Smart Buildings Research Articles

Distributed chaotic bat algorithm for sensor fault diagnosis in AHUs based on a decentralized structure

The effective operation of sensors in heating, ventilation and air conditioning (HVAC) systems to promote high-efficiency, energy-saving and low-risk intelligent buildings. However, most existing methods are either based on a centralized architecture or employ only pure digital simulation platforms for HVAC sensor fault diagnosis and verification, which are difficult to meet the complex and dynamic needs of HVCA systems. To overcome the difficulty in employment of centralized architecture involved multi sensor fault diagnosis and fault diagnosis methods limited by pure digital simulation platform, a fully distributed chaotic bat algorithm is proposed to diagnosis multiple-sensor faults and is verified by the hardware-in-the-loop simulation platform in HVAC systems. First, inspired by the bat swarm intelligence and multiple agents, we characterize the sensor fault diagnosis problem as an optimization problem of the bat predation behavior. Specifically, each sensor node is abstracted as a bat colony, and each bat colony communicates with directly connected neighboring bat colony in physical space. Second, to avoid the tendency of bat algorithm to fall into locally optimal solutions, the tent chaotic map is employed to enhance the bat colony search ability for improving the randomness of bat algorithm, involves escaping from local extreme points and increasing the diversity of the bat colony. Finally, we embed the temperature and humidity sensors into the hardware-in-the-loop simulation platform to construct the near-actual physical environment. The performance and convergence of the proposed method are verified using both a pure digital simulation platform and a built-in hardware-in-the-loop simulation platform. Compared with basic bat algorithm (BA), particle swarm optimization algorithm (PSO) and chaotic bat algorithm (CBA), the proposed fully distributed chaotic bat algorithm (DCBA) have higher accuracy and can achieve the 98.41 % accuracy in fixed bias (FB) sensor faults, 96.96 % accuracy in complete failure (CF), 95.84 % in drift bias (DB) and 96.08 % accuracy in accuracy drops (AD).

Enhanced multi-horizon occupancy prediction in smart buildings using cascaded Bi-LSTM models with integrated features

Accurate occupancy prediction in smart buildings is crucial for optimizing energy management, improving occupant comfort, and effectively controlling building systems, particularly for short- and long-term horizons. Recently, deep learning-based occupancy prediction methods have gained considerable attention. However, the full potential of these methods remains under explored in terms of model architecture variations and prediction horizons. This study introduces cascaded LSTM and cascaded Bi-LSTM models for multi-horizon predictions from 10 minutes to 24 hours, integrating a modified activation function, additional input features, and optimized hyper-parameters using OPTUNA. Traditional performance metrics and various other analyses were conducted to compare the models. Both models performed well for short- and long-term predictions, with minimal differences in the results. Nevertheless, analysis focusing on non-zero data errors (accounting for approximately 11% of occupied periods) and occupancy-wise errors showed a significant performance gap between the two models. The cascaded Bi-LSTM model demonstrated consistent performance across various prediction horizons and occupancy variations, with accuracy approximately 10-15% higher than the cascaded LSTM model, highlighting its superior capability in capturing complex dataset dynamics through a bidirectional process. This study highlights the importance of additional input features, data feature analysis, and multi-perspective result analysis to select the most suitable model for occupancy prediction, validated with pre- and post-modeling feature importance analysis.

Online model adaptation in Monte Carlo tree search planning

AbstractWe propose a model-based reinforcement learning method using Monte Carlo Tree Search planning. The approach assumes a black-box approximated model of the environment developed by an expert using any kind of modeling framework and it improves the model as new information from the environment is collected. This is crucial in real-world applications, since having a complete knowledge of complex environments is impractical. The expert’s model is first translated into a neural network and then it is updated periodically using data, i.e., state-action-next-state triplets, collected from the real environment. We propose three different methods to integrate data acquired from the environment with prior knowledge provided by the expert and we evaluate our approach on a domain concerning air quality and thermal comfort control in smart buildings. We compare the three proposed versions with standard Monte Carlo Tree Search planning using the expert’s model (without adaptation), Proximal Policy Optimization (a popular model-free DRL approach) and Stochastic Lower Bounds Optimization (a popular model-based DRL approach). Results show that our approach achieves the best results, outperforming all analyzed competitors.

