Real Office Building Research Articles

Multi-point temperature or humidity prediction for office building indoor environment based on CGC-BiLSTM deep neural network

The aim of this study is to predict the temperature or humidity changes at multiple relevant points in a building using a deep neural network architecture with multi-task learning to provide more reference information for the design and optimal operation of heating and ventilation systems. For this purpose, traditional multi-task prediction algorithm architecture is combined with Customized Gate Control and other neural networks to build deep neural network architectures for indoor environments with multi-point temperature or humidity prediction tasks. To test the prediction effectiveness of the architecture, a task of predicting temperature or humidity 24h in advance was designed on a real office building indoor environment dataset, and the prediction results were compared with other single-task and multi-task prediction models. Two experimental conditions were designed for this study, one using the complete training set and the other reducing the training set at a certain point. Through the final prediction results, it is found that the multi-task prediction architecture used in this paper shows better or nearly optimal results compared to other prediction models under both working conditions. This study provides some reference value for the application of multi-task prediction algorithms to the task of predicting indoor environments in buildings.

Reinforcement learning optimal control method for multi chiller HVAC system in an existing office building

Abstract Given that buildings consume approximately 33% of global energy, and HVAC systems contribute nearly half of a building’s total energy demand, optimizing their efficiency is imperative for sustainable energy use. Many existing buildings operate HVAC systems inefficiently, displaying non-stationary behavior. Current reinforcement learning (RL) training methods rely on historical data, which is often obtained through costly modeling or trial-and-error methods in real buildings. This paper introduces a novel reinforcement learning construction framework designed to improve the robustness and learning speed of RL control while reducing learning costs. The framework is specifically tailored for existing office buildings. Applying this framework to control HVAC systems in real office buildings in Beijing, engineering practice results demonstrate: during the data collection phase, energy efficiency surpasses traditional rule-based control methods from the previous year, achieving significantly improved energy performance (a 17.27% reduction) with minimal comfort sacrifices. The system achieves acceptable robustness, learning speed, and control stability. Reduced ongoing manual supervision leads to savings in optimization labor. Systematic exploration of actions required for RL training lays the foundation for RL algorithm development. Furthermore, by leveraging collected data, a reinforcement learning control algorithm is established, validating the reliability of this approach. This construction framework reduces the prerequisites for historical data and models, providing an acceptable alternative for systems with insufficient data or equipment conditions.

Fault prioritisation for Air Handling units using fault modelling and actual fault occurrence data

Fault impact analysis is an important step in developing Fault Detection and Diagnosis (FDD) tools for Heating, Ventilation and Air-Conditioning (HVAC) systems. It provides an insight into how HVAC systems react in the presence of operational faults and helps to prioritise which faults to focus on. The research in the existing literature has mostly focused on modelling several faults occurring in various HVAC systems and components to evaluate the effect of the fault on energy performance and the predicted thermal comfort of occupants. However, the real frequency of fault occurrence, an important factor which affects the overall impact a fault has on the system’s performance, has barely been addressed in the reviewed literature. This paper addresses this gap by considering real fault occurrence frequency data in addition to the effect of the faults on the energy performance of the building to assess their total energy impact. A real office building in the Netherlands was modelled using DesignBuilder and its energy performance was simulated using EnergyPlus. Air Handling Unit (AHU) faults were introduced using the native fault object and Energy Management System (EMS) features available in EnergyPlus. The fault occurrence data was extracted from AHU work orders using text mining. The fault modelling and text mining results were combined to get the total energy impact of the fault and, subsequently, the faults were prioritised using the Pareto principle. The research identified that fan failure, Heat Recovery Wheel (HRW) failure, fan stuck at 50% and Heating Coil Valve (HCV) stuck at 0% are the priority faults for winter and fan failure is the priority fault for summer.

