Energy saving targets! - Tested in households in the Swedish largest electricity saving experiment

Transcription

1 Pyrko - paper ID Energy saving targets! - Tested in households in the Swedish largest electricity saving experiment Jurek Pyrko, Lund University, Dept. of Energy Sciences, PO Box 118, SE Lund, Sweden, jurek.pyrko@energy.lth.se Abstract This paper presents the scientific evaluation of the results from the Swedish largest electricity saving experiment carried out during one year (February January 2013). 10,000 households from across the country, all of them customers of the grid company E.ON Sweden, have been involved in the experiment to find out how much electricity could be saved if the households had continual feedback on the usage and, above all, if it was possible to obtain energy use related behavioural changes in the households. At the beginning of the experiment, certain targets regarding expected electricity savings, compared to the year before, were decided by the participants. Electricity consumption and the cost was visualised via real-time meters with displays, which could. In an experimental computer and smart phone application, the participants could follow their electricity use visualized in different ways. Five special motivation-boosting actions were included in the experiment. Facebook and the customers personal web pages on the company s site were used as main communication channels. Scientific evaluation of the experiment included both quantitative and qualitative analyses of the households behaviour and experience (via web-questionnaires and in-depth interviews). At the end of the experiment, the overall temperature-corrected electricity saving was about 0.74% for the whole population, compared to 1.5% rise of electricity use for the control group. Active households as a whole saved 1.71% of power and those who saved power reached 8.43% savings. Those who reached their targets within this active group saved 14.88%. The participants declared that they hade made both technical and behavioural changes in their households thanks to the participation in the experiment. A majority of the participants would like to keep the possibility of getting energy feedback. A half of households would be willing to pay for this type of energy service in the future. Keywords: energy saving target, feedback, information, households, efficient energy use Introduction Between February 1, 2012 and January 31, 2013, Energy company E.ON Sweden AB carried out a project announced as Swedish largest energy saving experiment (called the Experiment in this paper). The aim of the Experiment was to make electricity visible for 10,000 customers with an annual electricity consumption of more than 10,000 kwh. A new dimension, compared to other experiments of this type, was that participants were obliged to sett up a clear electricity saving target to be achieved by their household during the Experiment. When choosing the level of savings, the households were informed what kinds of measures were needed to achieve the chosen goal. The primary objective of this Experiment was to investigate and evaluate with scientific methods whether visualisation of electricity consumption to households would lead to savings and behavioural changes. The study also aimed to examine how different factors as: family

2 structure, level of education, main heating system and price-zone location of the households affected their electricity use and their energy related behaviour. Method The energy company supplied all facts about the participating households and their monthly electricity consumption. These facts created the basis for the analysis of the results of the Experiment and how different user groups managed to reduce their electricity use. At last, about households were active participants of the Experiment. Unfortunately, about 100 of them had annual energy use below kwh and about 800 below 10,000 kwh. Electricity consumption during one calendar year (Feb 1, Jan 31, 2013) was compared with equivalent period For electrically heated households the data was corrected for outdoor temperature variations. The methodology used for temperature adjustment was developed by the author [5]. The percentage of electricity used for space heating was calculated as a difference between electricity use during winter (October -April) and summer (June-August) months. Degree-days values were obtained from the energy company and processed before they were allotted to the households. Four web-based surveys were conducted during the Experiment. The content and structure of each survey was designed in cooperation with representatives from the energy company. The responses from the questionnaires and the facts about participants of the Experiment were put together to enable the segmentation of different test groups. Statistical tests were then performed to ensure that any differences between groups were significant and not just a coincidence. A control group of 2,000 households was randomly chosen by the energy company among all the customers not participating in the Experiment and not aware that they were included in the control group. The Experiment All the participating households were equipped with displays connected to the smart electricity meters and showing electricity consumption, load demand and costs in real time. It was also possible to follow three appliances connected to the smart plugs. All the results could be followed on the Experiment s own website. Moreover, there was a specially developed application for both the iphone and Android where experiments participants could monitor their energy consumption and progress during the Experiment (see Figure 1). Figure 1. Different types of visualization during the Experiment [4].

