Power Engineering Seminars

Upcoming seminars

TBA


Past seminars

 

Stabilization of converter dominated DC microgrid using active damping method

Mr Mingfei Wu

Thursday, 8 November 2012, 1:00pm

Abstract: Loads connected to the bus in DC microgrid through tightly regulated converters are considered as constant power loads. The constant power load act as a negative incremental impedance in transient response, causing unstable oscillation in DC bus voltage. In my research, firstly, the relationship between tight level of regulation and unstable oscillation was studied. This relationship can provide the information for tradeoff between the fast response and oscillation level. An active damping method using soft starting is proposed. This method can damp the oscillation well by increasing the time response.


Two-switch ZCS Totem-pole Bridgeless PFC Boost Rectifier

Mr Khairul Muhammad

Thursday, 11 October 2012, 1:00pm

Abstract: A new two-switch zero current switching (ZCS) bridgeless boost rectifier with high power factor (PF) is proposed. The proposed bridgeless converter is based on a totem-pole configuration which allows the current to flow from high side to low side and vice versa during resonance. Hence, no auxiliary switch is needed to provide soft switching for all semiconductor devices. An average current mode controlled is employed to provide high PF and smooth input current waveform. A brief comparison between common soft-switching bridgeless boost converter topology and the proposed converter topology will be presented.


Smarter Voltage Control – multi-objective, co-ordinated, adaptive

Prof David Hill

Thursday, 16 August 2012, 1:00pm

Abstract: The topic of smart grids has received a lot of public attention, but from a scientific point of view it remains a developing concept beyond optimistic statements about integrating new generation, storage, load and ICT equipment. It remains to study the full capability and limits of such new structures for future power networks given they must provide the usual power balancing, voltage and frequency control, stability and recovery. This presentation rephrases the goals and possible implementation of the smart grid agenda in terms of control engineering for the whole (cyber-physical) system. It then applies this framework to the long-standing problem of voltage control now given new life by the uncertainty in renewable generation from rooftops to solar and wind farms.
Ultimately, at the core of the smart grid agenda, the goal must be to achieve a system which is more adaptive and resilient to changing power supply/demand, failures and attacks instead of relying on much more expensive redundancy in the physical system. The voltage control problem is essentially the coordination of massive numbers of control devices at different voltage levels and now gives challenges in uncertainty, scale and control granularity. Key ideas which can address this are hierarchical distributed control and use of learning. Optimisation techniques play a key role in their implementation.


A DG interface with active power filter capability using repetitive control

Mrs Anandalakshmi Thevampalayam Kaliappan

Thursday, 19 July 2012, 1:00pm

Abstract: The complexity of the power system keeps growing due to growing demand for electricity and a lot of efforts towards the integration of alternative sources of electric energy such as solar plant and wind farm. In order to overcome the challenges due to complexity the power grid needs to be intelligent in all aspects. The household appliances approximately account for 60 – 90 percent of total residential energy consumption, which is responsible for over seven tones of greenhouse gas emissions. The CO2 emissions of electricity during on-peak periods are significantly higher than during off-peak periods, emphasizing the need for managing peak loads. These carbon emissions can be significantly reduced through proper energy management techniques.
The proposed policy based framework will allow intelligent and flexible energy management of home appliances in a smart home which is complex and dynamic in ways that save energy and money automatically. A modern home with a wide range of automated appliances which results in more challenges for energy management is considered for this research work. The simulation results showed that this proposed system can efficiently manage energy in a smart home.


A DG interface with active power filter capability using repetitive control

Dr Xin Tang

Thursday, 21 June 2012, 1:00pm

Abstract: This seminar will present an expandable distribution generation (DG) interface with active power filter capability for three-phase three-wire system. The proposed system not only can control the active power flow, but also mitigate load unbalance, harmonics, and manage reactive power. The DG interface can simultaneously perform the two functions of power control station and active power filter (APF) or either one. To make the system more flexible, the grid current, instead of output current of the converter, is shaped to be a sinusoidal current in phase with the grid voltage. The reference currents are generated in multiplying the voltage loop controller output by the corresponding grid voltage waveforms. Consequently, harmonic analysis or dq transformation as required in the conventional design is not necessary. In this manner, controlling the DC voltage level is equivalent to performing an active power balance for the whole system. As a result, no additional hardware and interface are required. Experimental results verified the effectiveness of the proposed method.


