Search

The integration of solar photovoltaic (PV) systems into the distribution network creates various stability and reliability issues associated with the intermittency of solar PV power generation. Energy storage is a vital component required for overcoming the intermittency of solar PV. This study presents a priority-based demand response management (DRM) for loads with large time constants to create virtual energy storage. The virtual energy storage thus created can be used for partial levelling of intermittent output from solar PVs. The proposed DRM algorithm involves controlling loads with large time constants such as air conditioning systems and refrigerators based on the forecasted solar PV generation. The proposed method is evaluated using data-driven simulations, weather data and mathematical models. The proposed algorithm is highly suitable for megacities that have high number of multi-storey residential buildings. Utilising the virtual storage capacity available from the appliances will reduce the investment as well as the operation cost of renewable energy such as solar PV. Analyses on impact on temperature, percentage of interruptions, cost savings and impact on energy storage sizing are also presented for evaluating the performance of the proposed algorithm. NRF (Natl Research Foundation, S’pore) Published version

Real-time conditional monitoring and instinctive fault diagnosis plays an important role in the smart grid systems. Conventional data storage and management systems are facing the difficulties with the disseminated, heterogeneous and large volume of data. Conventional data management infrastructure utilizes centralized approach which comprises of large-scale server, disk array data storage hardware and relational database management system (RDBMS) for database services. This type of infrastructure may result in poor performance to support the requirement of smart grid applications for very large database with high data request rates at very low latency. Precise, fast, vulnerable and cooperative information system is the basis requirement for the smart grid applications. In this paper, we proposed a distributed data storage and management platform which guarantees high scalability and reliability data storage and management with the huge amount of information in the smart grid application. In order to evaluate and validate our scalable cloud-based data storage platform for smart grid, we deployed our proposed infrastructure in our cyber-physical test-bed which is a facility to test innovative technologies for building efficiency and urban sustainability.

As a major component of Internet of Things (IoT), pervasive wireless sensor networks (WSNs) are becoming ubiquitous in all aspects of life such as environmental monitoring, military, intelligent transportation. Data gathering is a crucial yet energy-consuming task for WSNs. More and more sensing devices are deployed, which results in a huge amount of data to be transmitted. Usually, the sensor node is equipped with tiny, non-rechargeable battery and limited computation capability. Consequently, it’s imperative to improve energy efficiency and prolong network lifetime for WSNs powered by non-rechargeable batteries. In addition to non-rechargeable battery, energy harvesting (EH) has emerged to be a promising power supply to WSNs, which may unlock the possibility of perpetual operation for WSNs. In practical environment, data redundancy and data correlation commonly exist, which motivates the development of data aggregation techniques for both battery powered and energy harvesting WSNs (EH-WSNs). Data aggregation can be categorized as function-based and correlation-based. This thesis aims to develop both function-based and correlation-based data aggregation algorithms for conventional WSNs powered by non-rechargeable battery to achieve energy efficiency and prolong network lifetime. Further, this thesis applies data aggregation into EH-WSNs to achieve perpetual operation and satisfying application requirements. First of all, this thesis proposes a function-based data aggregation algorithm for a single-hop sensor network. In certain applications such as alarm systems, only the maximum value is required. Therefore, function-based data aggregation can be used to reduce energy consumption as well as transmission delay. A priority-aware hybrid scheduling scheme is proposed to achieve fast decentralized energy-efficient MAX function computation in a single-hop network. In our proposed scheme, priority-aware TDMA scheme is used to schedule the transmission. CDMA is used to recover information whenever a collision occurs. The proposed scheme helps to reduce the number of transmissions and delay for MAX computation with good scalibility. Secondly, this thesis focuses on correlation-based data aggregation algorithms based on compressive sensing (CS). Leveraging the load balancing features of CS, a novel hybrid CS based data gathering algorithm with a dense measurement matrix for tree-based multi-hop data gathering network is developed. CS technique can guarantee the full recovery of sensor data while imposing simple computation load on the sensor. An optimization problem is formulated and solved by approximation algorithms. The proposed scheme significantly reduces total energy consumption and data gathering latency and dramatically improves network lifetime compared with benchmark algorithms. The sensor network may experience changing environment where deterministic routing path is vulnerable to attacks and is less energy-efficient. In this case, it would be beneficial to allow a certain amount of exploration in the routing path. A cost-aware stochastic compressive data gathering algorithm is developed for multi-hop WSNs. The mathematical Markov chain based model is used to analyze the stochastic compressive data gathering process. A sparse random measurement matrix with optimized compression probability is constructed to minimize expected total cost while guaranteeing the recovery accuracy. Extensive simulations on both synthetic and real data show that the proposed algorithm requires less total expected cost to achieve a certain level of accuracy compared with benchmark algorithm. Finally, the proposed algorithm prolongs network lifetime for a given recovery accuracy. Finally, this thesis proposes a harvesting-aware compressed data gathering algorithm, in which the heterogeneous EH rates of the nodes are taken into consideration in order to maximize information throughput while maintaining sustainable operation. An optimization problem is formulated and solved by approximation algorithms. Our proposed algorithm can significantly improve information throughput while maintaining sustainable operation. Doctor of Philosophy (EEE)