Wireless optimization for sensor networks using IoT-based clustering and routing algorithms.

Wireless sensor networks (WSN) are among the most prominent current technologies. Its popularity has skyrocketed because of its capacity to operate in difficult situations. The WSN market encompasses various industries, including building automation, security networks, healthcare systems, logistics, and military operations. Therefore, increasing the energy efficiency of these networks is of utmost importance. Hierarchical topology, which typically uses a clustering methodology, is one of the most well-known methods for WSN energy optimization. To achieve energy efficiency in WSN, hierarchical topology low-energy adaptive clustering hierarchy (LEACH) was first introduced, and this served as the foundation. However, conventional LEACH has several limitations, which have led to extensive research into improving LEACH's efficacy in its current form. The use of particular algorithms and strategies to enhance the functionality of the conventional LEACH protocol forms the basis of ongoing efforts. Utilizing this enhanced LEACH, performance in terms of throughput and network life may be enhanced by concentrating on elements such as cluster head formation and transmission energy consumption. The enhanced LEACH algorithm demonstrates significant improvements in both throughput and network lifetime compared with conventional LEACH. Through rigorous experimentation, it was found that the enhanced algorithm increases the throughput by 25% on average, which is attributed to its dynamic clustering and optimized routing strategies. Furthermore, the network lifetime is extended by approximately 30%, primarily because of enhanced energy efficiency through adaptive clustering and transmission power control.

Direct Illuminance-Contribution-Based Lighting Control for IoT-Based Lighting Systems in Smart Buildings

With the advent of low-voltage light-emitting diodes (LEDs) and advances in Internet of Things (IoT) technologies, smart buildings have recently become more energy efficient. Nevertheless, the lighting-control system is one of the major sources of electrical energy consumption in commercial buildings. This study proposes a direct illuminance-contribution-based lighting-control framework to reduce the energy of LED luminaires and ensure illuminance for user requirements in smart buildings. Specifically, we designed a direct illuminance-contribution-based lighting-control algorithm (DIC-LCA) using luminaires that are ideally axisymmetric with all light emitted below the horizontal plane and developed a WiFi lighting controller for the IoT-based lighting-control systems in smart buildings. The DIC-LCA can adjust the dimming level by calculating the illuminance based on the line of sight (LOS) distance for energy saving and user satisfaction. After simulation analysis, we prove that energy savings can be achieved by controlling the dimming levels of LED luminaires with high light contribution.

Energy management based on safe multi-agent reinforcement learning for smart buildings in distribution networks

The rapid urbanization and increasing use of distributed renewable energy resources have imposed a significant burden on power networks. Smart buildings equipped with artificial intelligence technology can play a pivotal role in energy management, ultimately enhancing energy efficiency and voltage quality. However, ensuring voltage stability within large-scale smart building systems presents challenges due to the coexistence of diverse energy sources and the fluctuating nature of renewable energy. This paper proposes a safe multi-energy management framework achieved by online decentralized execution and centralized training for large scale smart buildings in distribution networks. The energy management problem is formulated as a safety-augmented Markov decision process, presenting intractability for dynamic programming due to its extensive continuous state space. To solve this issue and improve the convergence speed and training process stability, a safety-augmented constrained multi-agent reinforcement learning algorithm based on reward extrapolation is proposed. In this algorithm, hazard values are introduced to enhance non-safe multi-agent reinforcement learning algorithms and meet safety constraints. A novel reward network is designed by imitating expert underlying intentions to ensure the rationality of the reward function for multi-objective tasks. Additionally, the loss function for estimating the Q-network is redesigned during training process to guarantee effective convergence. Theoretical analysis is conducted to provide the convergence guarantee. Numerical case studies based on actual data are performed to validate the effectiveness and scalability of our approach, showing that smart buildings can achieve superior energy management performance while ensuring voltage safety for distribution networks. The source code of the proposed algorithm is available at https://github.com/SYiyun/CMARL-EX.

