Article
Computer Science, Information Systems
Mohammad Ostadijafar, Anamika Dubey
Summary: The article introduces a tube-based model predictive controller to minimize the net cost of energy usage by a building's HVAC system while ensuring comfort level. This controller is robust against model uncertainty and exogenous disturbances, suitable for environments with complex models.
IEEE SYSTEMS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Mingchao Xia, Fangjian Chen, Qifang Chen, Siwei Liu, Yuguang Song, Te Wang
Summary: This paper proposes an adaptive optimal scheduling strategy for residential heating, ventilation, and air conditioning (HVAC) based on deep reinforcement learning. By learning the state transition probability, the method can reduce electricity costs while ensuring comfort. The effectiveness of the proposed method is demonstrated through experimental results.
JOURNAL OF MODERN POWER SYSTEMS AND CLEAN ENERGY
(2023)
Article
Computer Science, Information Systems
Iresha Pasquel Mohottige, Hassan Habibi Gharakheili, Arun Vishwanath, Salil S. Kanhere, Vijay Sivaraman
Summary: Buildings are required to follow standard operational procedures during emergency evacuation, including shutting down air handling units to prevent smoke spread. This study quantifies power excursions during evacuations and develops methods to optimize evacuation plans and minimize energy costs.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Microbiology
Changliang Nie, Xueyun Geng, Runqi Zhang, Lina Wang, Ling Li, Jianmin Chen
Summary: This study investigated the presence of cyanobacteria on the pre-filter of HVAC systems and found that the concentrations were highest in autumn and significantly higher compared to winter and summer. The airborne transportation of aquatic and terrestrial cyanobacteria contributed to their presence on the pre-filter. The dominant cyanobacteria families identified were Chroococcidiopsaceae, norank_cyanobacteriales, Nostocaceae, and Paraspirulinaceae. Some detected genera, such as Nodularia sp., Pseudanabaena sp., and Leptolyngbya sp., potentially produced cyanobacterial toxins that could pose a threat to human health. This study provides new insights into the distribution of cyanobacteria outside of aquatic habitats.
Review
Chemistry, Physical
Jason Woods, Nelson James, Eric Kozubal, Eric Bonnema, Kristin Brief, Liz Voeller, Jessy Rivest
Summary: Air conditioning contributes to greenhouse gas emissions, and the emissions associated with reducing humidity are larger than those associated with reducing temperature. With increasing cooling demand worldwide, humidity-related emissions will significantly increase.
Article
Construction & Building Technology
Abdoalnasir Almabrok, Mihalis Psarakis, Anastasios Dounis
Summary: This paper presents an FPGA implementation of a FOPID controller for HVAC systems. The controller parameters are optimized using the BB-BC evolutionary algorithm, resulting in reduced oscillations and improved convergence. The method has potential importance in reducing energy consumption.
ENERGY AND BUILDINGS
(2023)
Article
Thermodynamics
Dasheng Lee, Shang-Tse Lee
Summary: The application of artificial intelligence (AI) in heating, ventilation and air conditioning (HVAC) equipment has been extensively studied and numerous AI-equipped HVAC products have been launched. To design AI functions for energy-efficient HVAC systems, this study proposed a method that combined literature review with design thinking. However, the lack of cases of failed applications limited the depth of re-thinking. The study analyzed raw data from 88 research papers and concluded that AI application must be accompanied by necessary hardware improvements to achieve effective energy savings.
APPLIED THERMAL ENGINEERING
(2023)
Article
Green & Sustainable Science & Technology
Alperen Yayla, Kubra Sultan Swierczewska, Mahmut Kaya, Bahadir Karaca, Yusuf Arayici, Yunus Emre Ayozen, Onur Behzat Tokdemir
Summary: Buildings account for nearly half of the world's energy consumption, with 40% of that being consumed by HVAC systems. Traditional HVAC controllers are not efficient in responding to changes in occupancy and environmental conditions. This study develops an AI-based HVAC control mechanism that focuses on occupant comfort and improves energy efficiency. The results show that applying AI for HVAC operation can achieve a minimum of 10% energy savings while providing better thermal comfort to occupants.
Review
Construction & Building Technology
Mirza Rayana Sanzana, Tomas Maul, Jing Ying Wong, Mostafa Osama Mostafa Abdulrazic, Chun-Chieh Yip
Summary: Despite the promising results of deep learning research, its application in the construction industry, particularly in Facility Management and Maintenance (FMM), is still limited. This review explores the use of deep learning techniques in FMM, specifically in the field of Heating, Ventilation, and Air Conditioning (HVAC) for predictive maintenance. The paper highlights the importance of establishing relevant public datasets for the effectiveness of deep learning techniques and discusses the current challenges in this area.