Challenges to energy retrofitting of existing office buildings in high-rise high-density cities: The case of Hong Kong

Achieving carbon neutrality by 2050 has become a global goal, sparking concerns regarding energy consumption and carbon emissions in building operations. Office buildings in high-rise high-density cities serve as central business districts, contributing significantly to the city’s economic activity and consuming a lot of energy. The process of retrofitting existing office buildings for energy efficiency in high-rise high-density cities tends to be challenging. However, there is a lack of comprehensive understanding of the challenges involved in office buildings’ energy retrofitting, as they have not been thoroughly explored. This study aims to investigate the challenges to the existing office building energy retrofitting (EOBER) in high-rise high-density cities with real cases in Hong Kong. Initially, a systematic literature review was undertaken to identify 49 potential EOBER challenges and categorized into seven groups: technical, financial, institutional, social, environmental, regulatory, and other categories. Afterward, 23 EOBER challenges were identified through 24 semi-structured interviews with 36 real office building energy retrofitting cases in Hong Kong. Moreover, these challenges were quantified by the Z-numbers-based Delphi survey and analysis. Results show that regulatory challenges are the primary challenges, followed by financial challenges. The lack of government incentives, policies, legislation and regulations significantly hinders practitioners’ ability to engage in energy retrofitting initiatives. The long payback period of building energy retrofitting poses a critical financial concern for practitioners embracing such initiatives. In the end, this research proposed integrated strategies to tackle these challenges and increase building energy efficiency, including launching financial and regulatory incentives, shortening the interval for mandatory energy audits, disseminating knowledge, and diversifying finance channels of building energy retrofitting. The findings contribute to the body of knowledge by employing systems thinking to identify and evaluate EOBER challenges in high-rise high-density cities through empirical methodologies. Moreover, this study provides valuable references for practitioners in navigating these challenges and minimizing risks associated with the retrofitting process.

Multi-source domain generalization deep neural network model for predicting energy consumption in multiple office buildings

The effective prediction of building energy consumption can be used to optimize building operating modes and reduce the overall energy consumption and carbon emission of the building. To predict the energy consumption of the same type of building in the same area, it is often necessary to train separate prediction models for different buildings, which is usually costly computationally in order to get better prediction results. Therefore, this study will combine deep neural networks and multi-source domain generalization in an encoder-decoder architecture to train a model that can be directly applied to predict the energy consumption of multiple buildings in same type. In order to validate the predictive effectiveness of the model, it will be tested on real office building energy consumption dataset and different comparative experiments will be designed on the source and target domains. The results show that the constructed multi-source domain generalization model is able to accurately predict the energy consumption trend of different buildings 1 h in advance. It also has some energy consumption prediction ability for the unknow training set of buildings.

Innovative bio‐based steel envelope systems

AbstractThe INCSEB project is aimed at developing 5 ultra‐low carbon envelope systems made in steel and that incorporate innovative use of wood fiber insulation. The systems under development are: 3 sandwich panels (cladding, pitch & flat roofing), 2 claddings (double skin & façade with plank). INCSEB includes tests to study various behaviors: mechanical, fire, thermal calculation, acoustic insulation, air/water/vapor permeability. Interpretation and classification reports will be done for each type of test. 2 full scale prototypes will allow to estimate the durability in real life conditions. LCA indicators and comparison of carbon footprint (GWP) to conventional systems based on a real office building will be provided. An economic study of the projected price of the 5 systems compared to existing conventional systems is also planned. Besides, as far as the production of innovative panels is concerned, it appears that they can be integrated in current production lines. Finally, actions to facilitate the dissemination of results and industrial application will be organized: drafts of design and installation guides, publications, submission of 3 amendments to relevant EN standardization committees, workshop, BIM objects. Lastly, the manufacturers plan to apply for patents protection.