Energy and reserve procurement in integrated electricity and heating system: A high-dimensional covariance matrix approach based on stochastic differential equations

The integration of stochastic renewable energy sources into energy systems presents challenges due to flexibility insufficiency and uncertainty modeling inaccuracy. To address these challenges, this paper explores a high-dimensional covariance matrix approach based on stochastic differential equations (SDEs) for energy and reserve co-dispatch in integrated electricity and heating systems. By capturing the long-term correlation of wind power forecast errors, the SDE model simultaneously models point forecasts and high-dimensional positive-definite covariance matrices, resulting in a compact uncertainty set. To ensure computational efficiency, we employ duality theory to transform the stochastic co-dispatch problem with the high-dimensional uncertainty set into deterministic convex programming. Additionally, a virtual heat storage model is introduced for the district heating network, leveraging the thermal inertia of the network to enhance reserve capacity. Simulations validate the calibration and sharpness of the proposed high-dimensional uncertainty set. The results demonstrate that co-dispatch using the SDE-based uncertainty set reduces operational costs by 4.9% compared to a set that does not consider the long-term correlation of wind power, under a 95% coverage rate. Moreover, the storage capabilities of district heating pipelines and smart buildings provide cost-free reserves, saving the operational costs of the two test systems by 2.8% and 1.3%, respectively.

A real-time approach for smart building operations prediction using rule-based complex event processing and SPARQL query

A real-time approach for smart building operations prediction using rule-based complex event processing and SPARQL query

Strategic trading equilibria in local community considering co-existence of pool-based and peer-to-peer energy markets with free-rider problem in network usage cost

The peer-to-peer (P2P) energy trading market is considered a promising method for allocating prosumers’ locally produced power in smart buildings. However, the existing literature overlooks the coexistence of P2P and pool-based energy markets. Therefore, this paper establishes a bi-level energy market model that incorporates both P2P and pool-based markets. The prosumers’ individual profit maximization model considering their network usage costs in the P2P energy trading network at the upper level is first developed. A hybrid market-clearing framework determining the optimal trading amounts with the prosumers’ strategic offers at the lower level is then exploited. The distribution network power flow model in the market clearing framework is then convexified. The lower-level problem is replaced with its strong duality conditions. This bi-level problem can be reformulated as an equilibrium problem with equilibrium constraints (EPEC). This EPEC is then solved after being transformed into an equivalent mixed integer linear programming (MILP) model. Finally, a fifteen-bus system is employed to illustrate the characteristics of the hybrid market equilibrium depicted by this EPEC model. The influences of the prosumers’ strategic offers on the distribution locational marginal prices (DLMPs) and network usage costs are highlighted. Since the prosumers can choose to transmit their energy in the pool-based or P2P markets, the emission and pollution costs caused by power transmission can be minimized. A cleaner energy production is thus guaranteed. The free-rider problem is verified by comparing the network usage charges in the strategic and perfect competition scenarios.

Enhancing Energy Efficiency With Smart Building Energy Management System Using Machine Learning and IOT

Enhancing Energy Efficiency With Smart Building Energy Management System Using Machine Learning and IOT

On-device Online Learning and Semantic Management of TinyML Systems

Recent advances in Tiny Machine Learning (TinyML) empower low-footprint embedded devices for real-time on-device Machine Learning (ML). While many acknowledge the potential benefits of TinyML, its practical implementation presents unique challenges. This study aims to bridge the gap between prototyping single TinyML models and developing reliable TinyML systems in production: (1) Embedded devices operate in dynamically changing conditions. Existing TinyML solutions primarily focus on inference, with models trained offline on powerful machines and deployed as static objects. However, static models may underperform in the real world due to evolving input data distributions. We propose online learning to enable training on constrained devices, adapting local models toward the latest field conditions. (2) Nevertheless, current on-device learning methods struggle with heterogeneous deployment conditions and the scarcity of labeled data when applied across numerous devices. We introduce federated meta-learning incorporating online learning to enhance model generalization, facilitating rapid learning. This approach ensures optimal performance among distributed devices by knowledge sharing. (3) Moreover, TinyML’s pivotal advantage is widespread adoption. Embedded devices and TinyML models prioritize extreme efficiency, leading to diverse characteristics ranging from memory and sensors to model architectures. Given their diversity and non-standardized representations, managing these resources becomes challenging as TinyML systems scale up. We present semantic management for the joint management of models and devices at scale. We demonstrate our methods through a basic regression example and then assess them in three real-world TinyML applications: handwritten character image classification, keyword audio classification, and smart building presence detection. The results confirm the effectiveness of our approaches from various perspectives, such as accuracy improvement, resource savings, and engineering effort reduction.

