Peak Energy Demand and Demand Side Response

Jacopo Torriti

• Publisher: Routledge
• Year: 2015
• ISBN:
  9781317704478
• Edition: 1
• Topic: Economics
• Subtopic: Sustainable Development
• Index: Economics

With different intensities, depending on the season, every morning and evening of any weekday there are the same peaks in electricity demand. Peaks can bring about significantly negative environmental and economic impacts. Demand Side Response is a relatively recent solution in Europe which has the potential to reduce peak demand and ease impending capacity shortages.

Peak Energy Demand and Demand Side Response presents evidence on a set of Demand Side Response activities, ranging from price-based to incentive-based programmes and policies. Examples are drawn from different programmes for both residential and non-residential sectors of electricity demand, including Time of Use tariffs, Critical Peak Pricing Automated Demand Controllers and Ancillary Services. The book also looks at the actual energy saving impacts of smart meters, the activities which constitute peak demand and the potential opportunities associated with European smart grids and Capacity Markets.

This is the first book presenting comprehensive analysis of the impacts, cost benefits and risks associated with Demand Side Response programmes and policies. It should be of interest to students, scholars and policy-makers in the areas of energy, environmental economics and applied economics.

1 Peak demand

DOI: 10.4324/9781315781099-1

With different intensities, depending on the season, every morning and evening of any weekday there are the same peaks in electricity demand. Peaks are problematic for system balancing, utilities pricing and future grid development. The volume of energy demand depends on many factors, including weather, type of appliances, types of building, etc., but people, their routines and practices are the key to understanding what constitutes peak demand.

Introduction

It is 5.34pm on Tuesday 18th November 2014 in a crowded train of the London Underground. It is a day like any other. And this is what makes it special for peak electricity demand purposes. People have just left their offices and are rushing home. As soon as they get home they will all switch on the lights because it is already dark at this time of the year. Soon after, some will turn on their laptops, tablets, TVs or radios. They may run appliances, give a bath to children, turn on kettles for some tea, prepare meals or get ready to go out by having a shower and drying their hair. The homes they live in, the heating and cooling systems that change indoor temperatures, the appliances and devices they use will all differ, resulting in a variety of load intensities. However, the activities they carry out are very similar and all concentrated over the same period of time. The sum of these activities culminates in peak electricity demand.

There are some matters of fact in electricity supply that need to be taken into account when thinking about peak demand. First, generation and demand must be equal at any time. Second, grids do not store electricity. Third, the integration of renewables requires flexible power plants. Smart grids are the centrepiece of a future with more low-carbon and intermittent energy sources, more storage and more energy traded across borders. According to this vision, a vital role in the future smart grid will be played by flexible consumers: they will know how much power they are using, switch off appliances at peak times and manage demand by themselves with the aid of automated controllers and energy aggregators.

From an economic perspective, peak demand per se would not be a problem. Like in other markets where sudden increases in demand push prices up, for decades energy suppliers have been making money at times of peak demand thanks to higher marginal prices. Before deregulation of the electricity market occurred, utilities ran generators at ‘baseload power’ and at peak times they added to the supply mix plants, which could be easily powered up and down. With the deregulation of the industry, generation needs to take into account the marginal cost of electricity. In order to increase the penetration of renewables, legal incentives have been set in place so that System Operators prioritise renewables over other forms of generation. This means that renewables, whose generation is intermittent as it depends on weather, become baseload providers.

At times of peak demand, increasing the output of gas-fired generators may not be enough. This makes electricity prices more volatile. Electricity price spikes occur frequently when demand is very high because there is extremely high price volatility due to the higher volume risks. However, in the equation one has to consider also the risks of plant outages. At peak time less efficient or higher cost generating units are used to meet higher demand. In such circumstances, electricity prices in wholesale markets could fluctuate from less than #0.4/kWh to as much as #0.25/kWh on several days of the year.

