Showing posts with label greenhouse gas emissions. Show all posts
Showing posts with label greenhouse gas emissions. Show all posts
Monday, February 10, 2014
Fujitsu Australia ICT Sustainability Survey
Fujitsu Australia are conducting an "ICT Sustainability Survey". Lee Stewart, Head of Sustainability ANZ at Fujitsu Australia, would like to hear from CIOs and IT leaders about ICT energy efficiency and the role of ICT in reducing their organisation’s energy consumption and greenhouse gas footprint.
Tuesday, July 02, 2013
Solar Panels to Lower Electricity Prices
Dr. McConnell started by displaying a graph showing the rapid take-up in solar energy. He then cited a 2011 Victorian Auditor General's report suggesting that Victorian brown coal generators were worried they would be adversely effected by a reduction in electricity prices. It was not clear to me why researchers at the University of Melbourne would be worried about maintaining the profitability of overseas owned highly polluting brown coal power stations.
Dr. McConnell pointed out that solar power output peaks during the middle of the day, which coincides with the peak period for electricity demand, As a result more solar power will lower the price all generators get for their electricity. There is a very sharp spike in electricity price in the middle of the day. The result is that solar power will adversely black coal generators most, brown coal, gas and hydro electricity.
Dr. McConnell pointed out that the University of Melbourne's model does not include many factors and so does not exactly reflect real prices exactly. In particular the model does not sow the very high mid-day peak. In addition the model does not take into account distribution costs.
What this analysis prompts is the question of what effect solar has on the stability of the network and the amount of reserve generating capacity required. On a cloudy day, alternative generating will be required. More solar power might require more gas generation which can be started at short notice when there is no sun.
Also domestic rooftop generators in Australia are paid based on a fixed amount per KW Hour, not the market price. It would be interesting to see what effect paying based on the market price would be. While individual households might have difficulty selling into this market, the output could be aggregated and sold by an intermediary.
Also if the cost of solar power continues to drop it may be able to provide more of the daytime power. The output from solar power may be useful in supplying the increased demand for home air conditioning.
Dr. McConnell suggested that the reduction in electricity prices caused by solar power might be sufficient to cover the cost of the support schemes currently in place for home solar systems. But if solar power is profitable without a subsidy, then it should be possible to discontinue these schemes and allow home and commercial providers to cover the cost.
Dr. McConnell's paper "Retrospective modeling of the merit-order effect on wholesale electricity prices from distributed photovoltaic generation in the Australian National Electricity Market" with Patrick Hearps, Dominic Eales, Mike Sandiford, Rebecca Dunn, Matthew Wright and Lachlan Bateman in
http://dx.doi.org/10.1016/j.enpol.2013.01.052
Dr. McConnell pointed out that solar power output peaks during the middle of the day, which coincides with the peak period for electricity demand, As a result more solar power will lower the price all generators get for their electricity. There is a very sharp spike in electricity price in the middle of the day. The result is that solar power will adversely black coal generators most, brown coal, gas and hydro electricity.
Dr. McConnell pointed out that the University of Melbourne's model does not include many factors and so does not exactly reflect real prices exactly. In particular the model does not sow the very high mid-day peak. In addition the model does not take into account distribution costs.
What this analysis prompts is the question of what effect solar has on the stability of the network and the amount of reserve generating capacity required. On a cloudy day, alternative generating will be required. More solar power might require more gas generation which can be started at short notice when there is no sun.
Also domestic rooftop generators in Australia are paid based on a fixed amount per KW Hour, not the market price. It would be interesting to see what effect paying based on the market price would be. While individual households might have difficulty selling into this market, the output could be aggregated and sold by an intermediary.
Also if the cost of solar power continues to drop it may be able to provide more of the daytime power. The output from solar power may be useful in supplying the increased demand for home air conditioning.
Dr. McConnell suggested that the reduction in electricity prices caused by solar power might be sufficient to cover the cost of the support schemes currently in place for home solar systems. But if solar power is profitable without a subsidy, then it should be possible to discontinue these schemes and allow home and commercial providers to cover the cost.