3 On the Experiment website it was possible to follow how the various participants' electricity use had changed and see the total reduction of electricity consumption for the whole Experiment, see Figure 2. The participants could also see the changes in their own consumption, read the news and get advices and tips on efficient energy use. The Experiment also had an own Facebook page where the project, energy saving tips and questions were discussed and answered by the energy company staff. The experiment was divided into five different periods where visualization and motivation boosters for reduced electricity consumption varied. All the changes in daily electricity consumption for each household could be followed on the Experiment s website in real-time. Period 1 Economics - The Balance (March-April 2012) This first period was intended to visualise how the electricity cost varied day by day, showing their consumption pattern, and hopefully, raising awareness of the participants, encouraging them to explore possible ways to reduce their electricity consumption. When participants logged into the app, it showed how much the cost of their electricity consumption increased since last login (as a rain of falling coins ), and the total price of the electricity used by the household, see Figure 3 [4]. Period 2 Challenge - The Neighbour Feud (May-June 2012) The Neighbour Feud introduced a competitive element into the Experiment. The participants competed against each other trying to reduce their electricity consumption the most. During the challenge, electricity usage day by day was compared with the similar period of the year The participants were divided into groups of five and got points for every per cent of reduced electricity consumption. This challenge aimed primarily to prioritize the smaller energy efficiency measures and therefore could give maximum points for up to 20% reduction of electricity use. To reduce usage by over 20% was considered as possible only thanks to more radical changes (such as replacement of the heating system), which was not the focus in the Neighbour feud. During this period, the participants could follow the development of their own, and the neighbours, savings on smart phones. A house icon - the bigger the saving, the larger the house, represented each participant in each group, see Figure 4 [4]. Figure 2. The main graph from the Experiment home page on February 1st, 2013, showing mean total electricity use in kwh/day (day by day) compared to previous year (white line) [4]. The text in the right frame says: The use February 1 to February 1 has decreased by 10.73%

4 Figure 3. Visualization of The Balance [3]. the Neighbour Feud [3]. Period 3 Reward - The Carrot (July-September 2012) Figure 4. Smart phone display of This period focused mostly on energy savings, but also included some kind of competition. The participants were asked to contribute with their own energy-saving tips and the best and most innovative ones were rewarded with prizes such as electric bicycles. In addition, Swedish well-known cartoonist Henrik Lange illustrated many of the proposed energy saving tips in an unexpected way, as shown in Figure 5 - The Anonymous Energyholics. Figure 5. Example of energy saving tips from the motivation booster Carrot [4]. Text: Change your behaviour! Hej! My name is Gustav and I don t turn the light off when I leave a room. This was an unusual and innovative way to visualise and provide information on how energy efficiency measures can be implemented, see Figure 5. The participants were able to see all the energy efficiency advices on the website and in the smart phone application [4].

5 Period 4 Reminder - The General (October-November 2012) In contrast to the Carrot, the General can be compared with a whip or a tough personal trainer. Through constant reminders via text messages the "General" tried to motivate the experiment participants to reduce their electricity consumption, see Figure 6. The design of this motivation booster can be compared to a personal coach who constantly pushes and motivates trainees to raise their score [4]. Figure 6. Motivation booster the General. Text on the display says: Not bad! I ve checked your electricity use and it sinks. Is there something more you can do to lower it more? Think really properly now! [4]. Figur 7. Bongo on a smart phone [4]. Period 5 Empathy - Bongo (December 2012-January 2013) Bongo was a motivation booster during the last final period of the Experiment. This could be compared to 1990 s pet toy from Japan called Tamagotchi - a small teddy bear named Bongo, see Figure 7, that can be kept in a good mood by conserving electricity, which should send a clear signal to the participants of the Experiment about their achievements. As long as the household reduces the electricity consumption, Bongo is happy. But if the electricity use increases, Bongo becomes sour and sad [4]. Results Reduction in relation to the target The final result of the Experiment is based on active households. Of these, about 23% achieved or exceeded their stated energy saving targets. Of the remaining households, about