Data-Driven and Domain Adaption Framework for Optimizing Energy Usage Forecasting in Smart Buildings Using ARIMA Model

Building energy demand and energy system supply are increasingly balanced by energy storage and short-term DSM. This is necessary due to fluctuating renewable energy supplies and rising building power use. Worldwide, structures utilize tons of energy. Greenhouse gas emissions and operational expenses decrease with construction energy efficiency. Forecasting and optimization algorithms can solve challenges, including supply chains (inventory optimization), traffic, and sustainable energy system battery/load/production scheduling for carbon-free energy generation. We often solve optimization problems that need forecasting due to uncertain future values. Predicting and optimizing are challenging; therefore, little research has been done. Our method uses building energy modeling professionals' data to forecast neighboring building types for new or unknown building types. After training, we utilize the models to estimate energy usage for the k-closest building types and combine the predictions using a weighted average. We used time-series decomposition to detect uncertainty and a hybrid model to close this gap: The concepts encode static features and predictable patterns in time-series simulation results. The model learns latent performance differences and calibrates output using outcomes and history records. Historical data predicts public building energy ARIMA use. Our method covers data processing, training, validation, and forecasting. We measured our method. Mixed integer linear optimization and ARIMA projected most accurately.

Optimal layout scheme of fire detectors and operation condition monitoring technology in urban integrated substation

Aiming at the problem that it is difficult to obtain the life cycle law and detection coverage of fire detectors and the low reliability of fire alarm process in urban integrated substation, an optimal layout scheme and operation condition monitoring technology of fire detectors based on local outlier factor and neural network are proposed. Divide fire zones by equipment functions, calculate the number of fire detectors required by using ground area and indoor height, and integrate building automation system to create substation fire alarm system; The signal noise of fire detector equipment is eliminated by wavelet transform, the fuzzy characteristic analysis model of fire detector operation state is established, the local outlier factor is simulated as abnormal data, the past operation data is trained by BP(Back propagation) neural network, the difference between the current state and the normal operation state is obtained, and the detector operation state monitoring is completed. The operation reliability of fire detector is analyzed from the aspects of temperature and smoke perception. The simulation results show that the proposed technology can successfully obtain the life cycle law and detection coverage of fire detector, and ensure the safe use of urban integrated substation. In addition, through a series of experiments, the feasibility and effectiveness of this technology in practical applications are verified.The maximum time consumption of detector working condition monitoring in this method is only 3.3 s. The fitting degree of the method in this paper is 5.7 ml/m3, and the fitting degree of the actual smoke curve is 5.7 ml/m3. The temperature of the method in this paper is lower than 33 °C. In the future development, the introduction of more advanced self-diagnosis technology can be real-time monitoring and analysis of the status of the detector itself, timely detection of faults or anomalies, and automatic alarm or notification of maintenance personnel for repair and maintenance.

The Main Barriers Limiting the Development of Smart Buildings

Smart buildings play a key role in the complex ecosystem of cities and are often subject to barriers that limit their development. Although identifying these barriers is fundamental to creating an enabling environment for this segment’s expansion, few works aim to identify these challenges. This work has two main objectives: (1) to research the main barriers limiting the development of new smart building projects and (2) to prioritize these barriers from the perspective of professionals with experience in the field. We adopted an exploratory approach common in research that focuses on identifying and prioritizing variables related to a phenomenon, which is based on two main actions: obtaining information through a careful literature review and consulting professionals who work in the concerned field. The results showed that professionals assessed the 23 barriers identified through bibliographic research as important, with the most important being related to lack of qualified professionals, shortage of government policies, higher initial and construction costs, macroeconomic barriers and access to financing, high cost of intelligent systems and technologies, regulatory barriers, lack of knowledge about the current and potential benefits of smart buildings, and more complex design and construction.

