Seasons and Simulations

The energy yield from wind and solar generators is seasonal, the nature of the variation depends on the climate. In the UK the output of wind farms peaks in winter whilst solar parks make their greatest contribution during summer. Thus supplying a constant load or one with relatively small fluctuations is a challenge. There are two ways of looking at sustainable energy systems, they can either augment fossil/nuclear sources or they can displace them. These two approaches result in different strategies. In the former, fossil fuelled generators act as a back up for wind and solar source, in the UK it seems that at any given time the wind generation accounts for roughly 0 - 20% and on a calm summer's evening the output from PV devices will also approach zero. A combined wind and solar strategy when combined with storage and energy management has the potential to make a group of consumers largely independent of conventional energy sources. Taking the energy economy as a whole it is unrealistic to believe that fossil/nuclear sources will be displace in the foreseeable future, but it maybe possible to build new communities which are based on sustainable sources as an alternative to building additional coal, gas or nuclear fuelled plant.

The graph shows the result of a simplistic simulation of a wind farm and solar park, the parameters were arbitrarily set such that each unit had a peak output of around 1,000 MWh/month. The output of each source shows significant seasonal variation, but when combined the output is more or less constant.

This exercise did not take into account inconvenient issues that might arise from economic considerations.

Peak or Average

Over the past year I have been taking a cursory look at the relationship between the demand for energy and the ability of wind and solar systems to provide it.  On the supply side, wind energy comes in pulses which are stronger in winter than summer and the sun does not shine at night and is not too bright in winter.  On the demand side, the base load continues overnight, rises overnight as the day workers leave home and peaks in the evening when meals are cooked.  Adding storage to the system improves the relationship between supply and demand, but at a cost.  Whilst there are incentives to generate electricity, there is little to encourage people to be better consumers other than the price which is a crude and sometimes unfair, regulating device.

In general, there are two ways of designing a system.  A system designed for the average load will probably be smaller and cheaper than one designed handle the peaks.  The downside is that there will be times when frustration erupts around the smaller system.  Anyone who has had to specify office equipment such as printers, copiers, coffee machines will be familiar this dilemma.  One take on this is that if communication is better around the coffee machine than in meetings, scrap meetings and get a small coffee machine.

This post was in part, inspired by our washing machine.  I have no in depth knowledge of how this machine works, so I apologize if I have drawn the wrong conclusions.  Sometime back, I place an energy meter between the wall socket and the machine's power plug.  For most of the cycle, the energy consumed was significantly less than 1kw, during the early part of the spin cycle, it rose to around 2.5 kw, something, somewhere had to provide an extra 1.5 kw for a few minutes.  In the context of a washing machine, it might be possible to reduce this peak by increasing the time taken to get the drum up to a speed at which water and socks are separated.  This would not reduce the overall energy consumption and would increase the time taken to complete the wash, but it would make the machine a "better" consumer.  I guess this was what was behind the recent introduction of a maximum power rating for vacuum cleaners.

A lot of the discussion around sustainable energy focuses on technology, but behaviour is also an element.  Price alone is a crude regulator, but tariffs and business models could evolve to get a better outcome.  One possibility is a variant that  does not restrict consumption, but enforces a ceiling on the rate of consumption, this could encourage the development of storage and management solutions.  I've heard several references to tariffs in which the charge increases with consumption which provide an incentive to manage consumption.  These ideas are in conflict with the traditional business model which is based on sales volume and market share, what is needed is a business model which creates viable energy companies, but at lower levels of consumption.

Energy Conservation and Management

I suggest that energy management, conservation and storage are at least as important as generation in transitioning to a sustainable energy economy.

The development of wind farms and solar parks and other forms of sustainable generation is important, but it is easier to utilise energy from weather dependent sources if demand is low and managed and fluctuations in availability can be buffered with storage.

Reaching this conclusion (which I don't claim to be original) was a two stage process. I first became interested in wind and solar systems around 2007, (just before the financial crisis) partly to find a use for data and programming skills learnt in the oil and gas industry and partly a desire to be the energy equivalent of Tom and Barbara Good (characters in the 1975 BBC sitcom in which the lead characters attempt self-sufficiency in Surrey). Being cautious by nature I investigated solar PV by creating a very small off-grid system based on a 4.5 watt panel, 3 amp hour lead acid battery and an LED light. During the day the battery charged and by night the LED light illuminated alfresco dining. Over a year, I learnt that at 51N, solar does not do much in winter and even in summer clouds significantly reduce clear sky solar irradiance. Wandering around the garden with a wind speed meter convinced me that an urban valley is not a good place to put a small wind turbine, even if I could convince the neighbours and the council it was a good idea. The conclusion f this stage was that it was better to buy electricity from a company with wind turbines and solar panels in optimum locations, such as offshore and on ridges.