AUTOMATION IN CONSTRUCTION
(2022)
Article
Energy & Fuels
Mahmood Khatibi, Samira Rahnama, Pierre Vogler-Finck, Jan Dimon Bendtsen, Alireza Afshari
Summary: This paper proposes a hierarchical model-based scheme to aggregate the flexibility of small consumers in power markets. The scheme considers a power market mechanism where consumers remain committed to their bids without sacrificing comfort levels. A high-level control layer determines the energy budget for the building according to price signals, while a lower-level dispatch layer distributes the budget among different zones. Results show that the proposed scheme maintains commitment to the aggregator and desired comfort levels, with the centralized model performing better at the cost of comfort. Preliminary results on available regulating power for residential buildings are also reported for the first time.
Article
Chemistry, Multidisciplinary
Stephen Grigg, Zeyad Yousif Abdoon Al-Shibaany, Matthew Robert Pearson, Rhys Pullin, Paul Calderbank
Summary: Enhancing the passenger experience by reducing noise and improving sound quality in vehicles' interior spaces poses a challenge, especially with the rise of electric vehicles. Through the use of psychoacoustics and acoustic cameras, noise sources within a vehicle's interior space can be accurately identified and located, allowing for early modifications to be made to avoid unnecessary noise issues during the development of HVAC units.
APPLIED SCIENCES-BASEL
(2021)
Article
Construction & Building Technology
Kang Chen, Siliang Chen, Xu Zhu, Xinqiao Jin, Zhimin Du
Summary: This paper proposes an interpretable mechanism mining enhanced deep learning method for fault detection and diagnosis (FDD) model transfer among different HVAC systems. By conducting fault simulation experiments and training a one-dimensional convolutional neural network (1D-CNN), a general FDD model is obtained and verified on another type of chiller. The testing results indicate that the retrained transfer model has a good diagnostic effect for the target chiller.
BUILDING AND ENVIRONMENT
(2023)
Article
Engineering, Electrical & Electronic
Yaa T. Acquaah, Balakrishna Gokaraju, Raymond C. Tesioro, Gregory Monty, Kaushik Roy
Summary: Human-in-the-loop HVAC control methods have gained attention for addressing occupants' discomfort in buildings. Previous studies have focused on thermal sensation and preference using the ASHRAE Thermal Comfort Database II and machine learning. However, none have investigated merging environmental parameters and thermal images to predict thermal comfort. This study used a fusion of environmental sensors and thermal images to predict comfort indices, achieving high accuracy and 45% energy savings.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Information Systems
Yi Peng, Haojun Shen, Xiaochang Tang, Sizhe Zhang, Jinxiao Zhao, Yuru Liu, Yuming Nie
Summary: Finding the optimal energy-saving control strategy for HVAC systems has become crucial in realizing energy savings, emission reductions, and green buildings. Deep reinforcement learning (DRL) provides new ideas for HVAC energy consumption optimization. This study proposes a DRL-based framework for HVAC energy consumption optimization, which includes a CNN-LSTM model for energy consumption prediction and an enhanced DDPG algorithm for real-time energy consumption control.
Review
Construction & Building Technology
Saman Taheri, Paniz Hosseini, Ali Razban
Summary: Intelligent buildings use predictive technologies to optimize HVAC systems, reducing energy consumption while maintaining comfort. Model predictive control is an effective management method that can enhance energy efficiency by considering multiple objectives.
JOURNAL OF BUILDING ENGINEERING
(2022)
Article
Transportation Science & Technology
Kelvin Lee, Yu Jiang, Avishai (Avi) Ceder, Justin Dauwels, Rong Su, Otto Anker Nielsen
Summary: This study proposes a mixed integer linear programming model for the public transport schedule synchronization problem, considering both path transfer time and time-dependent travel time data. Novel valid inequalities are derived to improve computational performance. Numerical studies show that the use of time-dependent travel time data reduces path transfer times.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2022)
Article
Automation & Control Systems
Gang Chen, Yu Lu, Rong Su
Summary: This paper proposes a shapelet temporal logic to describe the temporal properties of shapelets, and an incremental algorithm is used to find the optimal logic expression for fault diagnosis. A case study on rolling element bearing fault diagnosis demonstrates the high accuracy of the proposed method.
Article
Engineering, Civil
Gammana Guruge Nadeesha Sandamali, Rong Su, Kalupahana Liyanage Kushan Sudheera, Yicheng Zhang
Summary: A framework for en-route air traffic flow management considering uncertainties in demand and capacity was proposed to ensure safety and optimize flight trajectories. The model utilizes various control actions to minimize delays, with computational complexity reduced by implementing two phases of ATFM. Experimental results demonstrate the effectiveness of the framework in realistic large-scale problems.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Yi Zhang, Rong Su, Yicheng Zhang, Bohui Wang
Summary: The paper proposes a multi-bus dispatching strategy combined with the boarding and holding control to optimize bus utilization, decrease passenger excess delay, and maintain high system reliability. The strategy involves adaptive bus capacity, adjusted dispatching time, holding and boarding limit strategies, and minimization of headway variation. Numerical examples show significant time reduction and efficiency improvement compared to traditional fixed schedules.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Automation & Control Systems
Kun Zhang, Rong Su, Huaguang Zhang
Summary: This article investigates the RL-based resilient control algorithm for a class of Markovian jump systems with completely unknown transition probability information. The control problem of the nonlinear Markovian systems is converted into solving a set of local dynamic games, where the control policy and attacking signal are considered as two rival players. The designed integral RL (IRL) algorithm combines potential learning and forecasting abilities, requiring less transmission and computation during the learning process.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Automation & Control Systems
Yan Wang, Rong Su, Mingsong Lv, Nan Guan
Summary: This paper addresses the optimal linear quadratic regulation (LQR) problem for discrete-time interconnected systems (ISs) over a weakly connected graph. A main challenge lies in designing an optimal controller for ISs with weakly connected topology due to the possible lack of information between subsystems. The paper proposes a multiple-step estimation approach to tackle this problem and explicitly derives the optimal value of the cost function. The effectiveness of the proposed method is demonstrated through simulations of a connected vehicle system.
