Past Seminars

Page 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Modelling Carbon Emissions Reduction from the Australian Electricity Sector
Ilya Pavlyukevich
Humboldt Universität zu Berlin
Start time: 3:15 pm
Date: Saturday 29 March 2008
Location: Theatre 3, ICT Bldg, 111 Barry Street, Carlton

We consider a one-dimensional dynamical system driven by a vector field -U', where U is a multi-well potential satisfying some regularity conditions. We perturb this dynamical system by a stable symmetric non-Gaussian L'evy process whose scale parameter decreases as a power function of time. It turns out that the limiting behaviour of the perturbed dynamical system is different for slow and fast decrease rates of the noise intensity. As opposed to the well-studied Gaussian case, the limiting probability is not concentrated in the global minimum of U.

Finally, we discuss simulated annealing of jump processes with a non-constant stability index and consider applications to non-local random search and stochastic optimisation.

---

Modelling Carbon Emissions Reduction from the Australian Electricity Sector
Deb Chattopadhyay
CRA International
Start time: 3:15 pm
Date: Friday 29 February 2008
Location: Theatre 3, ICT Bldg, 111 Barry Street, Carlton

Carbon emissions reduction has been a central part of public policy debates in recent years. Yet, there has been relatively little effort to explore the cost implications for alternative greenhouse gas (GHG) policies such as a mandatory introduction of renewable energy vis-à-vis an emissions trading scheme. We will first review the background GHG policy issues and choices within the electricity sector.

This study uses a set of mathematical programming models to analyse the optimal choice of technology and quantitatively address the implications for alternative GHG policies. There are two core models. The first model characterises the interaction among the players in the Australian National Electricity Market (NEM) as a non-cooperative Cournot game to determine the profit maximising generation strategy for the generators. This model is solved iteratively with a stochastic LP that mimics the centralised production scheduling performed by the electricity market operator. The Cournot game is formulated as a mixed complementarity problem (MCP) and is solved to obtain the equilibrium generation strategy for each generator over a number of years. The stochastic LP solves for the long-term investment and production schedule for all electricity generators in the system. An overview of these models will be provided.

Finally, a case study that CRA International had undertaken for the National Generators Forum in 2006/07 using these models, will be presented. Australia generates over 80% of its electricity from coal. Therefore, Australia faces a daunting task to substantially cut down its sectoral carbon emissions in the long term to achieve a 50% reduction from the 2000 emissions by 2050. The key study outcomes include the potentially very significant role that baseload low emission technologies such as nuclear and carbon capture and storage, need to play to achieve such a deep cut. It also highlights the ineffectiveness and sub-optimal nature of some of the other policies that rely solely on renewables to achieve the target.

---

Incentive-based control of ad hoc networks: a performance study
A.E. Krzesinski
University of Stellenbosch, South Africa
Start time: 11:00 am
Date: Tuesday 18 December 2007
Location: Old Geology Theatre 1, Parkville

Ad hoc networks are self-configuring networks of mobile nodes,connected by wireless links. If a destination node is beyond the transmission range of an origin node, then the nodes must cooperate to provide a multi-hop route. Any node can act as a sender, receiver or transit node. It is clear that it is in a node's interest to be a sender or receiver, but it is less clear what the value is of forwarding traffic on behalf of other nodes. The nodes should therefore be given incentives to act as transit nodes, because otherwise the network would fail to function. A way to do so is by introducing (for each node) a credit balance, where nodes use credits to pay for the costs of sending their own traffic, and earn credits by forwarding traffic from other nodes.

In this talk we focus on the situation where each node can move to improve its utility (expressed in terms of either credit balance or throughput). Here radio interference plays a pivotal role, as it defines an interesting trade-off: nodes may prefer to be close together in order to reduce the energy needed to transmit data, but on the other hand proximity increases interference, and has therefore a negative effect on connectivity. Simulation experiments reveal that the positions of the nodes converge to (non-trivial) optimal operating points.

---

Modeling price dynamics of emission allowances
Juri Hinz
National University of Singapore
Start time: 11:00 am
Date: Tuesday 11 December 2007
Location: Old Geology Theatre 1, Parkville

The climate rescue is on the top of the agendas today. To protect the environment, emission trading schemes are considered as one of the most promising tools. In a system of such type, a central authority allocates credits among emission sources and sets a penalty which must be paid per unit of pollutant which is not covered by credits at the end the period. This regulatory framework introduces a market for emission allowances and creates a need for risk management by appropriate emission-related financial contracts. In this talk we apply methodologies from stochastic analysis to address logical principles underlying price formation of tradable pollution certificates. Based on tools from optimal control theory, we characterize the equilibrium allowance prices and show the existence of the proposed price dynamics. Further, we illustrate the computational tractability of the resulting models. In the context of the least squares Monte Carlo method, we utilize fixed point arguments to derive appropriate numerical schemes, which are illustrated by examples.

---

Stochastic Geometry and Wireless Network Modeling
François Baccelli
INRIA - Ecole Normale Superieure, France
Start time: 3:15 pm
Date: Friday 30 November 2007
Location: Russel Love Theatre, Richard Berry Bldg, Parkville

The geometry of the location of mobiles and/or base stations plays a key role in several classes of wireless communication networks where it determines the signal to interference ratio for each potential channel and hence the possibility of establishing simultaneously some set of communications at a given bit rate. Stochastic geometry provides a natural way of defining (and computing) macroscopic properties of such networks, by some averaging over all potential geometrical patterns for example of the mobiles. The talk will survey recent results obtained by this approach for analyzing key properties of wireless networks such as coverage or connectivity, and for evaluating the performance of a variety of protocols used in this context such as medium access control or routing.

---

Page 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15