Occupancy Estimation from Blurred Video: A Multifaceted Approach with Privacy Consideration

Building occupancy information is significant for a variety of reasons, from allocation of resources in smart buildings to responding during emergency situations. As most people spend more than 90% of their time indoors, a comfortable indoor environment is crucial. To ensure comfort, traditional HVAC systems condition rooms assuming maximum occupancy, accounting for more than 50% of buildings’ energy budgets in the US. Occupancy level is a key factor in ensuring energy efficiency, as occupancy-controlled HVAC systems can reduce energy waste by conditioning rooms based on actual usage. Numerous studies have focused on developing occupancy estimation models leveraging existing sensors, with camera-based methods gaining popularity due to their high precision and widespread availability. However, the main concern with using cameras for occupancy estimation is the potential violation of occupants’ privacy. Unlike previous video-/image-based occupancy estimation methods, we addressed the issue of occupants’ privacy in this work by proposing and investigating both motion-based and motion-independent occupancy counting methods on intentionally blurred video frames. Our proposed approach included the development of a motion-based technique that inherently preserves privacy, as well as motion-independent techniques such as detection-based and density-estimation-based methods. To improve the accuracy of the motion-independent approaches, we utilized deblurring methods: an iterative statistical technique and a deep-learning-based method. Furthermore, we conducted an analysis of the privacy implications of our motion-independent occupancy counting system by comparing the original, blurred, and deblurred frames using different image quality assessment metrics. This analysis provided insights into the trade-off between occupancy estimation accuracy and the preservation of occupants’ visual privacy. The combination of iterative statistical deblurring and density estimation achieved a 16.29% counting error, outperforming our other proposed approaches while preserving occupants’ visual privacy to a certain extent. Our multifaceted approach aims to contribute to the field of occupancy estimation by proposing a solution that seeks to balance the trade-off between accuracy and privacy. While further research is needed to fully address this complex issue, our work provides insights and a step towards a more privacy-aware occupancy estimation system.

An ontology for automated fault detection & diagnostics of HVAC using BIM and machine learning concepts

This paper presents an ontology for AFDD (Automated Fault Detection and Diagnostics) of HVAC (Heating, Ventilation, and Air conditioning) systems in buildings called “AFDDOnto”. Presently, the AFDD models are mainly data-centric and often lack semantic information such as contextual information and spatial information; additionally, configuration information, analysis, and results used for model development are lost once developed. This impedes an effective mechanism for tracking changes and updating the model for future developments and use cases. The Proposed ontology can be used for AFDD model development, tracking changes, analytics, visualization, and digital twinning by enabling integration of BIM with BAS/BMS (Building Automation System/Building Management System) concepts and secondly to store AFDD configuration and analytics in the AFDDOnto. Select competency questions are constructed using SPARQL queries to access the proposed knowledge model. The proposed ontology has been tested against different measures using multiple metrics and a case study and further validated using a semi-structured survey of experts. Applied AI engineers, Facility managers, Asset managers, and building owners aiming to develop AFDD models for HVAC systems can benefit from adopting this ontology for HVAC maintenance, including analysis, model development, and knowledge management.