Redox flow based batteries are becoming increasingly popular because of the numerous advantages they offer. These systems are modular, highly efficient, and the cost is relatively low, making it easy to scale up. The battery, however, is not commercially widespread as lithium-ion or lead-acid batteries. One of the major reasons for this is not having enough ageing studies for this comparatively recent technology and lack of battery specific unique optimization techniques. This thesis focuses on the two important issues in batteries, namely capacity decay and charging methodology. The former is essential in the longevity and latter in short term efficiency and the rate of ageing. These are some of the key aspects that are less researched, but important in making the commercial implementation of Vanadium Redox flow Battery (VRB) viable. The capacity loss due to vanadium ion diffusion was simulated for multiple cycles of operation and the effect that flow rate has, on concentration flux in the cell was investigated. The concentration difference between two half cells is one of the major reasons for species crossover. As a consequence of diffusion, the charge transfer current density in the electrode varies. The relationship between charge transfer current density and the concentration of one species of vanadium was analysed at different temperatures. Further, the parameter dependence was analysed during the charging process of VRB for a system-aware building, which has distributed sources of energy production and amenable to flexible load scheduling. Preliminary studies on multi-modal VRB charging characteristics was done using an equivalent circuit model. The normalized energy and time efficiencies were studied for their operation in buildings. The results obtained show that there is a trade-off between the energy capacity and the charging time, depending on the charging current and flow rate. The concentration change of different species of vanadium ion was modelled with temperature influence for each of these vanadium ion variants. The concentration change of vanadium electrolyte was simulated for different temperatures. The change in capacity of VRB was calculated for different temperature profiles and the total loss was partially contributed by intrinsic loss and partly due to diffusion. It was also observed that the difference in the loss at different temperatures was higher during the initial 50% of the total life cycle and stabilizes as it proceeds. At the end of 200 cycles, the capacity loss at 40C was higher (14.23%) than that of 30C (13.37%) and 20C (12.67%) of which, a major part was contributed due to diffusion in the first 100 cycles of operation. Further, an extended dynamic model of the vanadium ion transfer was developed, in which, the effect of temperature and transfer of bulk electrolyte across the membrane was included. The model was used to simulate capacity decay for a range of different ion exchange membranes that are being used in the VRB. The model was made more comprehensive by including the effect of bulk electrolyte transfer. A volume change of 19% was observed in each half-cell for Nafion 115 membrane based on the simulation parameters. The effect of this change in volume directly affects concentration, and the characteristics were analysed for each vanadium species as well as the overall concentration in the half-cells. Following the initial feasibility study on multi-modal charging, a more detailed analysis of charging methodology and a profile based charging optimization for VRB was proposed. Individual over-potentials and the losses due to the pump were separately modelled for accuracy. A fuzzy logic controller was used to utilize the normalized loss values and the state of health of VRB from earlier studies on ageing to provide an optimal value for current and flow rate for the selected profile. The optimization was done in real-time to include flexibility of control due to external factors like temperature, humidity, etc., which can be added as a parameter based on the operator requirement. Three modes were proposed in this study namely, autonomous, time bias and energy bias. These modes, however, are a non-exhaustive fragment of the many control modes that can be achieved. Doctor of Philosophy (EEE)