The second stage started with a load of data, Excel, and Visual Basic, later to be replaced with Python. The objective was to understand the potential of wind and solar sources. A lot of time was spent fitting Weibull Distributions to wind speed data and a simple model of clear sky solar irradiance. The latter was based on simplified sun-earth geometry, the work of the Meinels from the early 70s which provided an estimate of direct normal irradiance and some crude observations of diffuse irradiance made in a local park. At this time I started trying to create a simple mode of cloud sky irradiance (this is still work in progress). There were several conclusions from this phase, the first being that offshore wind is a better source of energy than onshore wind, however, I know from my oil and gas days that working offshore is significantly more expensive than onshore. Economics may be inconvenient, but they can't be ignored. Secondly, was the problem of seasonality. It's a gross oversimplification, but at 51N, solar panels do not produce much energy in winter and summer winds are gentler than those of winter (the energy available from wind is proportional to the cube of its velocity). Thus a sustainable energy economy needs diverse sources, not an option for many urban households.

As an aside, much as I love software and data, its good to try and link what one sees on the screen with the effects of wind and sun which one feels in the face. I attempted to bridge this by cycling around the the town and surrounding countryside with a wind speed meter. If on the seafront, the wind was blowing smoothly at 10 m/s, on the roofs of multi story car parks in the town there would be long period gusts, depending on the direction my backyard could be calm and on the surrounding hills, the speed would still be 10 m/s, but with significant turbulence.

Having gently shelved the idea of putting PV panels on the roof, I concentrated on reducing household energy consumption. When we started, the house was lit by incandescent light bulbs and whilst we were not extravagant with energy, we did not think much about it. A quick win was swapping the incandescent lamps with CFL, this alone took consumption down from 20 - 25 kwh/day to less than 10 kwh/day. As things like computers have died, we opted for energy efficient replacements and in the last year, some CFL's have been replaced with LEDs with the result that we now consume less than 5 kwh/day and have relatively low electricity bills. For heating, we have opted for the counter intuitive solution of reverting to open fires. Our gas central heating system is not yet zoned, by only heating one room, albeit inefficiently, the the amount of energy consumption for space heating has dropped. None of this has required major capital investment, but has given a better outcome than a roof full of PV panels. Technology has also changed, once I used a big PC to do sums and software, this also acted as a space heater, increasingly I now use a Raspberry Pi, excluding the attached monitor, this take about 3 watts of energy. This is where the doctrine of unforeseen consequences kicks in, I sometimes light a fire to keep warm in winter.

Wind and solar are both climate weather dependent energy sources. Solar works well in winter under the clear skies of a hot desert, less so in the cloudy ones of a winter in a temperate maritime region. Wind energy arrives in pulses, sometimes days apart, sometimes weeks, the pulses being stronger in winter than summer. I have no experience of grid management, but my understanding of demand is that it is low at night, rises during the day and peaks in the early evening when families, cook meals, children do homework and many people watch television or interact with screens. It would seem that grid management would be easier and generation more efficient it the load was constant. Thus if every new building incorporated a few hours of storage capacity the peaks in demand would be flattened and it would be easier to incorporate sustainable energy sources such as wind and solar.

Similarly, demand can be managed, a trivial example is our washing machine, we always run this around 09:00 in the morning for no very good reason, however, if the washing machine is part of the internet of things then something could make a decision on when was the optimum time, say in the middle of the night when demand is low or sometime when the wind is blowing and wind farms are productive. I do not minimize the problems of this approach, not least of ensuring that the machine is loaded and the door shut. Also, it is important to define the boundaries of the systems being controlled. There are at least two configurations, the first is at the level of the home where the home accepts a maximum load that they will impose on the grid, maybe, in return for a lower tariff. Alternatively, there can be central control by the grid operator. This taken to its illogical conclusion could spark a debate on civil liberties and washing machines.

The answer is a cabbage!

The day job is messing with some energy management software, sustainable energy presents some inconvenient challenges, but there are some exciting technologies emerging (is it dangerous to use the word exciting when you are sixty one and a half?).  Not least of these is the internet of things.  Historically, energy distribution systems have been configured to supply the peak demand, the underlying assumption being that there is little intelligence in the system beyond thermostats and tariffs such as Economy 7 which encourages the use of off-peak electricity.  We only run the washing machine in the morning, because we've always done it and there is no timer switch.  During the spin part of the cycle, the washing machine draws about 2.5kw for a few minutes.  Taking this to its illogical conclusion, a large pile of dirty socks is causing a spike in demand which in turn causes a gas fired power station to start up.  If the nation did its dirty washing overnight and we might not need the gas fired power station.  If I can control the devices in my home, I can be a nicer energy consumer and possibly hand the task of energy management to my mobile phone.  I have a Raspberry Pi which is running an emulation of a system (not related to the stuff mentioned above), this imaginary system feeds its status to the Pi's web server which can be accessed by my phone.  Maybe one day the Pi and my phone argue about the best time to do the washing, it is the internet of things that lets the the washing machine join in the conversation.

