Back to results list
Show full item record
Please use this identifier to cite or link to this item:
|Title:||Optimal control of air-conditioning systems for enhanced indoor environment and energy efficiency using IoT-based smart sensors||Authors:||Li, Wenzhuo||Degree:||Ph.D.||Issue Date:||2021||Abstract:||Indoor environment affects both the mental and physical health of occupants. Air-conditioning systems are provided to maintain the expected indoor environment with the least possible energy use. Optimal control of air-conditioning systems is an effective means in solving the trade-off problem of ensuring acceptable indoor environment and minimization of energy use. The development of the Internet of Things (IoT) and IoT devices provides new possibilities regarding the optimal control of air-conditioning systems. As a type of IoT device, IoT-based smart sensors have advanced functions of communication, data analysis, computing and decision-making beyond their basic function of data collection. However, the use of the self-organization of a single IoT-based smart sensor and the coordination of multiple IoT-based smart sensors for optimal control of air-conditioning systems have not been fully considered. This PhD study therefore aims to develop optimal control strategies for air-conditioning systems to enhance indoor environment and energy efficiency by utilizing the abilities of self-organization and coordination of IoT-based smart sensors. A rule-based optimal control approach for a single-zone air-conditioning system with a single IoT-based smart sensor is proposed. It is conducted using a monitoring-diagnostic-intervention process that fully utilizes the data collection and decision-making functions of a single IoT-based smart sensor. Based on this approach and field data, rule-based optimal control strategies for enhanced indoor environment and energy efficiency are developed and validated. The results show that issues relating to indoor environment including excessive CO2 and dew condensation on the diffuser are identified and alleviated. Energy efficiency issues including resetting the temperature set-point and occupant-centric on/off control of air-conditioning systems in individual rooms are identified and solved.
A multi-agent based distributed optimal control strategy for multi-zone dedicated outdoor air systems (DOAS) with multiple IoT-based smart sensors is proposed. Two control objectives, indoor air quality (IAQ) and energy use, are considered in the multi-objective optimization problem for optimal control of DOAS. The control strategy decomposes the complex optimization problem into a number of simple optimization problems. Distributed agents, corresponding to individual rooms and the primary air handling unit (PAU), are assigned to handle these decomposed problems. A central coordinating agent coordinates these agents to find the optimal solutions. Two test cases under different outdoor weather conditions are conducted to validate the control strategy. Results of the distributed optimal control strategy can provide almost the same outputs as the expected optimum given by the traditional centralized optimal control strategy. The results of implementing the control strategy show its effectiveness in solving complex optimization problems and optimizing multi-zone DOAS, as well as demonstrating good scalability and reconfigurability. A multi-agent based distributed optimal control strategy for multi-zone VAV air-conditioning systems is proposed. The proposed strategy consists of three novel schemes. First, a temperature set-point reset scheme adopts a linear rule to correlate the resetting of the temperature set-points in individual zones to simplify the optimization problem while applying proper optimization to individual zones. Second, a multi-objective optimization scheme optimizes the fresh air ratio of the supply air and the temperature set-point in the critical zone by formulating the multi-objective optimization problem. Third, a multi-agent distributed optimization scheme is developed to solve the optimization problem in a distributed manner, facilitating the deployment on local control devices of limited capacity. Test results show that the strategy is effective in properly balancing thermal comfort, IAQ and energy use, while largely reducing programming challenges. For the test case with six rooms, the average absolute PMV was 0.05, the average CO2 concentration level was 818 ppm, the energy use of the VAV air-conditioning system was 315 kWh and the computational load of an optimization step was about 1 second only.
|Subjects:||Air conditioning -- Control
Internet of things
Hong Kong Polytechnic University -- Dissertations
|Pages:||xxi, 157 pages : color illustrations|
|Appears in Collections:||Thesis|
View full-text via https://theses.lib.polyu.edu.hk/handle/200/11088
Citations as of May 22, 2022
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.