In essence, peak energy demand has significant implications for system balancing, utilities pricing and future grid development. Issues of time and timing have not featured strongly in energy research and energy policy analysis as both have predominantly focused on estimating and reducing total average annual demand per capita both at the individual household and system levels. Traditionally, balancing demand and supply occurred via expansion of capacity base to deal with aggregate increases in energy demand. More recently, higher awareness over the greenhouse effects of fossil fuel generation implied that concerns regarding demand-supply matching cannot justify grid expansion. As a result, balancing energy supply and demand is becoming an increasingly complex challenge. It is this challenge which this book sets out to explore.

When I tell people that I carry out research on peak energy demand, many respond: ‘Ah, it is when there is an ad during EastEnders [a popular TV soap opera in the UK] and everyone puts their kettles on.’ This is more or less on point. What is on point is that peak demand is triggered by the simultaneous actions of many. Less on point is the fact that the kettles’ incidence is a very particular type of peak in demand, which for simplicity I will refer to as spikes. Spikes trigger sudden decreases in frequency across electricity generation and transmission networks. Frequency should always be between 49.5 and 50.5 Hz. Spikes are short in nature and are problematic when they occur infrequently. For instance, the highest spike in the history of the UK took place on the evening of the 4th July 1990 just before the World Cup semi-final penalty shootout between England and West Germany. This book does not focus on demand spikes, which are generally dealt with by the Transmission System Operator with the aid of pumped storage power stations. This book focuses on peak demand of longer duration (a couple of hours in the morning, four to six hours in the evening) which recur over time.

Many factors influence electricity demand. Several of them have been extensively analysed in research, including weather (Harris and Liu, 1993; Pérez-Lombard et al., 2008; Yu et al., 2009), building characteristics (Bartusch et al., 2012; Blom et al., 2011; Crawley et al., 2008), appliance design (Firth et al., 2008; Waide et al., 1997; Wood and Newborough, 2003), appliance control (Mohsenian-Rad and Leon-Garcia, 2010; Yamamoto, et al., 2008), interdependencies between energy services (Yoon, 1984), etc. Most of these factors have an impact on the intensity of demand. For instance, leaky homes require more energy to reach the same space heating temperature than better insulated homes. In general, older appliances are less efficient and end up consuming more electricity, gas and water than newer appliances. Weather is renowned for having a direct effect on heating and cooling loads. In certain seasons weather might also change occupancy trends to the point that in temperate countries in the summer electricity demand is significantly lower than in winter time. All of this clarifies how volumes of energy demand occur, but what about patterns? What constitutes peak energy demand? What shapes the load profile throughout the day? The answer is simple and complicated at the same. It is the people: what people do and when they do it. Peak demand is not determined by individuals’ desire to consume energy at a given point of the day, but by the way people’s days are structured, which is partly in their hands (routines and habits), but partly defined by the obligations and social structure of time (schedules and social practices). One very straightforward way of seeing this is to think about the substantial difference between load curves for weekdays and weekends. During the same season the weather can be equally good or bad at the weekend compared with the weekday. Everything else remains the same whether it is a day of the week or the weekend: building, appliances, fuel substitution, price of energy and appliance control, and even the moment of the day in which there is or there is not sunlight. The one thing that really changes is what people do. Peaks are a recurrent phenomenon during weekdays because life is structured according to routines and practices.

Routines, practices and times of the day

Since I have children in a country where I did not grow up, thanks to them I am catching up on the culture of the country in which they are growing up. One way of understanding a different culture is through children’s songs. Children seem to love repetition and routines and so many of their songs are written around everyday routines. These include songs around morning routines (getting up, breakfast, bathing, hand washing and getting dressed), toileting, going to school, lunch time, cleaning up time, brushing teeth, getting undressed, and finally (after much singing), sleep, rest, bedtime and ending the day. For a full list and authors the reader can refer to this website: www.songsforteaching.com/everyday/everydayroutines.htm.