Demand response via optimal pre-cooling combined with temperature reset strategy for air conditioning system: A case study of office building

Demand response (DR) through air conditioning (AC) systems in buildings is a promising way to balance the power grid. However, the majority of existing DR strategies are rule-based. The other ones are model-based but generally have adopted oversimplified AC system models. Consequently, the effect of model-based optimal control cannot be evaluated properly. To address these limitations, a co-simulation framework is developed to minimize power consumption and electricity costs by adopting both pre-cooling and temperature reset strategies. The co-simulation framework combines an EnergyPlus model that is calibrated by measured data and the genetic algorithm through Functional Mock-up Unit socket. Both time-of-use electricity prices and the constraints of thermal comfort are considered in the optimization problem. A real office building with a variable refrigerant flow (VRF) system is used to test the co-simulation framework. Compared with rule-based strategies, the power consumption of the VRF system following the optimal control strategy during the DR period is reduced by 80.12%, 74.70% and 55.60%, and the daily electricity cost is reduced by 14.98%, 12.11% and 8.35%, respectively. The quantitative analysis also reveals that 37.96% of the cold energy stored by the envelope in the pre-cooling period is released into the air during the DR period.

Prediction models of bioaerosols inside office buildings: A field study investigation

Bioaerosols formed by microorganisms in the air directly affect people’s health. The air quality in an office building in Shenzhen, China, is investigated and pollutant levels measured on 36 occasions; six times for each of six indoor spaces. A relationship between indoor bioaerosols and environmental factors was determined using both linear regression analysis and Poisson regression analysis. Our results and analysis indicate that linear regression is a poor predictor for the concentration of bioaerosols based on a single indicator. In contrast, Poisson regression can better predict the concentration of bioaerosols, and PM10 may be the indicator with the greatest impact on bioaerosols. As a result, a simple, fast, and low-cost online monitoring method for monitoring indoor bioaerosols is developed and reported. Our paper provides first-hand basic data to predict the indoor bioaerosol concentration and helps to formulate appropriate monitoring guidelines. The proposed method offers more practical values compared to existing studies as our prediction model facilitates estimation of the concentration of bioaerosols at low cost. Additionally, due to the current maturity and low cost of indoor environmental sensors, the proposed method is suitable for large-scale deployment for most buildings. Practical application Based on measurement data from a real office building, our investigation explores the relationship between indoor microorganisms and building environmental indicators through a combination of probability analysis and actual measurement. We establish a novel indoor microbial prediction model using the Poisson regression model. Our work presents an effective, low-cost, method for estimating the concentration of bioaerosols and discusses the possibility for large-scale deployment of microbial monitoring equipment inside buildings which may then support real-time monitoring of indoor microbial concentration to provide healthy indoor environments for personnel.

A framework using BIM and digital twins in facilitating LCSA for buildings

The building and construction industry generates 40% of annual global CO2 emissions, aside from other environmental and social impacts, while it contributes to economic and societal development as a whole. Life Cycle Sustainability Assessment (LCSA) is a proven methodology to assess the environmental, social and economic impacts of a product system over its lifecycle. Applying LCSA to the construction domain is not trivial, as it needs to gather and connect data across several fields. A clear research gap was identified in the lack of data integration and completeness to carry out LCSA. This paper addresses this gap and shows how the complementary use of Building Information Modelling (BIM) and Digital Twin (DT) sourced data can facilitate and improve the LCSA of buildings. It includes a review of the current state of the art; proposes a novel methodological framework based on existing standards, and emphasises the roles of BIM and DT for conducting LCSA in the building context; this is tested in a case study of a real office building. The outcomes of the study demonstrate that the inclusion of usually neglected social impacts can be considered under LCSA, and that BIM and DT data provides better impact estimates, which are necessary to evaluate the performance gap between design estimated impacts and measured ones. The challenges of implementing such a framework, as well as the limitations of existing technologies are also highlighted and discussed.

