with a specific behaviour in function of certain external variables. The EUs method obtains the total consumption by the addition of the forecasted end-uses. The inputs required for this method are the parameters that may affect consumption, such as temperature, type of day, etc. Historical data of the total consumption and the consumption of each end-use are also required. A model for prediction of the time temperature curve has been developed for the new forecast method (TEUs method). The temperature at each moment of the day is obtained using the prediction of the maximum and minimum daytime temperature. This provides various benefits when selecting the training days and in the training and forecasting phases, thus improving the relationship between expected consumption and temperatures. The method has been tested and validated with the consumption forecast of the Universitat Politècnica de València for an entire year. [Copyright &y& Elsevier]

Abstract: This paper described the development of regression models which are able to predict office building annual heating, cooling and auxiliary energy requirements for different HVAC systems as a function of office building heating and cooling demands. In order to represent the office building stock, a large number of building parameters were explored such as built forms, fabrics, glazing levels and orientation. Selected parameters were combined into a large set of office building models (3840 in total). As different HVAC systems have different energy requirements when responding to same building demands, each of the 3840 models were further coupled with five HVAC systems: VAV, CAV, fan-coil system with dedicated air (FC), and two chilled ceiling systems with dedicated air, radiator heating and either embedded pipes (EMB) or exposed aluminium panels (ALU). In total 23,040 possible scenarios were created and simulated using EnergyPlus software. The annual heating and cooling demands and their HVAC system''s heating, cooling and auxiliary energy requirements were normalised per floor area and fitted to two groups of statistical models. Outputs from the regression analysis were evaluated by inspecting models best fit parameter values and goodness of fit. Based on the described analysis, the specific regression models were recommended. [Copyright &y& Elsevier]

Abstract: This paper investigates the potential of using occupancy information to realize a more energy efficient building climate control. The study focuses on Swiss office buildings equipped with Integrated Room Automation (IRA), i.e. the integrated control of Heating, Ventilation, Air Conditioning (HVAC) as well as lighting and blind positioning of a building zone or room. To evaluate the energy savings potential, different types of occupancy information are used in a Model Predictive Control (MPC) framework, which is well-suited for this study due to its ability to readily include occupancy information in the control. An MPC controller, which controls the building based on a standard fixed occupancy schedule, is used as a benchmark. The energy use of this benchmark is compared with three other control strategies: first, the same MPC controller which uses the same schedule for control as the benchmark, but turns off the lighting in case of (an instantaneous measurement of) vacancy; second, the same MPC controller which uses the same schedule as the benchmark for control, but turns off lighting and ventilation in case of (an instantaneous measurement of) vacancy; and third, the same MPC controller as the benchmark but using a perfect prediction about the upcoming occupancy. This comparison is carried out for different buildings, HVAC systems, seasons and occupancy patterns in order to determine their influence on the energy savings potential. [Copyright &y& Elsevier]

Abstract: The commercial building sector has become the focus of many governmental energy reduction initiatives to achieve more sustainable development. Reducing building energy use starts by improving the design of buildings. To this end, energy modeling and simulation tools are used during the design phase to predict energy use and help designers choose and size the different building systems. Large discrepancies are however being observed between predicted and actual building energy performances. In order to determine the sources of errors and improve these predictions, the sensitivity of energy models to different input parameters needs to be evaluated. Studies in literature have extensively evaluated the sensitivity of models to the buildings’ technical design parameters. However, none considered the parameters related to the energy consumption behavior of occupants, leaving their impact on energy modeling unknown. This paper presents a comprehensive sensitivity analysis study performed on the occupancy behavioral parameters of typical office buildings of different size and in different weather zones. Significant sensitivity levels were observed, varying according to both building size and weather conditions. The highest sensitivity was obtained when varying the ‘heating temperature set point’ parameter in small-size buildings located in US weather zone 2 Dry. [Copyright &y& Elsevier]

Abstract: Model predictive control has become a widespread solution in many industrial applications and is gaining ground in the field of energy management and automation systems of buildings. A model with reasonable prediction properties is an ultimate condition for good performance of the predictive controller. This paper presents an approach in which a model of a building is selected by an iterative two stage procedure. In the first stage, a minimum set of disturbance inputs is formed so that the resulting model is the best with respect to a defined quality criterion; then the second stage comprises addition of the states to obtain the final minimum set of states maximizing the model quality. The procedure stops when it makes no sense to select more complex model as it brings no more quality improvements. Statistical tests such as the likelihood ratio test, the tests based on cumulative periodogram, the two-sample Kolmogorov–Smirnov test as well as others (fit factor and coefficient of determination) are used to evaluate the relationship between the addition of inputs/states and the model quality. Three identification approaches, namely model predictive control relevant identification, deterministic semi-physical and probabilistic semi-physical modeling are used for estimation of building parameters. Finally, a case study is provided where all the above mentioned approaches are investigated and tested. [Copyright &y& Elsevier]

