A and E

Last night I took one one of my mates to A and E, nothing serious. Since there was a long waiting time, it gave me an opportunity to read quite a bit of hot air. It seems that nuclear power is the future for us if we want to have a 'sustainable' energy consumption for the next 1000 years. There are obviously a lot of problems to overcome first before but it is not totally infeasible. Over the next decades we sill see a rising demand for electricity. This will be due because of a shift to cleaner energy. Electricity is a high grade form of energy which if generated in the right way can be very clean. Therefore a shift to the electrification of certain service such as public transport and heating will be experienced. The book talks about the joys of heat pumps compared to conventional gas fired boilers. They are much more efficient at producing energy (in the form of heat) then gas or combined heat and power (CHP) and thus should be used in the future to provide living spaces with heating (and in the summer with air conditioning) Therefore I will be researching ways in which these micro generators can be incorporated into microgrids.

BAE Systems Report

Today I read the BAE Systems report on smart grids, here is a summery:

This report gives a detailed overview of smart grids and there applications. It begins with defining what a smart grid is and what it consists of. It then goes on to define how it could be used to manage the electricity in a static or deployable army base.

It identifies three aspects of the system that it would implement: demand side management (DSM), supply side managment (SSM) and Network management (NM). DSM is concerned with understanding and manageing consumer power demands in order to minimise peak power across many different consumers.SSM is Resposible for ensuring there is sufficient electrical power available to meet the total real-time consumers electrical power requirements. Needs to balance the generating of power across different generators and facilities. Uses predicted power estimates. NM manages the transmission and distribution of electrical power from the electricity generators to the consumers. This report provides the advantages of using a smart gird to manage the electricy on an army base.

Reading this report gave me some interesting areas that I could look into:

power dispatch and scheduling
network load balancing

All in all I want to look at power flow management and how this could be achieved in a microgrid environment.

monday morning

Today I spent most of my time finishing off the reading of "factor graphs and the sum-product algorithm". Oli and I also went through some more examples for the power systems lectures. TODO (for tomorrow): Read the BAE systems report on Smart Grids, Make a gantt chart for time up until the 9 month report, Go through the last question for power systems analysis.

maths

Over the past few days I have been reading two papers that contain quite a bit of maths. I have been going over both papers trying to understand what is going on. It is slow progress for me. The second of the two papers "factor graphs and the sum-product algorithm" is easier to understand and get a grasp of. So far essentially what the algorithm does is to construct a factor graph from a particular problem (which contains functions and variables) and then compute each variable as a product of all the factors. This is essentially computing a functions as a product of local functions (which take less operations to compute). This is what I understand is going on. Other then this, I have been working through a number of power analysis problems consisting of load factors, peak power, cost for generation etc.

maths

Today I have been trawling through the paper "The Generalized Distributive Law" to understand the maths behind the Max-Sum algorithm. The GDL is based on the "humble" distributive law [i.e ab + ac = a(b + c)] but generalised to other things, such as trees. The idea is that you decompose a problem into a set of smaller problems such that the complexity to solve the smaller problems is less than the complexity to solve the original function. Example f(x1, x2, x3), the complexity to solve this is dependant on all the variables, whereas if f(x1, x2, x3) = f(x1, x2) + f(x2, x3) the complexity is dependant on the maximum number of variables in the functions (in this case 2) therefore reducing the number of operations required to solve the problem. Ruben helped me to understand the basic concept of the GDL algorithm. The paper did contain some difficult maths which I got the general gist of, however I will need to go over this in more detail. TODO (tomorrow) : Read the second paper that Max-Sum is based on "Factor Graphs and the Sum-Product Algorithm"

monday morning

This week I shall be focusing on a lot of reading, surprise surprise! I was given another book to read by my supervisor Gopal called Perfect Power which details "how the microgrid revolution will unleash cleaner greener, and more abundant energy". TODO: Finish reading the current paper, and read another, go to the 14:00 lecture on sustainable energy, read some hot air, look at notes on conventional generation techniques.