Dr. McConnell's paper "Retrospective modeling of the merit-order effect on wholesale electricity prices from distributed photovoltaic generation in the Australian National Electricity Market" with Patrick Hearps, Dominic Eales, Mike Sandiford, Rebecca Dunn, Matthew Wright and Lachlan Bateman in
Energy Policy, July 2013, Pages 17–27.
http://dx.doi.org/10.1016/j.enpol.2013.01.052
In 2007, the time frame for increasing the share of Victoria’s electricity consumption from renewable sources to 10 per cent was extended from 2010 to 2016. A range of factors, such as the interests of existing generators, the renewable energy industry and Victorian electricity consumers, were considered in extending the target. However, the extension occurred primarily to alleviate the concerns of brown coal generators that the 10 per cent target would deliver too much renewable energy generation too quickly,
which would reduce wholesale electricity prices and adversely affect existing
generators. ...
From: "Facilitating Renewable Energy Development", Victorian Auditor-General, April 2011
Monday, July 23, 2012
Sustainable Professional Education
CSIRO were short a speaker for their regular "Information Retrieval and Friends" and asked if anyone had something to present from a recent conference. I am just back from presenting a paper at the the 7th International Conference on Computer Science & Education (ICCSE 2012) in Melbourne, so volunteered to speak. My conference paper was "A green computing professional education course online: designing and delivering a course in ICT Sustainability using Internet and eBooks". The full text is available on-line, but the presentation notes are more readable.
About The Conference
ICCSE has been organized annually since 2006, by the National Research Council of Computer Education in Colleges and Universities of China (NRCCE). The first few conferences were held in China, then Singapore and most recently in Australia. The proceedings are published by IEEE.
While CS and engineering are international disciplines, some differences were noticeable for a conference organizes in China and held in Australia.
The standard IEEE Manuscript Templates for Conference Proceedings was used. However, the Yeedao paper submission system is primarily designed for Chinese users and I found the interface kept slipping from English into Chinese. Also the paper title and abstract were requested in both English and Chinense (I used Google Translate for the Chinese version).
Similarly, the conference arrangements were designed for traveler from China. As an example, payment of fees was by bank transfer to an Australian bank, but only the code for international, not domestic transfer was provided. I missed the welcoming lunch for the event as this assumed deligates were arriving via the international airport.
Papers were presented in English although most of the audience had this as a second language (and a Chinese to Hellish translator was required for the opening address).
I counted about two hundred delegates at the opening session, but by the last session of the second day there were about six delegates left. It may be that traditional oral presentations for such event should be abandoned and replaced with enhanced poster sessions, "un-conferecnes" and on-line sessions.
Real-time computer control as a metaphor for online education
The keynote address for ICCSE 2012 was from Professor Pedro Albertos,
Instituto Universitario de Automática e Informática Industrial Universidad Politécnica de Valencia (Spain) on "Interaction between computer control design and its implementation". He discussed the problems with the real-time computer control, arguing that real-time computing and control engineering need to be unified. Traditionally control engineering used analogue electronics with continuous measurements. Digital real-time control uses discrete time intervals. It is usually assumed that if there is a timing problem with a real time computer program, the solution is simply to speed up the processing so the time intervals are smaller. Professor Albertos argues that simply speeding up the process is not necessary, as where a feedback loop is used, most of the time the process is within parameters and no changes are required. Instead event based processing can be used, where processing is done when required.
It occurred to me that Professor Albertos' analysis might be a useful metaphor for online education. Face-to-face education, like control engineering assumes the tutor can see the class at all times and provide feedback. "Asynchronous" online education does not have all the students in the same place at the same time. The usually proposed solution to problems with online education is to provide quicker feedback, and ideally to use "synchronous" mode, where the student and tutor are online at the same time. However, this is inconvenient for the student and tutor (just as speeding up real-time computer processes wastes resources). Instead it might be better to have the online education event based, with time limited, so the student and tutor can be confident they will get a response when needed. Control theory has concepts such as "control effort", "open loop" and "closed loop", which could be applied to education.
Melbourne Model of Professional Education
Professor Iven Mareels, Dean of Engineering at University of Melbourne gave an overview of the changes to engineering and computer science programs brought about by the "Melbourne University Model". With this system, students choose from one of only six undergraduate degrees (engineering and computer sciecne students would normally undertake a Bachelor of Science).
Unlike undergraduate degrees from other Australian universities, the Melbourne degrees do not meet the requirements of professional bodies for entry. Students are expected to undertake a further 18 months study for CS or two years for Engineering to complete a masters.
Professor Mareels commented that this program was intended to meet the needs of international students. Many international students were undertaking the undergraduate degree in their own country and then coming to Melbourne University for the masters.
