Events Calendar
-
[ November 25, 2019]
Efficient Mechanisms for Socio-technical Systems
| Modern society is based on large-scale engineered systems, often at the service of human end-users, e.g., transportation and power networks. While the control of such systems is typically grounded on purely engineering principles, their performance greatly depends on how human users interact with them. A common is- sue arising in these settings is the performance degradation often incurred when the users’ interests are not aligned to the “greater good” (e.g., traffic routing). In this context, a natural question arises: how can we design behavior-influencing mechanisms to incentivize efficient use of the existing infrastructure? In this presentation, I answer this question in relation to the well-studied class of congestion games, of- ten used to model traffic assignment problems. More precisely, I show how to design mechanisms that utilize only local information, and robustly maximize the system efficiency. Surprisingly, optimal mechanisms designed using only local information perform closely to those designed using full information (1% difference for affine latency functions). Additionally, I show how the proposed approach recovers and generalizes a number of well-known results in the literature. Finally, I discuss how the marginal cost mechanism, known to be optimal in the continuous-flow approximation, results in a lower efficiency than that encountered if no mechanism was used. -
[ November 22, 2019]
AI Planning, Learning and Optimization Supported by Model Checking
| Model checking and decision problems, including discrete variables, are often modeled by transition systems and predicate logic. Combining these two frameworks results in a unified model based on modular Petri nets with shared variables. An incremental abstraction method is also introduced for complex temporal logic planning. Uncontrollable events, as well as mu-calculus, are then shown to be interesting alternatives to more common game formulations. Optimization of systems with discrete variables often integrates methods from AI and operation research. An interesting alternative is the model checker Z3, which now also includes optimization. For hybrid systems, a specific method is also presented for energy optimization of robot systems, resulting in up to 30% energy and 50% peak power reduction. Finally, a recent integration of temporal logic and reinforcement learning is discussed. To summarize, methods from model checking are shown to be very useful for AI planning, learning and optimization. -
[ November 18, 2019]
Data Analytics, Forecasting, and Optimization of Smart Power and Energy Systems
| To tackle the challenges of global warming, countries all around the world set aggressive goals to reduce carbon emission in different industries. As a major carbon emitter, the power and energy industry plays a vital role in decarbonization, where increasing renewable energy integration and improve energy efficiency are two effective approaches. However, high penetration of renewable energy integration challenges the reliability, economy, and flexibility of the power and energy systems. Fortunately, these challenges have come hand-in-hand with the advancements of Internet of Things (IoT), communication technology, and data science, which helps build the smart power and energy systems.
This talk will discuss three approaches to explore the flexibility and boost the efficiency of the power and energy systems. In the first part, this talk will introduce the concept of electricity consumer behavior model and then discuss how to make the best use of the fine-grained smart meter data available to process and translate them into actual information and incorporate into consumer behavior modeling and distribution system operations. In the second part, a systematic research on probabilistic load forecasting will be introduced by investigating how to model the uncertainties of the electrical load. In the third part, the talk will discuss how to model and optimize the gas, heat, and power integrated energy systems as whole so that the flexibility beyond power systems can be exploited. Future works for smart power and energy systems will be prospected. -
[ November 11, 2019]
Lipid metabolism in the green microalga Chlamydomonas reinhardtii
| Microalgae are fast-growing microorganisms that have developed efficient mechanisms to harvest and transform solar energy into energy-rich molecules such as lipids. They are thus potentially great renewable cell factories for production of fuels and biomaterials for the chemical industries. However, several biological as well as engineering challenges need to be addressed before the establishment of a profitable algal industrial biotechnology. The main goal of Li-Beisson’s laboratory is to investigate the molecular mechanisms involved in the conversion of solar energy and atmospheric CO2 to energy-dense compounds such as oil and hydrocarbon. The molecular limitations of these mechanisms are assessed and novel strategies are proposed for synthetic biology of microalgae. The further use of this knowledge will aid in the design and creation of strains for algal biotechnology. In this communication, she will present current projects and highlight major findings from her group. -
[ November 11, 2019]
Distributed Filtering for Cyber-Physical Systems: Consistency, Stability and Security