6 34% saved electricity but stayed below their saving targets. About 43% of the households even raised their consumption, in some cases as much as by 70%. Figure 8 shows all the households with power consumption higher than 5,000 kwh per year and with savings up to 40%. Savings below 0% mean higher consumption during the Experiment compared to the year before. The total electricity saving for the experiment group was about 0.74% for all households. The control group raised power consumption with 1.5% as a whole group, see Table 1. The group Active households comprises households participating actively in the Experiment by answering surveys (3985 households). This group as a whole saved 1.71% of power. In this group, totally 2203 households saved 8.43% of the electricity. Those who reached their targets within this active group (886 households) saved 14.88%. Table 1. The final energy savings from the Experiment. Whole Experiment group Total electricity saving Active households Control group Experiment group Number of households in the sample Active households All households 0.74% 1.71% 1.50% Only those with 0.70% 1.66% 1.50% power consumption of more than 5,000 kwh / year Only those with 0.81% 1.75% 0.74% power consumption of more than 5,000 kwh / year and savings +/- 70% Only those with 0.59% 1.41% 0.40% power consumption of more than 5,000 kwh / year and savings +/- 30% Saved power 8.43% 2203 Reached targets 14.88% 886 Control group Technical measures Approximately 66 % of the participants answered that they had carried out different technical measures during the Experiment in order to save energy. The responses also indicated that performing technical measures was preferred rather than change of the behaviour. Participants also seemed to use the real-time display 100Koll a lot to ensure that the measures they had undertaken really resulted in energy savings. Behavioural changes About 73 % of the participants said in the surveys that they had changed their behaviour as a way of achieving energy savings during the Experiment. Again, the answers indicate that the participants made bigger steps in behavioural adaptation than they would if they were not involved in the Experiment. The survey answers indicated also that the behavioural changes in the households did not affect the indoor comfort or family s living standard. Some respondents mentioned thou that the most negative impact of the measures was reduction of the indoor temperature. Respondents also indicated that both their partners and children often changed their energy related behaviour during the Experiment.

7 Figure 8. Achieved electricity savings in relation to the target for each household [2]. Usability of the system Although one-third of survey respondents strongly agreed with the statement that they only had few problems with the feedback system during the Experiment, many households used the opportunity to submit open answers about the problems they experienced. At least onethird of these comments indicated that the equipment did not work correctly. Sensors reading electricity consumption often lost contact with the display. This means that nearly 20 % of the households had this problem. This might also affect the results of electricity savings shown on the website because the estimation of electricity consumption made by the company was based on these readings. Many participants stated that their knowledge about household electricity consumption increased due to the visualization and feedback. This could possibly indicate that the idea of 100Koll-display was appreciated among the users. A majority of the participants would like to keep the possibility of getting energy feedback and a half of households would be willing to pay for this type of energy service in the future. References 1. Andersson, J & Larsson, P. Energy consumption and energy feedback - Evaluation of the Swedish largest energy saving experiment at E.ON - Phase 1 (In Swedish: Energianvändning och energifeedback - Utvärdering av Sveriges största energisparexperiment på E.ON Etapp 1), Department of Energy Sciences, Lund University, Hols, David & Taimor, Ashfaq. Energy consumption and energy feedback - Evaluation of the Swedish largest energy saving experiment at E.ON - Phase 2 (In Swedish: Energianvändning och energifeedback - Utvärdering av Sveriges största energisparexperiment på E.ON Etapp 2), Department of Energy Sciences, Lund University, E.ON. Website of the Experiment. Available at December 28, E.ON. Website of the Experiment. Available at January 6, Pyrko, Jurek. Electricity use data correction. Lund University, 2012.