This year's Bath heat was more competitive than usual with five entrants.
First up was Bryan with “3-D Face Recognition for Biometric Access Control”. He began with a brief overview of existing biometric techniques, 2-D images, fingerprints, retinal scans and the some of the problems that can arise when using them. Bryan described how he was involved in a research programme that sought to overcome these problems by working with 3-D images. He described two methods of obtaining these images using stereoscopy or by illuminating the subject from four directions in sequence.
UWE Centre For Machine Vision
---------
It is always interesting to see how image processing techniques have advanced over the years. Images intrinsically contain a daunting amount of data, the art is to extract the key information. Obtaining different views by changing the direction of illumination is an idea that goes back to at least the 1930s. There is a lesson there for solving problems in other fields – try reversing the viewpoint!
----------
Esther was next with “Data Diagnostics and Discoveries”. She told us that intelligent actuators have some ability to record information such as operating torque and that information would be interrogated by service personnel sent out to repair faulty plant. She was working on a project to improve the precision and amount of data collected by each actuator. The aim was to be able to differentiate between common-mode problems such as electrical supply failures and those particular to the actuator. These might be caused by the user abusing the actuator or faults developing within it. By collecting data from many actuators it should be possible to avoid unnecessary service visits.
Rotork Ltd.
----------
The 'obvious' solution here would be to collect the data 'real-time', to become part of 'the internet of things' (IoT). However there are justified concerns that the security of plants that use these intelligent actuators would be compromised should they be connected to an extended network. Hardware engineers build closed boxes, software engineers like to open them up!
----------
Tim followed with some shock treatment, £20 billion, the cost of air pollution, and then onto his presentation, “Is The Air We Breathe Killing Us?” He listed some of the earlier air pollution events, such as the Great Smog of London in 1952 and showed how the death rate had increased with the smog but had continued to be raised even after the smog had disappeared. Smogs were visible signs of pollution but he told us that one of the major concerns nowadays was particles of 2.5μm diameter or less. In the past pollution from the burning of fossil fuels had been a worry but attention was now turning to material abraded from tyres and brakes on road vehicles. Even the home was not safe. Indeed we were told that formaldehyde, pet dander and cooking bi-products put us at risk. To close we were assured that research was under way to improve matters. One proposal was to use filter towers using natural solar-powered convection.
UK air pollution 'linked to 40,000 early deaths a year' (BBC)
----------
Some commentators challenge these recent reports that warn of deaths caused by pollution. They suggest that studies carried out in a small part of the USA have been extrapolated and applied to the wider world, and to the extent that they are valid the real affect is to marginally reduce the life-span of many people rather than trigger sudden death of a few. Either way it must be a good thing reduce emissions of all sorts where possible but not to rush into adopting different techniques that might prove to be worse. (Diesel powered cars?)
University of Cambridge Winton Centre – Does Air Pollution Kill 40,000?
----------
Tom then looked to be continuing the disaster them with an illustration of an aircraft wing-tip to the ground. Fortunately “Condition Monitoring: Lessons from Aviation” only touched on the crash of Puma helicopter G-REDL and how systems that should have indicated the true condition of the aircraft were confounded by systems to contain the damage. He described how he was involved in a project to see if the experience of the aircraft industry in using condition monitoring could be applied to the operation of a planned power station. He explained that obtaining operational data wasn't really the problem, the difficulty was extracting the critical information from it.
AAIB Report on G-REDL
----------
I got the impression that our own experienced power engineers weren't great believers in condition monitoring. The consequences of in-service failure were too great so service and replacement intervals had been determined to ensure failure didn't happen. To my mind that does raise the question of how those intervals are determined and if they are too short plant availability is reduced and running costs are unnecessarily high. Collecting too much data has become endemic, the question is what do we need to know and how do we measure it. Just because we can measure it doesn't mean we should!
----------
Finally Oliver gave us “Zero Emission Motorcycles – Inverters for bikes”. He told us that the University of Bath had been involved in a project to produce an electrically-powered motorcycle that had successfully taken part in the Isle of Man TT Zero 2017 event. This motorcycle used an inverter to feed the synchronous drive motor. However the existing inverters, originally produced for cars were larger and heavier than would be ideal for bike use. He described how they had been working on developing their own inverter using SiC (Silicon Carbide) semi-conductors. Because they had lower switching losses than those used in their current inverter it should be possible to reduce the volume and weight from around 6 litres to 4 and 8.5 kg to 2 kg. Prototypes of the gate control circuitry had been made and the software written and production circuit boards were expected to be delivered soon.
Bath Zero Emissions Motorcycle
----------
The power handling of modern semi-conductors is quite remarkable. However keeping them cool can be a problem so what might be seen as quite small gains in overall efficiency can be very significant in reducing the cooling requirement. I can't say that I have ever been a fan of motorcycles but isn't the engine noise part of the fun? (Actually a modern bike can be very quiet and smooth).
----------
Usually after all the presentations have been made the audience is left to its own devices as the judges retire to make their decision. This year, as an experiment, we tried something new. Maurice Poole and I made our own presentations, something that has been called 'Present Around Our Worlds'. The thought was that many of us have a latent presentation within us, perhaps not enough to fill an hour, but also that shared experience within the local centres is often just limited to the chatter before a formal event and maybe we could do more. Given enough interest it might be possible to run a special multiple presentation event.
I started off with a talk about the “Trident”, the missile, in particular why it should be capable of having a 40 year service life.
Maurice touched on “Diabetes”, a condition that is not well understood by those lucky enough to not to suffer from or be related to those that do. The current BBC Four 'Nordic Noir' was held up as a public example of this misunderstanding.
Both Maurice and I struggled to keep within ten minutes so “well done” to our presenters in showing us it can be done! (As PAOW is supposed to be non-competitive there is no reason why those presentations should be of fixed length).
Present Around Our Worlds (IET Discussion)
----------
At last the results!
Tim was our winner and Bryan the runner-up. Esther, Oliver and Tom should take some consolation that the judges clearly struggled for quite a while to come to a decision.
May Tim continue 'on and upwards'!
