Air-balancing is a key technique in HVACs to achieve accurate air supply for better IAQ and energy-saving performance. The existing air-balancing methods are usually based on a hybrid duct system ...model, including a data-driven duct system model between terminal flow and static pressure, and an ASHRAE damper model between static pressure and terminal damper angle. Most of these methods only focus on how to improve the accuracy of the former part whereas neglecting the latter. However, the ASHRAE damper model cannot exactly reflect the actual dampers characteristics in practice and may lead to unavoidable modelling error. To fix this issue, this paper proposes a full data-driven duct system (FD3S) model to directly reveal the relationship between the terminal flow and terminal damper angle in the whole duct system. Therefore, the desired terminal damper angle under the design flow conditions can be more accurately predicted. Besides, since the proposed method directly models the relationship between terminal flow and terminal damper angle, it avoids installing extra pressure sensors and reduces the operation cost. Moreover, the complexity of the proposed FD3S model is greatly reduced compared with the previous hybrid model. Finally, the experimental results demonstrate the effectiveness of the proposed method.
•Airborne transmission characteristics of SARS-CoV-2 in enclosed spaces are provided.•Operation guidelines of HVAC systems issued during the pandemic are compared.•Energy impacts of COVID-19 on HVAC ...systems are analyzed quantitatively.•Future impacts on HVAC system innovations and research trends are discussed.
Heating, ventilation and air-conditioning (HVAC) system is favourable for regulating indoor temperature, relative humidity, airflow pattern and air quality. However, HVAC systems may turn out to be the culprit of microbial contamination in enclosed spaces and deteriorate the environment due to inappropriate design and operation. In the context of COVID-19, significant transformations and new requirements are occurring in HVAC systems. Recently, several updated operational guidelines for HVAC systems have been issued by various institutions to control the airborne transmission and mitigate infection risks in enclosed environments. Challenges and innovations emerge in response to operational variations of HVAC systems. To efficiently prevent the spread of the pandemic and reduce infection risks, it is essential to have an overall understanding of impacts caused by COVID-19 on HVAC systems. Therefore, the objectives of this article are to: (a) provide a comprehensive review of the airborne transmission characteristics of SARS-CoV-2 in enclosed spaces and a theoretical basis for HVAC operation guideline revision; (b) investigate HVAC-related guidelines to clarify the operational variations of HVAC systems during the pandemic; (c) analyse how operational variations of HVAC systems affect energy consumption; and (d) identify the innovations and research trends concerning future HVAC systems. Furthermore, this paper compares the energy consumption of HVAC system operation during the normal times versus pandemic period, based on a case study in China, providing a reference for other countries around the world. Results of this paper offer comprehensive insights into how to keep indoor environments safe while maintaining energy-efficient operation of HVAC systems.
•The physics of frost formation in HVAC/refrigeration systems is discussed, focusing on the thermodynamic state of water and frost development on cold surfaces.•A comprehensive review of frost ...prevention and control strategies is presented, focusing heavily on active frost prevention/frost removal methods.•Comparison of various frost prevention/control techniques is discussed, considering key parameters such as effectiveness, power consumption requirements, additional accessories requirement, and key advantages/limitations.•Relevant research gaps and opportunities for future research and development are discussed.
Frost accretion is a common problem in HVAC and refrigeration systems. Frost accretion strongly impacts the operating efficiency of HVAC equipment and leads to a considerable increase in energy consumption. Therefore, energy-effective frost control techniques can significantly enhance the coefficient of performance of HVAC systems while also enhancing the lifecycle durability of the equipment. More recently, passive and active frost control and defrosting techniques have been proposed to reduce/control frost growth on heat transfer surfaces and ultimately enhance the thermal/hydraulic performance of HVAC systems. This review paper is focused on active prevention and frost removal techniques in HVAC, heat pumps, and refrigeration systems. This work was conducted by categorizing the defrosting methods into frost prevention strategies and frost removal strategies. Frost prevention strategies include treating and conditioning the upcoming inlet airflow and vapor injection techniques. On the other hand, frost removal strategies include reverse cycle defrosting (RDC), oscillation and ultrasonic vibration, hot gas bypass, and the use of applied electric and magnetic fields, among other techniques. Despite extensive theoretical and experimental work on active frost removal and control techniques, no baseline for comparison has been established, making it challenging to compare different techniques. Finally, future prospects and conclusions are discussed in detail, and future research and development directions are proposed.