Abstract: Integrating passive systems such as cross and stack ventilation, a solar chimney, a wind tower or an earth-air tunnel into a building envelope has the potential to significantly reduce the ventilation and air-conditioning demand on buildings. However the application of combination of these features in low rise contemporary buildings is limited. Among the major reasons, is the lack of simple building modelling techniques preventing effective integrated passive systems. Existing coupled multi-zone airflow and thermal modelling software such as COMIS-TRNSYS, CONTAM-TRNSYS or TRNFLOW do not include the passive airflow components which require simultaneous prediction of temperature and airflow rate in the components such as solar chimneys and wind-induced earth-air tunnels. This paper develops an integrated model incorporating these passive airflow components into a coupled multi-zone ventilation and building thermal model. The model is validated against COMIS-TRNSYS software for a lightweight building with large openings. The prediction of each passive airflow component is validated with available published analytical and experimental findings. The solar chimney model is able to predict reverse flow and accurately predicts the air temperature in the chimney. The earth air tunnel (EAT) considers the transient heating up effect of the soil during operation and predicts the outlet air temperature. [Copyright &y& Elsevier]

• A prediction model of lighting and shading coupling control behavior is proposed. • The preferences and control habits of occupants were investigated and classified. • The indoor work plane illumination was predicted by a machine learning model. • The preference diversity of occupants greatly affect the prediction accuracy. Lighting control in office buildings is driven by occupant's demand for indoor light environment. The control behavior not only has a direct impact on occupants' visual comfort, but also relates with the building lighting energy consumption. However, due to the effect of glare, lighting control is often associated with shading adjustment. In this regard, this paper proposed a prediction model which can accurately describe the lighting and shading coupling control behavior by fully considering the difference and diversity of occupants. The light environment preferences and the usage habits of lighting and shading system of occupants were firstly investigated and classified by means of questionnaire. Markov model and log-logistic survival model were introduced to quantitatively describe the probability distribution of various shading and lighting control behaviors. On this basis, combined with the indoor workplane illumination prediction model, the behavior of occupant's lighting and shading coupling control can be predicted. By comparing the four models considering or not considering the diversity and coupling effect, it is found that the proposed coupling prediction model shows better performance, the maxium error rate is only 13.04% for the lighting energy consumption prediction. [ABSTRACT FROM AUTHOR]

• A novel integration model is proposed for building energy prediction. • The integration framework is developed via meta-features and layered structure. • Model performance is validated from accuracy, generalization, and robustness. • The novelty is demonstrated by comparison with existing models. • This study enriches the diversity of energy consumption prediction models. Building energy consumption prediction plays an irreplaceable role in energy planning, management, and conservation. Constantly improving the performance of prediction models is the key to ensuring the efficient operation of energy systems. Moreover, accuracy is no longer the only factor in revealing model performance, it is more important to evaluate the model from multiple perspectives, considering the characteristics of engineering applications. Based on the idea of model integration, this paper proposes a novel improved integration model (stacking model) that can be used to forecast building energy consumption. The stacking model combines advantages of various base prediction algorithms and forms them into "meta-features" to ensure that the final model can observe datasets from different spatial and structural angles. Two cases are used to demonstrate practical engineering applications of the stacking model. A comparative analysis is performed to evaluate the prediction performance of the stacking model in contrast with existing well-known prediction models including Random Forest, Gradient Boosted Decision Tree, Extreme Gradient Boosting, Support Vector Machine, and K-Nearest Neighbor. The results indicate that the stacking method achieves better performance than other models, regarding accuracy (improvement of 9.5%–31.6% for Case A and 16.2%–49.4% for Case B), generalization (improvement of 6.7%–29.5% for Case A and 7.1%-34.6% for Case B), and robustness (improvement of 1.5%–34.1% for Case A and 1.8%–19.3% for Case B). The proposed model enriches the diversity of algorithm libraries of empirical models. [ABSTRACT FROM AUTHOR]

Decarbonisation of the building sector is driving the increased use of heat pumps. As increased electrification of the heating sector leads to stress on the electricity grid, the need for district level coordination of these heat pumps emerges. This paper proposes a novel hierarchical coordination methodology, in which a price coordinator reduces the total instantaneous power demand of a building network below a power supply limit using a price signal. Each building has a model predictive controller (MPC) which maximises thermal comfort and minimises electricity costs. An additional term in the MPC objective function penalises the heat pump power demand quadratically, which when multiplied by a pseudo electricity price allows the price coordinator to reduce the peak power demand of the building network. A 2 building network is studied to analyse the price coordinator algorithm's behaviour and demonstrate how this approach yields a trade off between comfort, energy consumption and peak demand reduction. A 100 building network case study is then presented as a proof of concept, with the price coordinator approach yielding results similar to that of a centralised controller (less than 0.7% increase in energy consumption per building per year) and a roughly fourfold decrease in computation time. [ABSTRACT FROM AUTHOR]