Article
Automation & Control Systems
Yuting Zhu, Liyong Lin, Ruochen Tai, Rong Su
Summary: This paper provides an overview of networked supervisory control frameworks for discrete event systems in imperfect communication networks. It discusses state-of-the-art works on networked supervisory control, addressing channel imperfections such as delays and data losses. By presenting key concepts and results, the paper analyzes the advantages and disadvantages of different approaches. It also summarizes existing works, which follow two lines of thinking and result in different verification or synthesis approaches. The paper concludes by presenting future research topics.
DISCRETE EVENT DYNAMIC SYSTEMS-THEORY AND APPLICATIONS
(2023)
Article
Engineering, Civil
Yun Lu, Lingying Huang, Jiarong Yao, Rong Su
Summary: This paper proposes an intention prediction-based control method for vehicle platoons to handle the frequent cut-ins by human-driven vehicles. The method considers the tradeoff between platoon integrity and traffic safety, and includes prediction algorithms and a predictive control system. Driver-in-the-loop experiments demonstrate the real-time prediction of human drivers' cut-in intention and the prevention of cut-ins for vehicle platoons while ensuring road safety.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Shubham Gupta, Rong Su
Summary: In this study, an extended version of the differential evolution (DE) algorithm named multiple individual guided differential evolution (MGDE) is proposed. The MGDE algorithm introduces a novel mutation strategy based on multiple guiding individuals to manage diversity and convergence. Experimental results show that the MGDE algorithm performs well and is highly competitive with other metaheuristic algorithms.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Chemistry, Analytical
Liping Huang, Zhenghuan Li, Ruikang Luo, Rong Su
Summary: Despite the availability of sensing data in intelligent transportation systems, missing data is still a problem in traffic estimation. Existing studies have mainly focused on randomly missing data and neglected the distinction between missing data links. This paper proposes a general linear model based on probabilistic principal component analysis for imputing non-randomly missing traffic speed data.
Article
Automation & Control Systems
Ruochen Tai, Liyong Lin, Yuting Zhu, Rong Su
Summary: In this work, the problem of privacy-preserving supervisory control against an external passive intruder is investigated by co-synthesis of a dynamic mask, an edit function, and a supervisor. The goals include ensuring opacity of system secrets, covertness of dynamic mask and edit function, and satisfying safety and nonblockingness requirements. The approach models the co-synthesis problem as a distributed supervisor synthesis problem in the Ramadge-Wonham supervisory control framework, and proposes an incremental synthesis heuristic to synthesize a dynamic mask, an edit function, and a supervisor. The effectiveness of the approach is illustrated on an example about location privacy.
Article
Automation & Control Systems
Ruochen Tai, Liyong Lin, Rong Su
Summary: This work introduces a method for synthesizing optimal covert sensor-actuator attackers in the context of discrete-event systems (DES). It transforms the optimal covert sensor-actuator attacker synthesis problem into the optimal supervisor synthesis problem, building upon existing works in both areas. The study considers various optimization objectives, such as minimizing attack energy cost and minimizing time cost for damage infliction.
Article
Engineering, Civil
Xiaobei Yan, Maode Ma, Rong Su
Summary: An efficient authentication protocol for vehicle platoons in all handover scenarios is proposed in this paper, which utilizes certificateless aggregated signatures for mutual authentication, reducing signaling overhead and avoiding key escrow problems.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Proceedings Paper
Telecommunications
Xiaobei Yan, Maude Ma, Rong Su
Summary: With the emergence of 5G technology, vehicular communication has made significant progress. However, the current 5G mechanism faces challenges in terms of signaling overhead and security issues. In this paper, an efficient authentication protocol is proposed and evaluated to demonstrate its resilience against malicious attacks and its security functionality.
IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC 2022)
(2022)
Proceedings Paper
Automation & Control Systems
Gang Chen, Yu Lu, Rong Su
Summary: This paper proposes an algorithm to find the time optimal fault-tolerant controllable string for multi-process systems. It enforces specifications to the nonfaulty system to ensure efficient fault recovery and specification satisfaction through resource sharing. The algorithm checks the fault-tolerant property and sequentially finds the optimal fault-tolerant string based on system abstraction using a backtracking algorithm. The effectiveness of the proposed fault tolerant string synthesis algorithm is demonstrated through testing on a simulated manufacturing system.