I am an economist and much of this book is about ways to intervene, including through price, to modify peak energy demand. I also think that in order to understand what peak demand is made of, one needs to start from the activities which constitute peak demand. For this reason practices, routines, habits and everyday lives are essential starting points to the study of peak demand and to assess the potential of any forms of intervention. People leave traces of their electricity demand through bills (which tell us something about consumption for a period of time, generally a couple of months and how much people spend on electricity). Meters contain real-time information about electricity demand. When looking at this type of data one realises that there is plenty of variation between individual households. Beyond the physical loads and price, it is possible to find out something more about peak demand, some distinguishing features, and some signs of what people do. The point of peak demand lies not so much in individual’s choices as in the patterns they make. At the time of writing this book, I was very fortunate to be part of DEMAND, one of the End Use Energy Demand Research Centres funded by the Engineering and Physical Sciences Research Council in the UK. DEMAND starts from the observation that understanding when (peak) energy demand occurs, i.e. at what time of the day, is inextricably related to questions of where (peak) energy demand takes place, e.g. in the home, at work and on the move, and why, e.g. what activities underpin it (Walker, 2014). The DEMAND centre is led by Elizabeth Shove, a practice theorist who believes that social practices shape how and when we do things, and consequently when and where peak energy demand occurs. People do not consume or demand energy. Rather, they consume or demand the services that energy provides, e.g. laundry services, heating services, etc. (Wilhite et al., 2003). Individual behaviours and social practices are supported by the consumption of relevant energy services. Material arrangements, including energy services, only have meaning within, and in relation to, the practices in which they are enfolded, and through which they are reproduced (Schatzki, 2010). Changes to the timing of energy demand could only be triggered by non-discretionary factors, such as practices, levels of occupancy, location, number of occupants, type of tenancy and dwelling type (Shove, 2004; Warde, 2005).

Peak demand originates from a high level of synchronisation of activities happening in different households. High synchronisation can be seen as many people doing the same things at the same time. When synchronised activities are linked with appliances and devices which require energy, then these become of interest for peak demand purposes. Arguably the moment when we achieve the highest level of synchronisation is when people are sleeping in the middle of the night. We are all doing the same thing, but this has very low energy intensity implications. This is why we get troughs in electricity demand in the middle of the night, another interesting phenomenon with implications for energy systems which is addressed in this book. Thinking in terms of practice and synchronisation as underpinning peak demand does not mean that there is no variation across households or that routines dominate every day of our lives. This could also be seen as practices having different rhythms (Southerton, 2006).

In the media as well as in fictional literature, representations abound of time as being squeezed, more and more valuable, but also less and less available. Two distinct disciplinary perspectives epitomise how, in social theory, time – in terms of temporal allocation of human tasks, routines and activities – has been recognised as playing a major role in energy demand. First, in time geography, time budget is seen as a concept delimiting the time available for discretionary activities (Jenkins and O’Leary, 1997). Second, concepts of squeezing time and hotspots of energy consumption have also been considered through the lenses of social practice theories (Southerton, 2003). Rush hours, hotspots and experiences of time squeeze have been seen as temporal manifestations of relations between practices (Pantzar and Shove, 2010).

In essence, there is a lot of power in practice theory as a way to conceptually interpret the phenomenon of peak demand. Intuitively, what determines the timing of energy demand is not price, but the commitments we have with the outside world, especially work and school, but also shopping for, and preparing, meals which have their own set time. Regardless of what the real price of electricity might be, we still have to drop off children by a time that is set by the nursery or the school, go to work at a time set by our employer (unless one works from home of with flexible hour contracts, as explained in Chapter 3), and prepare meals which occur at times set either by the habit of the household or by the social norms of the country we live in.

Viewing the timing of energy demand as related to the social ordering of activities or practices implies moving away from seeing energy demand as consumption. This is a difficult mental exercise for economists like me, but they (we) c...

Table of contents

• Cover
• Halftitle Page
• Routledge Page
• Title Page
• Copyright Page
• Contents
• List of figures
• List of tables
• Preface and acknowledgements
• 1 Peak demand
• 2 Demand Side Response
• 3 People’s activities and residential peak demand
• 4 Smart meters
• 5 Time of Use pricing
• 6 Simulating Critical Peak Pricing
• 7 Automated Demand Controllers
• 8 Market-based ancillary services
• 9 Capacity Markets and Demand Side Response
• 10 Conclusions
• Index