Convincing building owners to retrofit advanced controls is a challenge due to the perceived complexity and unclear lack of benefits for a given building system. This paper describes an approach ...developed to demonstrate the benefits of Model Predictive Control (MPC) in an operating commercial building with access to historic data. A surrogate model of a building cooling delivery system and distributed energy resources, solar thermal fields, and thermal storage, has been developed to benchmark the building’s Business As Usual (BAU) operations. A combination of white, gray, and black box modeling techniques have been used to capture and validate the operation of the building system with operational data. To analyze costs and operational benefits of the MPC retrofit, we employ machine learning techniques to identify representative day profiles in the dataset that capture the typical behavior of the system across a range of weather conditions. We then compare the performance of the MPC algorithm with a baseline case in these representative day profiles. The results demonstrate that the predictive controller surpasses the BAU baseline performance by strategically shifting the operation cost to off-peak hours through optimal utilization of thermal storage and the chiller plant. These results satisfied the building manager to favorably consider implementing MPC in the building.
•An approach for evaluating benefits of advanced controls in a building with solar and storage.•Use of over two years of operational data to capture operation of building under different seasons.•Results demonstrate the value of using thermal storage for minimizing energy cost.•Historic data can be a powerful tool to convince building owners implement advanced controls.
•An encoder–decoder LSTM-based modeling strategy is developed for use within an EMPC framework. The modeling strategy is written and interpreted using standard control engineering conventions.•The ...end-to-end EMPC framework including model construction, model training, and closed-loop estimation and control are presented.•The EMPC framework is applied to an example building HVAC system that is simulated using EnergyPlus to demonstrate the approach.
Numerous studies have demonstrated the benefit of economic model predictive control (EMPC) applied to building heating, ventilation, and air conditioning (HVAC) systems. However, the construction and training of predictive models for building HVAC systems are widely recognized as a key technological barrier preventing large-scale adoption of EMPC for buildings. In this work, an encoder–decoder long short-term memory-based EMPC framework is developed. The key advantage of the approach is that a model may be automatically generated from a list of inputs and outputs. From the definition of inputs and outputs, the constructed model may be trained and automatically embedded into the EMPC framework for real-time estimation and control. The overall end-to-end EMPC framework from model training to on-line estimation and control are described. To this end, the encoder–decoder model provides a natural framework for state estimation (encoder), which is required to provide an initial condition for the predictive model of EMPC (decoder). Closed-loop simulations using EnergyPlus are performed to demonstrate the approach. The simulated closed-loop system consists of a building zone from a multi-zone building, which is served by an air handling unit-variable air volume HVAC system. For the HVAC example considered, the trained encoder–decoder model can predict the indoor air temperature and HVAC sensible cooling rate of a building zone over a two-day horizon with high accuracy. Considering a time-of-use electric rate structure, the EMPC, which manipulates the zone temperature setpoint, can reduce the HVAC power consumption cost relative to keeping the zone temperature setpoint at its maximum value (i.e., minimum energy approach).
Residential Heating, Ventilation, and Air conditioning (HVAC) systems are responsible for a significant amount of energy consumption, but their management is challenging due to the complexities of ...building thermodynamics and human activities. Reinforcement learning (RL) has been adopted to tackle this issue, but traditional RL methods require massive training data, long learning periods, and frequent equipment adjustments. To address these issues, we construct a new event-driven Markov decision process (ED-MDP) framework, which enables adjustments of control policies triggered by events, reducing unnecessary operations. Moreover, we propose an event-driven deep Q network (ED-DQN) method, which optimizes the action selection based on the triggered events. In the HVAC control problem, the proposed ED-DQN can effectively capture dynamic non-linear features of thermal comfort, and reduce the equipment damage caused by frequent adjustments. Our experimental results show that compared to three benchmark methods and three RL methods, our ED-DQN achieved state-of-the-art performance in both energy saving and thermal comfort violations. Moreover, our method demonstrates promising performance when applied to new test thermal environments, indicating its robustness and adaptability for optimizing residential HVAC controls.
•An event-driven Markov decision processes framework for optimal HVAC control.•Two types of events capture factors impacting performance.•Event-driven Deep Q network improves learning speed and minimizes decisions.•Comprehensive experiments show the superiority of the proposed method.
•Developed machine learning models for HVAC electricity consumption prediction.•Compared the performance of feed-forward back-propagation artificial neural network (ANN) with random forest (RF).•The ...ANN model performed marginally better than the RF model.•RF model can be used as a variable selection tool.