Apart from this blog, I don't advertise my interest in these things.  However, I have spent some time in the garden with a small PV cell and a multimeter observing the effect of clouds on a very small solar device and engaging in other equally bizarre pursuits.  This activity has not gone unnoticed and I occasionally get asked "why don't you have solar panels on your roof?".  This question always comes in summer and generally from people who have PV panels on their roof.  My response is that solar panels at latitude 51N  don't do much during winter when demand for energy is high.  In summer they work well but in winter sun-earth geometry and clouds can take the output down to well below 1 kwh/m2 compared to 8 kwh/m2.  Sustainable solutions need to take into account seasonality, solar panels might let you balance your consumption with generation over the period of a year, but in winter you still need the gas fuelled generators.

Conventional energy businesses have evolved a functional business mode, it's one that works because in the UK when you flick the switch, the lights go on (if they don't, the failure will probably be discussed in the media).  For sustainable systems, maybe agriculture is a better model because it is based on seasonality.  I'm a newbie gardener, but it seems to be that only difference between vegetables and solar panels is one of form and one tastes better than the other. There is a natural rhythm to the gardening year.  You start by planting seeds which will germinate when the soil is cold, like broad beans which will be harvested in late spring or early summer.  Next come plants like tomatoes which need the summer sun to grow.  When the spring and summer vegetables have been cleared, it is time to plant the winter vegetables like cabbage which will grow slowly as the days become shorter.  Thus on this bleak January day, I am going to pick a cabbage for the evening meal and later seek out a book on how to grow them properly, as those from my first attempt are a bit weedy. Part of the process of gardening is storing produce thus jam making hand hydrogen storage have a lot in common.

What are you doing with that soil in the kitchen?

At the age of eleven, I failed an exam called the 11+.  Passing it would have got me a place at the grammar school, which in turn might have led me directly to university.  As it was, I was placed in a secondary modern school.  The principal weakness of these schools was lack of aspiration, however, with hindsight, I realise the some of the teaching excellent and that I have reason to be grateful to at least three teachers.  Four years after leaving school at fifteen, I did secure a place at university to study engineering having passed some of the intervening four years as a merchant seaman (where I was England's most incompetent sailor) and as a shot blaster and painter.  Without the efforts of Mr. Wilson, Mr. Henschel and Mr. Stevenson, life could have been rather dull.

Mr. Wilson did not give me a love of maths, but he did provide a good foundation which later helped get into technical college to study for an Ordinary National Diploma.  In the 1960's many maths teachers had served in the Royal Artillery during the war, so in  addition to geometry, algebra and basics of calculus we were also taught some ballistics.  Mr. Henschel and Mr. Stevenson were both science teachers, Mr. Henschel retired in my fourth year and was replaced by Mr. Stevenson.  They were both good, but a little different.   Mr. Henschel taught in a methodical way which was not boring, I can remember many of his lessons including Hooke's Law, Ohm's Law, specific gravity, series and parallel circuits, make and brake mechanisms and much else.  With hindsight, I realise he was also quite wise, describing me as "sea lawyer" which is sadly true and an observation of which I should have taken more notice.  Mr. Stevenson was also good at "labs", but probably because he taught me in fifth form when young people should be open to ideas, he posed a few wider questions.  I remember asking a question about photosynthesis and how plants responded to different colour light, he suggested that I answer this myself by doing an experiment, I deeply regret not doing so.

For the past couple of years, sometime around sunset on Sunday afternoons I have been taking soil temperature measurements in my back yard.  The original reason for doing this was an attempt to work out when to sow seeds, in the way of these things, no obvious answer presented itself,  What was instructive was a few random measurements during the day which showed that there was considerable diurnal variation which creates a thermally turbulent environment for germinating seeds.  I appreciate that I could learned this in a book as I know this to be common knowledge amongst better gardeners than me, but one of the themes of this blog is how do you learn about stuff.

It was whilst poking a thermistor down a hole and hoping not to have to explain this bizarre activity to passers by, I remembered an experiment I had performed at school.  The objective was to measure the water and organic content of soil.  The method involved two steps, the first was warming a sample of soil to around 100 deg. C so that the water evaporated leaving dry soil.  In the second step, the sample was exposed to a bunsen burner and heated until it glowed a dull red after which it was assumed that all the organic matter had burned off.  At each stage the soil was weighed.  I adapted this process to utilise the facilities afforded by the kitchen scales and the living room fire.  It was my wife who posed the question which is the title of this post.

Approximately 200 gm of soil was placed in a an empty tin from which the label and remaining dog food had been removed.  In the morning, the previous evening's fire has burned out, but the ashes are still warm, so the tin is place amongst them until it is time to relight the fire in the evening.  At this point the soil is dry.  For the next stage, the coals are placed around the tin and the fire list as normal, the next morning when it has cooled the tin is weighed.

I'm hoping to create a time series of the water content measurements, as the winter recedes, it should be possible to place the tin in the sunshine which will cause it to dry, albeit at a slower rate than in the cooling ashes.

For a sample taken on 28-Dec-2014, the water content was 26.3% and the organic content 4.2%, the accuracy of these measurements is uncertain.

This is a long answer to a short question.