An inquiry into the effect of thermal energy meter density and configuration on load disaggregation accuracy

Initial and maintenance costs often prevent dense submeter installations that enable room-level thermal energy monitoring. Previous studies suggested that building automation system (BAS) trend data represents an untapped potential to disaggregate existing meter data for heating and cooling into device- and system-level end-uses. These techniques disaggregate meter data by analyzing trend data that provide contextual information regarding the operating status of energy-consuming equipment. However, the level of submetering required to enable end-use disaggregation has yet to be studied. To this end, this paper investigates the effect of submeter density and configuration on the performance of a regression-based disaggregation strategy using BAS trend data as predictors. The method was evaluated in two steps; first, using synthetic meter and BAS trend data generated by a building performance simulation (BPS) model of a government office building, and second, with submeter data from a real office building. The results highlight the factors affecting the minimum number of heating energy submeters needed to be installed in both buildings for accurate device- and system-level disaggregation. The methodology presented in the paper can also inform changes in building design codes and standards regarding the minimum density and appropriate configuration for submetering.

Grey-box model and neural network disturbance predictor identification for economic MPC in building energy systems

This paper presents a novel model identification procedure to treat the occupancy induced unmeasured heat disturbance in the building model identification step for a model predictive control (MPC) implementation. First, we present a grey-box identification framework to estimate a dynamic building model that systematically accounts for the large unmeasured disturbance in the training data. Then, we discuss an approach to compute the confidence intervals (CIs) on the parameters estimated in the building model. Next, we develop a framework to identify a neural network (NN) using historical operational data to give the feedforward predictions of the heat disturbance in the MPC optimization problem. Further, we discuss the application of the composite building model and the NN disturbance predictor for energy cost optimization using an economic MPC formulation. We present case studies using a simple two time scale based building model, and demonstrate the effectiveness of the proposed grey-box model and the NN disturbance predictor identification procedures. We perform closed-loop simulations to highlight that additional energy cost savings of up to 4–5% are provided by including the feedforward NN disturbance predictions in the MPC controller. Finally, we also demonstrate the suitability of the proposed two-step model identification procedure to estimate accurate and reliable models using an industrial data set collected from a real office building.

Model predictive control of a building renewable energy system based on a long short-term hybrid model

Considering solar photovoltaic (PV) usage in building energy systems (BES), energy systems that combine batteries and solar PV (PVB) have been widely used in buildings. Ensuring the safety of the battery and the operation requirements of other types of equipment to the extent possible while achieving the optimization goal has become an important challenge in the implementation of this system. In this case, model predictive control (MPC) has gained increasing attention in the operation control of BES owing to its strong fitting ability. However, the prediction models at the core of MPC are not adequately designed for the BES optimization problem, and numerous studies have not yet been modeled and tested in real buildings. In this study, we used a long- and short-term hybrid prediction model in the MPC framework. The proposed prediction models and MPC framework were validated using a measured dataset from a real office building in Japan. The results show that, compared with the currently used baseline control logic, the proposed hybrid prediction MPC framework can improve the battery safety index by 81.6%, while the combined heat and power continuous operation index can also be improved by 36.4%. Compared with the conventional MPC framework, the proposed hybrid prediction MPC framework improves the optimization of off-grid operation by approximately 69% while maintaining the remaining optimization objectives.

Analysis of daylight metrics based on the daylight autonomy (DLA) and lux illuminance in a real office building atrium in Tehran

A major part of an energy-efficient architecture approach is an atrium that can benefit from daylight as a controlling strategy to decrease energy loss. The reflectance characteristic of the surface is one of the main parameters for assessing daylight in an atrium building. Daylight factor was announced as the main variable for depicting different diagrams and deriving some crucial conclusions; meanwhile, other factors such as daylight autonomy could show some significant outputs. Office buildings are as important as residential places for spending time. Hence, saving energy and optical comfort should be managed based on the architectural aspects of a building. Accordingly, herein, the reflectance distribution and Well Index are simultaneously assessed. For obtaining the optimum illuminance, validation and simulation are done based on the real atrium office building in Tehran. The results revealed that the reflectance distribution on a uniform model and a striped model, and geometries of a building could have a significant effect on the vertical daylight outputs. When the reflectance distribution was horizontal with the black color at the top of the wall with a reflectance of 0.02, and one white band under with a reflectance of 0.85, the result is the optimized illuminance and DLA.