Technological advances in wireless electronics and algorithms have made it possible for us to have a plethora of assorted Wireless Sensor Network (WSN) nodes with diverse applications. Several recent developments have shown promise for the distinct nodes to be integrated and coordinate with each other despite having dissimilar hardware and protocol stacks. Some of the problems associated with the WSNs that have proven to be a major hurdle in their ubiquitous acceptance as the benchmark monitoring solutions are: energy consumption, scalability, reliability and security. All of these aspects demand thorough research; moreover, they also have a trade-off between each other, for instance, having a reliable system may demand an increase in energy consumption. Decreasing the energy consumption will in addition affect the range and scalability of the network. In this thesis the author has concentrated the research on integrating Radio-Frequency Identification (RFID) nodes with the traditional WSNs, placing more emphasis on reducing overall energy consumption of monitoring applications. Due to the diversity of the available nodes, their requirements and resources are drastically different. This feature introduces a challenge to reduce the energy consumption of an overall application composed of these hybrid platforms. In particular, some challenging in-pipe monitoring applications have been proposed for industries utilizing RFIDs and other sensor nodes, mainly because of the high range of low frequency RFID tags and their magnetic induction property made it suitable for metallic in-pipe monitoring. In order to reduce energy consumption and increase the network life-time of the proposed inspiring monitoring scenarios, a realistic estimate of the energy consumptions of various nodes deployed in one application with given MAC protocols is absolutely necessary. Experimenting with actual sensor networks to determine energy consumption is both time-consuming and costly, particularly when these diverse nodes are deployed in one application. Normally, simulators are used to estimate the performance of these nodes. Despite the abundance of WSN simulators offered today, none can simulate the energy efficiency given several types of nodes and situations in one application. Hence, a simulation platform is needed for a blended deployment of distinct types of WSNs and RFIDs to access application performance requirements while curbing energy consumption to improve the application's lifetime. In order to overcome this lack, a distinctive WSN simulator was developed, which is a dynamic, platform independent, rapid prototyping and interoperable multi-environment simulation platform for RFID and WSNs. Future works involve extending the simulator by improving it in terms of both its performance and functions. Finally, the techniques of Cognitive Radio (CR) and Cooperative Networks (CN) are investigated to enhance the range and performance of the monitoring applications. These techniques include Energy Detector, used in CRs spectrum sensing, and re-generative relays, used in CN. These techniques improve performance and reliability for long range in-pipe communications. MASTER OF ENGINEERING (EEE)

Cooperative localization in wireless ad-hoc networks deals with how to utilize local ranging measurements for locating remote nodes. In view of both the pervasive deployment of wireless ad-hoc networks and the incapability of wireless nodes for long-distance ranging, cooperative localization can be found in various applications. This thesis presents the research contributions to a number of theoretical and practical problems about cooperative localization, which can be grouped into two categories. The first contribution consists of using a simple Local Coordinate System for map construction, devising a Vector-addition Localization Scheme for anchor-free cooperative localization, and developing optimal path selection mechanisms for such vector-addition scheme to achieve better accuracy. Two mechanisms for optimal path selection are proposed in this thesis. The first mechanism is based on some geometric heuristics about local coordinate system, which is simple yet universal. Our second mechanism discloses some underlying properties of vector-addition localization scheme, and it makes use of Cramer-Rao Lower Bound (CRLB) as well as geometric principles for optimal path selection. Both mechanisms are able to produce, in term of localization accuracy, better solutions than conventional arbitrary shortest-path approach. The second part of our research work is focused on how to quickly disseminate the local ranging measurement information to the whole wireless ad-hoc network. Such fast message dissemination strategies are generic and thus applicable to many other general operations in wireless ad-hoc networks. Our main contribution in this area is to provide, for collision and collision-free cases, an approximate algorithm that maximizes the number of receptions in each dissemination wave. Our research work does not only produce faster dissemination strategies than previous methods as proved by numerical results, but also sheds light on the underlying theoretical aspects of the generic fast message dissemination problem. MASTER OF ENGINEERING (EEE)

The objective of this thesis is to develop new adaptive rate control schemes for the video encoder and channel encoder so that a better picture quality is achieved as compared to the existing rate control schemes. A novel joint source channel rate control scheme is proposed in this thesis, in which the bit error rate of the video bitstream at the video decoder is used as a control component to derive bit rate for the video encoder and channel encoder. The video bitstreams are protected unequally, so that the most important bitstreams can be delivered to the video decoder correctly. The joint source channel rate control scheme is adaptive to time-varying channel as the bit error rate control component is adaptive to the channel conditions. DOCTOR OF PHILOSOPHY (EEE)