The courses are part student directed problem based learning. However, Professor Mareels commented that traditional lecture based courses have to be used in part, due to their lower cost. I though this a curious comment, as it is possible to design cost effective student directed problem based-courses by using on-line techniques.
Professor Mareels commented that University of Melbourne had invested in new learning faciltiies for the new teraching techniques. It happened that I also attended a "Climate Services Think Tank" while in Melbourne. This was at the Melbourne Business School (MBS), and so I was able to experience the new facilities firsthand. UoM appears to be following the same model of learning facility as used by other Australian universities.
About The Conference
ICCSE has been organized annually since 2006, by the National Research Council of Computer Education in Colleges and Universities of China (NRCCE). The first few conferences were held in China, then Singapore and most recently in Australia. The proceedings are published by IEEE.
While CS and engineering are international disciplines, some differences were noticeable for a conference organizes in China and held in Australia.
The standard IEEE Manuscript Templates for Conference Proceedings was used. However, the Yeedao paper submission system is primarily designed for Chinese users and I found the interface kept slipping from English into Chinese. Also the paper title and abstract were requested in both English and Chinense (I used Google Translate for the Chinese version).
Similarly, the conference arrangements were designed for traveler from China. As an example, payment of fees was by bank transfer to an Australian bank, but only the code for international, not domestic transfer was provided. I missed the welcoming lunch for the event as this assumed deligates were arriving via the international airport.
Papers were presented in English although most of the audience had this as a second language (and a Chinese to Hellish translator was required for the opening address).
I counted about two hundred delegates at the opening session, but by the last session of the second day there were about six delegates left. It may be that traditional oral presentations for such event should be abandoned and replaced with enhanced poster sessions, "un-conferecnes" and on-line sessions.
Real-time computer control as a metaphor for online education
The keynote address for ICCSE 2012 was from Professor Pedro Albertos,Instituto Universitario de Automática e Informática Industrial Universidad Politécnica de Valencia (Spain) on "Interaction between computer control design and its implementation". He discussed the problems with the real-time computer control, arguing that real-time computing and control engineering need to be unified. Traditionally control engineering used analogue electronics with continuous measurements. Digital real-time control uses discrete time intervals. It is usually assumed that if there is a timing problem with a real time computer program, the solution is simply to speed up the processing so the time intervals are smaller. Professor Albertos argues that simply speeding up the process is not necessary, as where a feedback loop is used, most of the time the process is within parameters and no changes are required. Instead event based processing can be used, where processing is done when required.
It occurred to me that Professor Albertos' analysis might be a useful metaphor for online education. Face-to-face education, like control engineering assumes the tutor can see the class at all times and provide feedback. "Asynchronous" online education does not have all the students in the same place at the same time. The usually proposed solution to problems with online education is to provide quicker feedback, and ideally to use "synchronous" mode, where the student and tutor are online at the same time. However, this is inconvenient for the student and tutor (just as speeding up real-time computer processes wastes resources). Instead it might be better to have the online education event based, with time limited, so the student and tutor can be confident they will get a response when needed. Control theory has concepts such as "control effort", "open loop" and "closed loop", which could be applied to education.
Melbourne Model of Professional Education
Professor Iven Mareels, Dean of Engineering at University of Melbourne gave an overview of the changes to engineering and computer science programs brought about by the "Melbourne University Model". With this system, students choose from one of only six undergraduate degrees (engineering and computer sciecne students would normally undertake a Bachelor of Science).
Unlike undergraduate degrees from other Australian universities, the Melbourne degrees do not meet the requirements of professional bodies for entry. Students are expected to undertake a further 18 months study for CS or two years for Engineering to complete a masters.
Professor Mareels commented that this program was intended to meet the needs of international students. Many international students were undertaking the undergraduate degree in their own country and then coming to Melbourne University for the masters.
The courses are part student directed problem based learning. However, Professor Mareels commented that traditional lecture based courses have to be used in part, due to their lower cost. I though this a curious comment, as it is possible to design cost effective student directed problem based-courses by using on-line techniques.
Professor Mareels commented that University of Melbourne had invested in new learning faciltiies for the new teraching techniques. It happened that I also attended a "Climate Services Think Tank" while in Melbourne. This was at the Melbourne Business School (MBS), and so I was able to experience the new facilities firsthand. UoM appears to be following the same model of learning facility as used by other Australian universities.
Subscribe to:
Comments (Atom)