Multi-agent reinforcement learning dealing with hybrid action spaces: A case study for off-grid oriented renewable building energy system

With the application of renewable energy in building energy systems (BES), an increasing number of power grids require building energy systems coupled to realize off-grid operation which is one type of energy flexible and grid responsive operations. In this case, deep reinforcement learning (DRL) algorithms have gained more and more attention in the operation control of BES due to their strong fitting ability and model-free utilization characteristics. However, mainstream DRL algorithms cannot solve the reinforcement learning problem of hybrid action spaces, which also restricts its further application in BES including variety of energy sources. In this paper, we firstly use a multi-agent deep reinforcement learning algorithm (MADRL) to solve the RL problem of hybrid action spaces in the building controls domain. The proposed algorithm is validated on a measured dataset of a real office building in Japan. The results show that compared to the currently used baseline control logic, MADRL can achieve a 60% improvement in off-grid operation tasks. For battery safety, MADRL can reduce unsafe battery runtime by at least 80%. Furthermore, through experiments we find that the single-agent DRL algorithm cannot solve the reinforcement learning problem with hybrid action spaces. The MADRL framework achieves stable training and optimization by layering problems and agents.

Field demonstration and implementation analysis of model predictive control in an office HVAC system

Model Predictive Control (MPC) is a promising technique to address growing needs for heating, ventilation, and air-conditioning (HVAC) systems to operate more efficiently and with greater flexibility. However, due to a number of factors, including the required implementation expertise, lack of high quality data, and a risk-adverse industry, MPC has yet to gain widespread adoption. While many previous studies have shown the advantages of MPC, few analyzed the implementation effort and associated practical challenges. In addition, previous work has developed an open-source, Modelica-based tool-chain that automatically generates optimal control, parameter estimation, and state estimation problems aimed at facilitating MPC implementation. Therefore, this study demonstrates usage of this tool-chain to implement MPC in a real office building, discusses practical challenges of implementing MPC, and estimates the implementation effort associated with various tasks in order to inform the development of future workflows and serve as an initial benchmark for their impact on reducing implementation effort. This study finds that the implemented MPC saves approximately 40% of HVAC energy over the existing control during a two-month trial period and that tasks related to data collection and controller deployment activities can each require as much effort as model generation.

Mapping and modelling defect data from UAV captured images to BIM for building external wall inspection

With the increase of service years, external walls of high-rise buildings tend to suffer from a variety of defects which impose great safety risks. Traditional methods for inspecting high-rise building external walls require inspectors to work at height and identify defects manually, which is dangerous and inefficient. In recent years, there has been an increasing trend of using unmanned aerial vehicles (UAV) for inspecting building external walls, but how to manage the information obtained by the UAV is still a problem. In addition, although building information modelling (BIM) with rich geometric and semantic information has been applied in the construction engineering industry, BIM models usually lack updated condition data of facilities. Therefore, this paper presents a method for managing the inspection results of building external walls by mapping defect data from UAV images to BIM models and modelling defects as BIM objects. First, images of building external walls obtained by UAV are processed and useful information such as coordinates are extracted. Considering the small scale of single buildings, a simplified coordinate transformation approach is developed to transform location of real-world defects to coordinates in the BIM model. Meanwhile, a deep learning-based instance segmentation model is developed to detect defects in the captured images and extract their features. In the end, the identified defects are modelled as new objects with detailed information and mapped to the corresponding location of the related BIM component. To validate the feasibility, the proposed method has been applied to a real office building, which successfully mapped and integrated the defects of external walls with the BIM model. This study is applicable to both buildings and infrastructure, and is expected to facilitate structure inspection and decision making in maintenance with integrated data of as-is condition and as-built BIM.