interesting stuff

Read a few papers today and had some possible ideas about what to investigate: one paper Coalition formation in transmission expansion planning uses coalition formation to determine the optimal number of transmission lines to add to an existing network in order to supply forecasted load. This same technique could be applied to mircorgenerators to determine how many should be added to the network and in what locations in order to meet various flows. This could be based on certain factors like average power output or efficiency or something. I also talked to Harry in the labs today, he suggested Power Matcher (which I already new about) and also IDAPS which I had forgotten about, so will look at the various papers produced on these implementations in due course.

interesting lecture

Had a lecture on power systems analysis today. It was quite interesting and gave me a few ideas of what could be done. For instance load factor (average power / peak power) could be taken into consideration when controlling a microgrid. The idea is that you want the load factor to be as close to 1 as possible, that way the fluctuations between supply and demand is reduced meaning that high cost generators, which allows peak demand to be satisfied, do not have to be used as much! This is difficult to achieve in a microgrid that contains alot of renewable generators and many loads, but this could be solved using agents. I am interested in this and will look into this area of agent based microgrid management. TODO: Print off some more papers to read, read one paper and make notes, explore load factor and load loss factor, read notes on power systems analysis, go to lecture at 12:00 (sustainable energy), read some hot air (for background knowledge).

computer set up

Finally set up the new computer, its a beastly quad core with 12 GB of RAM and two 22" monitors, veeeerrrry nice! TODO: Read two papers and make comments, download and print some of the papers suggested by Gopal, read some Hot Air.

the start of the week

Tried to go to the conventional generation technologies lecture today at 9am but the lecture was full and there were no chairs. Sorted out my ID card and now in the office getting ready to read a bunch of stuff. TODO : Read some more hot air, Read one paper and make notes, Go to the lecture at 14:00. Over the weekend I have been thinking about what sort of area of smart grid I want to look at. I definitely want to look at integration of microgenerators in a microgrid environment. I need to find the papers that I read for my 4th year research project and make notes on them.

induction

Today is the first part of my induction to the university and postgraduate research. It involves some talks by various important people and a tour of the university campus (in case I have forgotten where everything is :p) TODO: Other then that I am going to read two papers today, they are Multi-Agent Systems for Power Engineering Applications Part I and II (written by the IEEE Power Engineering Society's (PES) Intelligent System Subcommittee). These should give a great deal of background knowledge about how to apply agents to the smart grid environment. I will also be reading my daily dose of "Sustainable Energy - without the hot air" I suggest anyone and everyone to read this, available online http://www.withouthotair.com/

lectures

I have a couple of lectures today, one was at 9:00 (power systems) and the other is at 12:00 (sustainable energy). Both should provide a good background knowledge to the area of Smart grids and energy. TODO: Read and provide comments on two papers, go to the lecture at 12, read some more of hot air, hand in demonstrator form and hopefully set up the new pc.

an interesting fact

One for the people against wind turbines because of bird deaths -
Annual average number of birds killed by wind turbines in denmark = 30,000
Annual average number of birds killed by cats in Britain = 55,000,000
Anyway I digress:
Today I read two papers on the AuRA-NMS which is currently being developed by a number of different Universities and companies. They have come up with a framework that can control a number of things using agents such as power flow and voltage levels, and integration of generators. I have also read a large amount of the book I mentioned previously, it is very interesting. From what I gather so far, the most feasible microgenerators that could be installed on every home and actually produce enough electricity to be worthwhile is photovoltaic panels. Also thermal panels which heat water are another viable way to decrease demands for electricity due to heating. Ways in which these technologies could be integrated and controlled by agents in a distributed system is of some interest to me.

tasks for the day

Got into the office about half an hour ago and have been getting ready to read all day. Got a selection of papers and a book on sustainable energy (without the hot air), so should be good! Hopefully the computer will come today so I can set up my working environment properly. Still, early days!

officially started

So I have now left the world of the intern and entered the realm of the postgraduate researcher! Had my first kickoff meeting today with Alex and Gopal, very daunting but expected. I have some papers to read on multi-agent based systems and energy, gonna get cracking with them over the next couple of days! My computer has apparently arrived but not yet with me, fingers crossed I get it soon.