Integration of wind power into power systems

Mr David Browne

Thursday, 24 May 2012, 1:00pm

Abstract: The recent growth of the connection of wind generation to the electricity network has given rise to new concerns in regards to system frequency response and regulation. In particular the penetration of wind generation in Australia is expected grow over the next decade, with penetration levels in South Australia and Tasmania expected to be among the highest in the world by 2020. This predicted large-scale connection of wind generation is likely to impact the regulatory, technical and market aspects of the National Electricity Market (NEM). The purpose of this seminar is to address the impact large-scale wind generation has on the frequency response of the power system. The IEEE-30 bus system has been used as a test case to analyse the frequency performance of a conventional islanded power system with increasing levels of wind energy. All of the modelling and simulation was conducted using the power system software DigSilent. The seminar will also introduce the area of research currently being pursued within the department in developing control methodologies to enable the participation of wind farms in the regulation of system frequency.


Drive systems – Silicon carbide CMOS, wind turbine protection and multi-phase machines

A/Prof John Fletcher, UNSW

Thursday, 24 November 2011, 1:00pm

Abstract: This seminar will review recent drive systems research including high-temperature silicon-carbide electronics, wind turbine drive design for fault ride-through and multi-phase drive systems. The presentation will review a recent collaborative research project on the development of a SiC CMOS process which developed PMOS and NMOS devices operating at 350oC with a view to developing high-temperature gate drive ICs for power electronics. This work has provided the industrial partner access to a $500M market in high temperature electronics. John will also present a modified protection scheme for wind turbines that utilises a dynamically-switched series resistance in the rotor circuit during grid faults. This limits both the over-voltage and over-current in the rotor-side converter and avoids having to disable the converter during these fault conditions thereby enhancing fault-ride through and improving the turbine’s contribution to fault current. The final part of the seminar will review some the research outputs from John’s research into multi-phase machine technology. An improved direct torque control for 5-phase induction machines will be presented which utilises the increased number of space-vectors available whilst eliminating auxiliary vector space voltages. A technique that avoids low-speed stator flux demagnetisation will be described which significantly improves low-speed stator flux control whilst maintain high-speed torque response.


Participation of wind turbines in system support

Dr Gregor Verbic

Thursday, 27 October 2011, 1:00pm

Abstract: The seminar will present a novel strategy for participation of variable speed wind turbines in primary frequency control. The proposed strategy is based on the deployment of the kinetic energy of the rotating masses to reduce the need for de-loaded operation while still being able to provide the required power reserve for the defined time frame. Steady-state de-loaded operating point is optimized with respect to the amount of kinetic energy stored in the rotating masses. The parameters of the algorithm are calculated off-line and stored in look-up tables. The proposed strategy is tested using a modified Nordic 32-bus test system for various operating scenarios. The results show that part of the primary frequency control power reserve of the thermal units can be replaced by the reserve provided by wind-power, while retaining the desired frequency response following a disturbance. A comparison with the conventional de-loaded strategy shows that between 2.21% and 3.84% energy savings can be achieved on a yearly basis, depending on the wind regime of the sites analyzed.


A High Step-Down Transformerless Single-Stage Single-Switch AC/DC Converter

Mr Tom Ki

Thursday, 29 September 2011, 1:00pm

Abstract: This seminar presents a high step-down tranformerless single-stage single-switch AC/DC converter suitable for universal line applications (90--270 Vrms). The topology integrates a buck type power factor correction (PFC) cell with a buck-boost DC/DC cell so that part of the input power is coupled to the output directly after the first power processing. With this direct power transfer feature and also sharing of capacitor voltages, the converter is able to achieve efficient power conversion, high power factor, low voltage stress on intermediate bus (less than 130V) and low output voltage without a high step-down transformer. The absence of transformer reduces the component counts and cost of the converter. Unlike most of the boost type PFC cell, the main switch of the proposed converter only handles the peak inductor current of DC/DC cell rather than the superposition of both inductor currents. The experiment results will be given to verify the proposed circuit.


