Richardt H. Wilkinson

Natural Balance of Multicell Converters: The Two-Cell Case

The multicell converter topology is said to possess a natural voltage balancing property. This paper is the first of a two-part series in which multicell converters are modelled for the general case of p-cells. This paper focuses on the development of the natural balancing theory for the two-cell case. An understanding of the two-cell case is fundamental to understanding the general balancing theory. The switching functions used in switching these converters are mathematically analyzed. Equivalent circuits are derived and presented. The switching and balancing properties of these converters are mathematically analyzed. The main conclusion of the analysis is that the natural balancing of these converters are influenced by three factors namely, the harmonic content of the reference waveform, the switching frequency and the load impedance. Mathematical tools are presented that can help designers to predict if balancing problems would occur for a particular set of operating conditions. As a result of the detailed understanding of the balancing mechanism that is gained through this theory it is shown that by adding a balance booster, the load impedance can be manipulated to improve the natural balancing of the converter. Simulation results are included to verify the presented balance theory and properties

Wireless sensor network gateway

Using wireless sensor networks is an effective way of gathering data and also opens up new sensing opportunities. Wireless sensor networks can be more useful if the data can be effectively retrieved from the sensor network. This paper describes the development of a gateway node for a wireless sensor network, implemented in TinyOS, using the AT91RM9200 ARM evaluation kit from Atmel. This node enables the user to remotely retrieve data from-, manage- and configure the network.

Energy harvesting for a condition monitoring mote

One application for wireless sensor networks (WSN) is the condition monitoring of high voltage power transformers to actively monitor, predict and identify causes of failures. This paper details a prototype design and testing of a hybrid power management system for a wireless sensor mote. The sensor mote, mounted on the outside of a high voltage transformer, utilizes a piezoelectric cantilever to generate electricity from the transformer tank vibration. Together with solar energy harvesting, the system allows for a battery-less self-sustained wireless sensor mote capable of autonomously monitoring its surroundings.

Voltage unbalance in the multicell converter topology

The multicell converter topology is said to possess a natural voltage balancing property. This paper models a 2-cell multicell converter. The switching and balancing properties of this converter is mathematically analyzed and presented. The balancing property of this converter is also proven mathematically from the presented properties and conditions for natural balancing are discussed.

An evaluation of compression techniques for Wireless Sensor Networks

Wireless Sensor Networks (WSN) typically consist of compact, low energy devices capable of sustaining a variety of ad-hoc topologies and operate for many years from a single, non-renewable, finite energy source such as a battery. This limitation in energy has led to much research in the field of energy conservation both through hardware and software techniques. The transmission of data incurs the highest energy cost in a WSN device and this study sets out to investigate different types of software compression techniques suitable for resource limited hardware, with the aim to minimize the energy required to transmit the data. An obvious trade-off exists between the computational energy used for compression versus the saving in energy associated with the transmission of compressed data instead of raw data. It is shown that lossless techniques could not sustain a fixed compression ratio when presented data with significant fluctuations and the study then considered lossy techniques for alternative compression methods. In addition, this study introduces a hybrid technique designed specifically for WSN devices and is shown to support other lossy techniques by further data reduction. The results from this study show that data reduction is possible when selectively implementing compression techniques best suited to application specific data, whilst being sensitive to the limited resources available.

One-Dimensional Spectral Analysis of Complex PWM Waveforms Using Superposition

This paper takes a new look at the mechanisms underlying the double-edge pulse-width modulation (PWM) process. It presents a novel way of deriving equations for the spectrum of double-edge PWM using basic mathematical techniques. In the process the underlying nonlinearities that generate the PWM sidebands are identified. Unlike the classical double Fourier series approach, the proposed method of deriving the PWM spectrum does not require the construction of the so-called unit cell. The interaction between this new model of the pulse-width modulator and the regular sampling process is studied, and generalized equations for the Fourier transforms of regularly sampled PWM waveforms are derived. A general solution to the important question of what happens to the PWM spectrum when the PWM reference consists of a summation of signals is presented. It is shown that the addition of reference signals in the time domain results in a double convolution of the PWM sidebands in the frequency domain. As an application of this result, it is shown how new analytic equations for the harmonics of third-harmonic injection PWM and space vector modulation can easily be derived. Finally, the new theoretical results are benchmarked against results from the well-established double Fourier series method.

Recursive Fourier Transform hardware

The Discrete Fourier Transform has played a fundamental role in signal analysis for radar systems, usually in its Fast Fourier Transform form. Traditionally, samples from a region of interest have been acquired, and processed in a block-processing fashion. The result is a delayed output equal to the time taken to acquire the data and then perform the Fourier Transform. To overcome this limitation, this work focuses on the development of a parallel framework for the execution of the Discrete Fourier Transform using a recursive algorithm, using many simple processing elements. The Fourier coefficients are updated every time a new sample is presented to the system, with a small latency. The system described in this work shows a 64 point Discrete Fourier Transform computation in parallel, which achieves a throughput of 817MSPS(29.42Gbit/s). The framework proposed promotes adjustable parallelism using the many simple processing elements. We show the engine is capable of dynamically correcting accumulated errors due to finite arithmetic, and allows simple adoption on existing systems.

A many processing element framework for the Discrete Fourier Transform

For the greater part of the last 30 years of computing history, computational processing has been primarily based on performance improvements of sequential processors. Research has shown that the future no longer lies in the direction of sequential processor performance improvements, but rather on the adoption of parallel processors and the development of suitable architectures. Computing in general has settled on generic suitable-for-all processing architectures, however recent research has since highlighted the importance of adopting many simpler processing elements, with a tailored focus on domain specific designs. In light of this, this work focuses on the development of a parallel framework for the execution of the Discrete Fourier Transform using a recursive algorithm using many simple processing elements. The system described in this work shows a 64 point Discrete Fourier Transform computation in parallel, which achieves a throughput of 2.19GSPS(39.49Gbit/s). The framework proposed promotes adjustable parallelism using the many simple processing elements, and allows simple adoption on existing systems.

DC-side harmonic compensation in DC traction applications

A problem associated with DC traction substations is the nonrobustness of the passive filters used for the filtering of the harmonics induced by the substations 12-pulse rectifiers. These harmonics cause problems with the communications and the signalling on the DC railway lines. Interference with the railway signalling system can have serious and sometimes fatal effects. Therefore, a method is being devised by which the problematic harmonics can be compensated for by means of an active power filter (APF) on the DC-side of the substations 12-pulse rectifiers.

Maximum energy harvesting from medium voltage electric-field energy using power line insulators

One essential requirement for a modern Smart Grid is continuous monitoring of the grid variables. This task is efficiently achieved using self-powered electronic devices deployed throughout the grid network. Within the vicinity of the grid power lines, the high voltages of these lines can provide a source of energy for such loads using capacitive coupling. This is known as Electric-Field Energy Harvesting (EFEH). However, the reported strategies that use this principle have not explored how to maximise their power transfer, which is essential for practical systems since the available energy from EFEH is invariably quite small. In this paper, a novel approach for optimum energy harvesting is proposed using the parasitic capacitance of medium-voltage power line insulators. A follow-on power electronic conversion system is then used to maintain the load conditions that maximise the extracted energy. The system performance has been analysed using theoretical studies and computer simulations to identify the operating conditions that maximise the extracted energy. Results obtained so far suggest that up to 100 mW can be harvested from 22 kV grid feeders, using only the capacitive coupling that exists between the line insulator harvester and the feeder conductor.