Development of sustainable conductive cementitious composite using graphite-coated spent catalyst waste

This study presents an innovative approach to developing sustainable conductive composites by coating graphite onto the surface of spent catalyst waste through nano-surface engineering techniques. The process ensures uniform adsorption of graphite onto the surface of the spent catalyst waste particles, followed by oven treatment and milling. This results in better integrity and effective bonding, leading to the production of graphite-coated spent catalyst waste (G-SCW). Scanning electron microscopy indicates the successful coating of spent catalyst waste with graphite. The research investigates the effect of G-SCW on the cementitious properties of paste and mortar. Incorporating G-SCW results in acceptable workability and setting time, while the compressive strength increases at early and later stages, with up to 20 % G-SCW content. The addition of G-SCW in the mortar significantly reduces the electrical resistivity, resulting in a 63 % reduction in resistivity compared to the reference mix, thereby enhancing the conductivity. Hydration studies confirm the presence of pozzolanic reaction in blended paste, as evidenced by a decrease in calcium hydroxide content. The sustainability assessment indicates a substantial reduction in embodied carbon and possibly producing mortar with lower cement content. These findings suggest great potential for developing sustainable conductive mortar with G-SCW, enabling smart building construction, and supporting sensor networks for structural health monitoring.

Detection and Determination of User Position Using Radio Tomography with Optimal Energy Consumption of Measuring Devices in Smart Buildings

The main objective of the research presented in the following work was the adaptation of reflection-radar technology in a detection and navigation system using radio-tomographic imaging techniques. As key aspects of this work, the energy optimization of high-frequency transmitters can be considered for use inside buildings while maintaining user safety. The resulting building monitoring and control system using a network of intelligent sensors supported by artificial intelligence algorithms, such as logistic regression or neural networks, should be considered an outcome. This paper discusses the methodology for extracting information from signal echoes and how they were transported and aggregated. The data extracted in this way were used to support user navigation through a building, optimize energy based on presence information, and increase the facility’s overall security level. A band from 5 GHz to 6 GHz was chosen as the carrier frequency of the signals, representing a compromise between energy expenditure, range, and the properties of wave behavior in contact with different types of matter. The system includes proprietary hardware solutions that allow parameters to be adjusted over the entire range and guarantee adaptation for RTI (radio tomography imaging) technology.

Comparative Analysis of the Impact of Advanced Information Technologies on the International Real Estate Market

This study examines the impact of advanced information technologies, such as big data, the Internet of Things (IoT), and artificial intelligence (AI), on the global real estate market. Through comparative analysis of real estate markets in North America, Europe, and Asia, and specific case studies from the US, UK, and China, we quantified the effectiveness of these technologies from 2018 to 2023. Our findings show that big data analytics improved market forecasting accuracy by 15.2%, IoT in smart buildings reduced energy consumption by 10.3% and enhanced security by 17.5%, and AI in property management increased operational efficiency by 17.6% and reduced maintenance costs by 13.2%. North America showed the highest technology acceptance, with a 20%-30% improvement in market efficiency, followed by Europe at 15%-25%, and Asia at 10%-20%. The successful application of these technologies depends on policy environment, market acceptance, and enterprise management. This research offers valuable insights and quantitative evidence for policymakers, industry practitioners, and researchers, highlighting the need for future studies on the integrated effects and long-term impacts of technology convergence.

A Review of Machine Learning

The ever-growing complexity of engineering systems and the vast amount of data generated necessitate advanced data analysis techniques like machine learning (ML). This review explores the motivations and advantages of integrating ML into real-world engineering applications. The paper highlights the limitations of traditional deterministic models in handling intricate interactions within modern engineering systems. ML offers significant benefits including improved efficiency, enhanced decision-making, advanced automation and control, adaptability, and cost savings. Real-world applications across various disciplines are explored, including image recognition and computer vision in self-driving cars, predictive maintenance for optimizing equipment lifespans, structural health monitoring for identifying potential damage, and signal processing for control systems in airplanes and traffic lights. The review concludes by summarizing key findings from case studies on autonomous vehicle navigation, smart grid optimization, wind turbine fault detection, and HVAC system optimization in smart buildings. These case studies showcase the power of machine learning techniques like deep learning, reinforcement learning, sensor fusion, and ensemble learning in achieving superior performance and efficiency in various engineering domains. Key Words: Machine Learning, Engineering Applications, Artificial Intelligence, Big Data, Predictive Maintenance, Robotics, Structural Design, Smart Grids, Fault Detection, Optimization, Control Systems.