Energy prediction models are used in buildings as a performance evaluation engine in advanced control and optimisation, and in making informed decisions by facility managers and utilities for enhanced energy efficiency. Simplified and data-driven models are often the preferred option where pertinent information for detailed simulation are not available and where fast responses are required. We compared the performance of the widely-used feed-forward back-propagation artificial neural network (ANN) with random forest (RF), an ensemble-based method gaining popularity in prediction – for predicting the hourly HVAC energy consumption of a hotel in Madrid, Spain. Incorporating social parameters such as the numbers of guests marginally increased prediction accuracy in both cases. Overall, ANN performed marginally better than RF with root-mean-square error (RMSE) of 4.97 and 6.10 respectively. However, the ease of tuning and modelling with categorical variables offers ensemble-based algorithms an advantage for dealing with multi-dimensional complex data, typical in buildings. RF performs internal cross-validation (i.e. using out-of-bag samples) and only has a few tuning parameters. Both models have comparable predictive power and nearly equally applicable in building energy applications.
Today, a high volume of operation interval data can be efficiently captured by a diverse range of measurements including sensors, control signals and meters, which are deployed in building automation ...systems (BAS). Hence, advanced data analytics tools such as fault detection and diagnostic (FDD) can be developed to analyze the operational performance of heating, ventilation and air conditioning (HVAC) systems. In the past, enormous efforts have been made to develop various FDD approaches, assuming interval data contains essential information for identifying fault signatures. However, a “data rich, but information poor” phenomenon exists due to the fact that not all measurements are sensitive to faults in HVAC systems. This highlights a significant research gap, the lack of systematic analysis of measurement sensitivity to different HVAC system faults, which is vital for FDD development, measurement deployment optimization, and control system design. To address this gap, this study introduces a novel approach to assess the sensitivity of BAS measurements in relation to various HVAC fault types. We propose two sensitivity indices (SI), the SI of fault (SI_fault) and the global measurement SI (SI_measurement_global) to quantify measurement sensitivities. The SI_fault quantifies the measurement's sensitivity to a particular fault, while the SI_measurement_global assesses its sensitivity across all fault types. These indices integrate probability distributions, enhancing the interpretability and scalability. Utilizing the HVACSIM+ fault simulation dataset, which includes 15 common faults at varying severity levels and 89 different measurements within an HVAC system, we conducted an extensive analysis of measurement sensitivities by looking at the proposed SIs.
Display omitted
•A novel method was developed to evaluate measurement sensitivities to faults in an HVAC system.•Two new sensitivity indices (SI) were proposed to quantify the measurement sensitivity under faults.•Data representing 15 common faults with different severity levels was used to analyze measurement sensitivities.•A use case is illustrated to indicate that the developed SI can be used to identify fault signatures.
Transactive control is a type of distributed control strategy that uses market mechanisms to engage self-interested responsive loads to achieve power balance in the electrical power grid. In this ...paper, we propose a transactive control approach of commercial building heating, ventilation, and air-conditioning (HVAC) systems for demand response. We first describe the system models, and identify their model parameters using data collected from systems engineering building (SEB) located on our Pacific Northwest National Laboratory campus. We next present a transactive control market structure for commercial building HVAC systems, and describe its agent bidding and market clearing strategies. Several case studies are performed in a simulation environment using building controls virtual test bed (BCVTB) and calibrated SEB EnergyPlus model. We show that the proposed transactive control approach is very effective at peak shaving, load shifting, and strategic conservation for commercial building HVAC systems.
This paper presents a novel framework for Offline Reinforcement Learning (RL) with online fine tuning for Heating Ventilation and Air-conditioning (HVAC) systems. The framework presents a method to ...do pre-training in a black box model environment, where the black box models are built on data acquired under a traditional control policy. The paper focuses on the application of Underfloor Heating (UFH) with an air-to-water-based heat pump. However, the framework should also generalize to other HVAC control applications. Because Black box methods are used is there little to no commissioning time when applying this framework to other buildings/simulations beyond the one presented in this study. This paper explores and deploys Artificial Neural Network (ANN) based methods to design efficient controllers. Two ANN methods are tested and presented in this paper; a Multilayer Perceptron (MLP) method and a Long Short Term Memory (LSTM) based method. It is found that the LSTM-based method reduces the prediction error by 45% when compared with a MLP model. Additionally, different network architectures are tested. It is found that by creating a new model for each time step, performance can be improved additionally 19%. By using these models in the framework presented in this paper, it is shown that a Multi-Agent RL algorithm can be deployed without ever performing worse than an industrial controller. Furthermore, it is shown that if building data from a Building Management System (BMS) is available, an RL agent can be deployed which performs close to optimally from the first day of deployment. An optimal control policy reduces the cost of heating by 19.4 % when compared to a traditional control policy in the simulation presented in this paper.
•Offline MARL can eliminate poor behavior during training while converging.•LSTM layers are an effective method for obtaining training models for RL.•Simulation model of an underfloor heating system supplied by a heat pump.•Simulations show cost is reduced by 19.4% when compared to traditional controllers.