Research on High-Voltage Transmission lines at Tsinghua University

Prof Mark MacAlpine

Friday, 2 September, 1:00pm

Abstract:  Prof MacAlpine has co-authored some 15 papers in this area in the past two years, 8 published and the rest in various stages of review and ‘polishing’ and, this talk will be based on 10 of them in the areas of corona from transmission lines, galloping and ice-shedding effects on transmission line conductors and RTV silicone coatings on porcelain insulators. China is building thousands of kilometres of transmission lines, many of them at 1000 kV AC and at high altitudes – and for these conditions there are many unanswered questions.
Corona on overhead lines causes radio interference and a significant amount of losses, which increase with ageing, altitude and humidity, but on-site measurements are not normally possible and laboratory measurements require large facilities.
Resonant-frequency galloping of conductors can lead to phase-to-phase faults. These can be initiated by aeolian effects or by an accumulation of ice breaking away from conductors. It is important to be able to design with these effects in mind, but again, full-scale experiments are difficult to arrange and expensive, but are necessary to confirm the results from simulations. One method of combating such oscillations is by installing insulating interphase spacers at the mid-point of the span. Again, to design the appropriate size, strength and flexibility for each case, an understanding of how they operate must be gained, which means careful simulations combined with some full-scale experiments to give confidence in applying the simulations to practical cases.
Applying a coating of RTV silicone rubber to porcelain insulators should combine the cheapness and rigidity of the one with the excellent pollution performance of the other, but how long do they last, how good is the resistance to pollution-flashover behaviour, and what about making it partially conducting to improve icing performance?


OPF with FACTS devices by AI method

Dr Andy Leung

Thursday, 4 August, 1:00pm

Abstract: This seminar presents a particle swarm optimization (PSO) method approach to solve the optimal power flow (OPF) in power systems incorporated with flexible ac transmission system (FACTS). The above approach integrates the power flow control needs into the OPF problem to fully utilize the characteristics of FACTS devices. A versatile FACTS device, UPFC (Unified Power Flow Controller), is considered. In the solution process, PSO coupled with full AC power flow, selects the best regulation to minimize the total cost and keep the power flows within their security limits. The optimization process with PSO is presented with case study examples using IEEE 30-bus system to demonstrate its applicability. The results have been compared with genetic algorithm (GA) to show the feasibility and the potential of this approach.


SPICE Steady State Modelling of Thermoelectric Generators involving the Thomson Effect

Mr Ian Laird

Thursday, 7 July, 1:00pm

Abstract: During operation thermoelectric generators (TEGs) are subject to the following thermal effects; Heat conduction according to Fourier’s law, Joule heating, Peltier heating and Thomson heating. Many SPICE-based models exist for TEGs however in the vast majority of them the Thomson effect is neglected due to its relatively small size compared to the other effects, as well as the complexity that results from including the Thomson effect in the model. This seminar presents a model that governs the steady state performance of a TEG that includes the Thomson effect whilst limiting the complexity of the SPICE model.


Fault Detection and Isolation in Power Systems

Dr Iman Shames, KTH, Stockholm, Sweden

Thursday, 12th May at 11:00am

Abstract: The talk consists of two parts; in the first part we are concerned about detecting and isolating those nodes that are not following the "protocol" governing the system. We particularly consider the problem of distributed fault detection and isolation in power systems. In the second part, we look into an edge-based operation to slow down the spread of a malfunction, that is called a "disease", in a network of interconnected nodes. More specifically, in the first part of this talk the existence of unknown input observers for networks of interconnected second-order linear time invariant systems is studied. We focus on two instances of such class of systems, which relate to practical applications. Particularly, the systems under different distributed control laws are considered and it is proved that for these systems one can construct a bank of unknown input observers, and use their output to detect and isolate possible faults in the network in a distributed fashion. Infeasibility results with respect to available measurements and relevant faults are provided, as well methods to remove the misbehaving agents from the network. Later the application of this family of fault detectors to fault detection in power networks and formations of mobile nodes is presented and further comments on the complexity of the proposed scheme are presented.

In the second part we look into the problem of disease propagation in interconnected systems. We are interested in modifying the network in a way that the disease propagation rate is slowed down. In the context of power systems a cascading failure can be considered as a disease. We address this problem by minimising the maximum eigenvalue of the adjacency matrix of the network graph via edge-based operations.