Analysis of Daylight Metrics Based on the Daylight Autonomy (Dla) and Lux Illuminance in a Real Office Building Atrium in Tehran

Analysis of Daylight Metrics Based on the Daylight Autonomy (Dla) and Lux Illuminance in a Real Office Building Atrium in Tehran

Energy Management Model for HVAC Control Supported by Reinforcement Learning

Heating, ventilating, and air conditioning (HVAC) units account for a significant consumption share in buildings, namely office buildings. Therefore, this paper addresses the possibility of having an intelligent and more cost-effective solution for the management of HVAC units in office buildings. The method applied in this paper divides the addressed problem into three steps: (i) the continuous acquisition of data provided by an open-source building energy management systems, (ii) the proposed learning and predictive model able to predict if users will be working in a given location, and (iii) the proposed decision model to manage the HVAC units according to the prediction of users, current environmental context, and current energy prices. The results show that the proposed predictive model was able to achieve a 93.8% accuracy and that the proposed decision tree enabled the maintenance of users’ comfort. The results demonstrate that the proposed solution is able to run in real-time in a real office building, making it a possible solution for smart buildings.

Automatic inspection data collection of building surface based on BIM and UAV

Conventional high-rise building surface inspection is usually inefficient and requires the inspectors to work at heights with high risk. Unmanned aerial vehicles (UAVs) carrying optical or thermal cameras are currently widely utilized as an effective tool for inspection. The UAV-based data collection, especially for unreachable inspection areas, is the basis of unmanned inspection of building surface. In addition, building information modeling (BIM) with rich geometric and semantic information can also be instrumental in building surface inspection. Therefore, this paper presented an automatic inspection method of building surface, especially for the inspection data collection, by integrating UAV and BIM. To minimize the length of UAV flight while collecting complete and high-quality image data considering the limited endurance capability, the coverage path planning problem is solved using genetic algorithm (GA). The required inspection areas are obtained from the BIM model of the target building to be inspected. To further enhance the automation of building surface inspection, the optimized UAV flight mission parameters are rapidly calculated based on the BIM model and proposed algorithm. A real office building in Shenzhen University campus is used to validate the presented automatic method. The quality of the collected inspection images using the UAV with optimized flight mission are evaluated. The results show that this method leads to time-efficient, accurate, and high-quality inspection data collection for building surface.

Analysis of Human-Building Interactions in Office Environments: to What Extent Energy Saving Boundaries can be Displaced?

Building energy consumption can be enhanced to a certain level, but human-oriented dimensions can open the doors to new opportunities. This article presents an assessment on the interactions between human-driven factors and energy consumption patterns in office buildings based on empirical evidence and a co-simulation. The objective of this work is to evaluate a range for saving capacities as a result of readjusting occupants’ prior habits and engaging them in building operation by determining internal and external factors associated with human-building drivers, needs, actions, and systems. The study is based on a real office building, and according to real behavioral data and subjective measurement to analyze human attitudes and interactions with the building end-uses such as cooling, lighting, and equipment along with factors that directly impact energy use such as fenestrations and window blinds. The survey results are extensively investigated and the correlations between demographic/socioeconomic traits and behavioral factors are examined. The survey results are incorporated into a co-simulation testbed to represent occupant attitudes and behaviors and model human-building interactions accordingly. Different scenarios are designed to perform an assessment on the role of human factors such as attitude, awareness of consequences, habits, ascription of responsibility and personal/social norms on building energy end-uses. The outcomes of this study demonstrate that by adjusting behavioral factors in an office building, an energy saving between 9 and 18% can be achieved without sophisticated technology interface or building retrofit/upgrade. The results are elaborated, and recommendations are explained in detail.