| In this talk, the distributed filtering problem over cyber-physical systems (CPS) is investigated. First, a scalable and fully distributed Kalman filter is provided and analyzed to achieve the consistent estimation for the states of potentially unstable systems. Then, we study the case with state equality constraints (SEC). We propose a distributed Kalman filter with guaranteed consistency and satisfied SEC Then, we establish the relationship between consensus step, SEC and estimation error covariance in dynamic and steady processes, respectively. Under an extended collective observability condition based on SEC, the stability of the filter is studied. Moreover, we consider a secure distributed filtering problem for linear time-invariant systems with bounded noises and unstable dynamics under compromised observations. We first propose a recursive distributed filter consisting of saturation-like observation update and consensus operation of state estimates. A sufficient condition is then established for the boundedness of estimation error. -
[ November 11, 2019]
Bio-Inspired Autonomy for Mobile Sensor Networks in Aquatic Environment
| There is a perceivable trend for robots to serve as networked mobile sensing platforms that are able to collect data in aquatic environment in unprecedented ways. The need for undisturbed operation of search and monitoring posts higher goals for sustainable autonomy. This talk reviews recent developments in autonomous collective foraging in aquatic environment that explicitly integrates insights from biology with models and provable strategies from control theory and robotics. One of the interesting insights is that sensing is used to enable implicit communications among agents, which is able to achieve coordinated behaviors while incorporating individual differences. The algorithms are rigorously analyzed and modified for mobile robots. Experimental effort with promising results demonstrates that bio-inspired autonomy might be preferred in aquatic environment that features severe limitation in communication. -
[ October 28, 2019]
Atomristors: Memory Effect and Applications in 2D Materials
| This talk will present the latest research progress on the atomic-level details of non-volatile resistance switching (NVRS) in 2D memory devices In particular, we will focus on memory characteristics and atomistic imaging and transport studies, together with the first principle calculations on the common point defects of 2D materials. For this investigation, we employ the STM to characterize the vertical metal-insulator-metal (MIM) memory device and examine the defect sites in atomic resolution to elucidate memory phenomenon. These studies provide one to one correlation between the structural and electronic properties of the heterogeneities and their role in the resistance switching mechanism. Applications from RF switches to neuromorphic computing will be highlighted. -
[ October 28, 2019]
Metal halide perovskites for optoelectronics: the interplay of physics and chemistry
| Metal halide perovskites have attracted enormous attention in the scientific community in recent years. This attention has been drawn by breakthroughs in perovskite optoelectronics, mainly in photovoltaics and LEDs Although there has been great progress in their use in optoelectronics, fundamentally understanding this class of materials is still challenging, in particular the interplay of several physical and chemical processes. Many fundamental questions are left unanswered, such as what makes them unique/excellent semiconductors; how to tune their optoelectronic properties; why some compositions are more stable than others; the consequence of their chemical instability on their optoelectronic properties; and how to stabilize them. We, the Computational Materials Physics Group, focus on addressing such questions using computer simulations and theoretical analyses based on the laws of Quantum Mechanics. In this talk, I will highlight our recent progress and give an outlook of our next challenges with a discussion of future strategy along this line of research. -
[ October 25, 2019]
High-Fidelity Simulation of Premixed Flames
| Lean premixed combustion is the desired combustion regime in industrial gas turbines. It results in a low level of NOX due to operating at lower temperatures and facilitates better combustion efficiency. However, the primary issue with operating gas turbines in this regime is thermoacoustic instability, commonly initiated by combustion-generated sound. To avoid this type of instability, the mechanism of sound generation by premixed flames needs to be fully understood. This presentation will provide a technical review of our understanding of the mechanism of sound generation by premixed flames. Examples from direct numerical simulation (DNS) studies undertaken at the University of Melbourne are provided. Furthermore, the implications of the findings for developing accurate models that can predict combustion-generated sound are discussed. -
[ October 10, 2019]
Nanostructured Electrocatalysts for Energy-relevant Conversion Processes
| Replacement of precious metal catalysts by commercially available alternatives is of great importance among both fundamental and practical catalysis research. Nanostructured carbon-based and transition metal materials have demonstrated promising catalytic properties in a wide range of energy generation/storage applications. Specifically engineering carbon with guest metals/metal-free atoms can improve its catalytic activity for electrochemical oxygen evolution reaction (OER) and hydrogen evolution reaction (HER), thus can be considered as potential substitutes for the expensive Pt/C or IrO2 catalysts in metal-air batteries and water splitting process. In this presentation, I will talk about the synthesis of nonprecious metal and metal free elements-doped graphene, and their application on electrocatalysis [1-6]. The excellent OER and HER performance (high catalytic activity and efficiency) and reliable stability indicate that new materials are promising highly efficient electrocatalysts for clean energy conversion. I will also present some research results of electrocatalytic CO2 reduction eraction (CRR) and N2 reduction reaction (NRR) conducted in my research group [7-10].